[House Hearing, 116 Congress]
[From the U.S. Government Publishing Office]



EXAMINING THE PROPOSED ABAWD RULE AND ITS IMPACT ON HUNGER AND HARDSHIP

=======================================================================

                                HEARING

                               BEFORE THE

    SUBCOMMITTEE ON NUTRITION, OVERSIGHT, AND DEPARTMENT OPERATIONS

                                 OF THE

                        COMMITTEE ON AGRICULTURE
                        HOUSE OF REPRESENTATIVES

                     ONE HUNDRED SIXTEENTH CONGRESS

                             FIRST SESSION

                               ----------                              

                             APRIL 3, 2019

                               ----------                              

                            Serial No. 116-2
                            
                            
                [GRAPHIC NOT AVAILABLE IN TIFF FORMAT]               


          Printed for the use of the Committee on Agriculture
                         agriculture.house.gov
                         
                         
                                 ________                         

                    U.S. GOVERNMENT PUBLISHING OFFICE
                    
36-188 PDF                 WASHINGTON : 2019
                        
                        
                        
                        
                        
                        COMMITTEE ON AGRICULTURE

                COLLIN C. PETERSON, Minnesota, Chairman

DAVID SCOTT, Georgia                 K. MICHAEL CONAWAY, Texas, Ranking 
JIM COSTA, California                Minority Member
MARCIA L. FUDGE, Ohio                GLENN THOMPSON, Pennsylvania
JAMES P. McGOVERN, Massachusetts     AUSTIN SCOTT, Georgia
FILEMON VELA, Texas                  ERIC A. ``RICK'' CRAWFORD, 
STACEY E. PLASKETT, Virgin Islands   Arkansas
ALMA S. ADAMS, North Carolina        SCOTT DesJARLAIS, Tennessee
    Vice Chair                       VICKY HARTZLER, Missouri
ABIGAIL DAVIS SPANBERGER, Virginia   DOUG LaMALFA, California
JAHANA HAYES, Connecticut            RODNEY DAVIS, Illinois
ANTONIO DELGADO, New York            TED S. YOHO, Florida
TJ COX, California                   RICK W. ALLEN, Georgia
ANGIE CRAIG, Minnesota               MIKE BOST, Illinois
ANTHONY BRINDISI, New York           DAVID ROUZER, North Carolina
JEFFERSON VAN DREW, New Jersey       RALPH LEE ABRAHAM, Louisiana
JOSH HARDER, California              TRENT KELLY, Mississippi
KIM SCHRIER, Washington              JAMES COMER, Kentucky
CHELLIE PINGREE, Maine               ROGER W. MARSHALL, Kansas
CHERI BUSTOS, Illinois               DON BACON, Nebraska
SEAN PATRICK MALONEY, New York       NEAL P. DUNN, Florida
SALUD O. CARBAJAL, California        DUSTY JOHNSON, South Dakota
AL LAWSON, Jr., Florida              JAMES R. BAIRD, Indiana
TOM O'HALLERAN, Arizona              JIM HAGEDORN, Minnesota
JIMMY PANETTA, California
ANN KIRKPATRICK, Arizona
CYNTHIA AXNE, Iowa

                                 ______

                      Anne Simmons, Staff Director

              Matthew S. Schertz, Minority Staff Director

                                 ______

    Subcommittee on Nutrition, Oversight, and Department Operations

                      MARCIA L. FUDGE, Ohio, Chair

JAMES P. McGOVERN, Massachusetts     DUSTY JOHNSON, South Dakota,  
ALMA S. ADAMS, North Carolina        Ranking Minority Member
JAHANA HAYES, Connecticut            SCOTT DesJARLAIS, Tennessee
KIM SCHRIER, Washington              RODNEY DAVIS, Illinois
JEFFERSON VAN DREW, New Jersey       TED S. YOHO, Florida
AL LAWSON, Jr., Florida              DON BACON, Nebraska
JIMMY PANETTA, California            JIM HAGEDORN, Minnesota

             Jasmine Dickerson, Subcommittee Staff Director

                                  (ii)
                             
                             
                             
                             C O N T E N T S

                              ----------                              
                                                                   Page
Johnson, Hon. Dusty, a Representative in Congress from South 
  Dakota, opening statement......................................     5
Fudge, Hon. Marcia L., a Representative in Congress from Ohio, 
  opening statement..............................................     1
    Prepared statement...........................................     3
    Submitted letters............................................   131
    Submitted proposed rule......................................   132
Hayes, Hon. Jahana, a Representative in Congress from 
  Connecticut:
    Submitted comment letter; authored by Marc Egan, Director of 
      Government Relations, National Education Association.......   158
    Submitted comment letter; authored by Lisa Davis, Senior Vice 
      President, No Kid Hungry Campaign, Share Our Strength......   159
Lawson, Jr., Hon. Al, a Representative in Congress from Florida, 
  submitted comment letter; authored by Center for Law and Social 
  Policy.........................................................   356
McGovern, Hon. James P., a Representative in Congress from 
  Massachusetts:
    Submitted letter.............................................   154
    Submitted article............................................   154
Panetta, Hon. Jimmy, a Representative in Congress from 
  California:
    Submitted comment letter; authored by Abby J. Leibman, 
      President and Chief Executive Officer, MAZON: A Jewish 
      Response to Hunger.........................................   379
    Submitted comment letter; authored by Keith Carson, Vice 
      President, Board of Supervisors, District 5; Chair, 
      Personnel, Administration and Legislation (PAL) Committee..   387
Peterson, Hon. Collin C., a Representative in Congress from 
  Minnesota, submitted comment letter; authored by Tony Lourey, 
  Commissioner, Minnesota Department of Human Services...........   151
Schrier, Hon. Kim, a Representative in Congress from Washington:
    Submitted comment letter; authored by Hon. Jay Inslee, 
      Governor, State of Washington..............................   162
    Submitted comment letter; authored by Stacy Dean, Vice 
      President, Food Assistance Policy, Center on Budget and 
      Policy Priorities..........................................   165
Van Drew, Hon. Jefferson, a Representative in Congress from New 
  Jersey; submitted comment letter; authored by Kate Leone, Chief 
  Government Relations Officer, Feeding America..................   346
Yoho, Hon. Ted S., a Representative in Congress from Florida, 
  submitted comment letter.......................................   392

                               Witnesses

Cunnyngham, Karen, Associate Director, Mathematica Policy 
  Research, Washington, D.C......................................     7
    Prepared statement...........................................     8
    Submitted questions..........................................   398
Adolphsen, Sam, Vice President of Executive Affairs, Foundation 
  for Government Accountability, Naples, FL......................    18
    Prepared statement...........................................    19
    Submitted questions..........................................   399
Hamler-Fugitt, Lisa, Executive Director, Ohio Association of 
  Foodbanks, Columbus, OH........................................    26
    Prepared statement...........................................    28
    Submitted questions..........................................   401
Shambaugh, Ph.D., Jay C., Director, The Hamilton Project, and 
  Senior Fellow, Economic Studies, Brookings Institution; 
  Professor of Economics, George Washington University, 
  Washington, D.C................................................    55
    Prepared statement...........................................    57
    Submitted questions..........................................   405

 

EXAMINING THE PROPOSED ABAWD RULE AND ITS IMPACT ON HUNGER AND HARDSHIP

                              ----------                              


                        WEDNESDAY, APRIL 3, 2019

                          House of Representatives,
      Subcommittee on Nutrition, Oversight, and Department 
                                                Operations,
                                  Committee on Agriculture,
                                                   Washington, D.C.
    The Subcommittee met, pursuant to call, at 9:00 a.m., in 
Room 1300 of the Longworth House Office Building, Hon. Marcia 
L. Fudge [Chair of the Subcommittee] presiding.
    Members present: Representatives Fudge, McGovern, Adams, 
Hayes, Schrier, Van Drew, Lawson, Panetta, Johnson, DesJarlais, 
Davis, Yoho, Bacon, Hagedorn, and Conaway (ex officio).
    Staff present: Jasmine Dickerson, Kellie Adesina, Alison 
Titus, Caleb Crosswhite, Ashton Johnston, Callie McAdams, 
Jennifer Tiller, Dana Sandman, and Jennifer Yezak.

OPENING STATEMENT OF HON. MARCIA L. FUDGE, A REPRESENTATIVE IN 
                       CONGRESS FROM OHIO

    The Chair. Good morning. This hearing of the Subcommittee 
on Nutrition, Oversight, and Department Operations entitled, 
Examining the Proposed ABAWD Rule and its Impact on Hunger and 
Hardship, will come to order.
    The purpose of today's hearing is to examine proposed 
changes to a long-standing USDA Able Bodied Adults Without 
Dependents, or ABAWD, policy that will impact a significant 
number of SNAP recipients. Such a change demands careful and 
deliberate consideration. Today, we will have this long overdue 
conversation.
    On February 1, I sent a letter to Secretary of Agriculture, 
Sonny Perdue, outlining my serious concerns with the 
Department's proposed rule on ABAWDs. The proposed rule 
included a 60 day comment period, which I now understand has 
been extended for a few days. However, given the seriousness of 
this topic, I requested an extension on the comment period so 
that there may be more time to explore its potential impacts. 
The Department rejected the request and, instead, Secretary 
Perdue responded to me by saying, and I quote, ``The proposed 
rule . . . would encourage broader application of the statutory 
ABAWD work requirement, consistent with the Administration's 
focus on fostering self-sufficiency and promoting the dignity 
of work. I believe these proposed changes support our mutual 
goal of improving the lives of those participating in SNAP.''
    Well, Mr. Secretary, I disagree. The goal of improving 
lives is mutual. Your methods, though, are harsh, arbitrary, 
and mean. There is no dignity in taking food from the poorest 
and most vulnerable of our citizens. It is dishonest and 
immoral for anyone to assume or suggest that poor people do not 
want to work, especially if that work only pays an average of 
$125 per month.
    And before we go any further, I want to make it very clear. 
People want a hand up, not a hand out, and it is insulting to 
suggest otherwise.
    The proposal before us fails to consider that unemployment 
is not the sole problem for ABAWDs. Many ABAWDs experience 
other hardships, including lack of housing, undiagnosed mental 
illness, learning disabilities, and poor health. The proposal 
before us makes clear this Administration does not understand, 
nor care, about the lack of access or barriers and hardships 
that keep many from finding and securing long-term employment. 
The proposal also tells me the Administration foolishly assumes 
everybody has the same access to resources needed to escape the 
cycle of poverty. If they just work 20 hours per week, it would 
solve their problems and move them out of poverty, magically. 
Lifting yourself up by your boot straps only works if you have 
boots.
    What I want to know is what USDA actually knows about those 
who will be affected by this rule? Based on the reports from 
our witnesses, Mathematica in particular, we are most likely 
dealing with the poorest of the poor. In fact, I am still 
waiting on my request for information during last month's 
hearing with the Secretary where I asked what percentage of 
ABAWD populations are veterans, homeless, have mental or 
physical limitations, or lack access to public transportation?
    Were any of these factors analyzed or data collected before 
the release of the proposed rule? Does the Department even 
internally track this kind of relevant information to better 
inform its rulemaking and policy decisions? If they were, 
please present it to us. It is time we call this what it is: a 
rush to accomplish a conservative political wish-list. If this 
was really about the dignity of work and efficiency of the 
program, we would wait to see the final results from the 2014 
Farm Bill, which provided $200 million for ten employment and 
training pilot projects. It is ill-advised to issue a rule 
without the supporting data or best practices learned from the 
pilots, to better serve the ABAWD population.
    USDA estimates that 755,000 people will lose benefits and 
predicts a savings in Federal spending on SNAP benefits of $7.9 
billion over 5 years. What will happen to the 755,000 people? 
If the Department is so eager to get people into jobs, will the 
Department hire them? The unemployment rate in my district is 
9.8 percent. Where are the jobs? My Republican colleagues love 
to talk about the surplus of jobs or low unemployment numbers, 
but we should remember that there is a skills gap at play 
within this population and many ABAWDs live in smaller, rural 
communities where jobs are not as readily available. Was the 
skills gap taken into consideration during formulation of this 
proposed rule? Low unemployment rates do little to tell us 
whether jobless individuals in a specific geographical area 
lack the necessary skills to obtain gainful work in the 
community. However, the Department proposes to limit existing 
state flexibility, to submit a variety of credible resources, 
and support materials to help tell the story a Bureau of Labor 
Statistics unemployment rate is unable to tell. A low 
unemployment rate does not erase the existence of significant 
barriers to unemployment in our nation's poorest communities.
    Without the skills necessary to obtain gainful employment 
and meet SNAP work requirements, what other options are there 
for these individuals to put food on the table?
    I am very concerned about the added burden these proposed 
cuts to SNAP place on other low-income services and charities 
like food banks. Every time Republicans trot out calls for 
welfare reform, they argue the private-sector will pick up the 
slack. Let me ask this, what does $7.9 billion in savings from 
SNAP mean if it increases the demand for other low-income 
programs or local charities that are already stretched thin? 
This proposed rule is nothing more than another attempt by the 
GOP and the Trump Administration to reintroduce the thoughtless 
House Republican SNAP provisions that were rejected in the 2018 
Farm Bill. We passed a bill. Please follow the law.
    The House and Senate passed a farm bill conference report 
by a historic 369 votes, and the President signed it without 
delay. Let's just follow the law. Rehashing failed policies is 
an affront to the democratic process and an utter waste of 
time. We have seen this Administration and my colleagues 
reciting the same negative talking points about people who are 
on SNAP time and again, and I am really very weary of it. 
Instead of proposing cruel and unsound ideas without merit, 
let's figure out how to help people in need.
    Our job is to do the most for those who have the least. 
Let's just follow the law.
    [The prepared statement of Ms. Fudge follows:]

    Prepared Statement of Hon. Marcia L. Fudge, a Representative in 
                           Congress from Ohio
    The purpose of today's hearing is to examine proposed changes to a 
long-standing USDA Able Bodied Adults Without Dependents, or ABAWD, 
policy that will impact a significant number of SNAP recipients. Such a 
change demands careful and deliberate consideration. Today, we will 
have this long overdue conversation.
    On February 1st, I sent a letter to Secretary of Agriculture Sonny 
Perdue outlining my serious concerns with the Department's proposed 
rule on ABAWDs. The proposed rule included a 60 day comment period. 
However, given the seriousness of this topic, I requested an extension 
on the comment period so that there may be more time to explore its 
potential impacts.
    The Department rejected that request and, instead, Secretary Perdue 
responded by saying, and I quote:

          ``The proposed rule . . . would encourage broader application 
        of the statutory ABAWD work requirement, consistent with the 
        Administration's focus on fostering self-sufficiency and 
        promoting the dignity of work. I believe these proposed changes 
        support our mutual goal of improving the lives of those 
        participating in SNAP.''

    Well Mr. Secretary, I disagree. The goal of improving lives is 
mutual--his methods are harsh, arbitrary and mean.
    There is no dignity in taking food away from the poorest and most 
vulnerable of our citizens.
    It is dishonest and immoral for anyone to assume or suggest that 
poor people do not want to work, especially if that work only pays an 
average of $125 a month.
    And before we go any further, I want to make it very clear: people 
want a hand up, not a hand out, and it is insulting to suggest 
otherwise.
    The proposal before us fails to consider that unemployment is not 
the sole problem ABAWDs face. Many ABAWDs experience other hardships, 
including lack of housing, undiagnosed mental illnesses, learning 
disabilities, and poor health.
    The proposal before us makes clear this Administration does not 
understand nor care about the lack of access or barriers and hardships, 
that keep many from finding and securing long-term employment.
    The proposal also tells me the Administration foolishly assumes 
everybody has the same access to the resources needed to escape the 
cycle of poverty. ``If they just work 20 hours per week, it would solve 
their problems and move them out of poverty.''!?
    ``Lifting yourself up by your boot straps'' only works if you have 
boots.
    What I want to know is what USDA actually knows about those who 
will be affected by their rule?
    Based on the report from our witness, Mathematica, we are most 
likely dealing with the poorest of the poor.
    In fact, I'm still waiting on my request for information during 
last month's hearing with Secretary Perdue, where I asked, ``what 
percentage of the ABAWD population are veterans, homeless, have mental 
or physical limitations, or lack access to public transportation?''
    Were any of these factors analyzed or data collected before the 
release of the proposed rule? Does the Department even internally track 
this kind of relevant information to better inform its rulemaking and 
policy decisions?
    If they were, please present it to us.
    It's time we call this what it is: a rush to accomplish a 
conservative political wish-list.
    If this was really about the dignity of work and efficiency of the 
program, we would wait to see the final results from the 2014 Farm 
Bill, which provided $200 million for ten Employment and Training pilot 
projects.
    It is ill-advised to issue a rule without the supporting data or 
best practices learned from the pilots, to better serve the ABAWD 
population.
    USDA estimates that 755,000 people will lose benefits and predicts 
a savings in Federal spending on SNAP benefits of $7.9 billion over 5 
years.
    What will happen to the 755,000 people? If the Department is so 
eager to get people into jobs, will they hire them?
    The unemployment rate in my district is 9.8 percent. Where are the 
jobs?
    My Republican colleagues love to talk about the surplus of jobs or 
low unemployment numbers, but we should remember that there's a skill 
gap at play within this population and many ABAWDs live in smaller, 
rural communities where jobs are not as readily available.
    Was the skills gap taken into consideration during formulation of 
this proposed rule? Low unemployment rates do little to tell us whether 
jobless individuals in a specific geographical area lack the necessary 
skills to obtain gainful work in their communities.
    However, the Department proposes to limit existing state 
flexibility to submit a variety of credible resources and support 
materials to help tell the story a Bureau of Labor Statistics 
unemployment rate is unable to tell. A low unemployment rate does not 
erase the existence of significant barriers to employment in our 
nation's poorest communities.
    Without the skills necessary to obtain gainful employment and meet 
SNAP work requirements, what other options are there for these 
individuals to put food on the table? I am very concerned about the 
added burden these proposed cuts to SNAP place on other low-income 
services and charities like food banks. Every time Republicans trot out 
calls for welfare reform, they argue the private-sector will pick up 
the slack.
    Let me ask this, what does $7.9 billion in savings from SNAP mean 
if it increases the demand for other low-income programs or local 
charities that are already stretched thin?
    This proposed rule is nothing more than another attempt by the GOP 
and the Trump Administration to reintroduce the thoughtless House 
Republican SNAP provisions that were rejected in the 2018 Farm Bill. We 
passed a bill--follow the law!
    The House and Senate passed a farm bill conference report by a 
historic 369 votes, and the President signed the bill without delay. 
Follow the law!
    Rehashing failed policies is an affront to the democratic process 
and an utter waste of time.
    We have seen this Administration and my Republican colleagues 
reciting the same negative talking points about people who are on SNAP 
time and again; I am tired of it.
    Instead of proposing cruel and unsound ideas without merit--let's 
figure out how to help people in need.
    Our job is to do the most for those who have the least.

    The Chair. I would now turn to my colleague, my friend, the 
Ranking Member, Mr. Johnson.

 OPENING STATEMENT OF HON. DUSTY JOHNSON, A REPRESENTATIVE IN 
                   CONGRESS FROM SOUTH DAKOTA

    Mr. Johnson. Thank you very much, Madam Chair, and I do 
appreciate Ms. Fudge convening this hearing, and I want to 
thank our witnesses for their participation.
    For me, an important foundation of all of this 
Subcommittee's work on nutrition is, first, that we all want to 
improve the lives of Americans who are facing hard times. I 
think that is obvious. Second, that SNAP is an important 
poverty program. That is something worth maintaining, something 
worth holding up. Third, that work and education are a 
critically important, a necessary part of helping people 
realize opportunities to move out of poverty.
    And so, each of us today on this dais is fighting for the 
same goal. Madam Chair mentioned that. We all want to see the 
lives of Americans improve. And a number of us on the broader 
Agriculture Committee on both sides of the aisle have 
experienced welfare programs, poverty programs, on a personal 
basis. But whether we have or we haven't, we all want to make 
sure that we maintain an effective and efficient social safety 
net.
    And so, we are going to disagree about the best way to do 
that, but the basic heart of the matter is intact. As a 
country, we spend $1 trillion a year on 80 social safety net 
programs, and we want to make sure they work. We are a nation 
of giving. We want to be a nation of opportunity.
    And so, able-bodied adults, the ABAWD population, we have 
been talking about this population for a long time. I know 
there have been a lot of hearings that have addressed this 
issue. I am excited to have our witnesses today hit on them 
even more. But from the welfare reform efforts of 1996, out of 
the farm bill discussions of 2018, for decades, this group and 
the broader Congress has been talking about ABAWDs, and we have 
made some progress. But nobody here would say that our work is 
done, and so, I am excited to work together to try to find a 
way to do even better, to find data-driven solutions for how we 
can improve the lives of these ABAWDs.
    You may have heard me say--because I say it a lot--that 
work has dignity. Work is opportunity. Work is an American 
value that we all need help to achieve. And I am excited to 
discuss today and in the future how we really can work with the 
Administration, work throughout Congress to try to make sure 
that able-bodied adults really do have a good pathway from 
welfare to work, and that is going to help us preserve these 
programs, this critically important SNAP Program, for our most 
in need friends and our most in need neighbors.
    As the Chair alluded to, my side of the aisle has talked a 
lot about a record economy, record job openings, and I do think 
that gives us a special opportunity to help people move out of 
poverty and into work. For that reason, I want to applaud 
Secretary Perdue and USDA in taking this regulatory action to 
make work an even more central component of this important 
program.
    These proposed rules, they really are intended, honestly 
intended to help work capable individuals seek new employment 
opportunities and be in a better position to realize their 
dreams. Now, some states have taken too much flexibility. They 
have taken too much liberty with the flexibility that Congress 
has given them, and I know that has been a bipartisan sentiment 
in the past, that both Democratic and Republican Members of the 
Agriculture Committee have said that, and so, I am looking 
forward to working with my colleagues to right size the amount 
of flexibility, to hold states accountable, and move more 
people off poverty.
    The term able-bodied, as we have talked about, is so key to 
this discussion, and I want to make sure that we are working to 
empower and not stigmatize the ABAWD population. Of course, 
Madam Chair is exactly right, that these folks have a certain 
number of challenges. There are barriers to unemployment. That 
isn't arguable. But despite those barriers, with help, they can 
still seek employment. They can stabilize their income. They 
can move to a place of even greater personal autonomy. That is 
the American dream.
    I think about during the farm bill discussions, there was a 
video that came out--and I was a private citizen at the time, 
but I was captivated by the video. It was about Latasha, and 
she was a former E&T participant here in Washington. She 
completed a certification program back in 2012, has been 
working successfully since. Her story is a story that should 
make us all proud. And with help, with additional 
accountability, with states doing a better job of managing 
their programs, there can be thousands more Latashas out there 
realizing a better life. And we all know that just ignoring the 
need for improvement, ignoring the need for a forum doesn't 
really improve anybody's life. That is not leadership. And so, 
let's work with SNAP recipients. Let's work with this 
Subcommittee. Let's work with the Administration to move even 
more people from welfare to work.
    And with that, Madam Chair, I yield and I welcome our 
witnesses.
    The Chair. Thank you very much, Mr. Johnson. I would ask 
that all Members submit their opening statements for the record 
so that we can begin with our witnesses as quickly as possible.
    I would like to introduce and welcome our witnesses. We 
would begin today with Ms. Karen Cunnyngham, Associate 
Director, Mathematica Policy Research, Washington, D.C. Mr. Sam 
Adolphsen, Vice President of Executive Affairs, Foundation for 
Government Accountability, Naples, Florida. Ms. Lisa Hamler-
Fugitt, Executive Director, Ohio Association of Food Banks, 
Columbus, Ohio; and Dr. Jay Shambaugh, Director of The Hamilton 
Project, Brookings Institution, Washington, D.C.
    Ms. Cunnyngham, you may begin. I would just bring your 
attention to the lighting system. The light will turn green 
when you begin. You will have 5 minutes to give your testimony. 
When you see the yellow light, it means you have 1 minute. When 
you see the red light, we would like you to conclude as quickly 
as possible.
    Thank you very much, and welcome.

STATEMENT OF KAREN CUNNYNGHAM, ASSOCIATE DIRECTOR, MATHEMATICA 
               POLICY RESEARCH, WASHINGTON, D.C.

    Ms. Cunnyngham. Chair Fudge, Ranking Member Johnson, and 
distinguished Members of the Subcommittee, thank you for 
inviting me to testify today. I am an associate director at 
Mathematica, and have been conducting research on SNAP for 
government agencies for 18 years.
    I currently direct a project commissioned by the Robert 
Wood Johnson Foundation to develop rigorous and objective 
estimates of the effects of proposed changes to SNAP. Much of 
what I will present today is based on findings from that 
project.
    SNAP participants who are ages 16 to 59 that do not have a 
disability, and are not working at least 30 hours per week must 
register for work unless they meet certain criteria, such as 
caring for an incapacitated person. Work registrants who are 
ages 18 to 49 and don't live with a child must work an average 
of at least 20 hours per week, or face a time limit of 3 months 
of benefits in a 3 year period. They are exempt from the time 
limit, however, if they participate in a qualifying employment 
and training program, or other meaningful work activity, have a 
percentage exemption from the state agency, or live in a waiver 
area, an area for which the state agency requested and received 
a Federal waiver from time limits because of high unemployment.
    USDA's proposed regulatory change would eliminate or modify 
some current waiver area criteria. For example, states would no 
longer be able to request a waiver for counties with overall 
unemployment rates less than seven percent. Table 1 in my 
written testimony summarizes the proposed changes.
    According to USDA's Regulatory Impact Analysis, among the 
SNAP participants who are ages 18 to 49 without a disability 
and childless SNAP households and in a waiver area in Fiscal 
Year 2016, about \3/4\ would be newly subject to the additional 
work requirement and time limit. USDA further estimates that 
under the proposed changes, between 755,000 and 851,000 of 
these people would not meet the work requirements in 2020, and 
would therefore lose eligibility after 3 months.
    Mathematica used Fiscal Year 2017 SNAP quality control data 
to examine the characteristics of SNAP participants who would 
be affected by the proposed changes. Specifically, we focused 
on the estimated 1.2 million SNAP participants who lived in a 
waiver area, could be newly subject to time limits, and were 
not working at least 20 hours per week. Among these SNAP 
participants, 97 percent lived in poverty, and 88 percent lived 
in deep poverty, compared with 39 percent of other SNAP 
participants living in deep poverty. Eleven percent were 
working, although less than 20 hours per week, and another six 
percent lived with someone else who was working. However, only 
\1/3\ were living in SNAP households with any reported income. 
Among those, the average household income was $557 a month, 43 
percent of the poverty level. The average monthly SNAP benefit 
was $181 per person. Finally, these SNAP participants were much 
more likely to live alone than other SNAP participants, 78 
percent compared with 23 percent.
    The potential impact on these individuals would vary by 
their circumstances and state. SNAP participants in the 17 
states without waiver areas would not be affected by the 
proposed changes. In other states, the state agency may offer 
slots in qualifying employment and training programs, or 
percentage exemptions to participants who would otherwise face 
a time limit.
    In many states, however, some SNAP participants would be 
newly required to work an average of at least 20 hours per 
week, or be subject to the time limit. Both SNAP participants' 
job readiness and the local labor market will affect SNAP 
participants' ability to find work.
    Although the Bureau of Labor Statistics estimates the 
national overall unemployment rate was 3.9 percent in 2018, 
some groups were less likely to find work. For example, the 
unemployment rate for young adults ages 20 to 24 was 6.9 
percent, and the rate for African American men was seven 
percent.
    Policy decisions should be informed by the best data 
available, and this proposed rule is no exception. Policymakers 
could gain a more complete picture of the likely effects of the 
proposed regulatory change if detailed information on the areas 
that would no longer qualify for a waiver were incorporated 
into state estimates of the people potentially affected. In 
addition, examining unemployment rates for subgroups of a state 
population would provide valuable insights to the availability 
of jobs for SNAP participants, and the potential for some 
groups to experience a disproportionate impact from proposed 
changes. New data collection on the circumstances of people who 
lose eligibility for SNAP because of time limits also could 
help policymakers understand whether and how well policy 
objectives are being achieved.
    Thank you. I look forward to your questions.
    [The prepared statement of Ms. Cunnyngham follows:]

Prepared Statement of Karen Cunnyngham, Associate Director, Mathematica 
                   Policy Research, Washington, D.C.
Addressing Proposed Changes to SNAP Waiver Area Criteria
    Good morning, Chairwoman Fudge, Ranking Member Johnson, and 
distinguished Members of the Subcommittee. Thank you for inviting me to 
testify at today's hearing, ``Examining the Proposed ABAWD Rule and its 
Impact on Hunger and Hardship.'' I am an associate director in 
Mathematica's Human Services Division and the director of a project, 
commissioned by the Robert Wood Johnson Foundation, to develop credible 
and objective estimates of the effect of proposed legislative and 
regulatory changes to the Supplemental Nutrition Assistance Program--or 
SNAP. My Mathematica colleagues and I are proud of this work, and 
appreciate the opportunity to apply our combined expertise in data, 
methods, policy, and practice to help enhance understanding of SNAP, 
refine strategies for its implementation, and ultimately improve the 
effectiveness of the program.
    SNAP, the largest of the domestic nutrition assistance programs 
administered by the Food and Nutrition Service of the U.S. Department 
of Agriculture (USDA), provides nutrition assistance to eligible, low-
income people in need. The proposed regulatory change we are here to 
discuss today would affect a subset of the overall SNAP population--
about three percent of the 41.5 million who participated in the program 
in Fiscal Year 2017. According to our analysis of Fiscal Year 2017 SNAP 
Quality Control (QC) data, the vast majority of SNAP participants who 
could be affected by the proposed rule are in deep poverty, and many 
live alone.
    In my testimony today, drawn from a research brief produced for the 
Robert Wood Johnson Foundation project, I will (1) outline the proposed 
regulatory changes, (2) discuss the estimated impacts, (3) summarize 
the characteristics of SNAP participants potentially impacted, and (4) 
suggest additional data collection and research to help inform this 
discussion.\1\
---------------------------------------------------------------------------
    \1\ Cunnyngham, Karen. ``Proposed Changes to the Supplemental 
Nutrition Assistance Program: Waivers to Work-Related Time Limits.'' 
Issue brief submitted to the Robert Wood Johnson Foundation. 
Washington, D.C.: Mathematica Policy Research, March 2019.
---------------------------------------------------------------------------
Understanding the Proposed Regulatory Changes
    Currently, SNAP participants ages 16 to 59 must register for work 
unless they are already working at least 30 hours per week; have a 
disability; or meet other criteria, such as caring for a young child or 
an incapacitated person. Work registrants who are ages 18 to 49 in 
childless SNAP households are subject to additional work requirements 
and a time limit: they must work an average of at least 20 hours per 
week to continue receiving SNAP benefits for more than 3 months in a 3 
year period. They are exempt from the time limits, however, if they (1) 
participate in a qualifying employment and training program or other 
meaningful work activity; (2) have a discretionary exemption from the 
state agency; or (3) live in a waiver area, an area for which the state 
agency requested and received a Federal waiver from time limits because 
of high unemployment.
    Table 1 shows how USDA's proposed regulatory change would eliminate 
or modify some current waiver area policies and leave others unchanged. 
In recent years, states based most of their requests for geographic 
waivers on an area qualifying for the extended unemployment benefits 
authorized during the Great Recession or experiencing an unemployment 
rate at least 20 percent above the national average. After SNAP time 
limits were reinstated following the Great Recession, some states have 
requested and received waivers for all or parts of the state, while 
others have not requested any time limit waivers at all. Table 2 
illustrates how the prevalence of state time limit waivers changed from 
2009 through 2018. Currently, 17 states have no waiver areas, either 
because no area in the state qualified or the state agency chose not to 
request a waiver (Table 3). Although states with the highest 
unemployment rates in 2018--Alaska and New Mexico--had statewide 
waivers, others with overall unemployment rates above the national 
average of 3.9 percent chose not to apply for a waiver for any areas of 
the state.

                      Table 1. Waiver Area Policies
------------------------------------------------------------------------
            Current policy                 Proposed regulatory change
------------------------------------------------------------------------
                   Criteria to establish waiver area
------------------------------------------------------------------------
The U.S. Department of Labor           Eliminated
 designated the area as a Labor
 Surplus Area based on a recent 24
 month average unemployment rate that
 is either (1) at least ten percent
 or (2) at least six percent and at
 least 20 percent above the national
 average
The Department of Labor determined     No change
 that the area meets the criteria for
 extended unemployment benefits,
 available to workers who have
 exhausted regular unemployment
 insurance benefits during periods of
 high unemployment
Data from the U.S. Bureau of Labor     No change
 Statistics (BLS) show the area had a
 recent 12 month average unemployment
 rate greater than ten percent
Data from BLS show the area had a      The unemployment rate also must
 recent 24 month average unemployment   be at least seven percent
 rate at least 20 percent above the
 national average
Alternate sources indicate a lack of   The alternate criteria will be
 sufficient jobs in an area,            applicable only to areas for
 including an unemployment rate         which data from BLS or a BLS-
 estimated with data from BLS and the   cooperating agency are limited
 Census Bureau's American Community     or unavailable, such as a
 Survey; a low and declining            reservation area or U.S.
 employment-to-population ratio; a      territory
 lack of jobs as a consequence of
 declining occupations or industries;
 or an academic study or other
 publication describing the area's
 lack of a sufficient number of jobs
------------------------------------------------------------------------
                       Other waiver area policies
------------------------------------------------------------------------
Waivers may be statewide               Only waivers based on extended
                                        unemployment benefits may be
                                        statewide
State agencies may combine data from   State agencies may combine data
 sub-state areas, such as counties,     only for areas collectively
 that are contiguous, share an          designated as Labor Market Areas
 economic region, or both               by BLS
Waivers may extend beyond the fiscal   Waivers based on a 24 month
 year                                   average unemployment rate may
                                        not extend beyond the fiscal
                                        year
Approval by governor not explicitly    Governor must approve waiver
 required                               request
------------------------------------------------------------------------


                      Table 2. Waiver Area Timeline
------------------------------------------------------------------------
 
------------------------------------------------------------------------
April 2009 to September 2010        Congress temporarily suspended the
                                     time limits through the American
                                     Recovery and Reinvestment Act.
October 2010 to December 2015       In Fiscal Year 2011, time limits
                                     continued to be waived based on
                                     extended unemployment benefits for
                                     45 states, the District of
                                     Columbia, Guam, the Virgin Islands,
                                     and some areas of five additional
                                     states.
                                    By the end of Fiscal Year 2015, time
                                     limits were re-implemented in nine
                                     states and in some areas of 13
                                     other states.
January 2016 to Fiscal Year 2017    Few areas still qualified for
                                     extended unemployment benefits, but
                                     many areas received time limit
                                     waivers based on other indicators
                                     of high unemployment, such as an
                                     unemployment rate at least 20
                                     percent above the national average.
                                     Seventeen states had no waiver
                                     areas for most of this time.
December 2018                       Seventeen states have no waiver
                                     areas; seven states, the District
                                     of Columbia, Guam, and the Virgin
                                     Islands have time limit waivers for
                                     their entire area; and the
                                     remaining states have waivers for
                                     some but not all areas of the
                                     state.
------------------------------------------------------------------------


                                       Table 3. Current State Waiver Areas
----------------------------------------------------------------------------------------------------------------
 
----------------------------------------------------------------------------------------------------------------
           No waiver areas            Some waiver areas                     Statewide waiver
----------------------------------------------------------------------------------------------------------------
Alabama            Missouri           Arizona            Massachusetts      Pennsylvania       Alaska
Arkansas           Nebraska           California         Michigan           Rhode Island       District of
                                                                                                Columbia
Delaware           North Carolina     Colorado           Montana            South Dakota       Guam
Florida            Oklahoma           Connecticut        Nevada             Tennessee          Louisiana
Indiana            South Carolina     Georgia            New Hampshire      Utah               New Mexico
Iowa               Texas              Hawaii             New Jersey         Vermont            Virgin Islands
Kansas             Wisconsin          Idaho              New York           Virginia
Maine              Wyoming            Illinois           North Dakota       Washington
Mississippi                           Kentucky           Ohio               West Virginia
                                      Maryland           Oregon
----------------------------------------------------------------------------------------------------------------
Source: The Food and Nutrition Service's ``ABAWD Waiver Status'' reports available at https://www.fns.usda.gov/
  snap/abawd-waivers.

Discussion of Estimated Impacts
    According to USDA's Regulatory Impact Analysis of the proposed 
rule, an estimated \3/4\ of ABAWDs currently living in a waiver area 
would be newly subject to a 3 month limit on their benefits.\2\ Some of 
them would increase their existing work to an average of 20 hours per 
week, find work, or meet the work requirements by participating in an 
employment and training program or workfare (that is, unpaid work 
through a state-approved program). But USDA estimates that between 
755,000 and 851,000 people in 2020, depending on future unemployment 
rates, would not meet the additional work requirements and would 
therefore lose eligibility after 3 months. For those living with others 
unaffected by the policy change, the SNAP household could continue to 
receive benefits, but the amount would be reduced; those living alone 
would lose all SNAP benefits. Nationally, the proposed regulatory 
changes would result in a 2.5 percent reduction in spending on SNAP 
benefits, according to USDA estimates.
---------------------------------------------------------------------------
    \2\ ABAWDs, or ``able-bodied adults without dependents'' are SNAP 
participants who are subject to work registration, ages 18 to 49, 
without a disability, and living in childless SNAP households.
---------------------------------------------------------------------------
    The potential impact would vary by state and depends on a variety 
of factors, including state agency policies, the local labor market, 
and the characteristics and circumstances of the participants. We used 
Fiscal Year 2017 SNAP QC data to estimate state percentages of SNAP 
participants ages 18 to 49, without a disability, and living in 
childless SNAP households who could be newly subject to a time limit 
(Figure 1). SNAP participants in the 17 states without waiver areas 
would not be affected by the proposed changes because they already face 
time limits unless they are engaged in meaningful work activities or 
are exempt for other reasons. In other states, the state agency may 
offer a slot in a qualifying employment and training program to 
participants who would otherwise face a time limit or use Federal 
``percentage exemptions'' to exempt some SNAP participants from the 
time limit.
Figure 1. Estimated impact by state

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Percentage of SNAP participants ages 18 to 49, without a 
        disability, and living in childless SNAP households who were 
        potentially subject to a time limit, lived in a waiver area, 
        and did not work 20 hours per week.
          Source: Fiscal year 2017 SNAP QC data.
          Notes: States with a white background did not have waiver 
        areas in Fiscal Year 2017. See appendix table for state 
        percentages.

    In many states with waiver areas, at least some SNAP participants 
living in those areas would be newly required to work an average of at 
least 20 hours per week to continue receiving benefits for more than 3 
months. Both the local labor market and SNAP participants' job 
readiness will affect their ability to find work. Although the national 
overall unemployment rate was 3.9 percent in 2018, according to BLS 
estimates, that rate represents an average, and some groups are much 
less likely to find steady work. For example, the unemployment rate for 
young adults ages 20 to 24 was 6.9 percent, and the rate for African 
American men was 7.0 percent. Access to a well-funded and robust SNAP 
employment and training program--which is not currently available in 
many areas--could help participants meet the work requirements.
    In addition, the characteristics and circumstances of SNAP 
participants will influence whether they lose eligibility for SNAP 
under the proposed change. For example, certain SNAP participants are 
not required to register for work because they care for an 
incapacitated person or meet other criteria; work requirements will not 
change for these participants. On the other hand, some participants who 
newly face a time limit might choose to forgo SNAP benefits and rely on 
other available resources, such as food banks or family members, rather 
than comply with work requirements.
Characteristics of SNAP Participants Potentially Impacted
    Mathematica used Fiscal Year 2017 SNAP QC data to examine the 
characteristics of SNAP participants who could face time limits on 
receiving SNAP benefits under the proposed regulatory change. In Fiscal 
Year 2017, eight percent of all SNAP participants (3.2 million people) 
were ages 18 to 49, did not have a disability, and did not live with a 
child. Twenty-one percent of this group were working an average of at 
least 20 hours per week, with the percentage ranging from nine percent 
to 36 percent across states. An estimated 1.2 million SNAP participants 
were not working an average of at least 20 hours per week and would 
have faced time limits but didn't because they lived in a waiver area. 
Among these SNAP participants who could be affected by the proposed 
regulatory changes:

   97 percent lived in poverty, compared with 80 percent of 
        other SNAP participants.

   88 percent had household income at or below 50 percent of 
        the poverty level, compared with 39 percent of other SNAP 
        participants.

   Among the \1/3\ living in SNAP households with reported 
        income, the average monthly household income was $557, or 43 
        percent of the poverty level.

   11 percent were working, although less than an average of 20 
        hours per week, and another six percent lived with someone else 
        who was working.

   5 percent lived with a person with a disability.

   The average monthly SNAP benefit was $181 per person, 
        compared with $120 for other SNAP participants.

   78 percent lived alone (Figure 2), compared with 23 percent 
        of other SNAP participants.
Figure 2. Living situation of those potentially affected 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: Fiscal year 2017 SNAP QC data.
Data-Driven Decision Making
    Objective, rigorously derived estimates of the potential impacts of 
proposed policy changes can provide additional insight for policymakers 
like you, who are faced with difficult decisions about how to allocate 
scarce resources in a way that helps the people who are most in need. 
To conduct the analysis I just described, we used the Fiscal Year 2017 
SNAP QC data available at https://host76.mathematica-mpr.com/fns/. 
Details about the small amount of data cleaning we did to ensure that 
state estimates aligned with state policy, and how we tabulated the 
data, are available upon request.
    Further analysis of existing data could provide additional insights 
into the likely effects of the proposed regulatory change. For example, 
state estimates of the number of people potentially affected could be 
refined using county-level data from state and Federal sources, 
incorporating more detailed information on which current waiver areas 
would not qualify under the proposed criteria. Examining unemployment 
rates for subgroups of a state population would also provide valuable 
insights into the availability of jobs for SNAP participants and the 
potential for some groups to experience a disproportionate impact from 
proposed changes. In addition, new data collection on the circumstances 
of people who lose eligibility for SNAP because of time limits could 
help policymakers understand whether and how well policy objectives are 
being achieved. Finally, Mathematica's evaluation of SNAP employment 
and training pilots for USDA will provide important information on 
innovative strategies for increasing employment and earnings among SNAP 
participants.
    I'm grateful for the opportunity to share this evidence, as well as 
the companion issue brief attached to my written statement, with you 
today. Thank you.

  Table A.1. Estimated state percentage of SNAP participants that could
   potentially be affected by proposed changes to waiver area criteria
------------------------------------------------------------------------
                                                SNAP participants 
                                         -------------------------------
                          Waiver areas      Number (in      Number (in
                                            thousands)      thousands)
------------------------------------------------------------------------
Alabama                 None                           0               0
4Alaska                 Statewide                      7              72
8Arizona                Some                          15             200
Arkansas                None                           0               0
4California             Statewide                    300              65
8Colorado               Some                           3              12
4Connecticut            Some                          26             620
Delaware                None                           0               0
4District of Columbia   Statewide                      8              53
Florida                 None                           0               0
4Georgia                Some                          81              66
6Guam                   Statewide                      1              37
8Hawaii                 Some                           *               1
Idaho                   Some                           *               1
4Illinois               Statewide                    178             770
Indiana                 None                           0               0
Iowa                    None                           0               0
Kansas                  None                           0               0
4Kentucky               Some                          32              54
Louisiana               Statewide                     56             730
Maine                   None                           0               0
6Maryland               Some                          18              31
Massachusetts           Some                          18              28
Michigan                Some                          78              51
8Minnesota              Some                           2              60
Mississippi             None                           0               0
Missouri                None                           0               0
6Montana                Some                           3             280
Nebraska                None                           0               0
6Nevada                 Statewide                     25              41
8New Hampshire          Some                           *               2
New Jersey              Some                           1               2
4New Mexico             Statewide                     27              53
6New York               Some                          84             390
North Carolina          None                           0               0
8North Dakota           Some                           *               4
Ohio                    Some                           4              40
Oklahoma                None                           0               0
6Oregon                 Some                          46              50
Pennsylvania            Some                          51              43
4Rhode Island           Statewide                     14             750
South Carolina          None                           0               0
6South Dakota           Some                           3              44
4Tennessee              Some                          69             670
Texas                   None                           0               0
8Utah                   Some                           *               1
Vermont                 Some                           *               5
6Virgin Islands         Statewide                      1              39
Virginia                Some                          20              46
4Washington             Some                          60              53
6West Virginia          Some                          15             470
Wisconsin               None                           0               0
Wyoming                 None                           0               0
------------------------------------------------------------------------
 SNAP participants ages 18 to 49, without a disability, and living in
  childless SNAP households who were potentially subject to a time
  limit, lived in a waiver area, and did not work 20 hours per week.
 
 81-27 percent
 628-52 percent
 453-77 percent0
 
* Less than 500.
Source: Fiscal year SNAP Quality Control data.

                               Attachment
                               
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

March 2019
Nutrition
Issue Brief
Karen Cunnyngham
Proposed Changes to the Supplemental Nutrition Assistance Program: 
        Waivers to Work-Related Time Limits
    A rule proposed (https://www.federalregister.gov/documents/2019/02/
01/2018-28059/supplemental-nutrition-assistance-program-requirements-
for-able-bodied-adults-without-dependents) by the U.S. Department of 
Agriculture (USDA) on February 2, 2019, would reduce the number of non-
disabled childless people age 18 to 49 who are receiving Supplemental 
Nutrition Assistance Program (SNAP) benefits. Currently, SNAP 
participants in this group must engage in meaningful work activity or 
face time limits on their benefits. However, if a geographic area has 
an unemployment rate that is at least 20 percent above the national 
rate or has other indicators of insufficient jobs, states can request 
that USDA waive the time limit for SNAP participants living in the 
area. The proposed rule would reduce the number of areas qualifying for 
a waiver by imposing stricter standards--for example, states would not 
be able to request a waiver for counties with unemployment rates less 
than seven percent.
    This issue brief, the third in a series of briefs analyzing the 
impact of proposed changes to SNAP, provides background on SNAP work 
requirements, time limits, and the proposed regulatory changes. The 
brief also sheds light on the characteristics of SNAP participants who 
could face time limits on receiving SNAP benefits under the proposed 
regulatory change. With support from the Robert Wood Johnson 
Foundation, Mathematica conducted this analysis using SNAP Quality 
Control (QC) data from Fiscal Year 2017, the most recent year for which 
data are available.
SNAP Participants Potentially Affected By Proposed Changes
    In Fiscal Year 2017, an estimated 1.2 million SNAP participants 
were not working an average of at least 20 hours per week and would 
have faced time limits but did not because they lived in a waiver area. 
Among these SNAP participants who could be affected by the proposed 
regulatory changes:

   88 percent had household income at or below 50 percent of 
        the poverty level.

   About \1/3\ lived in SNAP households with reported income; 
        the average monthly household income of this group was $557, or 
        43 percent of the poverty level.

   11 percent were working, although less than an average of 20 
        hours per week, and another six percent lived with someone else 
        who was working.

            A greater share of these SNAP participants lived in poverty 
        (97 percent) compared to other SNAP participants (80 percent).

   5 percent lived with a person with a disability.

   The average monthly SNAP benefit was $181 per person.
  
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: Fiscal year 2017 SNAP QC data.

    Under the proposed rule, an estimated \3/4\ of these SNAP 
participants would be newly subject to a 3 month limit on their 
benefits, according to USDA. Some of them would increase existing work 
to an average of 20 hours per week, find work, or meet the work 
requirements by participating in an employment and training program or 
workfare (unpaid work through a state-approved program). However, USDA 
estimates that \2/3\ (755,000 people in 2020) would not meet the 
additional work requirements and would therefore lose eligibility after 
3 months. For those living with others unaffected by the policy change, 
the SNAP household could continue to receive benefits, but the amount 
would be reduced; those living alone would lose all SNAP benefits.
SNAP Work Requirements and Current Waiver Policy
    Currently, SNAP participants age 16 to 59 must register for work 
unless they are already working at least 30 hours per week, have a 
disability, or meet other criteria, such as caring for a young child or 
an incapacitated person. Work registrants who are age 18 to 49 in 
childless SNAP households are subject to additional work requirements 
and a time limit: they must work an average of at least 20 hours per 
week to continue receiving SNAP benefits for more than 3 months in a 3 
year period. However, they are exempt from the time limits if they (1) 
participate in a qualifying employment and training program or other 
meaningful work activity, (2) have a discretionary exemption from the 
state agency, or (3) live in a waiver area, an area for which the state 
agency requested and received a Federal waiver from the time limits due 
to high unemployment (see waiver area timeline). In recent years, 
states based most requests for geographic waivers on the area 
qualifying for the extended unemployment benefits authorized during the 
Great Recession or experiencing a high unemployment rate. Currently, 17 
states have no waiver areas, either because no area in the state 
qualified or the state agency chose not to request a waiver (see map).

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
 Snapshot: Some SNAP Participants Age 18 To 21 Could Be Affected By the
                            Proposed Changes
 
    In 2017, about 498,000 SNAP participants were age 18 to 21, did not
 have a disability, and were in a childless SNAP household. Some of
 these young adults would newly face time limits under the proposed rule
 changes.
 
     One-third lived in a waiver area and did not work an
     average of at least 20 hours per week; these are the young adults
     who might lose their SNAP benefit because of the proposed changes.
 
     Slightly less than \1/2\ lived with a parent and ten
     percent lived with another relative, a spouse, or a peer; the
     remainder--about 40 percent--did not share food resources with
     another person.
 
     23 percent worked an average of 20 hours per week or more
     (enough to avoid time limits on their benefits), six percent were
     working fewer hours, and 17 percent were not working but lived with
     someone who was.
 
     The average monthly benefit was $142 per person.
------------------------------------------------------------------------
Source: Fiscal year 2017 SNAP QC data.

    USDA's proposed regulatory change would eliminate or modify some 
current waiver area policies and leave others unchanged, as shown in 
the table below.

                          Waiver area policies
------------------------------------------------------------------------
            Current policy                 Proposed regulatory change
------------------------------------------------------------------------
                    Criteria to establish waiver area
------------------------------------------------------------------------
The Department of Labor (DOL)          Eliminated
 designated the area as a Labor
 Surplus Area based on a recent 24
 month average unemployment rate that
 is either at least ten percent or at
 least six percent and at least 20
 percent above the national average
DOL determined that the area meets     No change
 the criteria for extended
 unemployment benefits, available to
 workers who have exhausted regular
 unemployment insurance benefits
 during periods of high unemployment
Data from the Bureau of Labor          No change
 Statistics (BLS) show the area had a
 recent 12 month average unemployment
 rate greater than ten percent
Data from BLS show the area had a      The unemployment rate also must
 recent 24 month average unemployment   be at least seven percent
 rate at least 20 percent above the
 national average
------------------------------------------------------------------------
                       Other waiver area policies
------------------------------------------------------------------------
Waivers may be statewide               Only waivers based on extended
                                        unemployment benefits may be
                                        statewide
State agencies may combine data from   State agencies may combine data
 sub-state areas, such as counties,     only for areas collectively
 that are contiguous, share an          designated as Labor Market Areas
 economic region, or both               by BLS
Waivers may extend beyond the fiscal   Waivers based on a 24 month
 year                                   average unemployment rate may
                                        not extend beyond the fiscal
                                        year
------------------------------------------------------------------------

Estimated Impact
    The proposed regulatory changes would result in a 2.5 percent 
reduction in spending on SNAP benefits nationally, according to USDA 
estimates. The potential impact varies by state and depends on a 
variety of factors, including state agency policies, the local labor 
market, and the characteristics and circumstances of the participants. 
For example, SNAP participants in the 17 states without waiver areas 
would not be affected by the proposed changes because they already face 
time limits unless they are engaged in meaningful work activities or 
are exempt for other reasons. In other states, the state agency may 
offer a slot in a qualifying employment and training program to 
participants who would otherwise face a time limit or use Federal 
``percentage exemptions'' to exempt some SNAP participants from the 
time limit.
    In many states with current waiver areas, at least some SNAP 
participants living in those areas will be newly required to work an 
average of at least 20 hours per week to continue receiving benefits 
for more than 3 months. Both the local labor market and SNAP 
participants' job readiness will affect their ability to find work. To 
provide some perspective, 21 percent of non-disabled childless SNAP 
participants age 18 to 49 worked an average of at least 20 hours per 
week, according to the Fiscal Year 2017 SNAP QC data. The percentage 
ranged from nine percent to 36 percent across states.
    In addition to job readiness, other characteristics and 
circumstances of SNAP participants will influence whether they lose 
eligibility for SNAP under the proposed change. For example, certain 
SNAP participants are not required to register for work because they 
are caring for an incapacitated person or meet other criteria; work 
requirements will not change for these participants. On the other hand, 
some participants who newly face a time limit may choose to forgo SNAP 
benefits and rely on other available resources, such as food banks or 
family members, rather than comply with work requirements.
Which states are more likely to be affected by the proposed changes?

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Percentage of non-disabled childless SNAP participants age 18 
        to 49 who were potentially subject to a time limit, lived in a 
        waiver area, and did not work 20 hours per week.
          Source: Fiscal year 2017 SNAP QC data.
          Note: States with a white background did not have waiver 
        areas in Fiscal Year 2017.
Differences in State Use of Waiver Areas
    Since SNAP time limits were reinstated after the Great Recession, 
some states have requested and received waivers for all or parts of the 
state while others have not requested any time limit waivers. The 
waiver area timeline illustrates how the prevalence of state time limit 
waivers changed from 2009 through 2018; the call-out box on the left 
shows state use of waiver areas in Fiscal Year 2017. While states with 
the highest unemployment rates in 2017--Alaska and New Mexico--had 
statewide waivers, others with overall unemployment rates above the 
national average of 4.4 percent chose not to apply for a waiver for any 
areas of the state.

------------------------------------------------------------------------
 
------------------------------------------------------------------------
                 State Waiver Areas in Fiscal Year 2017
------------------------------------------------------------------------
                             No waiver areas
------------------------------------------------------------------------
Alabama                              Missouri
Arkansas                             Nebraska
Delaware                             North Carolina
Florida                              Oklahoma
Indiana                              South Carolina
Iowa                                 Texas
Kansas                               Wisconsin
Maine                                Wyoming
Mississippi
------------------------------------------------------------------------
                            Some waiver areas
------------------------------------------------------------------------
Arizona                              New Jersey
Colorado                             New York
Connecticut                          North Dakota
Georgia                              Ohio
Hawaii                               Oregon
Idaho                                Pennsylvania
Kentucky                             South Dakota
Maryland                             Tennessee
Massachusetts                        Utah
Michigan                             Vermont
Minnesota                            Virginia
Montana                              Washington
New Hampshire                        West Virginia
------------------------------------------------------------------------
                            Statewide waiver
------------------------------------------------------------------------
Alaska                               Louisiana
California                           Nevada
District of Columbia                 New Mexico
Guam                                 Rhode Island
Illinois                             Virgin Islands
------------------------------------------------------------------------


                          Waiver area timeline
------------------------------------------------------------------------
 
------------------------------------------------------------------------
April 2009-September 2010           Congress temporarily suspended the
                                     time limits through the American
                                     Recovery and Reinvestment Act.
October 2010-December 2015          In Fiscal Year 2011, time limits
                                     continued to be waived based on
                                     extended unemployment benefits for
                                     45 states, the District of
                                     Columbia, Guam, and the Virgin
                                     Islands and in some areas of five
                                     additional states. By the end of
                                     Fiscal Year 2015, time limits were
                                     re-implemented in nine states and
                                     in some areas of 13 more states.
January 2016-Fiscal Year 2017       Few areas still qualified for
                                     extended unemployment benefits, but
                                     many areas received time limit
                                     waivers based on other indicators
                                     of high unemployment, such as an
                                     unemployment rate at least 20
                                     percent above the national average.
                                     Seventeen states had no waiver
                                     areas for most of this time.
December 2018                       Seventeen states have no waiver
                                     areas; seven states, the District
                                     of Columbia, Guam, and the Virgin
                                     Islands have time limit waivers for
                                     their entire area; and the
                                     remaining states have waivers for
                                     some but not all areas of the
                                     state.
------------------------------------------------------------------------

Sources
    Mathematica used Fiscal Year 2017 SNAP QC data to produce the 
estimates shown in the second half of page 1, the Snapshot on page 2, 
and the second paragraph and map on page 3. The underlying assumptions 
and key variables used are available upon request. USDA's estimated 
impact of the proposed regulatory changes, mentioned at the top of page 
2 and the first sentence of page 3, are drawn from the Regulatory 
Impact Analysis of the proposed rule. Finally, information on state 
waiver areas was compiled from FNS's ``ABAWD Waiver Status'' reports.

    This brief series was created by Mathematica in collaboration with 
the Robert Wood Johnson Foundation to analyze the impact of proposed 
changes to SNAP. Many individuals made important contributions, 
including Carmen Ferro, Sarah Lauffer, Joshua Leftin, Gwyneth Olson, 
and J.B. Wogan from Mathematica; Gina Hijjawi from RWJF; and Adam 
Zimmerman from Burness. Two other briefs in this series can be 
downloaded from Mathematica's website:

    Proposed Changes to the Supplemental Nutrition Assistance Program: 
Heating and Cooling Standard Utility Allowances and Earned Income 
(https://www.mathematica-mpr.com/our-publications-and-findings/
publications/proposed-changes-to-the-supplemental-nutrition-assistance-
program-heating-and-cooling-standard)
    Simulating Proposed Changes to the Supplemental Nutrition 
Assistance Program: Countable Resources and Categorical Eligibility 
(https://www.mathematica-mpr.com/our-publications-and-findings/
publications/simulating-proposed-changes-to-the-supplemental-nutrition-
assistance-program-countable-resources)

    For more information about Mathematica's work in this area, contact 
Senior Researcher Karen Cunnyngham at [email protected] 
or (202) 264-3480.

    The Chair. Thank you very much.
    Mr. Adolphsen--obviously, you are not getting a yellow 
light for some reason, so when you see the red light, just 
please try to wrap up.

         STATEMENT OF SAM ADOLPHSEN, VICE PRESIDENT OF
          EXECUTIVE AFFAIRS, FOUNDATION FOR GOVERNMENT
                   ACCOUNTABILITY, NAPLES, FL

    Mr. Adolphsen. Chair Fudge, Ranking Member Johnson, Members 
of the Nutrition Subcommittee, thank you for the privilege of 
testifying.
    I was brought up in a household that believed in hard work. 
My dad was a landscaper. My mom cleaned houses. A job was a 
point of pride, and I can still remember getting that first 
paycheck from a tough day raking blueberries in rural Maine.
    For many of us, that is our story. Work is central to our 
lives. It provides dignity and purpose. The growth of our 
communities is built on people living this experience, living 
the American dream. And work is key to achieving the long-term 
goal of the food stamp program, lifting people out of poverty. 
That is why Congress and President Clinton passed bipartisan 
work requirements for able-bodied adults on food stamps in 
1996. They recognized the power of work, and they were right. 
Where work requirements have been implemented, those leaving 
the program doubled their incomes in just 1 year. And they 
didn't just go to work in retail or fast food. They went back 
to work in more than 1,000 different industries.
    Now, these figures aren't extrapolations or anecdotes. Our 
experts studied the actual earnings of 600,000 able-bodied 
adults who left food stamps after work requirements were 
implemented in Florida, Kansas, and Arkansas.
    One young man in Arkansas--I will call him Nolan--reported 
no income while on welfare, $0. After work requirements were 
implemented, Nolan soon left the program. Then Nolan got a job. 
Within 1 year, he was earning $63,000, and by the end of 2 
years, he was making $93,000. Work requirements work.
    Unfortunately for millions of able-bodied adults on food 
stamps, this isn't the experience at all. And government bears 
a big part of the blame. When I was Chief Operating Officer of 
the food stamp agency in Maine, before we reinstated work 
requirements, I had 1,000 state employees helping fill out food 
stamp applications. But no one helping fill out job 
applications. We were letting people like Nolan down. 
Government should be giving a hand up, not just a hand out.
    The loopholes created at the Federal agency level have 
gutted the 1996 law, allowing work to be waived across the 
country by gerrymandering areas and using old economic data.
    I want you to remember two numbers, 2.6 million and 7.6 
million.
    First, 2.6 million. There are 2.6 million able-bodied 
adults on food stamps who will be waived from the work 
requirement this year, and three out of four don't work at all.
    Second, 7.6 million. There are 7.6 million available jobs 
today, and the lowest unemployment rate in 50 years. Employers 
are desperate for workers.
    To be clear, Federal law allows waivers only when there are 
not enough jobs, or unemployment is at least ten percent. But 
just 23 of the 1,100 counties and cities that waive work 
requirements have unemployment at or above ten percent. One 
California waiver county has 2.2 percent unemployment, and 
Ohio's waiver has more than doubled since 2017, even as its 
unemployment rate declined to near record low levels. Waivers 
from work shouldn't be so easy to get in the best economy in 
decades.
    Some have claimed that Congress rejected the type of 
changes proposed here by the Trump Administration, but the 
bipartisan 2018 Farm Bill, like every other farm bill since 
1996, reaffirmed the original work requirements, and it did not 
codify the current regulations that have allowed the waiver 
abuse.
    It is clear that the status quo does not reflect 
Congressional intent. Even Chairman Collin Peterson correctly 
pointed out that the loopholes have allowed states to 
``undermine Federal law.''
    The Trump Administration has the authority and the duty to 
fix the regulation and return waivers to their original purpose 
of exempting only those individuals in truly economically 
depressed areas. The track record of work requirements is 
clear. They work. And when this rule is implemented, we can all 
be confident that hundreds of thousands of Americans, people 
just like Nolan, will move from welfare to work and experience 
their own American dream.
    Thank you. I am happy to answer any questions.
    [The prepared statement of Mr. Adolphsen follows:]

   Prepared Statement of Sam Adolphsen, Vice President of Executive 
     Affairs, Foundation for Government Accountability, Naples, FL
Examining the Proposed ABAWD Rule
    Chair Fudge, Ranking Member Johnson, and Members of the Committee, 
thank you for the privilege of testifying. I am Sam Adolphsen, the Vice 
President of Executive Affairs at the Foundation for Government 
Accountability (FGA). FGA is a non-partisan research organization 
dedicated to helping millions of individuals achieve the American 
Dream.
    Prior to joining FGA, I served as the Chief Operating Officer of 
the Maine Department of Health and Human Services. In that role, I 
oversaw operations for Maine's welfare programs, including the food 
stamp program. My duties included direct oversight of the food stamp 
eligibility and policy office.
    I was fortunate to be brought up in a household that believed in 
hard work. My dad was a landscaper and my mom cleaned houses. I knew 
from a young age that work is not a dirty word--it is a good thing. A 
job was a point of pride, and I can still remember that first paycheck 
from a tough day raking blueberries in coastal Maine. I'm sure you 
remember your first job, too, and what it taught you.
    For so many of us that's our story-work is central to our lives. It 
provides us with dignity and purpose. The growth of our communities and 
our nation as a whole is dependent on people experiencing this--living 
their American Dream.
    And it is the key to achieving the long-term goals of the food 
stamp program: to help lift people out of poverty. Unfortunately, for 
millions of able-bodied adults on food stamps, this isn't the 
experience at all. Work isn't even in the picture and food stamp rules 
allow long-term dependency with no accountability.
    The law is clear: work requirements should be the standard for 
able-bodied adults with no children. And where the law is followed, 
work requirements have proven to move people from welfare to work and 
leave them better off. But despite an economy desperate for workers, 
loopholes in Federal food stamp rules continue to permit work 
requirements to be waived in states across the country, leaving 
millions of able-bodied adults with no kids on the sidelines.
Work Is Key to Achieving the Food Stamp Program's Goals
    In 1996, Congress passed--and President Clinton signed--
commonsense, bipartisan welfare reform. As part of that reform, most 
able-bodied, childless adults were required to work, train, or 
volunteer part-time as a condition of food stamp eligibility.\1\ These 
requirements applied to non-pregnant adults who are mentally and 
physically fit for employment, who are between the ages of 18 and 50, 
and who have no dependent children or incapacitated family members.\2\ 
Able-bodied adults who refused to meet these requirements were limited 
to just 3 months of food stamp benefits every 3 years.\3\
---------------------------------------------------------------------------
    \1\ 7 U.S.C.  2015(o) (2016), https://www.gpo.gov/fdsys/pkg/
USCODE-2016-title7/pdf/USCODE-2016-title7-chap51-sec2015.pdf.
    \2\ Ibid.
    \3\ Ibid.
---------------------------------------------------------------------------
    When it was first implemented in the 1990s, this commonsense work 
requirement moved millions of able-bodied adults from welfare to work 
and spurred rapid economic growth.\4\ Analyses of state-level 
implementation have reached similar conclusions.5-8 But this 
progress has been undermined by Federal loopholes that have allowed 
states to weaken and waive the requirements for millions of adults, 
even during periods of sustained economic growth.9-10 
States, which bear little of the cost for the program, continue to take 
advantage of these loopholes with regularity despite the booming 
economy. United States Department of Agriculture (USDA) Secretary Sonny 
Perdue recently noted in a hearing before Congress that the waivers, 
``were abused in Georgia,'' and he believes, ``are being abused in many 
places.'' \11\
---------------------------------------------------------------------------
    \4\ Kenneth Hanson and Karen S. Hamrick, ``Moving public assistance 
recipients into the labor force, 1996-2000,'' U.S. Department of 
Agriculture (2004), https://www.ers.usda.gov/webdocs/publications/
46832/49356_fanrr40.pdf.
    \5\ Jonathan Ingram and Nicholas Horton, ``The power of work: How 
Kansas' welfare reform is lifting Americans out of poverty,'' 
Foundation for Government Accountability (2016), https://thefga.org/wp-
content/uploads/2016/02/Kansas-study-paper.pdf.
    \6\ Jonathan Ingram and Josh Archambault, ``New report proves 
Maine's welfare reforms are working,'' Forbes (2016), https://
www.forbes.com/sites/theapothecary/2016/05/19/new-report-proves-maines-
welfare-reforms-are-working.
    \7\ Nicholas Horton and Jonathan Ingram, ``Work requirements are 
working in Arkansas: How commonsense welfare reform is improving 
Arkansans' lives,'' Foundation for Government Accountability (2019), 
https://thefga.org/wp-content/uploads/2019/01/Work-Requirement-are-
Working-in-Arkansas-How-Commonsense-Welfare-Reform-is-Improving-
Arkansans-Lives-1-9-19.pdf.
    \8\ Nicholas Horton and Jonathan Ingram, ``Commonsense welfare 
reform has transformed Floridians' lives,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/03/
FloridaTrackingStudyResearchPaper-3.1.19.pdf.
    \9\ Sam Adolphsen, et al., ``Waivers gone wild: How states have 
exploited food stamp loopholes,'' Foundation for Government 
Accountability (2018), https://thefga.org/wp-content/uploads/2018/06/
Waivers-Gone-Wild-6-5-18-update.pdf.
    \10\ Jonathan Ingram, et al., ``How the Trump administration can 
cut down on waivers gone wild,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/02/
LMA-Memo-FoodStampWaiversGoneWild-2.20.19.pdf.
    \11\ House Committee on Agriculture, ``Full House Agriculture 
Committee hearing with Secretary Perdue on the state of the rural 
economy,'' U.S. House of Representatives (2019), https://
www.youtube.com/watch?v=m8t4etV1X8g.
---------------------------------------------------------------------------
    As a result of these loopholes, most able-bodied adults receiving 
food stamps are not required to work. According to state data, nearly 
63 percent of able-bodied adults without dependents on the program--
some 2.6 million adults--will be waived from the work requirement in 
Fiscal Year 2019.12-13 With no work requirement in place, 
few able-bodied adults on the program actually work. Just two percent 
of able-bodied adults without dependents on food stamps work full-time, 
while roughly \3/4\ do not work at all.14-15
---------------------------------------------------------------------------
    \12\ Jonathan Ingram, et al., ``How the Trump administration can 
cut down on waivers gone wild,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/02/
LMA-Memo-FoodStampWaiversGoneWild-2.20.19.pdf.
    \13\ Jonathan Ingram and Nicholas Horton, ``How the Agriculture and 
Nutrition Act of 2018 would rein in workrequirement waivers gone 
wild,'' Foundation for Government Accountability (2018), https://
thefga.org/wpcontent/uploads/2019/01/How-the-Agriculture-and-Nutrition-
Act-of-2018-would-rein-in-work-requirement-waivers-gone-wild.pdf.
    \14\ Author's calculations based upon data provided by the U.S. 
Department of Agriculture on enrollment and work status of able-bodied 
adults without dependents. See, e.g., Food and Nutrition Service, 
``Supplemental Nutrition Assistance Program quality control database,'' 
U.S. Department of Agriculture (2016), https://host76.mathematica-
mpr.com/fns/PUBLIC_USE/2015/qcfy2015_st.zip.
    \15\ Council of Economic Advisers, ``Expanding work requirements in 
non-cash welfare programs,'' Executive Office of the President (2018), 
https://www.whitehouse.gov/wp-content/uploads/2018/07/ExpandingWork-
Requirements-in-Non-Cash-Welfare-Programs.pdf.
---------------------------------------------------------------------------
    These waiver loopholes have trapped millions of able-bodied adults 
in dependency. But these loopholes have also allowed state agencies to 
skip out on their duty to engage these adults and help put them back on 
the path to self-sufficiency. The work requirement was designed not 
just to require work or work activities by the recipient of the 
program, but also to require the administering agency to engage with 
able-bodied adults.\16\
---------------------------------------------------------------------------
    \16\ 7 U.S.C.  2015(d), https://www.gpo.gov/fdsys/pkg/USCODE-2016-
title7/pdf/USCODE-2016-title7-chap51-sec2015.pdf.
---------------------------------------------------------------------------
    In my role as chief operating officer at the Maine Department of 
Health and Human Services, I saw firsthand how--until we restored the 
work requirement statewide-agency bureaucrats would simply send out 
benefits on autopilot instead of engaging with adults to help reconnect 
them with their community. By waiving the work requirement for able-
bodied adults, the food stamp agency's responsibility to help people 
get back on their feet and move beyond welfare program dependency is 
also waived, making that important assistance more optional for the 
agency.
When Enforced, Work Requirements Promote Independence
    These commonsense work requirements have a proven track record of 
success. After Kansas restored these work requirements in 2013, the 
number of able-bodied adults without dependents on the program dropped 
by more than 75 percent.\17\ Those able-bodied adults went back to work 
in hundreds of diverse industries and their incomes more than doubled 
within a year.\18\ Better still, those higher incomes more than offset 
lost welfare benefits, leaving them financially better off.\19\
---------------------------------------------------------------------------
    \17\ Jonathan Ingram and Nicholas Horton, ``The power of work: How 
Kansas' welfare reform is lifting Americans out of poverty,'' 
Foundation for Government Accountability (2016), https://thefga.org/wp-
content/uploads/2016/02/Kansas-study-paper.pdf.
    \18\ Ibid.
    \19\ Ibid.
---------------------------------------------------------------------------
    Maine experienced similar successes after restoring the work 
requirement in 2014.\20\ The number of able-bodied adults without 
dependents on the program dropped by more than 90 percent and average 
wages more than doubled within a year.\21\
---------------------------------------------------------------------------
    \20\ Jonathan Ingram and Josh Archambault, ``New report proves 
Maine's welfare reforms are working,'' Forbes (2016), https://
www.forbes.com/sites/theapothecary/2016/05/19/new-report-proves-maines-
welfare-reforms-are-working.
    \21\ Ibid.
---------------------------------------------------------------------------
    When Arkansas followed suit in 2016, able-bodied adult enrollment 
dropped by 70 percent.\22\ Those adults saw their incomes more than 
double in the year after leaving the program and then more than triple 
in the second year.\23\ Higher wages more than offset lost food stamp 
benefits, leaving individuals better off than when they were trapped in 
dependency.\24\
---------------------------------------------------------------------------
    \22\ Nicholas Horton and Jonathan Ingram, ``Work requirements are 
working in Arkansas: How commonsense welfare reform is improving 
Arkansans' lives,'' Foundation for Government Accountability (2019), 
https://thefga.org/wpcontent/uploads/2019/01/Work-Requirement-are-
Working-in-Arkansas-How-Commonsense-Welfare-Reform-is-Improving-
Arkansans-Lives-1-9-19.pdf.
    \23\ Ibid.
    \24\ Ibid.
---------------------------------------------------------------------------
    These adults moved into many diverse industries, touching virtually 
every corner of the American economy. After Florida restored the work 
requirement in 2016, able-bodied adults without dependents found work 
far beyond the fast food or big box retail industries.\25\ In fact, 
these adults found work in more than 1,000 different industries.\26\ 
Better still, they used those initial jobs as stepping stones to other 
jobs in higher-paid industries. Nearly 70 percent of those who 
initially found work in the fast food industry or at temp agencies left 
those industries within a year, moving from lower-wage industries to 
higher-wage industries over time.\27\
---------------------------------------------------------------------------
    \25\ Nicholas Horton and Jonathan Ingram, ``Commonsense welfare 
reform has transformed Floridians' lives,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/03/
FloridaTrackingStudyResearchPaper-3.1.19.pdf.
    \26\ Ibid.
    \27\ Ibid.
---------------------------------------------------------------------------
    Work also provides powerful benefits far beyond the nominal value 
of earned wages. Work can help build new and positive social 
relationships, help individuals gain new skills, create new experiences 
that lead to future employment opportunities and higher incomes, and 
serves as the single best path out of poverty.\28\ It could even help 
solve major public health concerns like the opioid crisis.\29\ Work is 
a key predictor of success for someone recovering from substance abuse.
---------------------------------------------------------------------------
    \28\ Sam Adolphsen and Jonathan Ingram, ``Three myths about the 
welfare cliff,'' Foundation for Government Accountability (2018), 
https://thefga.org/wp-content/uploads/2018/02/Three-myths-about-the-
welfare-cliff-2-28-18.pdf.
    \29\ Jonathan Ingram and Sam Adolphsen, ``How moving able-bodied 
adults from welfare to work could help solve the opioid crisis,'' 
Foundation for Government Accountability (2019), https://thefga.org/wp-
content/uploads/2019/03/OpioidDeathsMemo-ResearchPaper-DRAFT4.pdf.
---------------------------------------------------------------------------
Employers, and the Economy, Desperately Need Workers
    At 3.8 percent, the nation's unemployment rate is hovering at its 
lowest point since 1969.\30\ The unemployment rate has stayed at or 
below four percent for 12 consecutive months, with some states seeing 
unemployment rates as low as 2.4 percent.31-32 Since June 
2017, 19 states have hit new record-low unemployment levels, including 
some who waive work requirements across their state.\33\
---------------------------------------------------------------------------
    \30\ Bureau of Labor Statistics, ``Labor force statistics from the 
current population survey,'' U.S. Department of Labor (2019), https://
data.bls.gov/timeseries/LNS14000000.
    \31\ Ibid.
    \32\ Bureau of Labor Statistics, ``Current unemployment rates for 
states and historical highs and lows, seasonally adjusted,'' U.S. 
Department of Labor (2019), https://www.bls.gov/web/laus/lauhsthl.htm.
    \33\ Ibid.
---------------------------------------------------------------------------
    More Americans are working today than at any point since the Bureau 
of Labor Statistics began tracking employment statistics.\34\ Average 
earnings have reached nearly $28 per hour--the highest level ever 
recorded.\35\ Nearly \3/4\ of all individuals now finding work were 
pulled off the sidelines and back into the labor force--a record 
high.\36\
---------------------------------------------------------------------------
    \34\ Bureau of Labor Statistics, ``Labor force statistics from the 
current population survey: Seasonally adjusted employment level,'' U.S. 
Department of Labor (2019), https://data.bls.gov/timeseries/
LNS12000000.
    \35\ Bureau of Labor Statistics, ``Employment, hours, and earnings 
from the current employment statistics survey: Average hourly earnings 
of all employees,'' U.S. Department of Labor (2019), https://
data.bls.gov/timeseries/CES0500000003.
    \36\ Council of Economic Advisers, ``Economic report of the 
President together with the annual report of the Council of Economic 
Advisors,'' Executive Office of the President (2019), https://
www.whitehouse.gov/wp-content/uploads/2019/03/ERP-2019.pdf.
---------------------------------------------------------------------------
    But even today's booming economy is not enough: employers are 
searching desperately to fill a record-high 7.6 million open jobs.\37\ 
At least \1/3\ of small businesses have unfilled job openings, the 
highest rate in 50 years.\38\ Employers are offering signing bonuses, 
student loan repayment, company cars, relocation fees, and more to find 
and retain talent--at all skill levels.\39\ For our economy to continue 
growing and thriving, we need the adults currently receiving food 
stamps and sitting on the sidelines to rejoin the workforce.
---------------------------------------------------------------------------
    \37\ Bureau of Labor Statistics, ``Job openings and labor turnover 
survey: Total nonfarm job openings,'' U.S. Department of Labor (2019), 
https://data.bls.gov/timeseries/JTS00000000JOL.
    \38\ Sam Adolphsen, ``There has never been a better time for 
welfare reform,'' Foundation for Government Accountability (2018), 
https://thefga.org/wp-content/uploads/2018/06/Its-Time-To-Get-To-Work-
FINAL.pdf.
    \39\ Ibid.
---------------------------------------------------------------------------
    Despite some concerns of a ``skills gap,'' the reality is that 
millions of jobs require little specialized education, training, or 
experience. In fact, according to the Bureau of Labor Statistics, 
nearly \3/4\ of the job openings that will occur over the next decade 
require a high school education or less.\40\ Nearly four out of five 
job openings require no training or less than a month's training on the 
job, while a whopping 87 percent require no prior experience.\41\
---------------------------------------------------------------------------
    \40\ Ibid.
    \41\ Ibid.
---------------------------------------------------------------------------
Loopholes Have Allowed States To Waive Work Requirements
    When Congress passed the food stamp work requirements into law in 
1996, it gave the Secretary of the United States Department of 
Agriculture the authority to waive work requirements in areas that had 
unemployment rates above ten percent or otherwise lacked job 
opportunities for these able-bodied adults.\42\
---------------------------------------------------------------------------
    \42\ 7 U.S.C.  2015(o)(4)(A) (2016), https://www.gpo.gov/fdsys/
pkg/USCODE-2016-title7/pdf/USCODE-2016-title7-chap51-sec2015.pdf.
---------------------------------------------------------------------------
    Despite these narrow parameters set forth by Congress, Federal 
rulemaking led to a regulation that is far more expansive than 
intended, creating loopholes and gimmicks for states to continue 
waiving work requirements for millions of able-bodied adults, even 
during periods of record economic growth.\43\ As a result, these 
commonsense requirements are waived wholly or partially in 33 states 
and the District of Columbia.\44\ As a result, nearly 2.6 million able-
bodied adults who would otherwise be required to work, train, or 
volunteer have those requirements waived altogether.\45\
---------------------------------------------------------------------------
    \43\ Sam Adolphsen, et al., ``Waivers gone wild: How states have 
exploited food stamp loopholes,'' Foundation for Government 
Accountability (2018), https://thefga.org/wp-content/uploads/2018/06/
Waivers-Gone-Wild-6-5-18-update.pdf.
    \44\ Jonathan Ingram, et al., ``How the Trump administration can 
cut down on waivers gone wild,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/02/
LMA-Memo-FoodStampWaiversGoneWild-2.20.19.pdf.
    \45\ Ibid.
---------------------------------------------------------------------------
    Although the statute specifies that the waivers should only apply 
to areas with high unemployment that lack a sufficient number of jobs, 
regulatory loopholes allow states to waive work requirements in areas 
with record-low unemployment by combining and gerrymandering them with 
areas with somewhat higher unemployment rates.\46\ These loopholes also 
allow states to use data from years ago, even when that data has no 
connection to current economic conditions.\47\ If that weren't bad 
enough, the regulation creates an alternative waiver option even in 
areas with unemployment rates below ten percent. Under this option, 
states can qualify for a waiver so long as their unemployment rates are 
20 percent above the national average during a 2 year period, no matter 
how low that rate is and no matter how many open jobs are 
available.\48\
---------------------------------------------------------------------------
    \46\ Sam Adolphsen, et al., ``Waivers gone wild: How states have 
exploited food stamp loopholes,'' Foundation for Government 
Accountability (2018), https://thefga.org/wp-content/uploads/2018/06/
Waivers-Gone-Wild-6-5-18-update.pdf.
    \47\ Ibid.
    \48\ Ibid.
---------------------------------------------------------------------------
    Of the more than 1,100 counties, towns, cities, and other 
jurisdictions where work requirements are currently waived, just 23 
have unemployment rates above ten percent.\49\ More than 800 of these 
jurisdictions have unemployment rates at or below five percent and 
nearly 200 have unemployment rates at or below three percent.\50\ The 
waived jurisdictions have unemployment rates as low as zero percent--
meaning work requirements are waived in areas with literally no 
unemployment.\51\ Despite claims that these areas are facing severe job 
shortages, the 33 states currently waiving the work requirement have 
more than a combined 3.7 million job openings posted online.\52\ These 
states are expected to experience nearly 13 million job openings per 
year over the next decade.\53\
---------------------------------------------------------------------------
    \49\ Jonathan Ingram and Sam Adolphsen, ``FNS-2018-0004-5999,'' 
Opportunity Solutions Project (2019), https://solutionsproject.org/wp-
content/uploads/2019/03/OSP-Comment-and-supplement.pdf.
    \50\ Ibid.
    \51\ Ibid.
    \52\ Author's calculations based upon data provided by Haver 
Analytics on February 2019 job postings gathered from more than 16,000 
Internet job boards, corporate boards, and other job sites.
    \53\ Author's calculations based upon data provided by state labor 
market information agencies on average annual projected job openings 
over the next decade.
---------------------------------------------------------------------------
Loopholes Have Expanded Work Requirement Exemptions
    Regulatory loopholes have also exempted hundreds of thousands of 
able-bodied adults from the work requirement in direct conflict with 
Congressional intent. Shortly before leaving office, the Clinton 
Administration created new exemptions for able-bodied adults who reside 
in households with children--regardless of whether they are parents or 
caretakers--as well as 50 year old able-bodied adults who would 
otherwise be required to work, train, or volunteer under the 
statute.54-55
---------------------------------------------------------------------------
    \54\ Jonathan Ingram, et al., ``Why the Trump administration should 
move able-bodied adult siblings from welfare to work,'' Foundation for 
Government Accountability (2019), https://thefga.org/wp-content/
uploads/2019/03/ABAWD-Siblings-to-Work-Research-Paper-DRAFT6.pdf.
    \55\ Jonathan Ingram, et al., ``Closing the food stamp loophole 
that allows 50-year-olds to avoid work,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/01/
50-Year-Old-Food-Stamp-Loophole-Memo-1.24.19.pdf.
---------------------------------------------------------------------------
    These exemptions conflict with the plain meaning of the food stamp 
statute, Congressional intent, prior interpretation by state agencies, 
and even Food and Nutrition Service's own interpretation of the same 
terms.56-57
---------------------------------------------------------------------------
    \56\ Jonathan Ingram, et al., ``Why the Trump administration should 
move able-bodied adult siblings from welfare to work,'' Foundation for 
Government Accountability (2019), https://thefga.org/wp-content/
uploads/2019/03/ABAWD-Siblings-to-Work-Research-Paper-DRAFT6.pdf.
    \57\ Jonathan Ingram, et al., ``Closing the food stamp loophole 
that allows 50-year-olds to avoid work,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/01/
50-Year-Old-Food-Stamp-Loophole-Memo-1.24.19.pdf.
---------------------------------------------------------------------------
The Proposed Rule Would Help Address Waiver Abuse
    The proposed rule represents a significant improvement over the 
status quo.58-59 By closing some of the most egregious 
loopholes that have led to widespread waiver abuse, the proposed rule 
brings waiver guidance more in line with statutory requirements that 
have been enshrined in law for more than 20 years. Under the proposal, 
states can continue to request waivers in areas that lack sufficient 
jobs but will not have as many avenues to abuse the process.
---------------------------------------------------------------------------
    \58\ Jonathan Ingram, et al., ``How the Trump administration can 
cut down on waivers gone wild,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/02/
LMA-Memo-FoodStampWaiversGoneWild-2.20.19.pdf.
    \59\ Jonathan Ingram and Sam Adolphsen, ``FNS-2018-0004-5999,'' 
Opportunity Solutions Project (2019), https://solutionsproject.org/wp-
content/uploads/2019/03/OSP-Comment-and-supplement.pdf.
---------------------------------------------------------------------------
    The first major area of change in the proposed rule is an attempt 
to reduce gerrymandering abuse. Federal law allows the Secretary to 
grant waivers in areas that lack sufficient jobs, but does not define 
``areas'' for waiver purposes.\60\ States have used this ambiguous 
language to gerrymander jurisdictions together to form ``areas'' solely 
to maximize the number of able-bodied adults waived from the work 
requirement.\61\ Illinois, for example, combines 101 of the state's 102 
counties into a single ``area,'' while California combines all but 
three counties into a single ``area'' for waiver purposes.\62\ These 
waived jurisdictions do not form a single, local region with a shared 
economy. Instead, they just happen to the jurisdictions that, when 
combining data, just marginally meet the current regulatory thresholds 
for waivers.
---------------------------------------------------------------------------
    \60\ Sam Adolphsen, et al., ``Waivers gone wild: How states have 
exploited food stamp loopholes,'' Foundation for Government 
Accountability (2018), https://thefga.org/wp-content/uploads/2018/06/
Waivers-Gone-Wild-6-5-18-update.pdf.
    \61\ Ibid.
    \62\ Jonathan Ingram, et al., ``How the Trump administration can 
cut down on waivers gone wild,'' Foundation for Government 
Accountability (2019), https://thefga.org/wp-content/uploads/2019/02/
LMA-Memo-FoodStampWaiversGoneWild-2.20.19.pdf.
---------------------------------------------------------------------------
    The proposed rule attempts to limit this abuse by only allowing 
states to combine jurisdictions together for waiver purposes if they 
form labor market areas.\63\ The purpose of this change is to ``target 
waivers to jurisdictions with a demonstrable lack of sufficient jobs,'' 
as required by the statute.\64\ But even this could be subject to 
abuse. States could still seek waivers in jurisdictions that have 
sufficient jobs and in areas where there are sufficient jobs within 
commuting distance.\65\
---------------------------------------------------------------------------
    \63\ Ibid.
    \64\ Ibid.
    \65\ Ibid.
---------------------------------------------------------------------------
    One solution the Trump Administration could take to solve this 
remaining problem--and better align the proposed rule with the food 
stamp statute--would be to prohibit states from combining jurisdictions 
for waiver purposes at all and to eliminate waivers for jurisdictions 
located in commuting zones with sufficient jobs.\66\
---------------------------------------------------------------------------
    \66\ Ibid.
---------------------------------------------------------------------------
    The second major change in the proposed rule sets a minimum 
unemployment floor for states seeking waivers. Although Federal law 
defines high unemployment as above ten percent, existing regulations 
allow waivers whenever an area's unemployment rate is 20 percent above 
the national average, with no minimum floor.67-68 This 
guarantees that at least some portion of the country will always be 
granted waivers, even during periods of unprecedented economic growth.
---------------------------------------------------------------------------
    \67\ 7 U.S.C.  2015(o)(4)(A) (2016), https://www.gpo.gov/fdsys/
pkg/USCODE-2016-title7/pdf/USCODE-2016-title7-chap51-sec2015.pdf.
    \68\ 7 CFR  273.24(f)(3)(iii) (2018), https://www.gpo.gov/fdsys/
pkg/CFR-2018-title7-vol4/pdf/CFR-2018-title7-vol4-sec273-24.pdf.
---------------------------------------------------------------------------
    The proposed rule attempts to address this abuse by setting a 
minimum floor of seven percent unemployment.\69\ But even this may not 
be enough to stop states from pursuing waivers in areas with sufficient 
jobs.
---------------------------------------------------------------------------
    \69\ Food and Nutrition Service, ``Supplemental Nutrition 
Assistance Program: Requirements for able-bodied adults without 
dependents,'' U.S. Department of Agriculture (2018), https://fns-
prod.azureedge.net/sites/default/files/snap/ABAWDtoOFR.pdf.
---------------------------------------------------------------------------
    A minimum unemployment rate of seven percent only truly matters 
during a period of near full employment, as the threshold would only 
activate when the national unemployment rate falls below 5.8 percent 
for a sustained 2 year window.\70\ This threshold is just slightly 
above the historical average ``natural'' unemployment rate--the level 
most economists agree is ``full employment''--and just below the 
average unemployment rate over the last 70 years.\71\
---------------------------------------------------------------------------
    \70\ Jonathan Ingram and Sam Adolphsen, ``FNS-2018-0004-5999,'' 
Opportunity Solutions Project (2019), https://solutionsproject.org/wp-
content/uploads/2019/03/OSP-Comment-and-supplement.pdf.
    \71\ Ibid.
---------------------------------------------------------------------------
    The Trump Administration could strengthen the rule even further--
and more closely align with the food stamp statute--by raising that 
threshold to ten percent. This would better target waivers to areas 
that have objectively high unemployment and lack sufficient jobs.
The Proposed Rule Better Reflects Congressional Intent
    Although some have claimed the proposed rule was ``specifically 
rejected'' by Congress in the 2018 Farm Bill, nothing could be further 
from the truth. The House-passed version of the farm bill made 
significant changes to the work requirement, but those changes were 
materially different from the proposed rule. The House-passed bill 
eliminated the time limit for able-bodied adults without dependents 
entirely, focusing instead on strengthening the work registration 
requirements for a broader group of able-bodied adults. It created new 
waivers and exemptions from the work registration requirements, but the 
qualifications for those waivers were materially different from those 
in the proposed rule. In short, the changes in the proposed rule were 
never even considered by Congress.
    Far from rejecting the changes proposed by the Trump 
Administration, the 2018 Farm Bill left in place the original work 
requirements first enacted in 1996. Those statutory requirements serve 
as the basis for the proposed rule, which simply seeks to close 
unlawful loopholes created through regulatory guidance. It is 
undisputed that the current regulatory framework does not reflect 
Congressional intent. Even Chairman Collin Peterson noted last year 
that the loopholes have allowed states to ``undermine Federal law'' by 
abusing these waivers.\72\
---------------------------------------------------------------------------
    \72\ Chuck Abbot, ``Food stamp revisions possible but not radical 
change, says key House Democrat,'' Fern's Ag Insider (2018), https://
thefern.org/ag_insider/food-stamp-revisions-possible-not-radical-
change-says-key-house-democrat.
---------------------------------------------------------------------------
    By leaving in place those statutory requirements exactly as first 
enacted in 1996, Congress signaled that it did not wish to codify the 
unlawful waiver expansions created through regulation. This left in 
place the authority--and the duty--of the Trump Administration to 
return these waivers to their original purpose.
Work Will Improve Lives and Boost the Economy
    The proposed rule represents a significant step forward in moving 
able-bodied adults from welfare to work and realigning Federal 
regulations with statutory requirements. It would not simply require 
millions of able-bodied adults without children to work--the rule will 
also encourage state agencies to do a better job of actually engaging 
with individuals and putting them back on the pathway to self-
sufficiency and better lives. The requirement will help connect able-
bodied adults who are out of work with employers who desperately need 
workers to fill open jobs. For those who cannot work immediately, it 
will connect individuals to available job training or educational 
opportunities. Whether through work, training, or volunteering, these 
adults will be better connected to their communities. This will 
ultimately move millions more able-bodied adults from welfare to work 
and from government dependence to independence.
                               Attachment
More Than 2,567,550 Able-Bodied Adults Have No Food Stamp Work 
        Requirements

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: U.S. Department of Agriculture.

    The Chair. Thank you. Ms. Hamler-Fugitt.

   STATEMENT OF LISA HAMLER-FUGITT, EXECUTIVE DIRECTOR, OHIO 
             ASSOCIATION OF FOODBANKS, COLUMBUS, OH

    Ms. Hamler-Fugitt. Thank you. Good morning, Chair Fudge, 
Ranking Member Johnson, and distinguished Members of the 
Subcommittee. Thank you for convening this hearing today and 
inviting me to testify on the Trump Administration's proposed 
rules related to unemployed or underemployed adults without 
dependents participating in the SNAP Program.
    My name is Lisa Hamler-Fugitt, and I serve as the Executive 
Director of the Ohio Association of Food Banks, Ohio's largest 
charitable response to hunger. We distributed over 200 million 
pounds of emergency food last year in an attempt to fill the 
gap for hungry Ohioans, but SNAP provides 12 times as much food 
while infusing resources into local communities.
    The Administration's proposed rule would limit access to 
SNAP for adults with very limited resources without improving 
their overall employment outlook or health outcomes. Based on 
my Association's firsthand experience operating the SNAP Work 
Experience Program, which provides services exclusively for 
clients required to find work under the current SNAP rule, I am 
here to provide you with my perspective on the impacts that 
this proposed rule would have in Ohio.
    Currently, 38 of Ohio's 88 counties have waived SNAP time 
limits due to high unemployment. If the proposed rule were to 
take effect today with the seven percent threshold for waiver 
eligibility, only three Ohio counties would qualify for the 
waiver. These three counties account for less than one percent 
of Ohio's current SNAP population, meaning that nearly all 
would be subject to the time limit if the proposed rule went 
into effect.
    Unfortunately, we know from our extensive experience that 
those subject to the time limit have profound barriers to 
employment. The Work Experience Program conducts in depth, 
comprehensive client assessments to determine the client 
employability and identify barriers to employment. Over the 
first 2 years of our program, we completed over 5,000 in depth 
interviews and gathered information on 5,500 self-reported 
employment and skills assessments. Our results represent the 
state's most comprehensive and up-to-date data available on 
this population.
    Our single largest and biggest takeaway is the term ABAWD 
is a complete misnomer for who this population is. One in three 
clients reported a physical or mental limitation ranging from 
back injuries to heart conditions to depression to PTSD. Many 
participants appear to be marginally or functionally 
illiterate, and likely experiencing significant learning 
disabilities. Additionally, many clients appear to have social 
and/or cognitive impairments, difficulty communicating, and a 
tendency to engage in repetitive behaviors, all signs of autism 
spectrum disorder. We believe that there are high levels of 
undiagnosed autism and other developmental disabilities in this 
population. One in three clients have no high school diploma or 
GED. Nearly \1/2\ reported that they do not have reliable 
transportation, whether through a personal vehicle, public 
transit, or ride sharing with family or friends. And 60 percent 
report that they do not have a current, valid driver's license. 
About \1/3\ of our clients had felony convictions, a stigma 
which can follow someone for a lifetime, even if their release 
is meant to suggest that they have been rehabilitated.
    Many of our clients are parents or caregivers with 
responsibilities that can serve as barriers to employment, and 
one in four of our clients had children that were not in their 
custody and many spent time parenting those children on a 
regular basis while the custodial parent works. Additionally, 
one in ten reported they are caregivers for family, friends, or 
relatives. In addition to these issues, many of our clients 
face other challenges which makes finding employment difficult.
    We serve hundreds of individuals who have aged out of the 
foster care system, only to find themselves living in homeless 
shelters, with friends, or on the street. Many other clients 
are experiencing challenges like homelessness and language 
barriers. These individuals face daunting challenges in finding 
employment, even when general unemployment rates are low, which 
is exactly why Congress gave states the option to waive the 
time limit in areas where there were insufficient jobs for 
those who were subject to the requirement.
    I would like to share just one story of a client, a 
Somalian refugee who relies on public transportation and 
requires an interpreter to fulfill his mandatory work 
requirements. Due to a paperwork error, he was mistakenly cut 
off his SNAP benefits and was sent to our local food pantry 
network to get food, until his case could be sorted out. Sadly, 
this case is not unique. Tens of thousands of real people like 
him are slipping through the cracks.
    We know all too well that harsh and arbitrary time limits 
are misguided and only increase hunger and hardship. The 
proposed rule would shift the burden of providing food from the 
Federal Government on to cities, states, and local charities 
like mine. It would be harmful to the local economies, grocers, 
retailers, and the agriculture community by reducing the amount 
of SNAP benefits and dollars available and economic activity.
    The Chair. Please wrap up for me.
    Ms. Hamler-Fugitt. Most importantly, the rule sidesteps the 
will of Congress, which rejected these changes when it enacted 
the 2018 Farm Bill.
    We hope that we can work together to stop these harmful 
policies from taking effect, and I would be happy to answer any 
questions you may have.
    [The prepared statement of Ms. Hamler-Fugitt follows:]

  Prepared Statement of Lisa Hamler-Fugitt, Executive Director, Ohio 
                 Association of Foodbanks, Columbus, OH
          The findings of our comprehensive assessment of able-bodied 
        adults without dependents can be found at our website at: 
        http://ohiofoodbanks.org/wep/WEP-2013-2015-report.pdf.

    Good morning, Chair Marcia L. Fudge, Ranking Member Dusty Johnson, 
and distinguished Members of the U.S. House Agriculture Subcommittee on 
Nutrition, Oversight, and Department Operations.
    My name is Lisa Hamler-Fugitt and I serve as the executive director 
of the Ohio Association of Foodbanks, Ohio's largest charitable 
response to hunger. My association represents Ohio's 12 Feeding America 
food banks and their more than 3,500 member hunger relief charities. 
Our mission is to provide food and resources to people in need and to 
pursue areas of common interest for the benefit of people in need. Last 
year, the association distributed 216 million pounds of food to more 
than two million low-income Ohioans--one in six of our hungry friends 
and neighbors.
    Thank you for convening this hearing today and inviting me to 
testify on the Trump Administration Proposed Rule: Supplemental 
Nutrition Assistance Program (SNAP): Requirements for Able-Bodied 
Adults without Dependents RIN 0584-AE57.
    This rule would limit the ability of states to waive the 3 month 
time limit that applies to unemployed and underemployed Able-Bodied 
Adults Without Dependents who receive benefits through the Supplemental 
Nutrition Assistance Program (SNAP).
    I'm here today to provide you with our association's firsthand 
experiences operating the SNAP Work Experience Program that serves only 
work-mandated unemployed and underemployed Able-Bodied Adults without 
Dependents in Franklin County, Ohio. The program began in SFY 2014, 
when the Administration of then-Governor John Kasich eliminated the 
statewide waiver and instead applied for a limited number of exemptions 
for only 16 predominantly rural, white counties. The Administration did 
not request exemptions for eligible cities where minority communities 
are concentrated and unemployment is high. Ohio had a statewide waiver 
that had been in place since mid-2000, when the Ohio General Assembly 
enacted legislation to compel the State of Ohio to apply for and 
implement the waiver.
Current Ohio Landscape

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
Ohio Counties Waived in FFY 2019
    Adams, Ashtabula, Athens, Belmont, Brown, Carroll, Clinton,
 Columbiana, Coshocton, Crawford, Cuyahoga, Erie, Gallia, Guernsey,
 Harrison, Highland, Hocking, Huron, Jackson, Jefferson, Lawrence,
 Lorain, Lucas, Mahoning, Meigs, Monroe, Morgan, Muskingum, Noble,
 Ottawa, Perry, Pike, Richland, Ross, Scioto, Trumbull, Vinton, and
 Washington
------------------------------------------------------------------------

    In FFY 2019, there are 38 counties in Ohio where the time limit has 
been waived due to high unemployment. Based on unemployment data 
obtained from the U.S. Bureau of Labor Statistics, the 24 month average 
unemployment rate in each of the counties was greater than 120 percent 
of the national unemployment rate during the same 24 month period.\1\
---------------------------------------------------------------------------
    \1\ Ohio Department of Job and Family Services FAL-171 Federal 
Fiscal Year 2019: Able-Bodied Adults without Dependents, http://
jfs.ohio.gov/ofam/FAL-171-FFY-2019-ABAWD-090718.stm.
---------------------------------------------------------------------------
    If the proposed rule were to take effect today with the seven 
percent threshold for waiver eligibility, only three Ohio counties 
would qualify for a time-limit waiver (according to BLS unemployment 
data over the most recent 24 month period available).\2\ These three 
counties--Adams, Meigs, and Monroe--account for less than one percent 
of Ohio's SNAP population. If the geographic distribution of ABAWDs 
matches that of the broader SNAP population, over 99 percent of Ohio's 
ABAWDs would now be subject to the SNAP time limit (up from 52 percent 
under current policy). In effect, the rule would add additional 
barriers blocking Ohioans in the poorest parts of the state from 
accessing basic nutrition.\3\
---------------------------------------------------------------------------
    \2\ Bureau of Labor Statistics. (2018). Local Area Unemployment 
Statistics, January 2017-December 2018.
    \3\ Supplemental Nutrition Assistance Program: Requirements for 
Able-Bodied Adults without Dependents [RIN 0584-AE57] The Center for 
Community Solutions, March 26, 2019.

 
 
 
           Current Policy             If Proposed Rule Took Effect Today
      Federal Fiscal Year 2019        Based on Most Recent BLS 24 Month
      (10-1-2018 to 9-30-2019)            Average Unemployment Data
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     
          Map by The Center for Community Solutions.
          Sources: U.S. Census Bureau. (2018) American Community Survey 
        5 year estimates, poverty status in the past 12 months. Feeding 
        America. (2018). Map the meal gap 2018: overall food insecurity 
        in Ohio by county in 2016. Bureau of Labor Statistics. (2018). 
        Local Area Unemployment Statistics, January 2017-December 2018. 
        Author's analysis, assuming waiver eligibility floor of seven 
        percent county unemployment rates.

          Americans want to work. The proposed SNAP able-bodied 
        restrictions will hurt many who want to work but can't for a 
        whole host of reasons--often because there are no jobs for 
        them.

    However, living in a county where the time-limit has been waived 
does not exempt ABAWDs from their obligation to participate in the 
labor force. Ohio administers a mandatory SNAP Employment and Training 
(SNAP E&T) program that is inclusive of ABAWDs. Under SNAP E&T, ABAWDs 
must participate in education/job training, job search/job readiness 
activities, or work experience or else be subject to a sanction, 
regardless of whether the individual lives in a county where the time-
limit has been waived.\4\
---------------------------------------------------------------------------
    \4\ http://jfs.ohio.gov/ofam/FAL-171-FFY-2019-ABAWD-090718.stm.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
Background: How Did We Get Here?
    Under the 1996 welfare law, adults aged 18-49 who are not physically
 or mentally unfit for work or caring for a minor child are eligible to
 receive Food Stamp/Supplemental Nutrition Assistance Program (SNAP)
 benefits for only 3 months in a 36 month period, unless the individual
 meets certain work requirements. These individuals are known as Able
 Bodied Adults Without Dependents (ABAWD) and are required to work at
 least 20 hours a week, participate in qualifying work or training
 program activities for at least 20 hours a week, or live in an area
 with high unemployment where the 3 month limit is temporarily waived.
    On the request of a state SNAP agency, the law also gives the USDA
 the authority to temporarily waive the time limit in areas that have an
 unemployment rate of over ten percent or a lack of sufficient jobs. The
 law also provides state agencies with a limited number of percentage
 exemptions that can be used by states to extend SNAP eligibility for
 ABAWDs subject to the time limit. The Department proposes to amend the
 regulatory standards by which the Department evaluates state SNAP
 agency requests to waive the time limit and to end the unlimited
 carryover of ABAWD percentage exemptions.
    When signing the welfare law in 1996, President Clinton singled out
 this as one of the bill's most harmful provisions and called for it to
 be substantially changed.
    The Administration's proposed rule RIN 0584-AE57 would encourage
 broader application of the statutory ABAWD work requirement and is
 intended to circumvent the will of Congress.[\5\]
[\5\] President William Clinton, Statement on Signing the Personal
 Responsibility and Work Opportunity Reconciliation Act of 1996, August
 22, 1996, http://www.presidency.ucsb.edu/ws/?pid=53219.
------------------------------------------------------------------------

SNAP Is Essential for Ohio
    The households served by our statewide emergency food assistance 
network represent diverse circumstances and challenges. Clients face a 
wide array of obstacles to food security, such as health issues, 
education levels, housing instability, unemployment/underemployment, 
disabilities, and insufficient income and resources.
    Our association recognizes that hunger is merely a symptom of 
poverty and we engage in other efforts to eradicate poverty and hunger. 
For more than a decade, we have provided services to connect low-income 
Ohioans with nutrition benefits and other work support programs. 
Knowing first-hand that hunger and health are directly linked, the 
association partners with the Ohio Department of Job and Family 
Services and the USDA Food and Nutrition Service as the state's SNAP 
outreach grantee. The association and our member food banks administer 
and conduct outreach and education on this critical food assistance 
program. We work on the front lines--reaching hungry Ohioans where they 
work, live, pray, play and learn.
    For more than 25 years, we have advocated for equitable public 
policy at the state and Federal levels to decrease hunger in Ohio. We 
work with local, regional, and national partners to inform 
policymakers, media, and other stakeholders about the issues facing 
Ohio's families.
    We know that SNAP is the first line of defense against hunger in 
our state and nation--in fact, our charitable network could never 
respond to the lack of adequate access to nutritious food on our own. 
In December 2018, Ohio SNAP issuance was $165 million, which provided 
supplemental food assistance benefits to 1.3 million Ohioans living in 
660,000 Assistance Groups. These households received an average of 
$124.48 in SNAP benefits per person, per month. Nearly \1/2\ (43 
percent) were children.\6\
---------------------------------------------------------------------------
    \6\ http://jfs.ohio.gov/pams/Case-Load-Summary-Report--December-
(002).stm.
---------------------------------------------------------------------------
    To get SNAP benefits, households must meet certain tests, including 
resource and income tests. Benefits are limited to a person with net 
income at or below 100% FPL (monthly net income of no more than $1,041 
per month for a household of one and $1,409 for a household of two 
people). The program also has work and work registration requirements 
for everyone 16 to 60 years of age.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
    In October 2013, 1.8 million Ohioans were receiving SNAP to help
 feed their families.i As of December 2018, enrollment had fallen to 1.3
 million, a decline of more than 26 percent.ii
 
        i ``Statement on the November 1st Cuts to the SNAP Program,''
     Food Research and Action Center. http://frac.org/statement-on-the-
     november-1st-cuts-to-the-snap-program/.
        ii Ohio Association of Foodbanks analysis of Ohio Department of
     Job and Family Services Public Assistance Monthly Statistics. http:/
     /jfs.ohio.gov/pams/index.stm.
------------------------------------------------------------------------

The Beginning and Approach of Ohio's Work Experience Program in 
        Franklin County, Ohio
    The association was approached in late 2013 by the Franklin County 
Department of Job and Family Services (FC[D]JFS) to assist them in the 
development of a process to screen and evaluate an estimated 12,000 
Franklin County SNAP recipients that would be affected by the state's 
decision to reimpose the ABAWD work requirement and time limit.
    The goals of this partnership, which began as a pilot program, were 
multifaceted, including not only assisting recipients in meeting the 
Federal work requirement in order to maintain their food assistance, 
but also providing them with meaningful work experience and job 
training and enhancing their ability to secure sustainable employment 
in order to become economically self-sufficient. To do that we needed 
to understand the barriers and challenges these Ohioans already face.
    The association developed and utilized a Work Experience Assessment 
Portal to conduct in-depth, comprehensive interviews and assessments 
designed to determine employability and identify barriers to 
employment. The data collected included: age and gender demographics, 
access to reliable transportation, methods of communication and 
identification, housing and living situations, criminal history, 
education completion, physical and mental health disabilities and 
limitations, employment history, and dependent and family 
relationships. These findings provided us with a deeper understanding 
of the issues and challenges participants face and provided us a 
framework for identifying and recruiting the types of community 
organizations that we needed to partner with that could help and host 
participants in order for them to meet the work requirements.
    Our recruitment process for developing new sites involved calling, 
mailing, e-mailing, and visiting numerous nonprofit and faith-based 
organizations in Franklin County. Each organization is required to sign 
a Memorandum of Agreement, establishing a strong partnership that also 
holds these organizations accountable for reporting hours for clients. 
The Work Experience Program Host sites (WEP) provided each participant 
with a volunteer assignment intended to provide training, education, 
and on-the-job work experience that would be beneficial in their search 
for future employment. Some sites even report hiring WEP participants 
at their organizations when they had open positions available.
    Prior to the participants being placed at a WEP host site, they 
were required to attend a three-part clinic to conduct an FBI/BCI 
background check and meet with possible employers and other employment 
service providers who helped secure identification, develop resumes, 
and demonstrate job search opportunities.
    After clients complete the assessment and attend the clinic, 
participants are placed at a qualified WEP host site to complete their 
monthly work requirement which allows them to maintain their SNAP 
benefit eligibility for the duration of their participation.
    Our interest in the ABAWD participants did not end when they exit 
our program. We are concerned about the well-being and long-term 
outcomes of our clients. The association conducted a post-WEP client 
study to examine the course of clients after they exited the program. 
The findings of this report provide information about post-
participation employment status and the most common causes of failure 
to comply with mandated ABAWD work requirements and WEP involvement.
    During the project's pilot period, from December 10, 2013 through 
September 1, 2015, WEP Assessment Specialists completed in-depth 
interviews with 4,827 ABAWD participants and gathered information from 
5,434 self-reported employability and skills assessments. Over the 
nearly 2 year pilot, the information obtained represents the most 
comprehensive and up-to-date information collected about this 
misunderstood population. These findings offer instructive, meaningful 
insight into who these individuals are and what is required in order to 
help address the barriers and challenges they face as they attempt to 
secure stable employment. These findings have provided the association 
with a framework that continues to guide our Work Experience Program 
partnership with the Franklin County Department of Job and Family 
Services that is now in its sixth year of operation.
ABAWD--``Able-Bodied''--Is a Complete Misnomer for Who This Population 
        Really Is
    ``Able-bodied'' indicates that clients are not medically certified 
and/or documented as physically or mentally unfit for employment. As 
part of the association's assessment, clients are asked to self-report 
disabilities or limitations, both physical and mental. Our findings 
identified elevated rates of participants with undiagnosed and 
untreated mental and physical limitations and disabilities. Clients who 
self-reported they were disabled with a physical or mental condition 
that rendered them unable to work required access to a doctor or 
medical professional who could provide the necessary documentation. 
Other clients were clearly disabled and required more intensive support 
services to complete an application for SSI or SSDI.

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Nearly one in ten clients requested special accommodations such as 
work assignments that require no heavy lifting, or no standing/walking 
for long periods of time.
    One in six clients reported that they had filed for Supplemental 
Security Income (SSI), or Social Security Disability Insurance (SSDI).
    Most Common Types of Physical and Mental Limitations Reported:

   18.3 percent--Back Injuries

   6.0 percent--Respiratory Difficulties

   5.9 percent--Knee Injuries

   3 percent--Diabetes

   2.8 percent--Shoulder Injuries

   2.5 percent--Arthritis

   2.3 percent--Heart Conditions

   10.1 percent--Depression

   9.3 percent--Bipolar Disorder

   8.1 percent--Anxiety

   3.1 percent--Post-Traumatic Stress Disorder (PTSD)

   1.5 percent--Schizophrenia

    According to the Ohio Department of Health, Adverse Childhood 
Experiences (ACEs) are a critical public health issue. ACEs are 
potentially traumatic experiences and events ranging from abuse and 
neglect to witnessing violent behavior and living with someone who has 
a problem with alcohol or drugs. Ohio is among five states where as 
many as one in seven children have experienced three or more ACEs--a 
significantly higher ratio than the national average.
    The association's WEP Assessment Specialist reported when 
conducting assessments that many participants appeared to be marginally 
and functionally illiterate, and likely experiencing significant 
learning disabilities. This prompts a deeper examination of social 
promotion policies that may exist in schools.
    Additionally, while assessing and observing clients, WEP 
Specialists noted that many clients appeared to have social and/or 
cognitive impairments, difficulty communicating, and a tendency to 
engage in repetitive behaviors, all signs of autism spectrum disorder. 
Since autism is a more recently identified disorder and has become a 
well-recognized ailment effecting one out of every 68 kids, it is 
highly likely that the ABAWD population may have high levels of 
undiagnosed autism, and certainly warrants further exploration.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
    Client Story: Mary is a 22 year old part-time college student who is
 studying to earn a Pharmacy Technician degree in hopes of one day
 becoming a Pharmacist. She is the first in her family to go to college
 and she has applied for and receives student loans that cover the cost
 of her tuition, books, and housing. She also receives SNAP and Medicaid
 benefits. Mary doesn't own a car and relies on public transportation
 and catches rides with family and friends or she walks. Mary also helps
 her mother care for younger sisters. Mary works for a large drug store
 chain which is on a bus line near the school she attends. When she was
 hired for the job, the store manager promised Mary she would work
 between 20 and 26 hours per week. Mary adjusted her class schedule to
 accommodate her work schedule, but unfortunately when the store sales
 began to lag behind projections, Mary's hours were cut in half, causing
 her to lose her SNAP benefits and leaving her with no way to feed
 herself. She has been pleading with the store manager to schedule her
 for additional hours, as this is a 24 hour/7 day a week store. Mary was
 told that she would need to be on call, but there are no guarantees
 that she will be called into work. The loss of SNAP benefits now
 threaten Mary's dreams and hopes and she is considering dropping out of
 school if she can't secure additional hours and regain her SNAP
 benefits.
------------------------------------------------------------------------

Employment
    There is limited employer demand for the ``hardest to employ'' 
groups, such as those with criminal records, lengthy periods of 
unemployment, or other barriers to works.
    Working 20 or more hours of paid employment per week, every week, 
qualifies an ABAWD to receive SNAP. Unfortunately, many clients were 
unable to identify how many hours they work per week because they are 
employed through a temporary employment agency (including day labor and 
labor pool agencies), which means clients may not have consistent work 
on a weekly basis.

------------------------------------------------------------------------
 
------------------------------------------------------------------------
      11.3%   Currently working
       8.3%   Working in-kind for rent or housing
        24%   Dismissed or fired from a job
------------------------------------------------------------------------

    While some have described this population as ``takers''--our 
research found that nearly eight in ten ABAWD clients have never been 
eligible for unemployment compensation benefits.
Education
    While the unemployment rate in Ohio is declining, clients in this 
population may not meet the educational standards for the jobs becoming 
available. Analyzing the statistics collected on education, we find how 
limited the prospects are for clients to enter the workforce in a 
position that will pay a sustainable living wage.
    Thirty percent of clients have no high school diploma or GED.
    Although 69.2 percent of clients have graduated from high school or 
have earned a GED, only 38.1 percent have attended college.
    A very small portion of clients (11 percent) who have attended 
college went on to earn a degree.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
    More than one million adults in Ohio do not have high school
 diplomas. Ohio's Adult Basic Education Programs only have the capacity
 to serve approximately 7,000 Ohioans each year.
    https://obm.ohio.gov/Budget/operating/doc/fy-20-21/
 BlueBook_BookOne_PBudgetRecommendations_FY20-21.pdf.
------------------------------------------------------------------------

Transportation
    Clients are supposed to receive a monthly travel stipend from their 
FCDJFS caseworker. Many clients report that they have not received the 
stipend. This could be due to an inaccurate mailing address, the 
inability to contact their caseworker, or a delay in dispersing of 
funds. Some clients report that the travel stipend is not enough to 
cover travel to and from work sites. Some clients do not have bank 
accounts and have to pay a service fee to cash the check they receive 
from FCDJFS, leaving an insufficient amount to purchase a monthly bus 
pass which the stipend should cover.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
Suspended Driver's Licenses
 
    In 2017, 1.1 million Ohioans had a suspended driver's license--
 nearly 12 percent of those old enough to drive in the state. Some
 suspensions have nothing to do with driving. If you don't pay your
 child support, you can lose your license. You can also lose it for
 dropping out of high school or getting caught smoking as a juvenile. It
 can be suspended if you miss a court date or fail to pay court fines on
 misdemeanor charges.
    https://www.daytondailynews.com/news/state--regional/ohio-fee-
 amnesty-for-suspended-drivers-has-started-but-only-lasts-six-months/
 5qQck20Vl2e3Mm EFRI1NTM/
------------------------------------------------------------------------

    Just 57 percent of clients report they have reliable access to 
transportation. This can be a personal vehicle, public transit, or 
utilizing friends and family members for transportation.
    Only 40 percent of clients have a valid driver's license, which 
indicates that clients are either using public transportation or are 
driving without a license. Some clients may not be able to obtain a 
driver's license if they owe child support and have had their driving 
privileges suspended, or if they have outstanding tickets or unpaid 
fines which they may be unable to resolve with their limited income.
    Fewer than one in five clients report having car insurance, 
inferring that some are driving without insurance which can be 
attributed to a variety of factors, including affordability.
    One in four clients do not live near a bus stop or bus line.
    About 15 percent of clients report they have been documented as 
Driving Under the Influence (DUI) or Operating a Vehicle Impaired 
(OVI). Having a DUI/OVI on an individual's driving record can affect 
their ability to obtain employment or housing, result in higher car 
insurance which they may be unable to afford, and/or lead to loss of 
driving privileges.
Criminal History
    As part of the assessment, clients are asked to complete an FBI/BCI 
background check. An overwhelming 96 percent of clients agreed to 
comply with this request. Clients who declined a background check do 
not qualify to participate in WEP with the Ohio Association of 
Foodbanks.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
Long-term impact of encounters with criminal justice system
 
    People with criminal justice (CJ) system involvement are more likely
 than the general population to face poverty, homelessness,
 unemployment, and poor health conditions, even before arrest. For
 example, people returning to their communities after incarceration are
 three to six times more likely to be diagnosed with a mental illness
 and about 50 percent experience chronic health conditions such as
 asthma and hepatitis.
    http://www.georgetownpoverty.org/wp-content/uploads/2019/02/
 Unworkable-Unwise-20190201.pdf.
------------------------------------------------------------------------

    Domestic violence can happen in any household regardless of 
socioeconomic status, race, age, or any other demographically defining 
factor. Studies show that domestic violence is three times as likely to 
occur when couples are experiencing financial strain. 11.2 percent of 
clients reported having domestic violence charges.
    A history of criminal activity or previous incarceration can have a 
tremendously negative impact on someone. They miss out on many 
opportunities, job related or otherwise. The stigma of a felony 
conviction can follow someone for a lifetime, even if their release is 
meant to suggest that they have been rehabilitated.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
    Client Story: At 15 years old, David was sentenced to 15 years in
 prison. Now, at 30 he has been released and was eager to start his life
 over. He was nervous during the assessment, but the WEP Specialist was
 able to get him to relax as he told his story. Later, he called our
 office to thank the Specialist for being so kind and understanding
 during the assessment and for also believing in him. He was thrilled to
 tell her that he learned to drive and is now enrolled at Columbus State
 Community College.
------------------------------------------------------------------------

    35.8 percent of the clients in our program have felony convictions; 
some clients have multiple felonies, or a combination of felonies and 
misdemeanors.
    12.8 percent of clients are on probation or parole which means they 
may not qualify for services offered through legal aid, such as record 
sealing.
    A recent report from the Kirwan Institute found that one in four 
people incarcerated in the State of Ohio were between the ages 18 to 
24. The incarcerated population from the 18 to 24 age group in Ohio has 
grown nearly 70 percent in recent years. Prison intake data from 
Franklin County indicate that the median age of first arrest for those 
entering the state correctional system in 2012 was 19 years old.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
                Other Issues Facing the ABAWD Population
------------------------------------------------------------------------
 
                Youth Aging Out of the Foster Care system
 
    5 percent of the clients had aged out of the foster care system and
 reported they were living with friends, in homeless shelters, or on the
 street.
------------------------------------------------------------------------
 
                        Homelessness and Housing
 
    Clients experiencing homelessness, health problems, language
 barriers and a lack of stable employment to fit their skill set make up
 nearly 12.7 percent of clients who reported other barriers standing in
 the way of employment.
------------------------------------------------------------------------
 
                  Non-Custodial Parents and Caregivers
 
    According to the USDA definition of an ABAWD, it is assumed that all
 clients do not have dependents. We found that clients with children,
 although not in their custody, still spend time parenting their
 children on a regular basis while the custodial parent works.
    One in four clients (23.5 percent) indicated that they had children
 not in their custody.
    Nearly one in five clients (18 percent) indicated that they owe
 child support.
    An under-employed or unemployed noncustodial parent who loses SNAP
 may need to divert his or her income from child support payments in
 order to stay afloat financially. This would be devastating given that
 child support represents more than \1/2\ of the income of the families
 in poverty who receive it.
    Having the status of caregiver to a relative should potentially
 exempt an individual from the work requirement. Caregivers can often
 replace the services of a Medicaid or Medicare home-healthcare
 provider. Nearly 13 percent of clients indicated that they are
 caregivers for a parent, friend, or relative.
------------------------------------------------------------------------

Employment & Job Seeking Needs

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
    Client Story: Dahman speaks only Somali and requires an interpreter
 or translator to fulfill his mandatory work activities and assignment.
 He has no transportation and relies on public transportation. Dahman
 returned to the JFS office attempting to find out about his food
 assistance benefit. Dahman had a large open wound on his arm that is
 draining, making it impossible for him to participate in any form of
 activity. Unfortunately, his County caseworker had not changed his
 employability plan or there had been an administrative delay in
 updating his care record, causing him to be sanctioned and to lose his
 SNAP benefits. Dahman was sent to a local food pantry to get food until
 his case could be sorted out and a new WEP placement could be located
 for him.
------------------------------------------------------------------------

Ohio Means Jobs Registration
    In an effort to offer more job seeking resources to clients, they 
are referred to Ohio Means Jobs (www.ohiomeansjobs.com). When asked if 
clients were already registered with Ohio Means Jobs 74.1 percent 
reported they were not registered, and most clients reported they have 
never heard of the website.
Additional Barriers
    To ensure a client is able to perform the duties assigned to them, 
we inquire about any supportive services they may need to successfully 
complete their work assignment. Over 15.7 percent of clients report 
needing supportive services. The most common services requested were 
language interpretation (especially for Somalian refugees) and help 
with transportation.
Churn Rates Are High
    When a client is no longer a participant in WEP due to a sanction, 
they may need to apply for a state hearing to overturn their sanction. 
Nearly 66 percent of clients reported taking this step to overturn 
their sanction, or reapplied for food assistance in another way after 
exiting WEP. It is estimated that there is a 3 month churn window, 
which is the average amount of time it takes for WEP participants to 
reenter SNAP after exiting the program.
    The amount of churn generated by the most common causes of 
noncompliance creates increased work as an average two out of every 
three participants, including those who identified some form of 
employment, must restart the entire process by reapplying through their 
case worker for SNAP benefits.

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Food Sourcing Strategies of Clients Who No Longer Received SNAP 
        Benefits
    If a client is not receiving food assistance due to a loss of SNAP 
benefits, they look for food elsewhere. When asked, ``How are you 
providing food for yourself in the absence of food benefits,'' clients 
gave multiple answers to the question, reflecting an increased demand 
on our emergency food network.
Conclusion
    Based on our experience, we know that harsh and arbitrary time 
limits are misguided and only increase hunger and hardship. This 
proposed rule is harsh and unfair. It denies vulnerable people food 
benefits at a time when they most need it and it does not result in 
increased employment and earnings. By time-limiting food assistance to 
this group, Federal law clearly intends to shift the burden of 
providing food to these unemployed individuals off of SNAP and onto 
states, cities, and local charities like ours. We can't meet the demand 
for emergency food assistance now--this rule will make a bad situation 
far worse. This rule will increase food insecurity among populations 
that are suffering from a lack of services, opportunities, and access 
to basic human needs.
    These individuals face daunting challenges in finding employment 
even when general unemployment rates are low. Our findings illustrate 
why Congress gave states the option to waive the time limit in areas 
where there are insufficient jobs for those subject to the rule. 
Without providing any evidence to the contrary, the rule proposes to 
limit the ways in which a state can demonstrate a lack of sufficient 
jobs for the individuals subject to the time limit. It does this by 
eliminating Labor Surplus Areas, low and declining employment-to-
population ratios, and seasonal unemployment, and requiring recent 
unemployment rates to be at least seven percent. But the Department 
fails to explain how it determined that the proposed new standards 
relate to employment opportunities for those subject to the rule, 
particularly given the significant barriers to employment facing this 
population that I've just shared with you.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
                  Proposed rule undermines existing law
 
    The proposed rule would:
 
     Take food away from 755,000 low-income Americans, cutting
     food benefits by $15 billion over 10 years (based on the
     Administration's own estimates).
 
     Not result in improvements in health or employment among
     the affected population (based on the Administration's own
     estimates).
 
     Fuel rates of hunger and poverty by denying vulnerable
     people nutrition assistance at a time when they most need it.
 
     Harm the economy, grocery retailers, and agricultural
     producers by reducing the amount of SNAP dollars available to spur
     local economic activity.
 
     Sidestep Congress, which rejected these changes when it
     enacted the 2018 Farm Bill.
------------------------------------------------------------------------

    The Department's commissioned reports as well as other research, 
including the association's WEP program results, paint a clear picture 
of individuals in this targeted group who have common characteristics 
that distinguish the group from other unemployed adults. These 
characteristics--including high poverty rates, health issues, and few 
supports--make finding and keeping employment a unique challenge. The 
Department simply asserts that the time limit will increase employment 
for this population but does not acknowledge its own research showing 
that this is not the case. While all aspects of the rule strike us as 
arbitrary, this disconnect between the agency's basic knowledge of the 
affected population and the assertions about how the proposed policy 
would increase employment is particularly surprising.
    Additionally, adequate work training slots do not exist even for 
the ABAWDs already impacted by the work requirements as currently 
imposed. This rule would subject hundreds of thousands of additional 
people to a requirement to fulfill work training if unable to secure 
paid employment, without acknowledging that availability of work 
training slots is grossly inadequate.
    In closing, the Department's proposed rule does not provide the 
analytical information needed to justify the policy change and to 
evaluate the proposed rule's likely impacts. Because of the 
deficiencies in reasoning and analysis, the proposed rule fails to 
answer basic questions related to the impact of the change and the 
people whom the proposed rule would affect, and so does not contain the 
information and data necessary to fully evaluate the proposed rule or 
to comment on key aspects on the Department's justification for the 
rule.
    The proposed rule would increase food insecurity and poverty in 
Ohio, as well as stifle economic activity. By scaling back one of the 
nation's most effective poverty-reduction programs, the rule would 
exacerbate hardship and reduce economic activity in areas that are 
already economically disadvantaged compared to the rest of the country.
    The proposed rule undermines states' ability to respond to economic 
hardship. By imposing artificial definitions of what it means for an 
area to ``lack sufficient jobs,'' the rule would undermine states' 
discretion to provide hunger relief in economically disadvantaged 
areas.
    The intent of the proposed rule is not supported by evidence. 
Though the USDA predicts that subjecting more SNAP recipients to work 
requirements would result in higher workforce participation rates, 
there is a lack of evidence to support this theory. In fact, existing 
evidence suggests that SNAP enrollment improves employment outcomes.
    The proposed rule would have a disparate impact on people of color 
in Ohio. The rule would make it even more unlikely that Ohio counties 
where people of color are concentrated would receive a time limit 
waiver.\7\
---------------------------------------------------------------------------
    \7\ The Center for Community Solutions: Public Comment to the U.S. 
Department of Agriculture, Food & Nutrition Service.

    The Ohio Association of Foodbanks requests that USDA consider each 
of these points and withdraw the proposed rule.
                               Attachment
Franklin County_Work Experience Program *
---------------------------------------------------------------------------
    * Work Experience Program, Ohio Association of Foodbanks, 101 E. 
Town St. Ste, 540, Columbus, OH 43215, www.ohiofoodbanks.org, 
614.221.4336.
---------------------------------------------------------------------------
Comprehensive Report_Able-Bodied Adults Without Dependents

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Table of Contents
  Executive Summary
  Assessment of ABAWDS in Franklin County
  Age & Gender
  Veteran Status
  Communication
  Criminal History
  Forms of Identification
  Transportation
  Disabilities & Limitations
  Children & Families
  Education
  Employment
  Work Experience Program
  WEP Volunteer Host Sites
  Recommendations
  Host Site Partner Organizations
Executive Summary
    For almost 2 years, the Ohio Association of Foodbanks has been 
assisting able-bodied adults without dependents (ABAWDs) receiving 
Supplemental Nutrition Assistance Program (SNAP) benefits in Franklin 
County with meeting the Federal work requirement to maintain their food 
assistance as part of an ongoing partnership with the Franklin County 
Department of Job and Family Services (FCDJFS). The association has 
been able to grow this Work Experience Program (WEP), offering more 
services and resources to ABAWDs in need. WEP provides work experience 
and job training for participants who are currently unemployed or 
underemployed, as a means to enhance their ability to secure 
sustainable employment.
    Prior to assigning a client in a job placement within our network 
of partner nonprofit and faith-based organizations, the association 
meets with each ABAWD to perform an in-depth assessment. To date, we 
have assessed close to 5,000 individuals. The data we have collected 
through these assessments continue to reinforce what we have been able 
to identify as key barriers for many of our clients as they seek 
gainful employment. Our findings indicate that many of our clients 
struggle with accessing reliable transportation, unstable living 
situations, criminal records, education, and both physical and mental 
health problems. Our deeper understanding of these issues has led us to 
partner with organizations that can help ABAWDs navigate through many 
of their challenges, giving our clients a better chance at improving 
their lives and supporting themselves.
    The data has prompted many recommendations to FCDJFS including but 
not limited to: providing additional funding for programs that support 
WEP participants and low-income households; expanding enrollment of 
nationally certified educational programs as well as programs for youth 
aging out of foster care; and creating an employment pipeline into 
strategic aspects of the job market.
Assessment of ABAWDS in Franklin County
    When Franklin County Department of Job and Family Services (FCDJFS) 
caseworkers make the determination that a client receiving SNAP 
benefits meets the criteria to be considered an able-bodied adult 
without dependents (ABAWD) and is required to work under Federal 
regulations, the client is referred to their local opportunity center 
to meet with an Ohio Association of Foodbanks Work Experience Program 
(WEP) assessment specialist. Each specialist completes a comprehensive 
interview with each client using a series of questions on the Work 
Experience Assessment Portal. The assessment is designed to determine 
employability and identify barriers to employment.
    The assessment process is part of an ongoing contract targeting 
clients who are subject to a strict, 3 month time limit in every 36 
month period for SNAP eligibility. As we approach the second 
anniversary of this program, we have closely examined the data 
collected from 4,827 ABAWDs and gathered from 5,434 self-reported 
employability and skills assessments that took place between December 
10, 2013 and September 1, 2015. Over the past 2 years the information 
obtained for this ongoing project represents the most comprehensive and 
up-to-date information collected about this misunderstood population. 
These findings offer instructive, meaningful insight into who these 
individuals are and what will be needed to address the barriers and 
challenges faced by these individuals as they attempt to secure stable 
employment.
Monthly Assessments

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    The chart depicts the number of ABAWD assessments performed by 
association staff for each month. Clients coming in for an initial 
assessment each month appear in blue, second time visits in any given 
month appear in orange, and clients who are completing the assessment 
for the third or more times appear in gray.
Age & Gender
Gender & Age Distribution

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    From the total population of 4,827 ABAWDs surveyed, 1,880 clients 
(38.9%) were female, and 2,945 clients (61.0%) were male. Two clients 
preferred to be identified as transgender.
    The chart represents a distribution of the ABAWDs based on age and 
gender. This distribution does not include the 507 clients (176 female 
and 331 male) for which there was no age listed, nor does it include 
the 83 clients (31 female and 52 male) who were over 50 at the time of 
the assessment and therefore exempted from the program.
Veteran Status
Percentage of Clients Reporting Military Service

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Only 156 clients (3.2%) reported that they were veterans. While 
veterans make up a relatively small percentage of all ABAWD clients, 
they represent a significant portion of the male population over the 
age of 35 as represented in the chart. As we encounter veterans, we are 
able to help them find resources designated to assist them with 
housing, employment, and shelter.
Communication
    Communication is critical to clients participating in WEP, and 
maintaining a reliable form of communication with clients has continued 
to be a challenge as FCDJFS and the association communicate with 
clients primarily by mail. Since we started collecting mailing 
information in April 2014, 65 clients have indicated that they do not 
have a mailing address, while 31 clients provided a mailing address and 
identified themselves as homeless. Additionally, 152 clients have 
provided a mailing address that is known to be a homeless shelter, 
check-in center, or mental health facility.

   Faith Mission (245 N Grant Ave ) 16 Clients

   Friends of the Homeless (924 E. Main St.) 21 Clients

   Open Shelter (61 E. Mound St.) 24 Clients

   Holy Family Soup Kitchen and Shelter (57 S. Grubb St.) 17 
        Clients

   Star House (1621 N. 4th) 4 Clients

   YWCA (595 Van Buren) 17 Clients

   YMCA (40 W. Long) 39 Clients

   Southeast Community Mental Health Center (16 W. Long St.) 10 
        Clients

   North Central Mental Health (1301 N. High St.) 4 Clients

    This indicates that at least 248 clients (5.1%) of our ABAWD 
clients are dealing with housing insecurity. These numbers do not 
capture the homeless clients who provide the mailing address of a 
relative or friend, and do not specifically identify that they are 
homeless.
Types of Communication Reported
Communication Avenues

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

   4,625 clients (95.8%) listed phone numbers

   1,800 clients (37.3%) listed e-mail addresses

   4,381 clients (90.8%) listed mailing addresses

   65 clients (1.3%) reported not having an address

   380 clients (7.9%) were assessed before address information 
        was asked

    While 95.8% of clients reported having phone numbers, this does not 
mean that they have continuous access to a phone. Clients using 
subsidized government provided cell phones often run out of wireless 
minutes before the end of the month, or in many other cases their 
personal phones have been disconnected, or phone numbers are frequently 
changed due to using prepaid cellular devices. We can only assume that 
if we are unable to contact clients via phone, potential employers are 
also unable to reach them.
    The association always offers clients the opportunity to register 
for an e-mail address as a viable, dependable alternative to a phone. 
Because most major employers require clients to fill out job 
applications online, having an e-mail address is critical to the 
application process. We encourage clients to visit their local 
libraries to check their messages, but find that some clients may not 
have reliable or readily available community-based access to the 
Internet. In this process, we also find that many clients struggle with 
using technology and computers.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
    Additional information gleaned from the 531 repeat ABAWD clients
 reinforces our findings, and provides insight into other forms of
 stable communication for this population. This 11% of ABAWD clients who
 have taken the assessment more than once shows:
 
     47% (253) have changed their phone number between
     assessments
 
     34% (181) have changed their addresses between assessments
 
    This transiency can have real consequences for ABAWD clients who are
 sanctioned (cut off from their benefits) because they did not receive
 an appointment or assignment notice from FCDJFS which required action
 to avoid a disruption in their benefits.
------------------------------------------------------------------------

Client Locations
    While the clients who have reported addresses represent 58 
different [ZIP C]odes in Franklin County, over 55% of clients come from 
nine [ZIP C]odes:

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

   43223: 141 clients (7.0%)

   43224: 140 clients (6.9%)

   43211: 137 clients (6.8%)

   43232: 133 clients (6.6%)

   43204: 123 clients (6.1%)

   43206: 117 clients (5.8%)

   43207: 116 clients (5.7%)

   43205: 112 clients (5.5%)

   43219: 104 clients (5.1%)
Criminal History
    As part of the ABAWD assessment, clients are asked if they are 
willing to complete an FBI/BCI background check. Over 96% of clients 
agree to comply with this request.
    A history of criminal activity or previous incarceration can have 
an incredibly damaging impact. The stigma of a felony conviction can 
follow someone for a lifetime, even if their release is meant to 
suggest that they have been rehabilitated. These restored citizens miss 
out on many opportunities, job related or otherwise.

   Over 35.8% of the clients in our program reported having a 
        felony conviction. Some clients have multiple felonies, or a 
        combination of felonies and misdemeanors.

   Close to 12.8% of clients are on probation or parole which 
        means they may not qualify for services offered through legal 
        aid, such as record sealing.

   541 clients (11.2%) have indicated that they have domestic 
        violence charges.

   709 clients (14.7%) reported having DUI or OVI violation. 
        These types of violations can severely limit a client's ability 
        to secure employment.
Percentage of Clients Reporting Felonies

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Forms of ID
    To apply for jobs, housing, and government benefits, to vote, or to 
obtain a driver's license, most agencies usually require two forms of 
Identification (ID). Because the association requires all participants 
to have an FBI and BCI background check to be placed at one of our host 
organizations we offer vouchers for clients to receive government 
issued state IDs when they indicate that they do not already have an 
ID.

   4,578 clients (94.8%) have some form of state 
        Identification.

     1,963 (40.7%) of clients have indicated that they have 
            a driver's license.

     2,615 have indicated that their primary form of 
            identification is a state ID.

     206 clients 4.3% indicated that they did not have any 
            form of state identification.

   4,369 clients (90.5%) reported having access to their Social 
        Security card.

     370 clients (7.7%) do not have access to their Social 
            Security card.

   3,969 clients (82.2%) reported having access to their birth 
        certificate.

     An additional 752 (15.6%) do not have a birth 
            certificate.
Forms of ID

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Transportation
    To assist with transportation, clients receive a monthly travel 
stipend from FCDJFS in the form of a $62 check. Many clients report 
that they have not received the travel stipend. This could be due to an 
inaccurate mailing address, the inability to contact their caseworker, 
or a delay in dispersing of funds. Some clients report that the travel 
stipend is not enough to cover travel to and from work sites. Some 
clients do not have bank accounts and have to pay a service fee to cash 
the check they receive from FCDJFS, leaving an insufficient amount to 
purchase a monthly bus pass which the stipend should cover.
    2,749 clients (57.0%) said they have access to reliable 
transportation, whether it is their own vehicle, the COTA bus system, 
or a ride from friends and family members. It is important to note that 
the use of a friend or family member's vehicle may not always be 
reliable. Owning a vehicle may pose its own challenges for low-income 
populations, as the car could break down and the client may not have 
the means to fix it.

   40% of clients said they do not have reliable 
        transportation.

   3,565 clients (73.9%) indicated that they live near a bus 
        stop.

   610 clients (12.6%) indicated that they did not live near a 
        bus stop.

   Only 40% of clients indicated that they have a valid 
        driver's license, which indicates that clients are either using 
        public transportation or are driving without a license.

     Some clients may not be able to obtain a driver's 
            license if they owe child support and have had their 
            driving privileges suspended, or if they have outstanding 
            tickets or unpaid fines which they may be unable to resolve 
            with their limited income.

   904 clients (18.7%) indicated that they did have car 
        insurance.

     An additional 3,232 clients (67.0%) indicated that 
            they did not have car insurance, inferring that some are 
            driving without insurance which can be attributed to a 
            variety of factors, including affordability. As it is the 
            law to maintain car insurance for any vehicles owned, some 
            clients could be making the tough choice to pay for 
            utilities, food, or medicine instead of car insurance.
Disabilities & Limitations
    ``Able-bodied'' indicates that clients should not be medically 
certified and documented as physically or mentally unfit for 
employment. As part of the assessment, clients are asked to self-report 
disabilities or limitations, both physical and mental.

   598 ABAWD clients (12.4%) have self-reported a disability. 
        Of these clients, 261 clients (44%) have indicated that they 
        are not able to work and earn $1,010 a month, which could make 
        them eligible for disability benefits.

     74 clients (12%) indicated that they are able to work 
            and earn $1,010 per month.
Percentage of Clients Reporting Disability

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

   1 in 3 ABAWD clients (32.5%) have self-reported some type of 
        physical or mental limitation. Of these clients, 25% (392) have 
        indicated that their condition limits their ability to perform 
        daily activities.

   70.3% (1,102) indicated some type of physical limitation.

   30.1% (471) indicated some type of mental limitation.

     Most Common Types of Physical and Mental Limitations Reported:
 
 
 
     Back Injuries 18.3%        Depression 10.1%
     Respiratory                Bipolar Disorder 9.3%
     Difficulties 6.0%
     Knee Injuries 5.9%         Anxiety 8.1%
     Diabetes 3%                Post-Traumatic Stress
     Arthritis 2.5%             Disorder (PTSD) 3.1%
     Shoulder Injuries 2.8%     Schizophrenia 1.5%
     Heart Conditions 2.3%
 

    Additionally, a small percentage of clients reported physical 
difficulties due to crimes of violence.

   27 reported physical difficulties as the result of gunshot 
        wounds.

   4 clients reported physical difficulties as the result of 
        stab wounds.
Physical or Mental Limitations

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Social Security and Health Care
    One in five ABAWD clients (18.6%) have reported filing for 
Supplemental Security Income (SSI) or Social Security Disability 
Insurance (SSDI). Of these clients, most have reported filing in the 
last 2 years:

   82 (9%) reported filing in 2015

   333 (37%) reported filing in 2014

   155 (17%) reported filing in 2013

   114 (13%) applied in 2012

   223 (25%) applied in 2011 or earlier

    One in four clients (25.0%) indicated said they were under a 
doctor's care, and 1,347 clients (27.9%) indicated that they were 
currently on medications.
    Nearly six in ten clients (58.2%) have reported already applying 
for Medicaid, although all clients may be eligible to receive this 
expanded necessary health coverage due to their low-income status. 
1,950 clients (40.4%) said they had not applied for Medicaid. As part 
of our outreach process, we invite health care navigators to our 
monthly WEP events to help clients sign up for health coverage.
Children & Families
    According to the USDA definition of an ABAWD, it is assumed that 
all clients do not have dependents. We found that clients with 
children, although not in their custody, still spend time parenting 
their children on a regular basis while the custodial parent works.

   1 in 4 clients (23.5%) indicated that they had children not 
        in their custody.

   868 clients (18.0%) indicated that they owe child support.

   86 clients (1.8%) indicated that they need childcare.

    Having the status of caregiver to a relative should potentially 
exempt an individual from participating in WEP. Caregivers can often 
replace the services of a Medicaid or Medicare home-healthcare 
provider. 618 clients (12.8%) indicated that they are caregivers for a 
parent, friend, or relative.
Education
Percentage of Clients Reporting Not Completing HS or GED

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Many of the clients in this population have not earned a degree or 
certification to work in industries that pay more than entry level 
wages.

   3,342 clients (69.2%) report having earned a high school 
        diploma or GED.

   1,424 (29.5%) of clients report never having graduated high 
        school.

    Of those students that did not earn a GED or high school diploma:

   121 (2.5%) report having attended last in the 12th grade

   404 (8.4%) report having attended last in the 11th grade

   316 (6.5%) report having attended last in the 10th grade

   190 (3.9%) report having attended last in the 9th grade

   86 (1.8%) report having left school before high school

   5 clients (0.1%) report never having attended school before
College Education
    Of the students who earned either a high school diploma or GED, an 
additional 1,324 (28%) attended college, and an additional 520 (11%) 
earned some type of degree or certification.
Highest Level of Education of ABAWD Clients

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Employment
    Working 20 or more hours of paid employment per week, every week 
can exempt an ABAWD from participating in WEP.

   547 clients (11.3%) indicated that they are currently 
        working.

     16 clients (2.9%) indicate that they are working less 
            than 10 hours per week

     62 clients (11.3%) indicate that they are working 10-
            20 hours per week

     75 clients (13.7%) indicate that they are working 20-
            30 hours per week

     34 clients (6.2%) indicate that they are working 30-40 
            hours per week

     23 clients (4.2%) indicate that they are working over 
            40 hours a week

     337 clients (61.1%) did not indicate how many hours 
            they were working

    At least 91 clients (1.9%) reported that they generally work for 
temporary employment agencies (including day labor and labor pool 
agencies). These clients may be unable to identify how many hours they 
work per week due to inconsistent scheduling and availability of 
consistent job assignments. Because of this, clients may not be able to 
regularly fulfill the 20 hour work requirement to qualify for an 
exemption.
Most Common Employment Industry

   Warehouse Work (including pick/pack, forklift)

   Customer Service

   Food Service (including fast food, restaurants, cooking, and 
        food preparation)

   Janitorial and Cleaning

   Construction (including carpentry, masonry, drywall, and 
        electric)
Employment History
    Having gaps in a resume can influence an employer's decision in the 
hiring process, which can negatively impact a client's chances of 
obtaining employment. Of the 4,284 clients who reported the time since 
they were last employed, 1,579 (36.8%) reported working last sometime 
within the current year. An additional 1,216 clients (28.4%) reported 
working last in the previous year, 665 clients (15.5%) reported working 
last within the last 2-3 years, 429 (10.1%) reported working last 
within 4-6 years, 204 (4.8%) reported working last within the last 7-10 
years, 109 clients (2.5%) reported working last between 11-15 years, 34 
clients (0.7%) reported working last within the last 16-20 years, 12 
clients (0.3%) reported working last over 20 years ago, and 36 clients 
(0.8%) reported having never worked before.
Year Client was Last Employed

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

In-Kind Work
    Just as traditional employment can exempt a client from 
participating in WEP, in-kind work may qualify clients from an 
exemption as well. 402 clients (8.3%) reported working in-kind for food 
or housing.

   67 clients (16.7%) reported working less than 10 hours per 
        week

   84 clients (20.9%) reported working 10 to 19 hours per week

   82 clients (20.4%) reporting working 20 to 29 hours per week

   21 clients (5.2%) reported working 30 to 39 hours per week

   28 clients (7.0%) reported working 40 or more hours per week

   120 clients (29.8%) did not report the number of hours they 
        were working per week
Employment Assistance
    The ABAWD assessment screens for additional assistance or equipment 
clients may need to perform tasks at their worksite.

   435 clients (9.0%) indicated that they needed special 
        accommodations at their worksite in order to do a job. The most 
        commonly requested accommodations were no heavy lifting and no 
        standing or walking for long periods of time.

   757 clients (15.7%) indicated that they need supportive 
        services to obtain employment. The most commonly requested 
        services were language interpretation (especially for Somalian 
        refugees) and help with transportation.
Workforce Development
    In an effort to offer more job seeking resources to clients, they 
are referred to Ohio Means Jobs (www.ohiomeansjobs.com). 7 in 10 
clients indicated that they were not registered to work through Ohio 
Means Jobs website. This shows that the outreach for the Ohio Means 
Jobs website has been ineffective in reaching this population.
    We assist clients with creating resumes so they are able to take 
them to career fairs and apply for jobs that require resumes.

   2,594 clients (53.8%) indicated that they did not have a 
        current resume.

   2,183 clients (45.2%) indicated that they would like help to 
        write or update their resume.

   2,410 clients (49.9%) indicated that they were not 
        interested in help to write or update their resume.
Unemployment Compensation Benefits
    Many job applications ask if applicants have ever been fired or 
dismissed from a previous position. One in four clients (24.0%) 
reported having been previously fired or dismissed from a job. When 
this question appears on a job application it can be a deterrent for 
employers to hire an applicant.
    We inquire if clients have ever received unemployment compensation 
benefits, as this can qualify them for an exemption in participating in 
WEP if they are still receiving it. Nearly eight in ten clients (78.3%) 
reported that they have never received unemployment compensation 
benefits.

   886 clients (18.4%) reported that they are receiving or have 
        received unemployment compensation, ranging in time from 1984 
        to February 2015.
Work Experience Program
    Immediate program goals for WEP participants are to actively ensure 
viable work opportunities for ABAWDs in Franklin County to fulfill the 
work requirement to maintain their SNAP benefits and prepare ABAWDs for 
reentry into the workforce. The long-term goals and objectives for WEP 
participants are focused on decreasing unemployment among Franklin 
County ABAWDs to break systemic cycles of poverty and hunger and ensure 
clients can become economically self-sufficient.
Consistent Outreach

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    During the initial ABAWD assessment at the FCDJFS opportunity 
centers, clients are given information about job openings and job fairs 
in Franklin County. When we find that one of the many barriers the 
assessment is meant to capture is stifling a client in their attempt to 
secure employment, we refer them to clothing banks, resources for 
homelessness, mental health facilities, educational opportunities, and 
food pantries.
    All new clients are required to attend a WEP employment and 
resource fair their first month in the program. We bring together 
employers (with assistance from FCDJFS Workforce Development and 
Franklin County Economic Development), health care navigators and 
certified application counselors, Legal Aid Society of Columbus 
lawyers, workforce development agencies, GED and adult education or 
vocational training organizations, and many more stakeholders to ensure 
we are able to offer clients a variety of valuable services.
    At this event, clients also receive a required background check for 
their job placements. They participate in hands-on activities and 
receive assistance with filling out job applications and creating or 
updating resumes, assistance with using computers, and referrals to 
obtain suiting for job interviews.
WEP Volunteer Host Sites
Type of Host Sites

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    The recruitment process for developing new sites involves calling, 
mailing, e-mailing, and visiting numerous nonprofit and faith-based 
organizations in Franklin County. Each organization is required to sign 
a Memorandum of Agreement, establishing a strong partnership that also 
holds these organizations accountable for reporting hours for clients.
    Each volunteer experience through WEP is intended to give 
participants training, education, or experience that would be 
beneficial in an ABAWD's search for future employment. Some sites even 
report hiring WEP workers when they have open positions available.
    A list of possible volunteer roles could include but is not limited 
to:

   Janitorial Work

   Painting

   Grounds Maintenance & Landscaping

   Warehouse Positions

   Office and Clerical Work

   Manual Labor

   Customer Service

   Food Preparation and Service

          ``One of our WEP clients began working at the Broad Street 
        Food Pantry in October 2014 as part of the Ohio Association of 
        Foodbanks Work Experience Program. From the time she started, 
        she demonstrated excellent work ethics--never missing a day, 
        always working hard and making sure that customers were served 
        efficiently, the shelves kept full, and the pantry kept clean 
        and neat. Last winter when our assistant moved on to another 
        job, our WEP client was one of the first candidates we 
        identified. After a thorough search, we hired her for the 
        permanent position.''
                   Kathy Kelly-Long, Broad Street Food Pantry Director.
                   
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          WEP participants paint a mural at Fusion Bakery and Cafe.
Placements
    Our network of nonprofits, workforce development partners, and 
faith-based organizations make it possible for Franklin County ABAWDs 
to obtain their required work hours through volunteer service or job 
readiness activities, while also offering work experience. Placements 
are made at these organizations after clients have completed a 
background check at the WEP monthly employment and resource fair.
    The Ohio Association of Foodbanks requires clients to have a 
background check to ensure that we are not placing clients in 
situations that may compromise the integrity of our partners, and to 
protect their clients and staff in the event of a known conflict of 
interest. Clients are not eligible to be placed at a volunteer host 
site until their FBI/BCI background check is received.
    Through the assessment process we gather an inventory of job skills 
from each clients. We are able to determine what jobs would best suit 
that client, and strategically place them at sites where we believe 
they will thrive. We do make accommodations for any client that is 
already volunteering in the community, and make an attempt to bring 
their volunteer site on as a host organization so that the client can 
maintain their relationship with that organization.
AB[A]WD Placement Compliance
    At times, it can be very difficult to place clients at a volunteer 
site. If the host location is not on the bus line or if it is not 
easily accessible by public transportation, clients can have a hard 
time getting to their placement. Some host sites even require a college 
education or degree, which many of our clients do not have. Some sites 
have a list of restricted felonies which would limit a large portion of 
our clients from volunteering with those sites. The same is true for 
workforce development programs. Many clients do not meet the minimum 
education requirements to enroll in such programs, or struggle with 
passing an entrance exam.
    The Ohio Association of Foodbanks placement specialist makes every 
effort to place all clients, no matter how limiting their personal 
situations may be. Even with the best effort to make sure that a 
client's skills match the site's needs, and that the location is less 
than an hour bus ride from their address, not all clients report to 
their assigned placements each month. In order for a client to remain 
compliant with WEP they must report to their worksite for 23 hours per 
month. When a client fails their work requirement hours they are 
sanctioned and at risk of losing their monthly SNAP benefits.
ABAWD Placement Compliance

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Recommendations
    As we bring light to the situations this population faces, we are 
able to make the following insightful recommendations which are 
supported by the findings of the WEP assessment data. These 
recommendations have been presented to FCDJFS after the first analysis 
of this information. They are meant to encourage other government 
organizations to consider a further examination of the implication of 
programs like WEP.
Program Next Steps
    The specific program needs of the Ohio Association of Foodbanks 
will enhance the overall client experience while strengthening 
relationships with our partners.

   Coordinate with other Departments of Job and Family Services 
        statewide in an effort to replicate the positive results we 
        have seen in Franklin County, to expand this program to other 
        metro and rural areas.

   Increase the efficiency of our program in order to enhance 
        client satisfaction and success while working with very limited 
        resources.

   Coordinate with Franklin County to offer more opportunities 
        for clients to connect with available employment and training.

   Improve quality assurance measures and outcomes as well as 
        communication channels between the Ohio Association of 
        Foodbanks, clients, host sites, and Franklin County Department 
        of Job and Family Services.
Increase Oversight To Improve Effectiveness
   Analyze the expenditures of Workforce Development Programs 
        funded by FCDJFS compared to outcomes. WEP at the Ohio 
        Association of Foodbanks has proven a 24% success rate, 
        compared to a 16% success rate of similar government funded 
        workforce programs in Franklin County.
Provide Additional Funding to Organizations Supporting WEP
   When clients fail a WEP assignment and do not have access to 
        their food stamp benefits, they may begin utilizing the 
        services of their local emergency food programs. This warrants 
        more emergency funding to be provided to Mid-Ohio Foodbank to 
        support the purchase, acquisition, and distribution of 
        additional food for Franklin County food pantries, soup 
        kitchens, shelters, and churches who are feeding the 
        individuals affected.

   Ut[il]ize banked months of exemptions (estimated at 405,000) 
        to re-enroll participants in the food assistance program while 
        Departments of Job and Family Services work to establish 
        additional work experience program infrastructure.

   Provide additional funding to the Ohio Association of 
        Foodbanks to support the cost of emergency vouchers for 
        transportation, travel vouchers, and basic needs.

   To increase interest in becoming a part of the host site 
        network, there needs to be more incentive for organizations to 
        serve ABAWDs through WEP. By offering operating support to the 
        nonprofit and faith-based organizations that are providing WEP 
        services and slots, we can motivate more sites to partner with 
        the Ohio Association of Foodbanks, while current sites may be 
        able to effectively increase their capacity to serve more 
        ABAWDs.

   Provide supplemental support for the continuation, 
        expansion, and analysis of workforce development programs 
        operated by the Ohio Association of Foodbanks for young adults 
        aging out of the foster care system. All youth who successfully 
        complete these programs either enroll in school or start 
        working, which in many cases exempts them from partic[i]pating 
        in WEP as ABAWDs.

   Improve the funding and training of a specialized unit 
        dedicated to the implementation of this work requirement and 
        the ABAWD population's specific needs.
Study the Social and Economic Impact of WEP
   Monitor and report on the impacts to well-being, health, and 
        safety of clients, WEP host site staff/volunteers, and the 
        community at large.

   Conduct an Economic Impact Analysis on the loss of food 
        assistance/SNAP benefit issuance on the Franklin County 
        economy.

   Provide funding for comprehensive case-management, 
        longitudinal tracking of employment, wages, public assistance 
        participation, and well-being of the ABAWD population.
Provide More Work Support Opportunities for ABAWDs
   Expand enrollment, participation, and successful completion 
        of nationally certified programs such as the FastPath program 
        at Columbus State Community College, including ServSafe, 
        customer service, advanced logistics, and STNA.

   Create an employment enterprise or pipeline into strategic 
        aspects of the job market. This will help harder-to-employ 
        individuals find opportunities to gain sustainable employment.

   Prioritize Workforce Investment Act funding to provide 
        education, training, and supportive services to ensure a 
        seamless delivery of services.

   Establish a relationship with the Ohio Department of 
        Rehabilitation and Correction in order to address the specific 
        concerns of the employer community in regard to the future 
        employment of felons.

   Examine opportunities to secure additional USDA/SNAP 
        Employment and Training funds to enhance service delivery.
Examine and Evaluate the Needs of Special Populations
   Provide support and funding for a study on the mental and 
        physical health status and outcomes of the ABAWD population and 
        their utilization of Medicaid.

   Fund person-centered, community-based case management of 
        ABAWDs applying for SSI/SSDI, and supportive services including 
        Legal Aid assistance to non-custodial parents and individuals 
        with criminal charges and felony convictions.

   Convene a study group to examine the impact of temporary and 
        day labor employment services and its effects on this 
        population.

   The Ohio Association of Foodbanks will continue to analyze 
        assessments and data including current and previous encounters 
        with the criminal justice system, community impact, and these 
        associated costs.
Host Site Partner Organizations
    Without the support of our wonderful network of nonprofit and 
faith-based organizations we could not offer so many meaningful 
volunteer opportunities to ABAWDs in Franklin County. We extend our 
sincere gratitude to each organization for their continued partnership 
and dedication to serving the community.

 
 
 
Agora Ministries                     J. Ashburn, Jr. Youth Center
Authority of the Believers           King Arts Complex MLK
Beatty Recreation Center             Kingdom Alive Word Church
Brice UMC                            Libraries for Liberia Foundation
Bridge Community Center              Long Lasting Community Development
Broad Street Food Pantry             Loving Hands Learning Center
Broad Street UMC                     Lutheran Social Services Ohio
                                      Benefit Bank--South
Calhoun Memorial Temple              Lutheran Social Services Ohio
                                      Benefit Bank--West
Cat Welfare Association              Magic Johnson Bridgescape Academy--
                                      New Beginnings
Catique                              Mock Rd University for Children
Center for Family Safety             National Parkinson Foundation
                                      Central & Southeast OH
Chalmers P. Wylie VA Ambulatory      New Salem Baptist Church and
 Care Center                          Community Development
Charitable Pharmacy of Central       NNEMAP, Inc.
 Ohio, Inc.
Child Development Council of         Ohio Association of Foodbanks
 Franklin County
Christ Harvest Church                Ohio Business Development Center
City of Whitehall                    Ohio Empowerment Coalition
Clintonville Beechwold               Pri-Value Foundation
Colony Cats (& dogs)                 Project Redeem
Columbus Arts Technology Academy     R.F. Hairston Early Learning Center
Columbus Chosen Generation           Reeb-Hossack Community Baptist
 Ministries                           Church
Columbus Growing Collective          Seven Baskets Community Development
                                      Corp
Columbus Humanities Arts &           Shiloh Christian Center
 Technology Academy
Columbus Urban League                Short North Stage at The Garden
                                      Theater
Community Kitchen, Inc.              Society Of St. Vincent De Paul
Core Resource Center, Inc.           Soldiers of Life Food Pantry
East Columbus Development Company    Somali Bantu Youth Community of
                                      Ohio
EL Hardy Center                      Southeast Friends of the Homeless
Family Missionary Baptist Church     Southeast, Inc.
Franklinton Gardens                  St. Dominic Roman Catholic Church
Genesis of Good Samaritans           St. Marks United Methodist Church
 Ministries
Glory Praise & Help Center           St. Philip Episcopal Church Food
                                      Pantry
Greater Ebenezer Cathedral of        St. Stephens Community House
 Praise and Kingdom Kids Daycare
Habitat for Humanity's ReStore       Stoddart Avenue Community Garden
Hands On Central Ohio                Temple Israel
Heart Food Pantry                    Trinity Assembly
Heart of Christ Community Church     United House of Prayer
Helping Hands Health And Wellness    Unity of Columbus
 Center, Inc.
Holy Family Soup Kitchen             Welcome Home Ohio
House of Refuge for All People       Wesley Church of Hope UMC
HUB Community Development
 Corporation
 


    The Chair. Thank you very much. Dr. Shambaugh.

 STATEMENT OF JAY C. SHAMBAUGH, Ph.D., DIRECTOR, THE HAMILTON 
    PROJECT, AND SENIOR FELLOW, ECONOMIC STUDIES, BROOKINGS 
                   INSTITUTION; PROFESSOR OF 
           ECONOMICS, GEORGE WASHINGTON UNIVERSITY, 
                        WASHINGTON, D.C.

    Dr. Shambaugh. Chair Fudge, Ranking Member Johnson, and 
Members of the Subcommittee, thank you for inviting me to join 
in this important discussion. My name is Jay Shambaugh. I serve 
as the director of The Hamilton Project, the Senior Fellow of 
Economic Studies at the Brookings Institution, and as a 
professor of economics at George Washington University. I am 
here to provide evidence regarding SNAP, a program that lifts 
millions of Americans out of poverty, reduces food insecurity, 
improves economic security, and acts as a crucial fiscal 
automatic stabilizer.
    Research shows that SNAP is a highly effective program. It 
also shows that work requirements keep people out of the SNAP 
Program, but have little or no impact on work. The proposed 
rule takes a number of steps to reduce the flexibility of 
states in using waivers or exemptions from work requirements. 
The proposed rule and its impact analysis are correct, that the 
changes will reduce SNAP participation, but provide literally 
zero evidence that the changes would increase employment.
    Agencies may change regulations when there is compelling 
public need and when benefits outweigh costs. In my remaining 
time, I would like to highlight three areas where the proposed 
rule fails to meet this standard.
    First, in theory, work requirements are in place to 
motivate those who do not want to work to do so. But very few 
ABAWDs on SNAP, 1.4 percent, are ``not interested'' in working. 
The vast majority are, in fact, in the labor force. However, 
their labor market experience, as is true for many low paid 
workers, is highly unstable as participants tend to cycle in 
and out of full-time employment.
    In the research I have conducted with my Brookings 
colleague, Lauren Bauer, which has been provided to the 
Committee, we find that 75 percent of ABAWDs over 2 years are 
labor force participants. Over \1/3\ of those in the labor 
force would satisfy the work requirements at some points in 
time, but not at other points in time over that 2 year window, 
almost as many would consistently satisfy the work requirement.
    Of those who generally work but sometimes do not, the 
majority are not working due to ``work related reasons.'' That 
is, they lost a job or couldn't get enough hours in a given 
month to satisfy the work requirement. We also find that the 
title ``able-bodied'' is a misnomer for some of this group, as 
80 percent of ABAWDs who were not in the labor force at all 
over the 2 year window list health and disability as the reason 
they are not working. These are people who should be eligible 
for exemptions but could fail to receive them.
    Based on the characteristics of the targeted population, 
the Federal Government should not be impeding states' ability 
to apply for waivers from work requirements in areas where 
there is evidence of a lack of sufficient jobs or limiting 
states' ability to use exemptions to address individual cases.
    Second, the proposed rule fails to consider the effect of 
the proposed changes in the face of a deteriorating economy. 
Consider that when the economy was shedding 300,000 jobs a 
month in 2008, states successfully applied for waivers to work 
requirements statewide or for distressed regions using 
geographies and indicators that USDA would deem invalid under 
the proposed rule. Our analysis provided to the Committee 
demonstrates that the rule would have reduced waiver 
eligibility early in the Great Recession.
    In 2008, the State of Ohio was granted a work requirement 
waiver for the entire state for 2 years. By the proposed rule, 
Ohio could not apply for the statewide waiver, the 20 percent 
rule they used would be compromised by an excessively high 
unemployment rate floor, and the extended time period granted 
would be denied. Our submitted analysis shows the proposed rule 
takes a waiver system that is already too slow to respond to an 
economic downturn and makes it worse.
    Last, the goal of the proposed rule is to incentivize work, 
but the consequences of the rule is to, in fact, incentivize 
ABAWDs to reside in distressed economies if they want to avoid 
time limits. Work requirements are applied to places of 
residence. Individuals wanting to move to places with a 
stronger economy would risk their food resources because they 
would suddenly face work requirements. Reducing the statewide 
or geographic grouping waivers could lower labor mobility.
    In conclusion, the evidence recommends against expanding 
work requirements, whether through restricting states' ability 
to apply for waivers or extending exposure to sanction to 
parents or older Americans. There are better ways to encourage 
work within the SNAP Program, such as adjusting the earnings 
disregard, expanding wrap-around services, and improving 
training and placement. There are also better ways to improve 
waiver eligibility, such as automatically granting waivers in 
the event Congress authorizes emergency unemployment 
compensation. These reforms would strengthen and support SNAP 
as well as the economy.
    I am also happy to take any questions.
    [The prepared statement of Dr. Shambaugh follows:]

 Prepared Statement of Jay C. Shambaugh, Ph.D., Director, The Hamilton 
 Project, and Senior Fellow, Economic Studies, Brookings Institution; 
 Professor of Economics, George Washington University, Washington, D.C.
    Chair Fudge, Ranking Member Johnson, and Members of the Committee:

    Thank you for inviting me to join this important discussion 
regarding the U.S. Department of Agriculture's Proposed Rule: SNAP 
Requirements for Able-Bodied Adults Without Dependents.
    My name is Jay Shambaugh, and I serve as the Director of The 
Hamilton Project and as a Senior Fellow in Economic Studies at the 
Brookings Institution and a Professor of Economics at George Washington 
University. I am here to provide evidence regarding SNAP, a program 
that lifts millions of Americans out of poverty, reduces food 
insecurity, improves economic security, and acts as a crucial fiscal 
automatic stabilizer.
    Research shows that SNAP is a highly effective program. It also 
shows that work requirements keep people out of the SNAP program but 
have little or no impact on work. The proposed rule takes a number of 
steps to reduce the flexibility of states in using waivers or 
exemptions from work requirements. The USDA's Notice of Proposed 
Rulemaking and its Regulatory Impact Analysis are correct that the 
changes will reduce SNAP participation, but provide no evidence that 
the changes would increase employment.
    Agencies, such as USDA, may issue regulations when there is a 
compelling public need and when the benefits outweigh the costs. In my 
remaining time, I would like to highlight three areas where the 
proposed rule fails to meet this standard.

  (1)  The proposed rule ignores the reality of the population that 
            receives SNAP and the volatility they face within the labor 
            market.

    In theory, work requirements are in place to motivate those who do 
not want to work to do so. But very few ABAWDs on SNAP, 1.4 percent, 
are ``not interested in working.'' The vast majority are in the labor 
force. However, the labor market experience of SNAP participants--as it 
is for many low-paid workers--is highly unstable, and participants tend 
to cycle in and out of full-time employment.
    In research that I have conducted with my Brookings colleague 
Lauren Bauer, which has been provided to the Committee, we find that 75 
percent of ABAWDs are labor force participants. Over \1/3\ of those in 
the labor force would satisfy the work requirements at some points but 
not at others over a 2 year window, almost as many as would 
consistently satisfy them. Of those who generally work but sometimes do 
not, the majority don't work due to ``work related reasons.'' That is, 
they lost a job or couldn't get enough hours. We also find that the 
title ``Able-bodied'' is a misnomer given that 80 percent of ABAWDs who 
were not in the labor force said it was due to health and disability; 
these are people who should be eligible for exemptions but could fail 
to receive them.
    Based on the characteristics of the targeted population, the 
Federal Government should not be impeding states' ability to apply for 
waivers from work requirements in areas where there is evidence of a 
lack of sufficient jobs or limiting states' ability to use exemptions 
to address individual cases.

  (2)  The proposed rule fails to consider the effect of proposed 
            changes in the face of a deteriorating economy.

    USDA's proposed rule and Regulatory Impact Analysis also fail to 
weigh the detrimental effect of their proposal during economic 
downturns. Consider that when the economy was shedding 300,000 jobs a 
month in 2008, states successfully applied for waivers to work 
requirements state-wide or for distressed regions using geographies and 
indicators that the USDA would deem invalid under the proposed rule. 
Our analysis shows the rule would have reduced waiver eligibility early 
in the Great Recession.
    For example, in 2008, the State of Ohio was granted a work 
requirement waiver for the entire state for 2 years. By the proposed 
rule, Ohio could not apply for a statewide waiver, the 20 percent rule 
they used would be compromised by an excessively high unemployment rate 
floor, and the extended time period granted based on evidence of dire 
economic conditions would be denied.
    Our submitted analysis shows the proposal takes a waiver system 
that is already too slow to respond to an economic downturn and makes 
it even worse.

  (3)  This proposed rule could reduce labor mobility and trap people 
            in areas with less economic opportunity.

    The goal of the proposed rule is to incentivize work, but the 
consequence of the rule is to incentivize ABAWDs to reside in 
distressed economies if they want to avoid time limits. Work 
requirements are applied to the place of residence. Individuals wanting 
to move to places with a stronger economy would risk their food 
resources because they would suddenly face work requirements. Reducing 
statewide or geographic grouping waivers could lower labor mobility.
    In conclusion, the evidence recommends against expanding work 
requirements, whether through restricting states' ability to apply for 
waivers or extending exposure to sanction to parents or older 
Americans. There are better ways to encourage work within the SNAP 
program, such as adjusting the earnings disregard, expanding wrap-
around services, and improving training and placement. There are also 
better ways to improve waiver eligibility, such as automatically 
granting waivers in the event that Congress authorizes Emergency 
Unemployment Compensation. These reforms would support and strengthen 
SNAP as well as the economy.
                               Attachment
    Good Afternoon:

    Thank you for inviting me to testify before the Nutrition 
Subcommittee on the topic ``Examining the ABAWD Rule and its Impact on 
Hunger and Hardship.'' My written testimony is attached.
    For your reference, you will also find recent Hamilton Project 
research regarding this issue that we submit for the record, including:
    Comment on USDA's Proposed Work Requirement Rules: In response to 
the U.S. Department of Agriculture's Notice of Proposed Rulemaking, 
Lauren Bauer, Jana Parsons, and Jay Shambaugh analyze the effect of 
changing eligibility for work requirement waivers on coverage over time 
and describe the characteristics and employment statuses of Able-Bodied 
Adults without Dependents. In this comment, we provide evidence and 
analysis that the USDA has proposed a rule that is arbitrary, that the 
rule runs counter to the compelling public need for waivers to work 
requirements during economic downturns, and that the rule fails to 
consider much less prove that the benefits to participants and the 
economy outweigh the costs.
    Work Requirements and Safety Net Programs: In this paper, Lauren 
Bauer, Diane Whitmore Schanzenbach, and Jay Shambaugh describe who 
would be impacted by an expansion of work requirements in SNAP and an 
introduction of work requirements into Medicaid. We find that most SNAP 
and Medicaid participants who would be exposed to work requirements are 
attached to the labor force, but that a substantial share would fail to 
consistently meet a 20 hours per week threshold. Among persistent labor 
force non-participants, health issues are the predominant reason given 
for not working. There may be some subset of SNAP and Medicaid 
participants who could work, are not working, and might work if they 
were threatened with the loss of benefits. This paper adds evidence to 
a growing body of research that shows that this group is very small 
relative to those who would be sanctioned under the proposed policies 
who are already working or are legitimately unable to work.
    For more than a decade The Hamilton Project has produced evidence-
based policy proposals on how to create a growing economy that benefits 
more Americans. We believe this can be accomplished by promoting 
strong, sustainable, long-term economic growth; recognizing the 
mutually reinforcing roles of economic security and economic growth; 
and, embracing a role for effective government in making needed public 
investments.
    We welcome the opportunity to share more of our research and policy 
proposals with you. Your staff can contact me at [Redacted] or 
[Redacted] as well as The Hamilton Project's Managing Director Kriston 
McIntosh at [Redacted] or [Redacted].
            Warm regards,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Jay Shambaugh,
Director, The Hamilton Project;
Senior Fellow, the Brookings Institution.
                              attachment 1
March 28, 2019

  Food and Nutrition Service, U.S. Department of Agriculture
  7 CFR Part 273
  Docket Number: FNS-2018-0004
  Docket RIN 0584-AE57

  Certification Policy Branch, Program Development Division,
  Food and Nutrition Service, U.S. Department of Agriculture,
  3101 Park Center Drive,
  Alexandria, Virginia 22302

    To whom it may concern:

    We are writing in response to the Department of Agriculture's 
notice of proposed rulemaking (NPRM) regarding Supplemental Nutrition 
Assistance Program: Requirements for Able-Bodied Adults with Dependents 
(Docket ID FNS-2018-0004).
    Given that SNAP is a highly effective automatic stabilizer, 
proposals that change the conditions by which economically distressed 
places become eligible for work requirement waivers should be held to 
the highest evidentiary standards.
    This comment summarizes and provides evidence relevant to the 
rulemaking. The USDA's Proposed Rule does not meet an evidentiary 
standard and would weaken SNAP's responsiveness to an economic downturn 
without increasing labor force participation rates.
    Based on the research produced and attached herein, we find no 
evidence of a compelling public need for regulation nor that the 
benefits outweigh the costs. We ask that the USDA review and address 
each evidentiary point herein, as well from the research attached, as 
part of the notice and comment process. The existing rules should be 
sustained.
            Sincerely,

  Lauren Bauer,
  Fellow, Economic Studies, The Brookings Institution;
  Jana Parsons,
  Senior Research Assistant, The Hamilton Project;
  Jay Shambaugh,
  Director, The Hamilton Project; Senior Fellow, Economic Studies, The 
    Brookings Institution; Professor, The George Washington University.
Table of Contents
  I. Introduction
  II. Supplemental Nutrition Assistance Program

    A. SNAP and Incentives to Work
    B. SNAP Effectiveness
    C. Macroeconomic Stabilization

  III. Modeling Waiver Eligibility

    A. Regulatory Impact Analysis

      1. Waiver Take-up
      2. Statewide Waivers and Geographic Areas
      3. 20 Percent Rule
      4. Labor Surplus Areas
      5. Effect on Society and Uncertainties

    B. Analysis Based on Eligibility

      1. Work Requirement Waiver Eligibility during the Great Recession
      2. Modeled Eligibility Versus the Proposed Rule
      3. Eligibility Versus The Proposed Rule

  IV. Employment Status Changes
  V. Conclusion
  VI. References
  VII. Appendix
I. Introduction
    The goals of safety net programs are to provide insurance 
protection to those who are experiencing poor economic outcomes and to 
support those who are trying to improve their situation. The 
Supplemental Nutrition Assistance Program (SNAP, formerly the Food 
Stamp Program) ensures that eligible participants and families have 
access to food when they have no or low income. SNAP does so by 
providing participants with resources to raise their food purchasing 
power and, as a result, improve their health and nutrition. SNAP lifts 
millions out of poverty and supports work while reducing food 
insecurity. Evidence shows that SNAP increases health and economic 
security among families in the short term as well as economic self-
sufficiency in the long-term.
    SNAP is designed to expand as unemployment rates rise and household 
income falls, and in fact, caseloads increase as the unemployment does 
(Ganong and Liebman 2018). SNAP, Medicaid, and Unemployment Insurance 
provide the majority of automatic spending fiscal stabilization during 
economic downturns (Russek and Kowalewski 2015) and SNAP's 
responsiveness to downturns has increased over time (Bitler and Hoynes 
2010). Studies show that when SNAP payments increase to a local area in 
response to an economic downturn, they serve as an effective fiscal 
stimulus to the local area (Blinder and Zandi 2015; Keith-Jennings and 
Rosenbaum 2015).
    In accordance with the law, including the recently reauthorized 
farm bill, Congress authorizes states to manage the work requirement 
for so-called able-bodied adults without dependents (ABAWDs) in 
accordance with the needs of their state. After 1996, certain non-
disabled SNAP participants ages 18-49 without dependent children are 
limited to 3 months of benefits out of 36 months if they do not work or 
participate in a training program at least 20 hours per week or 
participate in workfare. States have had the option to impose work 
requirements on certain beneficiaries since the 1980s. See Rosenbaum 
(2013) and Bolen, et al. (2018) for a detailed description of SNAP work 
requirements. States are not required to assign these participants or 
provide slots in training programs, so for many participants, this 
provision functions as a time limit rather than a work requirement.
    Exempt from ABAWD work requirements are those outside the age 
range, those who are medically certified as unfit for employment, those 
with dependents or who reside in a household with a minor, those who 
are pregnant, and those who are otherwise exempt. States must exempt 
certain individuals, such as those who are ``unfit'' for work, and are 
permitted to exempt a share of individuals for other reasons.
    States are permitted to apply to the U.S. Department of Agriculture 
(USDA) for waivers to the time limit provisions for the entire state as 
well as sub-state geographic areas if their economic conditions meet 
certain standards. The state must be able to provide evidence that the 
state or a state-determined sub-state area has (1) a recent 12 month 
average unemployment rate over ten percent, (2) a recent 3 month 
average unemployment rate over ten percent, (3) a historical seasonal 
unemployment rate over ten percent, (4) is designated as a Labor 
Surplus Area (LSA), (5) qualifies for Extended Benefits to Unemployment 
Insurance (EB), (6) has a low and declining employment-to-population 
ratio, (7) has a lack of jobs in declining occupations or industries, 
(8) is described in an academic study or other publications as an area 
where there is a lack of jobs, (9) has a 24 month average unemployment 
rate 20 percent above the national average for the same period, 
starting no earlier than the start of the LSA designation period for 
the current fiscal year.
    The intent of the work requirement waivers is to ensure that 
participants are not penalized for not working when it is difficult to 
find a job. As there is no one way to measure job finding difficulty, 
there are a variety of ways to measure labor market weakness in the 
current rules. The current waivers can be at the county, regional, or 
state level. They are both absolute (above certain levels of 
unemployment) and relative (compared to national average) as both may 
be an important signal to a state that economic conditions warrant 
waiving work requirements.
    The USDA proposes to disallow states from applying for statewide 
waivers except on the basis of the state qualifying for EB (option 5) 
and from making regional determinations. USDA proposes to maintain 
options 1 and 5 and eliminate waiver eligibility options 3, 4, 6, 7, 
and 8 of the preceding paragraph with regard to counties or Labor 
Market Areas (LMAs). It proposed to modify option 2 (an unemployment 
rate of ten percent in a recent 3 month period) to only be used in 
support of ``an exceptional circumstance (p. 983),'' ``the rapid 
disintegration of an economically and regionally important industry or 
the prolonged impact of a natural disaster (p. 985).'' The USDA 
proposed to modify option 9 (the so-called ``twenty-percent rule'') 
such that ``an area must have an average unemployment rate at least 20 
percent above the national average and at least seven percent for a 
recent 24 month period (p. 984).'' USDA also requests feedback on using 
six and ten percent unemployment as rate floors.
    The proposed rule also reduces states' ability to use exemptions 
for individuals by limiting states ability to accumulate those 
exemptions. The exemptions allow states to shield individuals from work 
requirements if state administrators feel the work requirements are 
inappropriate for that individual, for example due to temporary 
problems with hours, health, caregiving, or other issues that restrict 
their ability to work.
    The USDA proposes new rules that are arbitrary. The USDA and its 
Regulatory Impact Analysis (RIA) fail to fully consider the costs and 
benefits of the proposed rule, including the costs and benefits under 
alternative economic conditions.
    The proposed rule limits a state's ability to apply for work 
requirement waivers when its economy is weak or relatively weak 
compared to the overall national economy. The USDA and the RIA do not 
consider the benefits to program participation for individuals nor 
SNAP's role as an automatic stabilizer when weighing proposed changes. 
The rule is likely to push a considerable number of current 
beneficiaries who are either in the labor market or unable to work off 
the SNAP rolls while failing to expand for newly eligible participants 
at the onset of a recession. It does so absent evidence that labor 
force attachment among ABAWDs would increase as a result of this 
proposal even in a strong economy and without consideration to the 
costs both for individuals and the economy in any circumstance.
    The analyses reported in this comment show that the proposed rule 
would weaken one of the strongest automatic stabilizers in the fiscal 
policy toolkit. The analysis presented below that fewer counties would 
be eligible for waivers at the start of a recession relative to current 
rules. Instead of SNAP participation expanding promptly, rapidly, and 
expansively as the unemployment rate rises, the proposed rule would 
slow eligibility for geographic waivers, and in fact, could cause the 
program to contract. The proposed rule undermines the role that SNAP 
plays at the onset of a recession, during poor economic times, and in 
mitigating the effects of recessions. While the stated goal is to limit 
waiver eligibility in a strong economy, the proposed rule fails to 
ensure waivers are available to states in a weak economy. The USDA and 
its RIA have failed to consider this critical issue, much less weigh 
the costs and benefits to these changes.
    The proposed work requirements would make regional waivers more 
difficult to obtain and state-wide waivers difficult to obtain in the 
absence of EB. By making it more difficult for states to apply for a 
statewide waiver and by limiting state's ability to determine 
economically-linked areas, USDA reduces the geographic mobility of 
program participants and ties their benefit receipt to maintained 
residency in an area that, by its own definition, is economically 
lagging. This seems likely to reduce employment and labor force 
participation of SNAP program participants as it effectively traps them 
in lagging economic areas. No analysis in the RIA is presented to 
consider these costs.
    While proposing to eliminate evidentiary standards that are not 
based on federally-produced data, the USDA proposes eliminating two 
that are (LSAs and seasonal unemployment) and introduce uncertainty 
into what is currently a standard with clear and universal applications 
(3 month unemployment rate over ten percent.) No analysis in the RIA is 
presented to consider these costs.
    The analyses reported in this comment suggest that the proposed 
changes to work requirement regulations will put at risk access to food 
assistance for millions who are working, trying to work, or face 
barriers to working. We find the USDA provides no evidence that 
limiting waivers from work requirements makes this population more 
likely to work or more self-sufficient. Our analysis shows that the 
overwhelming majority of SNAP participants subject to work 
requirements, ABAWDs, are in fact in the labor force; but, most have 
volatile employment experiences that would leave them failing the work 
requirements from time to time. Our analysis also shows that labor 
force participants experiencing a gap in employment do so for work-
related reasons outside their control. Furthermore, the vast majority 
of ABAWDs not in the labor force are not in fact able-bodied, but 
suffer from serious health problems or have a disability. By further 
proscribing the individual waiver eligibility pool and the use of 
exemptions, the proposed rule limits state's discretion to provide food 
assistance. No analysis in the RIA is presented to consider the work 
experiences and health conditions of ABAWDs, the benefits to them for 
SNAP program participation, and the costs to them and to society of 
time limits.
    This comment summarizes and provides evidence relevant to the 
rulemaking. Based on the research produced and attached herein, we find 
no evidence of a compelling public need for regulation nor that the 
benefits outweigh the costs. We ask that the USDA review and address 
each evidentiary point herein, as well from the research attached, as 
part of the notice and comment process. The existing rules should be 
sustained.
II. Supplemental Nutrition Assistance Program
    In this section we review published evidence on SNAP and work 
requirements.
A. SNAP and Incentives to Work
    SNAP is the most near universal of means-tested transfer programs 
in the United States. Certain households' SNAP eligibility is 
determined by meeting a gross income test whereby all sources of income 
fall below 130% of the Federal poverty level (FPL) for its household 
size. The net income test requires that a household's net income, i.e., 
gross income minus the earnings disregard and other deductions, is 
below 100% FPL.
    Subject to meeting the income and asset limits, benefits are 
allocated to households through the following formula:

        Household SNAP benefit = maximum benefit^0.3 * net income.

    Households without any net income receive the maximum benefit for 
their household composition. Those with positive net income see their 
benefit levels reduced by 30 cents on the dollar of net income.
    While one might worry that providing income support decreases the 
incentive to work, SNAP currently addresses work disincentives in a 
variety of ways. SNAP has an earnings disregard of 20 percent as part 
of the net income calculation, meaning that the value of the earnings 
disregard increases as income does and that those with earned income 
receive larger SNAP benefits than those with no earned income (Wolkomir 
and Cai 2018). This means that when a person moves from being a labor 
force non-participant to working while on SNAP, total household 
resources will increase; as a beneficiary earns more up to the 
eligibility threshold, total household resources continue to increase. 
The combination of the earnings disregard and a gradual phase-out 
schedule--that states have the option to further extend and smooth--
ameliorate but do not eliminate work disincentives.
    Work requirements in SNAP are meant to force work-ready individuals 
to increase their work effort and maintain that work effort every month 
by threatening to withhold and subsequently withholding food assistance 
if a person is not working a set number of hours. In practice, the 
application of work requirements sanctions many groups: those who are 
unable to work, those who are able to work but who do not find work, 
those who are working but not consistently above an hourly threshold, 
and those who are meeting work or exemption requirements but fail to 
provide proper documentation.
    During the Food Stamp Program's introduction in the 1960s and 
1970s, reductions in employment and hours worked were observed, 
particularly among female-headed households (Hoynes and Schanzenbach 
2012). But in general, there is little evidence that SNAP receipt 
itself depresses work effort substantially (Fraker and Moffitt 1988; 
Hoynes and Schanzenbach 2012). Whether work requirements could offset 
the small work disincentive would depend on their targeting and whether 
those who are not working could readily increase their labor supply. In 
fact, the evidence suggests that work requirements decrease SNAP 
participation, including at times when roll expansion is aligned with 
automatic stabilization (Ganong and Liebman 2018; Harris 2019; Ziliak, 
Gundersen, and Figlio 2003). Recent analysis published as a working 
paper suggests that SNAP participation by ABAWDs is substantially 
reduced by work requirements but that increase in work is minimal 
(Harris, 2019). Even the specifications that find the largest increases 
in work suggest five participants would lose SNAP benefits for every 
one that becomes employed due to work requirements.
    The USDA and RIA provided no evidence that there would be any 
increase in labor supply resulting from a change in what areas would 
qualify to apply for a waiver. Projections for increased labor supply 
are tied to the 2019 President's Budget projections for an ever-
decreasing national unemployment rate. In fact, because there is no 
evidence that ABAWDs will increase their labor supply in response to 
work requirements, USDA also ``estimated the impacts under an alternate 
scenario that assumes instead that rate of employment remains at 26 
percent (p. 26).'' Failure to prove that labor supply would increase as 
a result of the proposal in good economic times, much less bad, 
suggests that there is no compelling public need for new regulation.
B. SNAP Effectiveness
    Several studies have found that SNAP reduces the likelihood that a 
household will experience food insecurity or very-low food security 
(Collins, et al., 2014; Kreider, et al., 2012; Mabli, et al., 2013; 
Nord and Prell 2011; Ratcliffe, McKernan, and Zhang 2011; Shaefer and 
Gutierrez 2013; Schmidt, Shore-Sheppard, Watson 2016). Moreover, 
evidence from safety net expansions--such as the temporary benefit 
increase under the American Recovery and Reinvestment Act of 2009 
(ARRA) and a pilot program that provided additional benefits to 
families of children during the summer months when school meals were 
not available--shows reductions in rates of food insecurity and very-
low food security (Collins, et al., 2013; Schanzenbach, Bauer, and 
Nantz 2016; Smith and Valizadeh 2018). Recent studies have shown that 
SNAP improves health outcomes and households' financial well-being, and 
even improves the later-life outcomes of individuals who had access to 
the program as children (Almond, Hoynes, and Schanzenbach 2011; 
Hinrichs 2010; Hoynes, Schanzenbach, and Almond 2016; Shaefer and 
Gutierrez 2013).
    For example, a recent study by Hoynes, Schanzenbach, and Almond 
(2016) finds long-term positive effects from consistently providing 
access to the Food Stamp Program (now called SNAP) during early life. 
Taking advantage of the relatively long rollout period when the program 
was originally introduced, the study compares children who lived in 
different counties within a state and who were born at different times 
to measure the long-term impacts of access to the program. Access to 
the Food Stamp Program at early ages--starting before birth in cases 
where the mother received food stamps during pregnancy, and continuing 
through age five--leads to a number of positive long-run health and 
economic outcomes.
    As shown in figure 1, access to the Food Stamp Program over this 
age range has substantial positive impacts on later health, lowering 
women's and men's incidence of metabolic syndrome--a health measure 
that includes diabetes, high blood pressure, obesity, heart disease, 
and heart attack--by 0.3 and 0.5 standard deviations, respectively. 
Women are also 34 percentage points more likely to report excellent or 
very good health if they had access to food stamps from before birth 
through age 5.
    These gains also extend to economic outcomes. Women with access to 
the Food Stamp Program over the full early life period have much higher 
economic self-sufficiency--a measure that includes completed education, 
employment status, earnings, and financial success--than those who did 
not. Furthermore, access to food stamps increased high school 
graduation rates by more than 18 percentage points.
Figure 1. Impact of Access to Food Stamps During Early Life on Adult 
        Health and Economic Outcomes
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Hoynes, Schanzenbach, and Almond 2016.
          Note: Hollowed bars are not statistically significant.

    In addition to reducing food insecurity, SNAP participation may 
also reduce households' risk of suffering financial hardships (Figure 
2). Shaefer and Gutierrez (2013) use variation in state-level policies 
that affect SNAP access to study the impact of SNAP participation on a 
variety of outcomes. They find that receiving SNAP reduces the 
likelihood of food insecurity by 13 percentage points.
    SNAP also has spillover impacts on other aspects of families' 
financial well-being. Households have more resources available for 
other essential expenses, such as housing, utilities, and medical 
bills. Shaefer and Gutierrez estimate that SNAP participation reduces 
the risk of falling behind on rent or mortgage payments by seven 
percentage points and on utility bills (gas, oil, and electricity) by 
15 percentage points. Participants are also less likely to experience 
medical hardship: SNAP participation decreases the likelihood of 
forgoing a necessary visit to a doctor or hospital by nine percentage 
points.
Figure 2. Impact of SNAP Participation on Food Insecurity and Other 
        Financial Hardships
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Shaefer and Gutierrez 2013.
          Note: Sample includes low-income households with children. 
        Medical hardship is measured as whether the interviewee 
        reported that in the past 12 months someone in the household 
        chose not to see a doctor or go to the hospital when needed 
        because of cost.

    The USDA and RIA fail to consider the costs and benefits to 
restricting access to SNAP on food security, economic security, and 
health. While labor force attachment is a path to economic self-
sufficiency as the rule states, the evidence shows that SNAP benefit 
receipt also leads to economic self-sufficiency, household budget 
stabilization, and improved health. The rule states that imposing 
additional work requirements ``would also save taxpayers' money (p. 
982)'' but does not provide an analysis that considers the 
countervailing costs to limiting access to SNAP. The USDA and RIA fail 
to consider the costs to nonparticipation on both individual households 
and, as we will show throughout, the economy as a whole.
C. Macroeconomic Stabilization
    While the safety net should expand to provide resources to 
households experiencing firsthand economic losses, governments may use 
fiscal policy--additional government spending or tax cuts--to stimulate 
the economy during a recession. A fiscal multiplier is an estimate of 
the increased output caused by a given increase in government spending 
or reduction in taxes. Any multiplier greater than zero implies that 
additional government spending (or reduced taxes) adds to total output. 
Fiscal multipliers greater than one indicate an increase in private-
sector output along with an increase in output from government 
spending. This can occur because the additional spending can turn into 
increased employment or wages which subsequently increase output.
    Although there is disagreement among economists over the exact size 
of various fiscal multipliers (see Auerbach, Gale, and Harris [2010] 
for a discussion), multipliers are generally believed to be higher 
during recessions than they are under normal economic conditions when 
the economy is near its full potential, and they are in particular 
thought to be higher when the central bank is not raising rates in 
response to economic fluctuations (Auerbach and Gorodnichenko 2012; 
Fazzari, Morley, and Panovska 2014; see Ramey and Zubairy 2014 for a 
dissenting view). This is likely because downturns are characterized by 
slack in both labor and capital markets (i.e., available resources are 
not fully employed), thereby allowing fiscal stimulus to increase total 
output. Multipliers are also higher when the spending program or tax 
cut targets lower-income people, who are more likely to spend the 
stimulus (Parker, et al., 2013; Whalen and Reichling 2015).
    Not all spending or tax cuts are created equal, as indicated by the 
variation in fiscal multipliers shown below in Figure 3. But during the 
depths of the recession, each spending multiplier analyzed by Blinder 
and Zandi (2015) was greater than one, indicating that spending on 
these programs raised output by more than their costs. Note that the 
multipliers reported here are broadly similar to those estimated by CBO 
(Whalen and Reichling 2015).
    As shown in the below figure, the most stimulative type of spending 
during the recession was a temporary increase in the SNAP maximum 
benefit: for every $1 increase in government spending, total output 
increased by $1.74. Work-share programs and UI benefit extensions were 
also relatively stimulative. Consistent with economic theory, the 
programs with the largest multipliers were those directed at low-income 
or newly unemployed people. More recently, as the economy has improved, 
the multipliers have diminished. However, the multipliers for SNAP 
benefits, workshare programs, and UI benefits remain above one, 
indicating that these programs remain highly effective as forms of 
stimulus, generating additional private-sector economic activity. SNAP 
multipliers were also estimated to be greater than 1 in 2015Q1, well 
after the recession had ended.
Figure 3. Fiscal Stimulus Multipliers (Spending Programs), 2009 and 
        2015
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Blinder and Zandi 2015.

    Poverty and economic hardship typically increase in recessions and 
decrease in economic expansions. In particular, households with few 
resources are especially affected by the business cycle. Among poor 
households, the effect of the Great Recession was particularly severe 
relative to previous recessions. The unemployment rate rose notably 
more for lower education workers. This is a typical feature of 
recessions: less-educated workers face larger employment losses when 
the economy turns down (see Aaronson, et al., 2019 for a review). The 
safety net plays an important role in mitigating these effects, partly 
by automatically expanding during economic downturns as eligibility for 
safety net programs increases.
    Over the course of the Great Recession, SNAP rightly expanded to 
provide more benefits to eligible and newly eligible participants, 
including ABAWDs. Part of this expansion was the result of Bush 
Administration, Congressional, and Obama Administration action at 
several points over the course of the recession to expand waiver 
eligibility because existing policy was not sufficient to meet economic 
goals. These actions were necessary for macroeconomic stabilization and 
because the existing rules for ``lack of sufficient jobs evidence'' in 
applying for ABAWD work requirement waivers insufficiently responded to 
economic circumstances.
    The USDA and its RIA fail to model and consider the costs and 
benefits to the proposed rule during any alternate economic conditions. 
USDA proposes making changes to existing policy that would weaken 
responsiveness to indicators of an economic downturn (statewide waiver; 
20 percent rule; 3 month lookback), its persistence (statewide waiver; 
3 month lookback), and sluggish recoveries in particular places 
(statewide waiver; 20 percent rule). Our analysis provides evidence 
that existing policy (20 percent rule without a floor, ten percent rule 
with two lookback periods) provided coverage more in keeping with the 
economic conditions at various points in time than the proposed 
changes. Furthermore, the USDA fails to offer proposals, such as 
linking waiver eligibility to Emergency Unemployment Compensation (EUC) 
in the event that EUC is authorized, that would make waiver eligibility 
more responsive during the onset of a recession.
III. Modeling Waiver Eligibility
A. Regulatory Impact Analysis
    The USDA's proposed rule makes several changes for which the 
Regulatory Impact Analysis must account. The USDA proposes to disallow 
states from applying for statewide waivers except in the case of EB 
eligibility and to define regions at their discretion. The USDA 
proposed maintaining eligibility for geographic areas qualifying for EB 
and with 12 month unemployment rates above ten percent. USDA proposes 
to modify eligibility for those places with an unemployment rate of ten 
percent in a recent 3 month period to only be used in support of ``an 
exceptional circumstance p. 985.'' USDA proposes to put an unemployment 
rate floor of seven percent to the 20 percent rule. We provide evidence 
that the RIA does not properly analyze the effects of these proposed 
changes, thus substantially underestimating the impacts.
1. Waiver Take-up
    The Regulatory Impact Analysis (RIA) is based on areas that have 
taken up a waiver in a single contemporaneous period. The RIA failed to 
consider eligibility for waivers in its analysis for the single time 
period it did analyze. The RIA did not consider the effect of their 
proposal under alternative macroeconomic conditions, either in actual 
take-up or in eligibility. In doing so, it materially underestimates 
the number of program participants who would be subject to time limits 
during recessions.
    The RIA writes that they chose to model actual waiver take-up 
rather than eligibility because ``States do not always seek waivers for 
eligible areas. Some States seek no time limit waivers; others only 
seek waivers for a portion of qualifying areas within the State. 
Therefore, the Department assumed that if a county was not currently 
waived, the State would not seek a waiver for that area under the 
revised criteria (p. 20).''
    This logic is faulty and false by recent evidence. States that have 
declined to take-up waivers for which they are eligible are assumed to 
have made a choice that they would never make differently--even if 
economic conditions in their state deteriorated. Similarly, states that 
have applied for waivers for which they are eligible are assumed to be 
the only places where the impacts of more stringent eligibility would 
be felt in perpetuity.
    By this logic, we could look at waiver status in any preceding year 
as the expression of a state's policy preference--preferences that 
change based on economic conditions. As USDA noted in the RIA, in July 
2013, 44 states and D.C. applied for statewide waivers and six states 
had waivers for part of their state. Had the RIA used recently 
expressed actual preferences for rather than the single time period 
that they considered or modeled the effect based on eligible areas, 
they would have found larger impacts in Federal spending and the number 
of individuals denied access to resources to purchase food.
    In 2017,\1\ each of the 17 states that did not avail themselves of 
time limit waivers had at least one county that was eligible. This does 
not mean they would always choose to decline to use waivers. In fact, 
of the 17 states currently eligible for waivers that are not using 
them, 14 were using waivers to cover counties not individually eligible 
in 2008 (shown later in figure 8) and every state received waivers for 
at least a part of the state in 2009. The existing waiver process 
allows states to determine when it makes sense to apply for them based 
on their understanding of their local economy. It is incumbent on the 
NPRM to explain why limiting that discretion furthers program goals.
---------------------------------------------------------------------------
    \1\ 2017 is the last year for which there is publicly available 
data on waiver take-up by county.
---------------------------------------------------------------------------
2. Statewide Waivers and Geographic Areas
    The RIA does not consider the effect of eliminating statewide 
waivers except as downstream to other policy changes. It does not model 
whether a state would ever qualify for a waiver based on each 
underlying geographic unit's qualification. It does not model state 
eligibility for EB, or in relation to EUC. In doing so, the NPRM and 
RIA fail to justify proposed restrictions on statewide waivers.
    Statewide waivers are particularly critical during serious economic 
downturns. Any heterogeneity in the use of waivers impedes the 
geographic mobility of program participants. Unlike in UI, where 
individuals retain benefits if they move to a better labor market, SNAP 
ties benefit receipt to their place of residence. In order to maintain 
benefits, participants are incentivized not to move to find work, but 
to maintain residency in an area that is economically lagging but 
waiver eligible. This reduces employment and labor force participation 
of SNAP program participants and does not increase economic self-
sufficiency. USDA does not provide analysis to consider these costs.
    The RIA does not consider Labor Market Areas (LMAs) in its 
analysis, though a county can become eligible for a waiver due to being 
a part of an LMA. It writes, ``Because a small number of areas 
estimated to lose eligibility may actually qualify as part of a larger 
LMA, the Department rounded the impact from 77.4% down to 76 percent 
(p. 22).'' There is no justification for this rounding nor for 
excluding counties eligible as part of an LMA from their analysis. The 
result is misattribution of some counties otherwise eligible to the 
state-selected geographic group or statewide standard and an 
unspecified effect on policy impacts. Our analyses show that more than 
five percent of counties qualify only through being a part of an LMA.
    The RIA does not include most of New England--Connecticut, Maine, 
Massachusetts, New Hampshire, and Vermont--in its analysis. Failure to 
do so both affects the validity of the estimates and calls into 
question whether counties and LMAs are an appropriate level of 
geography as is argued.
3. 20 Percent Rule
    The Regulatory Impact Analysis failed to correctly model the 20 
percent rule. They write: ``The Department obtained monthly 
unemployment and labor force data from BLS . . . for the 24 month 
period from January 2016 to December 2017 for 3,077 counties and 
county-equivalents (p. 21).'' The Department therefore determined that 
any waived county that was waived but not by the 20 percent rule was 
part of a contiguous state-determined geographic group or through a 
statewide waiver.
    This is incorrect because the RIA fails to accurately model the 20 
percent rule or consider other paths to eligibility. The 20 percent 
rule states that the first month of the 24 month period used to 
identify whether an area's unemployment rate is 20 percent above cannot 
be earlier than the first month BLS uses to determine LSAs. The RIA 
does not say what period it is calculating 20 percent eligibility, but 
it does so using only one 24 month period. Within a window for 
applications, there are in fact ten distinct 24 month periods against 
which a state can submit a waiver application.
    The RIA states that in linking the 20 percent rule to LSA 
designations states will be prevented ``from using older data (p. 
16).'' This is false. The proposed rule does not make any changes with 
regard to the time period over which data can be taken, only that the 
waiver expiration date would be proscribed.
    The NPRM defends a six percent unemployment rate floor by noting 
that if there is agreement the ``natural rate'' of unemployment hovers 
near five percent, then 20 percent above that would be six percent. 
But, the Department does not choose a six percent floor, instead 
preferring a seven percent floor (in part because of a concern that 
``too few individuals would be subject to ABAWD work requirements'' 
without explaining why the number would be too few.) In addition, the 
Administration's forecast suggests the unemployment rate will stabilize 
at 4.2 percent and never rise above it this decade. Twenty percent 
above that rate would be a floor of five percent. No attempt to justify 
a higher floor like seven percent is made beyond noting it will subject 
more people to work requirements.
4. Labor Surplus Areas
    By failing to provide sufficient evidence for the seven percent 
floor to the 20 percent rule, the USDA consequently fails to justify 
removing Department of Labor (DOL) designation as a Labor Surplus Area 
(LSA) as a waiver qualification. Essentially, LSAs are also determined 
by the 20 percent rule and the ten percent rule, but have a floor of 
six percent unemployment. A city with a population of at least 25,000, 
a town or township of at least 25,000, counties, balances of counties, 
and county-equivalents can all qualify as LSAs. Under exceptional 
circumstances, civil jurisdictions, Metropolitan Statistical Areas and 
Combined Statistical Areas are geographies that could qualify as LSAs. 
The justifications for removing LSAs run counter to stated goals: high-
quality and federally-produced data and clear standards for areas with 
insufficient jobs should determine waiver eligibility. The USDA and RIA 
fail to provide sufficient evidence for removing waiver eligibility 
based on LSA designation.
5. Effect on Society and Uncertainties
    The RIA acknowledges that it fails to consider actual impacts under 
any alternative economic conditions, ``(including cyclical (p. 29).'' 
They also acknowledge that meeting work requirements is a function of 
both the availability of jobs and the ``extent that States offer 
qualifying E&T or workfare opportunities (p. 29).''
    The RIA acknowledges that ``there may be increases in poverty and 
food insecurity (p. 28)'' for those who fail to meet work requirements, 
``those ABAWDs who become employed will likely see increased self-
sufficiency and an overall improvement in their economic well-being (p. 
28),'' and that ``a number of those affected by strengthened work 
requirements are able to secure employment in a wide range of different 
industries (p. 28).''
    The effect of the proposed regulatory changes were inadequately 
analyzed, failing to take into account the costs and benefits of 
restricting access to the program. The RIA does not provide estimates 
for increases in rates of poverty or food insecurity and its attendant 
costs. In particular, it does not engage with the evidence of the long-
run benefits of SNAP, the effect of SNAP on reductions in food 
insecurity and poverty, nor with the concerns regarding reducing 
resources to the children of non-custodial parents. It does not provide 
evidence for increased labor supply among ABAWDs, and in fact the RIA 
acknowledges elsewhere that employment rates may not increase at all as 
a result of the policy change. ``A number of those affected (p. 28)'' 
is not a specific analysis on which to base a regulatory change.
    Without evidence that any affected program participant would become 
employed as a result of the policy, it remains unclear whether there 
are any benefits to the proposed rules.
B. Analysis Based on Eligibility
    In this section, we provide evidence for the share of counties that 
would have been eligible for a waiver based on each trigger from 2007 
to the present (1) in existing regulation, (2) through policy changes 
throughout the Great Recession, and (3) in the proposed rules including 
for each unemployment rate floor to the twenty-percent rule. Modeling 
eligibility and take-up over time is appropriate for identifying 
program effects.
    The geographic unit considered in each of the following models are 
the share of counties eligible for a waiver. These counties can gain 
eligibility individually, as a county in a labor market area (LMA) that 
is eligible, or because the county is in a state that has a statewide 
waiver.\2\
---------------------------------------------------------------------------
    \2\ To understand maximum eligibility, we look at county 
eligibility based on the county-level data as well as the LMA-level 
data. Because the LMAs in New England States are made up of minor civil 
divisions and not counties, eligibility in counties in ME, MA, NH, VT, 
and CT is only modeled on county data.
---------------------------------------------------------------------------
    We are unable to show the share eligible based on state-selected 
geographic areas under current rules.\3\ We do not model triggers based 
on the following rules: a historical seasonal unemployment rate above 
ten percent; Labor Surplus Area designation by the Department of 
Labor's Employment and Training Administration; a low and declining 
employment-to-population ration; a lack of jobs in declining 
occupations or industries; or, is described in an academic study or 
other publication as an area where there is a lack of jobs.
---------------------------------------------------------------------------
    \3\ The Regulatory Impact Analysis conducted by USDA also does not 
model sub-state groups for eligibility optimization.
---------------------------------------------------------------------------
    In our model, a geographic unit can be eligible for a waiver based 
on three unemployment rate thresholds (in addition to other policy 
mechanisms discussed below). First, a geographic unit is eligible if it 
has a 24 month average unemployment rate that is 20 percent above the 
national average for the same 24 month period.\4\ Second, a geographic 
unit is eligible for a waiver if it has a 12 month average unemployment 
rate above ten percent. Third, a geographic unit is eligible for a 
waiver if it has a 3 month unemployment rate above ten percent. A state 
can generally request a 12 month waiver and specify the implementation 
date on the waiver request.\5\
---------------------------------------------------------------------------
    \4\ We follow the USDA guidance and rounded national and local 
unemployment rates to the nearest tenth.
    \5\ The window for a waiver application based on the 20 percent 
rule is based on Section V of the USDA guidance. We assume that states 
will apply for waivers on the last possible application date, i.e., the 
end of a fiscal year period as defined in the guidance. The guidance 
states that ``For example, the 24 month period for the Fiscal Year 2017 
LSA list runs from January 1, 2014 through December 31, 2015. Thus, a 
waiver that would start in Fiscal Year 2017 could be supported with a 
24 month period beginning any time after (but not before) January 1, 
2014.'' Therefore, if a geographic unit has a 24 month average that 
starts on January 1, 2014 and ends on January 1, 2016, the latest they 
could apply for the waiver would be September 30, 2017. The waiver 
period extends 12 months from the application date. We therefore assume 
that the geographic unit in question is eligible for a waiver from 
January 1, 2016 through September 30, 2018.
---------------------------------------------------------------------------
    If a state qualifies under any of these triggers or if a state's 
unemployment insurance extended benefits program triggers on, then the 
state is eligible for a statewide SNAP waiver. In this analysis, we 
model EB eligibility based on the first date that a state is shown to 
be eligible on a Department of Labor EB trigger notice.\6\ We also 
model eligibility based on EUC and ARRA.
---------------------------------------------------------------------------
    \6\ We follow USDA guidance with regard to EB-based eligibility. A 
state is eligible for a work requirement waiver based on EB if a state 
has (1) a 13 Week Insured Unemployment Rate (IUR) of five percent and 
120 percent of each of the last 2 years; (2) an IUR of six percent; (3) 
a 3 Month Total Unemployment Rate (TUR) of 6.5 percent and 110 percent 
of either of the last 2 years.
---------------------------------------------------------------------------
    In the following sections, we model waiver eligibility and waiver 
take-up as a share of counties from 2007 to present.
1. Work Requirement Waiver Eligibility during the Great Recession
    Work requirement waivers in a recession are important for two 
reasons. First, job finding rates fall in recessions and difficulty 
finding work may mean many individuals who are trying to be labor force 
participants will be sanctioned for failure to work the required number 
of hours. This is counter to program goals. It is well-known that 
recessions strike marginalized populations in the labor force more 
harshly than higher income, higher education individuals. Because 
during a recession more people become eligible for and would benefit 
from program participation due to recent job or income loss as well as 
the inability to find sufficient work, it is particularly important to 
waiver time limits for the SNAP-eligible population. Second, removing 
individuals from SNAP during a recession shrinks SNAP's role as an 
automatic stabilizer by providing spending in depressed areas during a 
downturn.
    In order to expand access to geographic waivers in response to the 
recession, executive and Congressional action was necessary. None of 
the automatic triggers were sufficient to turn on the waivers for much 
of the country promptly. The Bush and Obama Administrations, Congress, 
and states took action throughout the Great Recession to increase 
geographic eligibility for waivers, directly and through clarifying 
ties to Unemployment Insurance (UI).
    During the Great Recession, Congress enacted Emergency Unemployment 
Compensation (EUC), a temporary program that extended the amount of 
time during which an eligible UI participant could retain benefits. 
Congress authorized EUC on June 30, 2008, extending the expiration date 
to January 1, 2014 (American Taxpayer Relief Act of 2012).
    Additionally, the Bush Administration clarified on January 8, 2009 
that eligibility for Emergency Unemployment Compensation (EUC) also 
qualified states for SNAP waivers.\7\ EUC established several tiers of 
additional weeks of UI benefits, with each tier contingent on a state 
having a total unemployment rate that exceeded a given threshold. EUC 
tier qualifications interacted in different ways with SNAP Waiver 
eligibility over the EUC period. Importantly, states were eligible for 
SNAP waivers if they were eligible for particular tiers of EUC, and not 
just if they took EUC (see table 1 for eligibility thresholds and the 
interaction of SNAP waivers and EUC tiers).
---------------------------------------------------------------------------
    \7\ EUC trigger notices are issued on a weekly basis. Our analysis 
is on a monthly basis. If a state was eligible for EUC in at least 2 
weeks in a month, we consider it to be eligible for EUC in that month.
---------------------------------------------------------------------------
    ARRA was enacted on February 17, 2009. It stated that for the 
remainder of FY2009 and through FY2010 ABAWDs were waived from work 
requirements to maintain access to the program. While a few localities 
declined this authorization, every county in the U.S. was eligible for 
waiver from February 17, 2009 to September 30, 2010.
    Figure 4 models each component of work requirement waiver 
eligibility that was operational from 2006 to present. The unit is the 
share of counties eligible for a waiver, whether individually, as part 
of an LMA, or as part of an eligible state. The set of triggers and 
eligibility standards are based on standing regulation as well as 
policy changes made over the course of the Great Recession to increase 
waiver eligibility. The criteria that did not change over the course of 
the Great Recession were eligibility based on EB, the twenty-percent 
rule, and the ten percent unemployment rate by two look-back period 
rules.
Figure 4. Counties Eligible for A Work Requirement Waiver by Trigger, 
        2007-present
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Local Area Unemployment Statistics, BLS (2000-2018); 
        EB and EUC Trigger notices (DOL); Bureau of Labor Statistics 
        (2000-2018).

    The 20 percent rule (light blue) slowly increases the availability 
of waivers at the start of the recession in the absence of 
Congressional action as some parts of the country had its unemployment 
rate rising before the rest. This analysis shows that the vast majority 
of areas waived from the rules in the third quarter of 2008--a period 
when the economy was losing over 300,000 jobs a month--was due to the 
20 percent rule. Since 2016, the vast majority of counties eligible for 
an ABAWD waiver is due to qualifying under the 20 percent rule. Still, 
it is not a perfect trigger. If the entire country is facing rising 
unemployment rates, the waivers would not be available anywhere until 
local 3 month (or 12 month) unemployment rates exceed ten percent or 
EB-based triggers come on under standing rules. This analysis shows 
that the 20 percent rule plays a critical part in SNAP's role as an 
automatic stabilizer and should not be weakened.
    Standing policy with regard to statewide waivers would have 
provided wider coverage in the event that eligibility based on EUC and 
ARRA did not occur. The Extended Benefit (EB) trigger for UI in-law has 
failed to trigger on during recessions without Congressional and state 
action since its enactment (Wandner 2018), though work requirement 
waivers are based on eligibility by USDA-determined thresholds that 
ameliorate this issue. For a short period of time in late 2008 and the 
first week of 2009, EB eligibility provided the widest amount of 
coverage, but its acceleration in 2008 was not sufficiently early or 
fast enough to reduce the value of the 20 percent rule. USDA proposes 
to maintain EB-based eligibility, and the evidence presented here shows 
this is a necessary but not sufficient waiver eligibility condition.
    During the Great Recession, Emergency Unemployment Compensation was 
authorized in June 2008 but it was not until January 2009 that the Bush 
Administration clarified that states eligible for a particular tier of 
EUC were also eligible for SNAP work requirement waivers. About 90 
percent of counties became eligible based on this measure, and through 
ongoing memorandums linking work requirement waiver eligibility to 
different EUC tiers, a high level of waiver eligibility was maintained 
through 2016. Given that roughly 35 percent of counties were already 
eligible based on the 20 percent rule in 2008, the expansion of waiver 
eligibility based on EUC dramatically expanded waiver eligibility. Had 
waiver eligibility been tied to EUC upon enactment, work requirement 
waivers would have been an even more effective counter-cyclical tool. 
An improvement to the rules would be to clarify that in the event EUC 
is authorized, states become immediately eligible for work requirement 
waivers.
    Combining these indicators into three bins--eligibility based on 
standing policy as of 2006, additional eligibility based on EUC, and 
additional eligibility based on ARRA--we can model the effect of 
existing waiver policy and of the policy preferences of Administrations 
of both parties and Congress with regard to waiver eligibility (figure 
5).
Figure 5. Counties Eligible for ABAWD Work Requirement Waiver, 2007-
        present
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Local Area Unemployment Statistics, BLS (2000-2018); 
        EB and EUC Trigger notices (DOL), Bureau of Labor Statistics 
        (2000-2018).

    Current policy with regard to waiver eligibility provided all the 
coverage until the Bush Administration linked waiver eligibility to 
EUC. Existing and recession-responsive policy functioned to provide 
close to 100 percent waiver eligibility from 2009 to 2014. The scope of 
coverage was driven by policy actions taken at the Federal and state 
levels to increase eligibility for EUC and EB, which had downstream 
effects on SNAP work requirement waivers. In the absence of such 
actions, the 20 percent rule is the most effective of standing rules at 
providing waiver eligibility at the start of the recession and EB is 
the most effective during recovery. No standing rules provide coverage 
of the scale and speed instigated by policy actions taken during the 
Great Recession.
2. Modeled Eligibility Versus the Proposed Rule
    We compare existing standing policy (purple) for waiver eligibility 
with the proposed rules including three options proposed by the USDA 
for the 20 percent rule as they would have performed not just ``now,'' 
as the RIA showed, but over the course of the Great Recession (figure 
6). The model for the proposed rule also maintains eligibility for 
areas having an unemployment rate above ten percent over a recent 12 
month period and for areas in which EB would have triggered 
eligibility.
    Because eligibility based on EB is consistent across standing and 
proposed rules, we focus on how the different floors to the 20 percent 
rule (no floor, six, seven, and ten percent unemployment floors) affect 
access to SNAP at the onset and during the Great Recession before 
discussing considerations of when, whether, and how to have waivers 
trigger off.
    USDA's preferred modification is to implement a seven percent floor 
for the 20 percent rule and eliminate the 3 month lookback and 
statewide waivers (light green).\8\ Had this rule been in place in the 
first quarter of the Great Recession, when the economy was losing 
300,000 jobs a month and when SNAP rolls should be expanding, waivers 
would have been limited to less than 20 percent of counties. The ten 
percent floor (teal) would have performed worse, with less than ten 
percent of counties eligible. The six percent standard (dark green) 
covered less than 30 percent of counties.
---------------------------------------------------------------------------
    \8\ The new regulations state that for the 20 percent rule, the 
period of eligibility for a state will only last through the end of the 
fiscal year in which a state applied, as opposed to 1 year from the 
date of the application. We have assumed that the waiver application 
limits are the same as the current regulations, and have extended the 
period of waiver through the end of the fiscal year. Additionally, we 
have applied the same rounding standards to the respective floors as to 
the 20 percent cutoff above.
---------------------------------------------------------------------------
Figure 6. Counties Eligible for ABAWD Work Requirement Waiver, Existing 
        and Proposed Regulations
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Local Area Unemployment Statistics, BLS (2000-2018); 
        EB and EUC Trigger notices (DOL), Bureau of Labor Statistics 
        (2000-2018).
          Note: Because eligibility for waivers due to EB status is 
        included in each line, the lines converge once widespread EB 
        status occurs in early 2009.

    By standing and proposed rules, waiver eligibility dissipate 
measurably in 2013. But, the revealed preference of the policymakers at 
the time was that the current rules were too restrictive and needed to 
be relaxed. A number of decisions were made to expand and extend waiver 
eligibility, both early in the recession and afterwards. Figure 7 
highlights the difference between the revealed preferences of 
policymakers working to stabilize the economy during the recession and 
how waiver eligibility would have worked based on the proposed rules.
    The purple line shows eligibility for work requirement waivers 
based on standing regulations, EUC, and ARRA. This line contrasts with 
eligibility for the proposed rules: EB eligibility, ten percent 
unemployment with a 12 month lookback, and the 20 percent rule with 
varying floors. The revealed preferences on policymakers during the 
Great Recession was to use policy tools relevant to identifying areas 
with insufficient jobs to expand SNAP work requirement waiver 
eligibility, in part because existing rules were insufficient to the 
task. Both at the start of the recession and in the event of a sluggish 
recovery, the proposed rules diminish SNAP's role as an economic 
stabilizer and safety net.
Figure 7. Counties Eligible for ABAWD Work Requirement Waiver, Proposed 
        Regulations versus Actual Eligibility during the Great 
        Recession
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Local Area Unemployment Statistics, BLS (2000-2018); 
        EB and EUC Trigger notices (DOL); Bureau of Labor Statistics 
        (2000-2018).

    As this analysis emphasizes, if there is a problem with the current 
rules, it is in the beginning of a recession because existing rules do 
not allow states to respond promptly to a recession. The proposed rule 
does not address or fix waiver responsivity to the onset of an economic 
downturn. Thus, the fact that the proposed rule would make the waiver 
process less responsive to an economic downturn and less able to 
accomplish the goals of the program is absent from considerations of 
costs and benefits. It is incumbent on the proposed rule to ensure that 
it does not make responsiveness to an economic downturn worse.
3. Eligibility Versus The Proposed Rule
    In the preceding sections, we have modeled waiver eligibility to 
the extent possible and clearly articulated the ways in which we would 
not be able to model legitimate features of the existing rules. Most 
notably, we were unable to model regional eligibility and were unable 
to model eligibility based on Labor Surplus Areas. By adding to these 
models data from publicly available maps produced by the Center on 
Budget and Policy Priorities, we are able to identify counties that are 
eligible for work requirement waivers by triggers that we were unable 
to model through our method for those states that implemented these 
standards.\9\
---------------------------------------------------------------------------
    \9\ The data on work requirement waiver eligibility can be found at 
https://www.cbpp.org/research/food-assistance/States-have-requested-
waivers-from-snaps-time-limit-in-high-unemployment. The data on county 
eligibility was copied by hand and duplicated by a second researcher 
using mapchart.net to produce a JSON, which converted the visualization 
into data used to produce the analyses. We did not have access to any 
waiver application information or to USDA-produced information 
regarding waiver eligibility. If any area of a county received a 
waiver, we counted the entire county as receiving a waiver due to an 
inability to be more precise. These maps are predicated on waiver take-
up; we continue to be unable to identify waiver eligibility based on 
regional eligibility or LSAs for states that chose not to apply.
---------------------------------------------------------------------------
    For focal years 2008 and 2017, we produce maps of the continental 
United States to identify differences in waiver eligibility by the 
proposed rule, the existing rules as modeled, and the existing rules as 
waived. Figures 8 and 9 are maps showing which counties would be 
eligible for work requirement waivers under both current rules (which 
do not model eligibility based on grouping of contiguous areas) and the 
proposed rules (EB, ten percent rule with a 12 month lookback, 20 
percent rule with the seven percent unemployment rate floor [purple]), 
which counties would lose eligibility due to changes in standing rules 
(blue), and which counties would lose eligibility because they are 
regionally eligible or eligible by one of the criteria (like LSAs) that 
we are unable to model (orange).
    In 2008, during the Great Recession, most states used the 
flexibility afforded to them by standing rules to quickly respond to 
changing economic conditions and cover areas that would not be 
individually eligible--either by applying for statewide waivers or 
through regional eligibility. For example, Ohio applied for and was 
granted a 2 year statewide waiver in June of 2008 to cover July 1, 2008 
to June 30, 2010 based on the state qualifying under the 20 percent 
rule (Ohio Job and Family Services 2008) and parts but not all of 
Pennsylvania qualified regionally (Pennsylvania Department of Human 
Services 2008). As economic conditions deteriorated, existing 
flexibility with regard to both geographic unit and economic indicators 
allowed states to respond more quickly to the recession than Congress 
or the Executive Branch.
Figure 8. Waiver Eligibility by Standing and Proposed Rules, 2008
2008

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    The NPRM states ``a significant number of states continue to 
qualify for and use ABAWD waivers under the current waiver standards 
(p. 981).'' Based on the USDA waiver status notifications, over the 
course of 2017, eight states and D.C. were approved to receive a 
statewide waiver, 26 states had a partial waiver, and 16 states were 
implementing time limits statewide. Figure 10 shows that six states 
would have no eligible areas for work requirements under the proposed 
rules, of which three (Vermont, New Hampshire, Connecticut) have 
currently eligible areas that would lose coverage. The states who 
submitted waiver applications, in doing so expressing their preference 
for waiver flexibility, and who would have seen coverage reduced based 
on the proposed rules had they been implemented in 2017 are Alabama, 
Arizona, California, Colorado, Connecticut, Georgia, Idaho, Kentucky, 
Maryland, Massachusetts, Michigan, Minnesota, Montana, Nevada, New 
Hampshire, New Jersey, North Dakota, Ohio, Oregon, Pennsylvania, Rhode 
Island, South Dakota, Tennessee, Utah, Vermont, and Washington. 
According to USDA and affirmed in our analysis, 17 states declined to 
submit a waiver for eligible areas: Alabama, Arkansas, Delaware, 
Florida, Indiana, Iowa, Kansas, Maine, Mississippi, Missouri, Nebraska, 
North Carolina, Oklahoma, South Carolina, Texas, Wisconsin, and Wyoming 
(USDA 2017a).
Figure 9. Waiver Eligibility and Take-up, 2017
2017

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Table 1 shows where in the U.S. and through which eligibility 
trigger would counties have lost eligibility in 2017. We show RIA Table 
3 for comparison and assume that Table 3 refers to 2018. By our 
calculations, in 2017, 1,322 counties were eligible and 1,012 counties 
took up a waiver.
Table 1. Impact of Rule Provisions

    RIA Table 3. Impact of Rule Provisions on Currently-Waived Areas
------------------------------------------------------------------------
                            Areas still
Currently waived areas =  qualifying for   Reduction in       Percent
           975                waivers      waived areas      reduction
------------------------------------------------------------------------
Eliminate other                      621            ^354            ^36%
 eligibility criteria
Eliminate statewide                  582             ^39             ^4%
 waivers
Implement 7% UR                      220            ^362            ^37%
 threshold
                         -----------------------------------------------
  Total                              220            ^755            ^76%
------------------------------------------------------------------------
            Impact of Rule Provisions on Take-up Areas, 2017
------------------------------------------------------------------------
Current areas taking up     Areas still    Reduction in       Percent
 waivers = 1,012             taking up     waiver areas,     reduction
                              waivers         take-up
------------------------------------------------------------------------
Eliminate 10% UR, 3                1,011              ^1              0%
 month lookback
Implement 7% UR                      853            ^158            ^16%
 threshold
Eliminate EUC                        853               0              0%
Eliminate ARRA-related               853               0              0%
 triggers
Eliminate statewide                  820             ^33             ^4%
 waivers
Eliminate other                      574            ^246            ^30%
 eligibility criteria
                         -----------------------------------------------
  Total                              574            ^438            ^43%
------------------------------------------------------------------------

    Looking first at counties that would qualify individually or as 
part of an LMA, one county would lose eligibility due to the 
elimination of the 3 month lookback on ten percent unemployment and 158 
counties would lose eligibility based on the implementation of a seven 
percent floor to the 20 percent rule. This is substantially smaller 
than the 362 counties that the RIA states would lose eligibility due to 
the implementation of a seven percent unemployment rate floor to the 20 
percent rule. This is evidence that the RIA incorrectly modeled the 20 
percent rule and that failing to account for LMA-based eligibility has 
substantially affected their estimates.
    Next, we look at the effect of eliminating statewide waivers on 
eligibility. In 2017, the following states had statewide work 
requirement waivers: Alaska, California, District of Columbia, 
Illinois, Louisiana, Nevada, and New Mexico. Alaska would maintain 
statewide eligibility based on EB, but 33 counties would lose 
eligibility because of the loss of these statewide waivers.
    Like the RIA, we do not directly model the remaining eligibility 
criteria. Unlike the RIA, we assign the remainder of take-up counties 
to this category, rather than starting with it. We find that 246 
counties taking up waivers would lose eligibility by eliminating the 
remaining eligibility criteria, compared with 354 for the RIA. We find 
that 574 counties among those actually waived in 2017 would retain 
eligibility, while the RIA finds that 220 counties would.
    The NPRM has misspecified the justification for the NPRM and has 
failed to properly analyze the regulatory impact. This analysis finds a 
deleterious effect of the new rules at the onset of a recession and 
less reduction in coverage ``today.'' For these reasons, the current 
rules should be maintained.
IV. Employment Status Changes
    When an area is not subject to a waiver, work requirements subject 
Able-bodied Adults without Dependents (ABAWDs) to a time limit for 
receiving SNAP benefits under the law. The exemptions to this rule are 
at the participant level, for example, those receiving disability 
income or who are ``unfit'' for employment based on a physical or 
mental disability, those who have dependent minor children, and those 
outside the targeted age range are not subject to the work 
requirements.
    This section provides evidence that suggests waivers from work 
requirements at both the individual and geographic area should be more 
readily available. We show that economic conditions beyond the control 
of program participants are driving whether they can meet the 20 hour a 
week standard consistently, as work-related reasons explain a 
substantial share of gaps in working for pay. ABAWDs also appear to be 
in substantially poorer health than non-SNAP recipients. Furthermore, 
about 20 percent of ABAWDs are non-custodial parents, potentially 
exposing children to benefit loss from which the law protects them.\10\
---------------------------------------------------------------------------
    \10\ 20 percent of ABAWDs in the SIPP reported having a child under 
the age of 21 who lived in a different household or who reporting being 
a parent but who did not have a child living at home.
---------------------------------------------------------------------------
    The proposed rule would make it more difficult for geographic areas 
to qualify to apply for waivers. This will mean that some areas where 
states have weak enough economies to warrant the waivers would not be 
able to use them. We show that during 2013 and 2014, when only seven 
states and the District of Columbia had annual unemployment rates above 
seven percent:

   A plurality of ABAWDs experience labor force status 
        transitions over an extended period of time that would expose 
        workers to benefit loss even though they are in the labor 
        force;

   More than \1/3\ of workers who experienced a period of not 
        working said that it was due to a work-related reason, such as 
        failure to find work or being laid off while less than \1/2\ of 
        one percent of ABAWDs were not working due to lack of interest; 
        and,

   Four out of five ABAWDs who are out of the labor force are 
        not in fact able-bodied: while they do not receive disability 
        income, they report health or disability as the reason for not 
        working.

    The decline in labor force participation--especially among prime-
age males--has drawn extensive attention in academic and policy circles 
(e.g., Abraham and Kearney 2018; Juhn 1992; White House 2016). Some 
recent academic work has emphasized the fact that participation may be 
declining in part because an increasing number of labor force 
participants cycle in and out of the labor force: a pattern with direct 
relevance to proposed work requirements. The most comprehensive look at 
the behavior of people cycling through the labor force is Coglianese 
(2018). He documents that, among men, this group he refers to as ``in-
and-outs'' take short breaks between jobs, return to the labor force 
fairly quickly (within 6 months), and, crucially, are no more likely 
than a typical worker to take another break out of the labor force. See 
also Joint Economic Committee (2018) for a discussion of the in-and-out 
behavior of nonworking prime-age men and reasons for their non-
employment.
    SNAP participants who are employed but who work in jobs with 
volatile employment and hours would be at risk of failing work 
requirements. This group includes those who lose their job, sanctioning 
those who were recently employed and are searching for a new job. 
Similarly, those who work in jobs with volatile hours would be 
sanctioned in the months that their average hours fell below 20 hours 
per week, whether due to illness, lack of hours offered by the 
employer, or too few hours worked by the participant if they fail to 
receive a good-cause waiver. By making it more difficult for states to 
provide waivers when they feel conditions warrant, the proposed rule 
will cause more people to lose SNAP benefits.
    Low-wage workers in seasonal industries such as tourism would 
potentially be eligible for SNAP in the months when they are working, 
but not in the months without employment opportunities. In other words, 
while benefits are most needed when an individual cannot find adequate 
work, under proposed work requirements these are the times that 
benefits would be unavailable. Disenrollment could make it more 
difficult for an individual to return to work--for example, if a person 
with chronic health conditions is unable to access needed care while 
they are between jobs. Any work requirement that banned individuals 
from participation for a considerable amount of time after failing the 
requirements would be even more problematic for those facing churn in 
the labor market.
    In a set of analyses, Bauer (2018), Bauer and Schanzenbach (2018a, 
2018b) and Bauer, Schanzenbach, and Shambaugh (2018) found that 
although many SNAP beneficiaries work on average more than 20 hours a 
week every month, they frequently switch between working more than 20 
hours and a different employment status over a longer time horizon.
    For this comment, we examine labor force status transitions and the 
reasons given for not working among ABAWDs over 24 consecutive months, 
January 2013-December 2014. The data used are from the first two waves 
of the Survey of Income and Program Participation (SIPP). By using a 
data set that allows us to track workers over time, we identify the 
share of program participants who are consistently out of the labor 
force, the share who would consistently meet a work requirement, and 
the share who would be at risk of losing benefits based on failing to 
meet a work requirement threshold.
    We assume that to comply with a program's work requirement, 
beneficiaries would have to prove each month that they are working for 
at least 20 hours per week, or at least 80 hours per month, which is 
the typical minimum weekly requirement among the SNAP work requirement 
proposal. We calculate the share of program participants who would be 
exposed to benefit loss because they are not working sufficient hours 
over the course of 24 consecutive months. Among those who would be 
exposed to benefit loss and who experienced a gap in employment, we 
describe the reasons given for not working to help quantify potential 
waiver eligibility.
    We remove from the analysis all those who have a categorical 
exemption, excluding those outside the targeted age range, those with 
dependent children, full- or part-time students, and those reporting 
disability income. Program participants are those who reported 
receiving SNAP at any point between January 1, 2013, and December 31, 
2014. The vast majority of states over time period covered by the 
analysis had unemployment rates below seven percent in either 2013, 
2014, or both.\11\ The preponderance of evidence presented shown here 
is thus occurring in labor markets that the proposed rule says has 
sufficient jobs available to ABAWD SNAP participants.
---------------------------------------------------------------------------
    \11\ The states which had an unemployment rate above seven percent 
in both 2013 and 2014 were: Georgia, Illinois, Michigan, California, 
Mississippi, Rhode Island, Washington D.C., and Nevada.
---------------------------------------------------------------------------
    We categorize each individual in each month into one of four 
categories: (1) employed and worked more than 20 hours a week, (2) 
employed and worked less than 20 hours a week, (3) unemployed and 
seeking employment, or (4) not in the labor force. If a worker was 
employed at variable weekly hours but maintained hours above the 
monthly threshold (80 hours for a 4 week month and 120 hours for a 5 
week month) then we categorize them as (1) employed and worked more 
than 20 hours a week for that month. Individuals are considered to have 
a stable employment status if they do not change categories over 2 
years, and are considered to have made an employment status transition 
if they switched between any of these categories at least once. There 
is no employment status transition when a worker changes jobs but works 
more than 20 hours a week at each job.
    Among working-age adults, SNAP serves a mix of the unemployed, low-
income workers, and those who are not in the labor force (USDA 2017b 
(https://www.fns.usda.gov/snap/facts-about-snap)). Figure 10 describes 
employment status of ABAWDs. Those receiving SNAP benefits who are in 
the demographic group currently exposed to work requirements--adults 
aged 18-49 with no dependents--generally participate in the labor 
market, with just 25 percent consistently not in the labor force 
(discussed below). While 58 percent worked at least 20 hours per week 
in at least 1 month over 2 years, 25 percent were over the threshold at 
some point but fell below the 20 hour threshold during at least 1 month 
over 2 years. Very few are always working less than 20 hours a week or 
always unemployed (less than two percent in either case), and 14 
percent move across these categories.
    These findings give a markedly different impression than a snapshot 
in time--1 month. When we compare the 1 month (December 2013) against 
24 months (January 2013-December 2014), we find that using 1 month of 
data, more program participants appear to be labor force non-
participants and more appear to meet the work requirement threshold. 
That is, looking only at 1 month of data, an observer would both think 
there is a bigger problem of labor force non-participation in SNAP than 
there really is, and would think that fewer labor force participants 
would lose benefits in a state or county with work requirements.
    There is a meaningful portion of SNAP participants in the labor 
force and working, but not all are working above the monthly work 
requirement threshold consistently. Coglianese's (2018) finding that 
workers who are in and out of the labor force are not more likely to 
take another break later on suggests it is unclear how much more 
consistently work requirements would attach these people to the labor 
force. In our work, too, we find that frequent movement between labor 
status categories over time increases the number of people exposed to 
losing benefits for failing to consistently meet a work requirement and 
decreases the number of people who are entirely out of the labor 
market.
Figure 10. Employment Status in One Month versus Two Years, SNAP 
        Participants 18-49 with No Dependents
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Survey of Income and Program Participation; authors' 
        calculations.

    It is helpful to consider specifically what types of individuals 
would be affected by proposed work requirements and why they are not 
currently working if they are not in the labor force to better 
understand the possible impacts of expanded work requirements. It is 
clear that some people face barriers to working outside the home and as 
such, many work requirements exempt people receiving disability income, 
people with young dependents, or students; but, accurately exempting 
all those who are eligible can be challenging and is likely to result 
in terminating coverage for many people with health conditions or 
caregiving responsibilities that fall outside of states' narrow 
definitions.
    We next examine the reasons ABAWDs gave for not working over the 2 
year period (figure 11). Those in solid green were in the labor force 
but experienced at least one spell of unemployment or labor force 
nonparticipation. Among the labor force participants who were asked why 
they were not working for pay during at least 1 week, we report the 
reason for not working in months they were not working. For 
perspective, the share of the population that worked consistently over 
the 2 years and therefore was never asked why they were not working, 
are shown in the green crosshatch. Those in the blue were out of the 
labor force for the entire 2 year period. Each person is assigned one 
reason--their most frequent reason--for not working.
Figure 11. Most-Frequent Reason for Not Working for Pay, SNAP 
        Participants 18-49 with No Dependents
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Survey of Income and Program Participation; authors' 
        calculations.

    Focusing first on the 25 percent of the SNAP ABAWD population that 
was not in the labor force over the full sample, we find almost 85 
percent reported that the reason that they were not working was poor 
health or disability (this is about 20 percent of all ABAWDs). Another 
quarter of the sample is in stable work. The remaining 50 percent, 
though, were in the labor force at some point, but at other times not 
working. Among that group, more than \1/2\ (28 percent of all ABAWDS) 
reported that a work-related reason, such as not being able to find 
work or being laid off, was their reason for not working for pay.
    As shown in figure 12 below, a substantially larger share of adult 
SNAP participants were not working due to work-related reasons than the 
overall population, even during this time period (2013-14) when the 
economy was on an upswing. More than a quarter of ABAWDs experienced a 
period of not working for pay or nonparticipation due to labor market 
conditions outside their control. This share is 80 percent larger than 
the share of work-related reasons among the overall population. That 
is, even when the economy is improving, SNAP participants may be in 
particularly vulnerable occupations and find themselves frequently 
unable to work due to their local job markets. This is the group that a 
waiver for economic reasons is most directly intended to help, and this 
evidence shows that even when the economy is over 4 years after a 
recession, this group may still be at risk of losing benefits not 
because they do not want to work, but because they are unable to either 
find a job or get the requisite number of hours.
Figure 12. Share Not Working for Pay for Work-Related Reasons Overall 
        versus SNAP, by Demographic Characteristics
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Survey of Income and Program Participation; authors' 
        calculations.

    This evidence presented thus far shows that those who are most at 
risk to losing benefits under the proposed rules are workers 
experiencing normal labor market fluctuations and those who should be 
eligible for exemptions but often fail to receive them. Among 
persistent labor force non-participants, we find that health issues are 
the predominant reason given for not working even though the analysis 
excludes program participants who reported disability income because 
they would be eligible for a categorical exemption from a work 
requirement. This group would also lose SNAP benefits if work 
requirement waivers were removed.
    Some have questioned whether survey respondents are likely to 
provide accurate information about their health. This criticism stems 
from social desirability bias; survey respondents might feel pressure 
to report a more publicly acceptable reason for not working than what 
might actually be true. In this case, a respondent who simply does not 
want to work would say that they are not working because of a health 
condition; a health problem is a socially acceptable reason for not 
working, but the real reason is not.
    In this analysis, we show that those reporting health as a reason 
for not working do appear to be in poor health. We investigate the 
prevalence of reported health conditions among ABAWD SNAP 
participants.\12\
---------------------------------------------------------------------------
    \12\ Those who were not working due to health or disability 
reported that they were not working for pay because they were unable to 
work because of chronic health condition or disability, temporarily 
unable to work due to injury, or temporarily unable to work due to 
illness. Those in the stable work category did not experience a period 
of unemployment or nonparticipation over the 2 year period. Those in 
the period of unemployment or nonparticipation group were at least once 
not working for pay during the 2 year period. Labor force non-
participants did not work for pay at all during the 2 year period. 
Those in the labor force non-participant due to health group did not 
work for pay at all during the 2 year period and the most frequent 
reason given for their nonparticipation was health.
---------------------------------------------------------------------------
    Using the information from the prior analyses, we divide the SNAP 
participants into five groups:

   Stable work--those who worked consistently for 2 years;

   Transitioned in and out of work due to health--those who 
        were in the labor force but experienced a period of 
        unemployment or nonparticipation due to a health condition or 
        disability;

   Transitioned in and out of work, other--those who were in 
        the labor force but experienced a period of unemployment or 
        nonparticipation for a reason other than health or disability;

   Labor force non-participant due to health--those who did not 
        work at all for 2 years due to a health condition or 
        disability; and,

   Labor force nonparticipation, other--those who did not work 
        at all for 2 years for a reason other than health or 
        disability.
Figure 13. Health Characteristics of ABAWDs, by Employment Status

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: Survey of Income and Program Participation; authors' 
        calculations.

    We look at whether SNAP participants who would be exposed to work 
requirements are in self-reported fair or poor health, take a 
prescription medication daily, respond affirmatively to at least one in 
a battery of questions about disability, or spent more than 30 days 
over a 2 year period in bed due to ill health.\13\ These questions 
about health are self-reported, but are considerably less subject to 
the social desirability bias that may affect how a respondent answers 
the question as to why they are not working. In fact, these questions 
are asked in the survey long before the respondent is asked about their 
labor force status, reducing the likelihood they are manipulating their 
response to justify not working.
---------------------------------------------------------------------------
    \13\ Those in self-reported poor health responded ``poor'' to the 
question ``what is your health status?'' Those in the daily 
prescription medication group responded affirmatively to the question 
``Did you take prescription medication on a daily basis?'' Those in the 
any disability responded affirmatively to at least one of the following 
questions: Do you have serious difficulty walking or climbing stairs; 
do you have difficulty dressing or bathing; do you have serious 
difficulty concentrating, remembering, or making decisions; do you have 
a serious physical or mental condition or a developmental delay that 
limits ordinary activity; do you have difficulty doing errands alone; 
do you have difficulty finding a job or remaining employed; are you 
prevented from working; are you deaf or do you have serious difficulty 
hearing; are you blind or do you have serious difficulty seeing? Those 
who spent more than 30 days in bed responded to the question ``How many 
days did illness or injury keep you in bed more than half of the day'' 
for at least 30 days over the 2 year period.
---------------------------------------------------------------------------
    Ninety-nine percent of ABAWD labor force non-participants who 
reported the reason for their nonparticipation was due to health in 
fact reported health problems; 91 percent reported a disability, 86 
percent reported taking medication daily, 82 percent reported being in 
self-reported fair or poor health, and 39 percent reported spending 
more than 30 days in bed. For those labor force non-participants 
reporting a different reason for their nonparticipation, three in five 
reported a health problem. More than \1/3\ reported a disability, 
almost \1/2\ took daily medication, and 15 percent spent more than 30 
days in bed. Among those who were labor force participants but 
experienced a period of unemployment or nonparticipation due to health, 
nine out of ten reported a health condition. About seven in ten 
reported a disability and taking a daily prescription, about 60 percent 
were in self-reported fair or poor health, and a quarter spent more 
than 20 days in bed.
    The prevalence of health conditions among ABAWD labor force non-
participants as well as labor force participants working unstably due 
to health contrasts with those working stably. But to be clear, even 
among this group, a quarter report a disability, 44 percent are taking 
a daily prescription medication, \1/5\ are in self-reported fair or 
poor health, and six percent spent substantial time in bed.
    Those who are SNAP participants with health issues who are unable 
to work and who would be exposed to work requirements would be required 
to obtain documents verifying their health problems frequently in order 
to retain an exemption. These people could lose access to the program 
due to paperwork requirements unless administrative capacity were 
expanded greatly to monitor and adjudicate these health concerns. Even 
then, administrative failures could lead to loss of access to food 
benefits.
    There may be some SNAP participants who might join the labor force 
if they were threatened with the loss of benefits. Recent evidence 
shows that this group is very small relative to those who would be 
improperly sanctioned by work requirements who are already working or 
are legitimately unable to work. This evidence is directly relevant to 
claims in the NPRM and RIA that exposing more areas to work 
requirements would increase self-sufficiency. The USDA has failed to 
provide evidence that this would be the case, and the evidence produced 
in this section make it clear that work requirement would harm labor 
force participants who experience market volatility and labor force 
non-participants, the vast majority of whom have a health condition.
V. Conclusion
    Executive Order 12866 states that agencies, such as USDA, may issue 
regulations when there is a compelling public need and when the 
benefits outweigh the costs in such a way as to maximize net benefits. 
We find that both the NPRM and its RIA insufficiently analyze the 
proposed rule and fail to consider the costs and benefits under 
alternate economic conditions or to the participants in any 
circumstance. In this comment, we have provided evidence and analysis 
that the USDA has proposed a rule that is arbitrary, that the rule runs 
counter to the compelling public need for waivers to work requirements 
during economic downturns, and fails to consider much less prove that 
the benefits outweigh the costs. The existing rule should be sustained.
VI. References

 
 
 
    Aaronson, Stephanie R., Mary C. Daly, William Wascher, and David W.
 Wilcox. 2019. ``Okun Revisited: Who Benefits Most From a Strong
 Economy?'' Brookings Papers on Economic Activity, Conference Draft.
    Abraham, Katharine G., and Melissa S. Kearney. 2018. ``Explaining
 the Decline in the U.S. Employment-to-Population Ratio: A Review of the
 Evidence.'' Working Paper 24333, National Bureau of Economic Research,
 Cambridge, MA.
    Almond, Douglas, Hilary W. Hoynes, and Diane Whitmore Schanzenbach.
 2011. ``Inside the War on Poverty: The Impact of Food Stamps on Birth
 Outcomes.'' The Review of Economics and Statistics 92 (2): 387-403.
    American Recovery and Reinvestment Act of 2009 (ARRA), Pub. L. 111-5
 (2009).
    American Taxpayer Relief Act of 2012, Pub. L. (2012).
    Auerbach, Alan J., William G. Gale, and Benjamin H. Harris. 2010,
 Fall. ``Activist Fiscal Policy.'' Journal of Economic Perspectives 24
 (4): 141-63.
    Auerbach, Alan J., and Yuriy Gorodnichenko. 2012. ``Measuring the
 Output Responses to Fiscal Policy.'' American Economic Journal:
 Economic Policy 4 (2): 1-27.
    Bauer, Lauren. 2018. ``Workers could lose SNAP benefits under
 Trump's proposed rule.'' The Hamilton Project, Brookings Institution,
 Washington, D.C.
    Bauer, Lauren, and Diane Whitmore Schanzenbach. 2018a, July 27.
 ``Employment Status Changes Put Millions at Risk of Losing SNAP
 Benefits for Years.'' Blog. The Hamilton Project, Brookings
 Institution, Washington, D.C.
    ___. 2018b, August 9. ``Who Loses SNAP Benefits If Additional Work
 Requirements Are Imposed? Workers.'' Blog. The Hamilton Project,
 Brookings Institution, Washington, D.C.
    Bauer, Lauren, Diane Whitmore Schanzenbach, and Jay Shambaugh. 2018,
 October 15. ``Work requirements and safety net programs.'' Report. The
 Hamilton Project, Brookings Institution, Washington, D.C.
    Bitler, Marianne, and Hilary Hoynes. 2010, Fall. ``The State of the
 Social Safety Net in the Post-Welfare Reform Era [with Comments and
 Discussion].'' Brookings Papers on Economic Activity. Brookings
 Institution, Washington, D.C.
    Blinder, Alan S., and Mark Zandi. 2015. ``The Financial Crisis:
 Lessons for the Next One.'' Center on Budget and Policy Priorities,
 Washington, D.C.
    Bolen, Ed, Lexin Cai, Stacy Dean, Brynne Keith-Jennings, Catlin
 Nchako, Dottie Rosenbaum, and Elizabeth Wolkomir. 2018. ``House Farm
 Bill Would Increase Food Insecurity and Hardship.'' Center on Budget
 and Policy Priorities, Washington, D.C.
    Coglianese, John. 2018, February. ``The Rise of In-and-Outs:
 Declining Labor Force Participation of Prime Age Men.'' Working Paper,
 Harvard University, Cambridge, MA.
    Collins, Ann M., Ronette Briefel, Jacob Alex Klerman, Anne Wolf,
 Gretchen Rowe, Ayesha Enver, Christopher W. Logan, Syeda Fatima, Marina
 Komarovsky, Julia Lyskawa, and Stephen Bell. 2014. ``Summer Electronic
 Benefits Transfer for Children (SEBTC) Demonstration: 2013 Final
 Report.'' Nutrition Assistance Program Report, Food and Nutrition
 Service, Office of Policy Support, U.S. Department of Agriculture,
 Washington, D.C.
    ___. 2013. ``Summer Electronic Benefits Transfer for Children
 (SEBTC) Demonstration: 2012 Final Report.'' Nutrition Assistance
 Program Report, Food and Nutrition Service, Office of Policy Support,
 U.S. Department of Agriculture, Washington, D.C.
    Fazzari, Steven M., James Morley, and Irina B. Panovska. 2014.
 ``State-Dependent Effects of Fiscal Policy.'' Research Paper No. 2012-
 27C, Social Science Research Network, Rochester, NY.
    Fraker, Thomas, and Robert Moffitt. 1988. ``The effect of food
 stamps on labor supply: A bivariate selection model.'' Journal of
 Public Economics 35 (1): 25-56.
    Ganong, Peter, and Jeffrey B. Liebman. 2018. ``SNAP Enrollment:
 Disentangling Business Cycle Fluctuations and Policy Changes.''
 American Economic Journal: Economic Policy 10 (4): 153-76.
    Harris, Timothy F. 2019. ``Do SNAP Work Requirements Work?'' Upjohn
 Institute working paper 19-297. W.E. Upjohn Institute for Employment
 Research, Kalamazoo, MI.
    Hinrichs, Peter. 2010. ``The Effects of the National School Lunch
 Program on Education and Health.'' Journal of Policy Analysis and
 Management 29 (3): 479-505.
    Hoynes, Hilary W., Diane Whitmore Schanzenbach, and Douglas Almond.
 2016. ``Long Run Impacts of Childhood Access to the Safety Net.''
 American Economic Review 106 (4): 903-34.
    Hoynes, Hilary Williamson, and Schanzenbach, Diane Whitmore. 2012.
 ``Work Incentives and the Food Stamp Program.'' Journal of Public
 Economics 96 (1): 151-62.
    Joint Economic Committee. 2018. ``Inactive, Disconnected, and
 Ailing: A Portrait of Prime-Age Men Out of the Labor Force.'' SCP
 Report 2-18, Joint Economic Committee, Washington, D.C.
    Juhn, Chinhui. 1992. ``Decline of Male Labor Market Participation:
 The Role of Declining Market Opportunities.'' The Quarterly Journal of
 Economics 107 (1): 79-121.
    Keith-Jennings, Brynne, and Dottie Rosenbaum. 2015. ``SNAP Benefit
 Boost in 2009 Recovery Act Provided Economic Stimulus and Reduced
 Hardship.'' Center on Budget and Policy Priorities, Washington, D.C.
    Kreider, Brent, John V. Pepper, Craig Gundersen, and Dean Jolliffe.
 2012. ``Identifying the Effects of SNAP (Food Stamps) on Child Health
 Outcomes When Participation Is Endogenous and Misreported.'' Journal of
 the American Statistical Association 107 (499): 958-975.
    Mabli, James, Jim Ohls, Lisa Dragoset, Laura Castner, and Betsy
 Santos. 2013. ``Measuring the Effect of Supplemental Nutrition
 Assistance Program (SNAP) Participation on Food Security.'' Prepared by
 Mathematica Policy Research for the U.S. Department of Agriculture,
 Food and Nutrition Service.
    Nord, Mark and Mark Prell. 2011. ``Food Security Improved Following
 the 2009 ARRA Increase In SNAP Benefits.'' Economic Research Service:
 Economic Research Report 116.
    North Carolina Department of Health and Human Services. 2008. DSS
 Administrative Letter No. Economic Family Services 7-2008, Continued
 and Modified 2007 Regional ABAWD Two-Year Waiver. https://
 www2.ncdhhs.gov/info/olm/manuals/dss/ei-30/adm/EFS_FNS_AL-7-2008.pdf.
    Ohio Job and Family Services. 2008. Office of Family Stability
 Letter #71. http://jfs.ohio.gov/ofam/pdf/OFS-Letter--71-ABAWDS1.pdf.
    Parker, Jonathan A., Nicholas S. Souleles, David S. Johnson, and
 Robert McClelland. 2013. ``Consumer Spending and the Economic Stimulus
 Payments of 2008.'' American Economic Review 103 (6): 2530-53.
    Pennsylvania Department of Human Services. 2008. Operations
 Memorandum--Food Stamps OPS080301. http://services.dpw.state.pa.us/
 oimpolicymanuals/snap/OPS080301.pdf.
    Ramey, Valerie A. and Sarah Zubairy, 2014. ``Government Spending
 Multipliers in Good Times and in Bad: Evidence from U.S. Historical
 Data.'' Working Paper #20719, National Bureau of Economic Research,
 Cambridge, MA.
    Ratcliffe, Caroline, Signe-Mary McKernan, and Sisi Zhang. 2011.
 ``How Much Does the Supplemental Nutrition Assistance Program Reduce
 Food Insecurity?'' American Journal of Agricultural Economics 93 (4):
 1082-98.
    Rosenbaum, Dorothy. 2013. ``The Relationship between SNAP and Work
 among Low-Income Households.'' Center on Budget and Policy Priorities,
 Washington, D.C.
    Rothstein, Jesse. 2011. ``Unemployment Insurance and Job Search in
 the Great Recession.'' Brookings Papers on Economic Activity,
 Washington, D.C.
    Russek, Frank and Kim Kowalewski. 2015. ``How CBO Estimates
 Automatic Stabilizers.'' Working Paper. Congressional Budget Office.
    Schanzenbach, Diane Whitmore, Lauren Bauer, and Greg Nantz. 2016,
 April 21. ``Twelve facts about food insecurity and SNAP.'' Report. The
 Hamilton Project, Brookings Institution, Washington, D.C.
    Schmidt, Lucie, Lara Shore Sheppard, and Tara Watson. 2016. ``The
 Effect of Safety-Net Programs on Food Insecurity.'' Journal of Human
 Resources, University of Wisconsin Press 51 (3): 589-614.
    Shaefer, H. Luke, and Italo A. Gutierrez. 2013. ``The Supplemental
 Nutrition Assistance Program and Material Hardships Among Low-Income
 Households with Children.'' Social Service Review 87 (4): 753-79.
    Smith, Travis A., and Pourya Valizadeh. 2018. ``Aging out of WIC and
 Child Nutrition: Evidence from a Regression Discontinuity Design.''
 Unpublished Working Paper.
    Tax Relief, Unemployment Insurance Reauthorization, and Job Creation
 Act of 2010, Pub. L. (2010).U.S. Department of Agriculture (USDA).
 2009. ``SNAP--ABAWD Statewide Waivers--New Criteria for Unemployment
 Insurance Extended Benefits Trigger.'' https://fns-prod.azureedge.net/
 sites/default/files/snap/ABAWD%20Statewide%20Waivers.pdf.
    ___ 2016. ``Supplemental Nutrition Assistance Program--Guide to
 Supporting Requests to Waive the Time Limit for Able-Bodied Adults
 without Dependents (ABAWD.)'' https://fns-prod.azureedge.net/sites/
 default/files/snap/SNAP-Guide-to-Supporting-Requests-to-Waive-the-Time-
 Limit-for-ABAWDs.pdf.
    ___ 2017a. ``Characteristics of USDA Supplemental Nutrition
 Assistance Program Households: Fiscal Year 2016 (Summary).'' U.S.
 Department of Agriculture, Washington, D.C.
    ___ 2017b. ``Supplemental Nutrition Assistance Program (SNAP): Facts
 About SNAP.'' U.S. Department of Agriculture, Washington, D.C.
    U.S. Department of Labor (DOL). N.d. ``Expiration of Unemployment
 Insurance Programs.'' U.S. Department of Labor, Washington, D.C. https:/
 /oui.doleta.gov/unemploy/docs/supp_act_eb-euc-expired.pdf.
    ___ N.d. Trigger Notice No. 2004-1--2019-9. State Extended Benefit
 (EB) Indicators. https://oui.doleta.gov/unemploy/claims_arch.asp.
    ___ N.d. Local Area Unemployment Statistics. https://www.bls.gov/lau/
 .
    Wandner, Stephen A. 2018. ``Unemployment Insurance Reform: Fixing a
 Broken System.'' W.E. Upjohn Institute for Employment Research,
 Kalamazoo, MI.
    Whalen, Charles J., and Felix Reichling. 2015, February. ``The
 Fiscal Multiplier and Economic Policy Analysis in the United States.''
 Working Paper 2015-02, Working Paper Series, Congressional Budget
 Office, U.S. Congress, Washington, D.C.
    Wolkomir, Elizabeth and Lexin Cai. 2018. ``The Supplemental
 Nutrition Assistance Program Includes Earnings Incentives.'' Center on
 Budget and Policy Priorities, Washington, D.C.
    Ziliak, James P., Craig Gundersen, and David N. Figlio. 2003. ``Food
 stamp caseloads over the business cycle.'' Southern Economic Journal 69
 (4): 903-19.
 

VII. Appendix

      Appendix Table 1. Interactions between Emergency Unemployment
                Compensation and SNAP Waiver Eligibility
------------------------------------------------------------------------
                EUC threshold for
  Date range       SNAP  Waiver       Tier eligibility        Source
                   eligibility
------------------------------------------------------------------------
January 8,     Tier II              3 month seasonally   The Bush
 2009-Novembe                        adjusted total       Administration
 r 5, 2009                           unemployment rate    clarified \2\
                                     (TUR) of at least    that EUC
                                     six percent; or 13   counted for
                                     week insured         SNAP waivers
                                     unemployment rate    on January 8,
                                     (IUR) of at least    2009. Any
                                     4.0 percent (CRS     state that was
                                     2014 \1\)            eligible for
                                                          Tier II EUC
                                                          was eligible
                                                          for SNAP
                                                          waivers based
                                                          on EUC
                                                          eligibility.
                                                          From January
                                                          9, 2009 to
                                                          November 6,
                                                          2009,
                                                          eligibility
                                                          for Tier II
                                                          was
                                                          conditional on
                                                          having a TUR
                                                          of at least
                                                          six percent or
                                                          an IUR of at
                                                          least four
                                                          percent. Tier
                                                          II was not
                                                          universal
                                                          among states
                                                          before
                                                          November 6,
                                                          2009 (Table 1
                                                          in Rothstein
                                                          2011 \3\).
November 6,    Tier III             3 month seasonally   When all states
 2009-May 31,                        adjusted TUR of at   were eligible
 2012                                least six percent;   for Tier II
                                     or 13 week IUR of    benefits,
                                     at least 4.0         states had to
                                     percent (CRS 2014    additionally
                                     \1\)                 qualify for
                                                          Tier III
                                                          benefits in
                                                          order to be
                                                          eligible for a
                                                          SNAP waiver
                                                          application
                                                          (CBPP \4\
                                                          2018). State
                                                          eligibility
                                                          for EUC tier
                                                          II became
                                                          unconditional
                                                          on November 6,
                                                          2009
                                                          (Rothstein
                                                          2011 \3\).
June 1, 2012-  Tier II              3 month seasonally   On June 1, 2012
 Dec. 31 2013                        adjusted TUR of at   \5\ Tier II
                                     least six percent    qualifications
                                     (CRS 2014 \1\)       go back to a 3
                                                          month
                                                          seasonally
                                                          adjusted TUR
                                                          of at least
                                                          six percent
                                                          and therefore
                                                          Tier II is no
                                                          longer a
                                                          universal
                                                          tier.
                                                          According to
                                                          DOL, as of
                                                          January 12,
                                                          2014 EB is not
                                                          currently
                                                          available in
                                                          any state (DOL
                                                          \6\).\14\
\14\ Because
 we round to
 the nearest
 month, we
 end the EUC
 eligibility
 period in
 December
 2013.
 Waivers
 based on EUC
 were granted
 through
 2016.
------------------------------------------------------------------------
\1\ https://fas.org/sgp/crs/misc/R42444.pdf.
\2\ https://fns-prod.azureedge.net/sites/default/files/snap/
  ABAWD%20Statewide%20 Waivers.pdf.
\3\ https://www.brookings.edu/wp-content/uploads/2011/09/
  2011b_bpea_rothstein.pdf.
\4\ https://www.cbpp.org/sites/default/files/atoms/files/3-24-17fa.pdf.
\5\ https://en.wikipedia.org/wiki/Unemployment_extension.
\6\ https://oui.doleta.gov/unemploy/docs/supp_act_eb-euc-expired.pdf.

                              attachment 2
                              
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Economic Analysis D October 2018
Work Requirements and Safety Net Programs *
---------------------------------------------------------------------------
    * Lauren Bauer, The Hamilton Project and the Brookings Institution; 
Diane Whitmore Schanzenbach, The Hamilton Project, the Brookings 
Institution, and Northwestern University; and Jay Shambaugh, The 
Hamilton Project, the Brookings Institution, and The George Washington 
University.

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

---------------------------------------------------------------------------
    Acknowledgments

          We thank reviewers John Coglianese, Jason Furman, Heather 
        Hahn, Kriston McIntosh, Ryan Nunn, and the Center on Budget and 
        Policy Priorities. We also thank the following who provided 
        research assistance: Patrick Liu, Jimmy O'Donnell, Jana 
        Parsons, Becca Portman, and Areeb Siddiqui.

    Mission Statement

          The Hamilton Project seeks to advance America's promise of 
        opportunity, prosperity, and growth.
          We believe that today's increasingly competitive global 
        economy demands public policy ideas commensurate with the 
        challenges of the 21st Century. The Project's economic strategy 
        reflects a judgment that long-term prosperity is best achieved 
        by fostering economic growth and broad participation in that 
        growth, by enhancing individual economic security, and by 
        embracing a role for effective government in making needed 
        public investments.
          Our strategy calls for combining public investment, a secure 
        social safety net, and fiscal discipline. In that framework, 
        the Project puts forward innovative proposals from leading 
        economic thinkers--based on credible evidence and experience, 
        not ideology or doctrine--to introduce new and effective policy 
        options into the national debate.
          The Project is named after Alexander Hamilton, the nation's 
        first Treasury Secretary, who laid the foundation for the 
        modern American economy. Hamilton stood for sound fiscal 
        policy, believed that broad-based opportunity for advancement 
        would drive American economic growth, and recognized that 
        ``prudent aids and encouragements on the part of government'' 
        are necessary to enhance and guide market forces. The guiding 
        principles of the Project remain consistent with these views.
Abstract
    Basic assistance programs such as the Supplemental Nutrition 
Assistance Program (SNAP, formerly the Food Stamp Program) and Medicaid 
ensure families have access to food and medical care when they are low-
income. Some policymakers at the Federal and state levels intend to add 
new work requirements to SNAP and Medicaid. In this paper, we analyze 
those who would be impacted by an expansion of work requirements in 
SNAP and an introduction of work requirements into Medicaid. We 
characterize the types of individuals who would face work requirements, 
describe their labor force experience over 24 consecutive months, and 
identify the reasons why they are not working if they experience a 
period of unemployment or labor force nonparticipation. We find that 
the majority of SNAP and Medicaid participants who would be exposed to 
work requirements are attached to the labor force, but that a 
substantial share would fail to consistently meet a 20 hours per week-
threshold. Among persistent labor force non-participants, health issues 
are the predominant reason given for not working. There may be some 
subset of SNAP and Medicaid participants who could work, are not 
working, and might work if they were threatened with the loss of 
benefits. This paper adds evidence to a growing body of research that 
shows that this group is very small relative to those who would be 
sanctioned under the proposed policies who are already working or are 
legitimately unable to work.
Introduction
    Basic assistance programs such as the Supplemental Nutrition 
Assistance Program (SNAP, formerly the Food Stamp Program) and Medicaid 
ensure families have access to food and medical care when they are low-
income. These programs lift millions out of poverty while reducing food 
insecurity and increasing access to medical care. They also support 
work, and increase health and economic security among families in the 
short term as well as economic self-sufficiency in the long-term.
    Today, some policymakers at the Federal and state levels intend to 
add new work requirements in order for beneficiaries to receive SNAP 
benefits and participate in the Medicaid health insurance program. In 
general, those exposed to a work requirement would be required to prove 
that they are working or participating in a training program for at 
least 20 hours per week each month. Failure to prove that they have met 
the work requirement or are eligible for an exemption would mean that a 
program participant would lose food assistance benefits or health 
insurance for a time, or until they met the standard.
    Work requirements are meant to force work-ready individuals to 
increase their work effort and maintain that work effort every month by 
threatening to withhold and subsequently withholding food assistance or 
health coverage if a person is not working a set number of hours. The 
strategy presumes that the reasons that many low-income individuals are 
not working or meeting an hourly threshold every month is either due to 
their own lack of effort or to work disincentives theoretically 
inherent to means-tested programs. It is clear that some people face 
barriers to working outside the home and as such, many work 
requirements exempt people that receive disability income, people with 
young dependents, or students; but, accurately exempting all those who 
are eligible can be challenging and is likely to result in terminating 
coverage for many people with health conditions or caregiving 
responsibilities that fall outside of states' narrow definitions. 
Proponents of work requirements would ideally only like to sanction 
individuals who are able to work, but choose not to. But in practice 
strict enforcement of proposed work requirements will sanction many 
groups, including: those who are unable to work, those who are able to 
work but who do not find work, those who are working but not 
consistently above an hourly threshold, and those who are meeting work 
or exemption requirements but fail to provide proper documentation. 
Evidence suggests that the vast majority of those exposed to proposed 
work requirements for SNAP and Medicaid fall into these groups.
    In this paper, we analyze those who would be impacted by an 
expansion of work requirements in SNAP and an introduction of work 
requirements into Medicaid. Our principal contribution is to 
characterize the types of individuals who would face work requirements, 
describe what their work experiences are over a 2 year period, and 
identify the reasons why they are not working if they experience a 
period of unemployment or labor force nonparticipation. We find that 
most of those who fail the new work requirements are either those who 
are in the labor force already but who experience unstable employment, 
or those who might be eligible for hardship exemptions, such as those 
with health problems who are not already receiving disability income. 
The compositional and labor market analyses reported below suggest that 
the proposed work requirements will put at risk access to food 
assistance and health care for millions who are working, trying to 
work, or face barriers to working.
    Adding explicit work requirements to assistance programs must be 
analyzed in the context of program goals and from many angles. Who 
would be impacted by an expansion of work requirements? What are the 
administrative costs and challenges of managing the work requirements? 
How do the requirements interact with the realities of the low-wage 
work experience? And how would the requirements impact the health and 
economic benefits to program participation? For example, removing 
Medicaid coverage may have little positive work-incentive effect for 
the currently healthy but may undermine public health goals and reduce 
the labor supply of those who do encounter health problems and have 
lost their coverage. Removing SNAP benefits from working-age adults may 
impact resources available not just to them, but also to any seniors 
and dependents in the household. Finally, tight work requirements can 
undermine the automatic stabilizer aspect of these programs. Instead of 
SNAP expanding as the unemployment rate rises, the work requirements 
would cause the program to contract, resulting in more people losing 
benefits when work becomes difficult for them to find.
    There may be some subset of individuals who could work, are not 
working, and might work if they were threatened with the loss of 
benefits. This paper adds evidence to a growing body of research that 
shows that this group is very small relative to those who would be 
sanctioned under the proposed policies who are already working or are 
legitimately unable to work (Bauer and Schanzenbach 2018a, 2018b; 
Garfield, et al., 2018; Goldman, et al., 2018).
    The goals of safety net programs are to provide insurance 
protection to those who are experiencing poor economic outcomes and to 
support those who are trying to improve their situation. Our analysis 
suggests that work requirements will harm more individuals and families 
than they would help the small share who might increase their labor 
supply.
SNAP, Medicaid, and Incentives to Work
    The social safety net is intended to provide insurance against bad 
outcomes. But, for means-tested benefit programs, economic theory 
suggests it may reduce the incentive to work because (1) individuals 
are only eligible for a program when their income remains below a given 
threshold and (2) participants stand to lose benefits as income 
increases or reaches the eligibility threshold. In addition, any time 
someone receives unearned income of sufficient size, it may 
theoretically reduce the amount of work that an individual wants to 
supply to the market. In some cases, worries about work disincentives 
have led to the implementation of time limits or work requirements for 
a set of individuals as a condition for program eligibility.
    Such work requirements can undermine the insurance value of the 
programs, though, if people who are not working either cannot work due 
to individual limitations or are unable to find steady work due to 
economic fluctuations. Evaluating whether work requirements are an 
appropriate policy lever--as opposed to addressing work disincentives 
through other means--thus depends on the goals of the program overall, 
the characteristics of the target population, the design of the work 
requirements, the cost of administering the program, the likelihood of 
erroneously limiting access, and the strength of the incentive effects.
    Work requirement policies often have difficulty distinguishing 
between those who are able to work and those who are unable to work, 
because both groups can be hard to observe and verify. As a result, 
strict enforcement of work requirements will sanction those who are 
unable to work, as well as those who could work but do not obtain 
employment in response to the requirements. They may also sanction some 
who are able to work but who are not able to find work, as well as 
those who are working but fail to provide proper documentation.
    In order to evaluate whether a work requirement is in keeping with 
the purpose of a means-tested program, there are a number of dimensions 
by which a proposal should be evaluated. One would want to exempt those 
whom society does not feel should be forced to work, accommodate 
changes in the business cycle that make work more difficult to find, 
and have a system of verification and exemption that does not raise 
barriers to entry or remove program participants who should maintain 
access. But, one would have to ensure that work requirements do not 
punish those who cannot obtain a job due to economic conditions in 
their area, penalize those who are actually working but have 
temporarily lost hours, limit access to programs for an extended period 
of time after failing a work requirement, or, compromise the insurance 
goals of the program in question. These parameters can be quite 
difficult to meet and they set the criterion by which policymakers can 
determine whether work requirements are inappropriate for the program 
in question.
    There is an extensive literature on whether work requirements can 
in fact push people into the labor force, principally studying the 
impacts of the 1996 Temporary Assistance for Needy Families (TANF) 
reform (see Blank 2002 and Ziliak 2016 for reviews). The labor supply 
of the TANF population did in fact rise, but this took place amidst a 
strong economy and support from the Earned Income Tax Cred[i]t (EITC) 
expansion as well (Schanzenbach 2018). For example, Fang and Keane 
(2004) find that while work requirements were the most important factor 
driving the decline in participation in welfare programs, the EITC 
expansion and macroeconomic factors were more important in driving the 
increase in work participation (they find work requirements had a 
positive impact as well, but the contribution was smaller). Work 
requirements often come with a variety of supports and involve 
different enforcement mechanisms and levels of stringency. See 
Hamilton, et al., (2001) for a detailed review as part of the National 
Evaluation of Welfare-to-Work Strategies. Many of the work requirement 
programs that have generated positive results also had substantial 
education and skills training components (Pavetti and Schott 2016). 
Other studies, such as Meyer and Rosenbaum (2001) and Grogger (2004) 
suggest a smaller or negligible role for the TANF reforms compared with 
other factors, especially the EITC expansion.
    In this analysis, we focus more on the people who would be impacted 
by new work requirements and the reasons why they are not working, as 
opposed to the question of the labor supply response. Given the extent 
to which the labor market conditions--in particular for potentially 
impacted populations--are different than those in the 1990s (Black, 
Schanzenbach, Breitwieser 2017; Butcher and Schanzenbach 2018), it is 
helpful to consider specifically what types of individuals would be 
affected by proposed work requirements and why they are not currently 
working to better understand the possible impacts of expanded work 
requirements. In this section we describe the SNAP and Medicaid 
programs, the structure of their work incentives, and evidence of the 
programs' incentive effects on labor supply.
SNAP
    Since the 1960s SNAP has provided resources to purchase food for 
millions of low-income households. The goal of the program is to 
provide beneficiaries with resources to raise their food purchasing 
power and, as a result, improve their health and nutrition. Households 
are eligible for SNAP if they meet an asset and income threshold, or if 
they receive assistance from programs like Supplemental Security 
Income. SNAP benefit levels are targeted based on a given household's 
income and expenses.
    SNAP currently addresses work disincentives in a variety of ways. 
Similar to the EITC, SNAP addresses work disincentives through an 
earnings disregard of 20 percent and a gradual benefit reduction 
schedule. This means that the size of the earnings disregard increases 
as income increases and that those with earned income receive larger 
SNAP benefits than those with no earned income (Wolkomir and Cai 2018). 
When a person moves from being a labor force non-participant to working 
while on SNAP, total household resources will increase; as a 
beneficiary's earnings approach the eligibility threshold, total 
household resources continue to increase. The combination of the 
earnings disregard and a gradual phase-out schedule--that states have 
the option to further extend and smooth--ameliorate but do not 
eliminate work disincentives.
    States have had the option to impose work requirements on certain 
beneficiaries since the 1980s. Most SNAP participants between the ages 
of 18 and 59 without dependents under 6 are required to register for 
work, accept a job if one is offered to them, and not reduce their work 
effort. States are required to operate an employment and training 
program, and may require some SNAP recipients to participate or suffer 
sanctions. See Rosenbaum (2013) and Bolen, et al., (2018) for a 
detailed description of SNAP work requirements. After 1996, SNAP work 
requirements and benefit time limits were imposed on individuals aged 
18-49 without dependents under the age of 18, requiring them to 
register for work and accept a job if one is offered to them. If they 
work or participate in a training program for at least 20 hours per 
week, they can maintain access to the program. This population is 
allowed to receive 3 months of benefits out of 36 months if they do not 
work or participate in a training program. States are permitted to 
exempt a share of individuals and apply to the U.S. Department of 
Agriculture (USDA) for a waiver to the time limit provisions, an 
essential capacity for SNAP's function as an automatic stabilizer. 
Studies show that when SNAP payments increase to a local area in 
response to an economic downturn, they serve as an effective fiscal 
stimulus to the local area (Blinder and Zandi 2015; Keith-Jennings and 
Rosenbaum 2015). Among other changes, the proposed work requirements 
would make these regional waivers more difficult to obtain.
    SNAP improves health and economic outcomes in both the near and 
long terms (see Hoynes and Schanzenbach 2016 for a review), but had a 
negative effect on employment in the past. During the Food Stamp 
Program's introduction in the 1960s and 1970s, reductions in employment 
and hours worked were observed, particularly among female-headed 
households (Hoynes and Schanzenbach 2012). Whether work requirements 
could offset this disincentive would depend on their targeting and 
whether those who are not working could readily increase their labor 
supply.
Medicaid
    Since 1965, the Medicaid program has been administered in 
partnership between Federal and state governments to provide medical 
assistance to eligible individuals. The core goal of the program is to 
provide health services and to cover health-care costs in order to 
improve health. Under the Patient Protection and Affordable Care Act 
(ACA), the eligible population expanded to include low-income adults 
under the age of 65 who previously did not qualify.
    Although some SNAP beneficiaries have been subject to work 
requirements since the 1980s, Medicaid work requirements are being 
rolled out for the first time in certain states. The ACA does not allow 
work requirements to be imposed as a condition for program 
participation in Medicaid, but states may apply for a waiver under 
Section 1115 of the Social Security Act to introduce work requirements 
if the Department of Health and Human Services determines doing so 
advances program objectives. Though the Obama Administration and the 
U.S. District Court for the District of Columbia (which rejected 
Kentucky's proposal for work requirements in Medicaid) did not view 
work requirements as supporting core program goals, the Trump 
Administration has expressed its conviction that work requirements are 
allowable (Centers for Medicare & Medicaid Services 2018; Garfield, 
Rudowitz, and Damico 2018; Stewart v. Azar).
    In the case of Medicaid, there are societal costs to taking health 
insurance away from an otherwise eligible person due to work 
requirements. For example, since there are rules requiring hospitals to 
provide medical care to those experiencing life-threatening emergencies 
regardless of the individual's ability to pay, those without insurance 
will in many cases seek and receive treatment in ways that are more 
expensive for society (Institute of Medicine 2003). Second, care 
delivered via insurance may include preventive care, check-ups, and 
other care that is more efficient than delaying care until a medical 
problem becomes severe enough to be treated in an emergency room. Thus, 
denying insurance may not reduce costs for society. Finally, evidence 
suggests that health insurance is valued by participants at less than 
its cost, making proposed work requirements less effective at raising 
employment (Finkelstein, Hendren, and Luttmer 2015).

                                 Box 1.
------------------------------------------------------------------------
 
-------------------------------------------------------------------------
              Trends in Prime-Age Labor Force Participation
 
    For a number of decades labor force participation in the United
 States rose. This was especially true for prime-age (25-54) workers,
 whose participation rose from 65 percent in the middle of the 20th
 century to a peak of 84 percent in 1999. This persistent trend obscured
 an offsetting force: Prime-age men were steadily working less while
 prime-age women were working more. In 1949 97 percent of prime-age men
 were in the labor force, but only 36 percent of women were. By 1999
 those figures were 92 percent for men and 77 percent for women.
    Although women's labor force participation rose in the 1980s and
 early 1990s, policymakers were concerned about the low labor force
 participation for single women with children, which remained relatively
 flat over that period. But for the past 20 years single women who head
 households with children have participated in the labor market at
 nearly the same rate as single women without children or married women
 without children. In fact, for the first time, in 2017 the labor force
 participation rate of single women with children was higher (79.09
 percent) than single women without dependents (79.06 percent.) Married
 women with children are still more likely to be out of the labor force
 (box figure 1). More recently, overall labor force participation has
 declined, in part due to the aging population. Older working-age
 Americans (55-64) are less likely to work, with a labor force
 participation rate in 2017 around 72 percent for those aged 55-59 and
 57 percent for those aged 60-64, compared to the current 82 percent for
 those aged 25-54.
    These trends provide context for who is not currently working that
 society might prefer to work. Most prime-age men work, though nearly
 ten percent do not. Most unmarried prime-age women with children also
 work. A much smaller share of older Americans work.
------------------------------------------------------------------------

Box Figure 1.
Prime-Age Women's Labor Force Participation, by Marital Status and 
        Presence of Children under Age 18
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Current Population Survey Annual Social and Economic 
        Supplement (ASEC) (Bureau of Labor Statistics [BLS] 1977-2017); 
        authors' calculations.

    Evidence of the effect of Medicaid participation on employment for 
childless adults is decidedly mixed, with population differences and 
prevailing economic conditions as potential explanations for why 
studies have shown positive, negative, and no effects on employment 
(Buchmueller, Ham, and Shore-Sheppard 2016). Nevertheless, in the years 
since Medicaid expansion through the ACA, the preponderance of evidence 
suggests that Medicaid receipt has had little or positive effects on 
labor supply (Baicker, et al., 2014; Duggan, Goda, and Jackson 2017; 
Garthwaite, Gross, and Notowidigdo 2014; Gooptu, et al., 2016; 
Kaestner, et al., 2017), with notable exceptions (e.g., Dague, DeLeire, 
and Leininger 2017).
    While there is no research evidence regarding the effect of work 
requirements in Medicaid, last month, as the first state to implement a 
plan, Arkansas disenrolled program participants for failing to comply 
with work requirements. Arkansas terminated coverage for 4,353 citizens 
for failing to qualify for an exemption or to meet work requirements, 
while an additional 1,218 reported 20 hours per week of work activities 
and 2,247 reported an exemption in the month of August (Rudowitz and 
Musumeci 2018).
    For these programs to accomplish their goals, eligible people 
should not be dissuaded from applying for or improperly prevented from 
receiving those benefits. Evidence suggests that, under a variety of 
scenarios, the vast majority of those losing access to Medicaid would 
not lose access because they failed to meet a work requirement, but 
because they failed to successfully report their work/training activity 
or exemption (Garfield, Rudowitz, and Musumeci 2018; Goldman, et al., 
2018). For example, in Arkansas, the only state currently implementing 
a work requirement in Medicaid, beneficiaries are required to report 
through an online portal, Access Arkansas (Arkansas Department of Human 
Services n.d.), despite a large number of program-eligible Arkansans 
who lack Internet access (Gangopadhyaya, et al., 2018).
Characteristics of Those Who Would Face New Work Requirements
    Potential loss of access to SNAP and Medicaid on the basis of a 
work requirement is a function of whether the person is qualified for 
and verified as exempt from working and, if not, whether the person 
works sufficient hours each month to meet the requirement. Those who 
have a categorical exemption from work requirements--students, for 
example--are not required to work unless their status changes. 
Exemptions from work requirements can be applied individually for a 
variety of reasons, including temporary health problems, or, more 
broadly, when the unemployment rate for a location is high. Certain 
educational or training activities can also qualify for meeting hourly 
thresholds.

                                 Box 2.
------------------------------------------------------------------------
 
-------------------------------------------------------------------------
                 Proposed Expansion of Work Requirements
 
    In April 2018 President Trump issued an Executive Order requiring
 that all means-tested programs be reviewed for the presence of current
 work requirements, the current state of enforcement and exemption, and,
 for those programs without current work requirements, whether such
 requirements could be added (White House 2018).
    This Executive Order builds on executive action to implement work
 requirements in Medicaid for the first time. In letters to governors
 (Price and Verma 2017) and state Medicaid directors (Neale 2018), the
 U.S. Department of Health and Human Services (HHS) has offered guidance
 for states considering submitting a waiver request to apply work
 requirements for those receiving Medicaid. Since the Centers for
 Medicare & Medicaid Services offered guidance to the states with regard
 to Medicaid in 2017, 14 states have submitted work requirement
 proposals to HHS. HHS has approved four states' plans, though
 Kentucky's plan was vacated. The state of Arkansas has begun to enforce
 work requirements (Urban Institute 2018). State proposals vary in terms
 of the age range and household composition of exposure, who is exempt,
 and the hours required for work or approved activities.
    Additionally, in reauthorizing the farm bill, in June 2018 the House
 voted to expand the scope of who is required to work in order to
 receive SNAP benefits to include adults 18-59 with dependent children
 aged 6-18 as well as those aged 50-59 without dependents under the age
 of 6. As of publication, the conference committee is considering this
 proposal.
------------------------------------------------------------------------

    To highlight one difficulty in designing a work requirement policy, 
consider the group of SNAP and Medicaid participants who usually are 
not working. Many individuals in this group are not expected to work, 
including the elderly, disabled, children, students, caregivers, and 
the infirm. In fact, nearly \2/3\ of individuals who participate in 
SNAP are elderly, disabled, or children (USDA 2017a).
    Some of these characteristics are straightforward to observe and 
verify, such as age, school enrollment, and receipt of disability 
benefits. Other characteristics are difficult to observe and costly to 
verify, such as those with temporary medical conditions that make it 
impossible for them to work, those who have a chronic health condition 
but do not meet the high standard set for disability benefits (or have 
not applied for disability benefits), and those who do not have the 
skills, childcare, or transportation to obtain a job in their local 
economy at present. Another share of this group might be capable of 
employment but not willing to work; in that case the work requirements 
might or might not provide enough incentive for them to get jobs.
    Using data from the Current Population Survey Annual Social and 
Economic Supplement (ASEC), we quantify exposure to work requirements 
in 2017 based on broad demographic characteristics. To do so, we 
separate those who would likely qualify for a categorical exemption 
from those who would be required to work or who would qualify for a 
waiver to maintain eligibility. To be clear, while we model who is 
eligible for a categorical exemption, evidence suggests that not 
everyone in these groups will successfully navigate the system and 
obtain the exemption; in fact, estimates suggest that most people who 
lose coverage under this policy will be eligible for an exemption or 
already be working. For SNAP we followed the Federal guidelines for 
categorical exemption; for Medicaid we created a composite from among 
the different plans put forth by the states based on how frequently 
such groups are exempt.
    For SNAP, minors, those who are older than 59 years, students, 
those receiving disability benefits, and those with a child under the 
age of 6 are exempt from both current and new, proposed work 
requirements. The samples are further limited to U.S. citizens and non-
active military. For simplification, we describe those aged 18-49 
without dependents as being currently exposed to work requirements and 
those aged 18-59 with a dependent between the ages of 6 and 17 
(inclusive) as well as those between the ages of 50 and 59 with no 
dependents under the age of 6 as newly required to meet work 
requirements or to participate in a training program in order to 
receive SNAP benefits. For the current group, some may live in places 
exempt from work requirements or have an unobserved good-cause 
exemption.
Figure 1.
Exposure to Work Requirements among Adult SNAP Participants, 2017

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: ASEC (BLS 2018); authors' calculations.

    How many adult SNAP participants are--or would be--exposed to work 
requirements? Figure 1 shows the entire adult population (18 or older) 
who reported SNAP participation in 2017. Each rectangle represents a 
share of the total population and whether the individuals in that share 
were eligible for a categorical exemption to work requirements (teal), 
were in a population currently exposed to a work requirement (green), 
or would be newly exposed to work requirements under the House proposal 
(purple). The shaded rectangles sum to 100 percent, the total adult 
SNAP participant population.
    Under the House bill parameters (described in box 2), combined with 
current work requirements, \1/3\ of all adults who reported receiving 
SNAP benefits during 2017 would be exposed to work requirements, though 
a portion of those impacted could apply for exemptions based on 
verified health- or work-related concerns. Some already face work 
requirements, but 22 percent of all participants would be newly exposed 
to work requirements under the House bill (purple).
    Figure 1 also shows the reasons some participants would be exempt 
from new requirements. The majority (67 percent) of adults currently 
receiving SNAP benefits would still be exempt from work requirements 
based on age, having a dependent under the age of 6, or having student 
or disability status. Some would be exempt for multiple reasons; we 
group them first by age, then by the presence of dependents, and then 
by student or disability status. For example, while figure 1 shows just 
14 percent exempt due to disability, 24 percent of all adult SNAP 
recipients report receipt of disability benefits.
    In 2017, 2.2 million people who reported SNAP benefit receipt were 
exposed to work requirements during the year based on their demographic 
characteristics. Under the House proposal and based on 2017 numbers, 
this would more than double with 2.5 million adults aged 18-49 with 
dependent children aged 6-17 and 1.6 million adults aged 50-59 who 
would be exposed to work requirements nationally for the first time.
    In any household, there may be others who rely on the benefits, and 
not just the individual facing work requirements. The solution to 
concerns for other individuals in the household has typically been to 
waive work requirements for those who likely cannot work or who reside 
with those for whom shielding from benefit loss is a priority. Any 
reduction in SNAP benefits to adults would reduce the total amount of 
resources available to them to purchase food, including food for 
children. There are 3.5 million children and 710,000 seniors in these 
households that would be exposed to possible benefit loss due to work 
requirements.
Figure 2.
Exposure to Work Requirements among Adult Medicaid Participants, 2017

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: ASEC (BLS 2018); authors' calculations.

    We perform the same exercise to show the share of Medicaid 
beneficiaries who are targeted by the policy based on potential new 
rules (figure 2). Minors, seniors (those over the age of 64), students, 
those receiving disability benefits or Medicare, and those with a child 
under the age of 6 are those who are generally eligible to be exempt 
from work requirements based on the plans that states submitted, though 
there is variation across states. We apply these categories to the 
entire adult Medicaid population, acknowledging that not every state 
has submitted a work requirement proposal and that the affected 
population varies by state plans. A nationwide expansion of these rules 
would target 22.4 million Americans for a possible loss of Medicaid 
coverage.
    Almost \1/2\ of all adult Medicaid beneficiaries would be targeted 
by work requirements if the composite rules were applied nationwide. 
The largest share of those exempt from work requirements are parents 
with young children (22 percent) followed by those reporting disability 
income (13 percent) and Medicare/Medicaid dual enrollees (12 percent). 
About six percent of Medicaid participants are students.
Volatility in the Low-Wage Labor Market
    The decline in labor force participation--especially among prime-
age males--has drawn extensive attention in academic and policy circles 
(e.g., Abraham and Kearney 2018; Council of Economic Advisers [CEA] 
2016; Juhn 1992). Some recent academic work has emphasized the fact 
that participation may be declining in part because an increasing 
number of labor force participants cycle in and out of the labor force: 
a pattern with direct relevance to proposed work requirements. The most 
comprehensive look at the behavior of people cycling through the labor 
force is Coglianese (2018). He documents that, among men, this group--
which he refers to as ``in-and-outs''--takes short breaks between jobs, 
returns to the labor force fairly quickly (within 6 months), and, 
crucially, is no more likely than a typical worker to take another 
break out of the labor force. See also Joint Economic Committee (2018) 
for a discussion of the in-and-out behavior of nonworking prime-age men 
and reasons for their non-employment.
    SNAP or Medicaid participants who are employed but who work in jobs 
with volatile employment and hours would be at risk of failing work 
requirements. This group includes those who lose their job; for 
example, the House bill sanctions participants for months they are not 
working or in training for at least 20 hours per week, even if they 
were recently employed and are searching for a new job. Similarly, 
those who work in jobs with volatile hours would be sanctioned in the 
months that their average hours fell below 20 hours per week, whether 
due to illness, lack of hours offered by the employer, or too few hours 
worked by the participant if they fail to receive a good cause waiver.
    Low-wage workers in seasonal industries such as tourism would 
potentially be eligible for SNAP in the months when they are working, 
but not in the months without employment opportunities. In other words, 
while benefits are most needed when an individual cannot find adequate 
work, under proposed work requirements these are the times that 
benefits would be unavailable. Disenrollment could make it more 
difficult for an individual to return to work--for example, if a person 
with chronic health conditions is unable to access needed care while 
they are between jobs. Any work requirement that banned individuals 
from participation for a considerable amount of time after failing the 
requirements would be even more problematic for those facing churn in 
the labor market.
    In a set of analyses, Bauer and Schanzenbach (2018a, 2018b) found 
that although many SNAP beneficiaries work on average more than 20 
hours a week every month, they frequently switch between working more 
than 20 hours and a different employment status over a longer time 
horizon. Using the ASEC, those authors found that, over the course of 
16 months between 2016 and 2018, about 20 percent of individuals aged 
18-59 without a dependent child under age 6 switched between working 
more than 20 hours a week and working fewer than 20 hours per week, 
seeking employment, or being out of the labor force.
    In this economic analysis we examine labor force status transitions 
and the reasons given for not working among those targeted for work 
requirements over 24 consecutive months, January 2013-December 2014, 
using the first two waves of the Survey of Income and Program 
Participation (SIPP).\1\ By using a dataset that allows us to track 
workers over time, we identify the share of program participants who 
are consistently out of the labor force, the share who would 
consistently meet a work requirement, and the share who would be at 
risk of losing benefits based on failing to meet a work requirement 
threshold.
    We assume that to comply with a program's work requirement, 
beneficiaries would have to prove each month that they are working for 
at least 20 hours per week averaged over the month, which is the 
typical minimum weekly requirement among the SNAP and Medicaid work 
requirement proposals. Looking first at SNAP and then at Medicaid, we 
calculate the share of program participants who would be exposed to 
benefit loss because they are not working sufficient hours would be 
exposed to benefit loss and who experienced a gap in employment, we 
describe the reasons given for not working to help quantify potential 
waiver eligibility.
    We remove from the analysis all those who have a categorical 
exemption. For SNAP and Medicaid, we exclude those outside the targeted 
age range, those with children under 6, full- or part-time students, 
and those reporting disability income. Those receiving Medicare are 
additionally excluded from the Medicaid analysis. As an instructive 
example, the labeled group ``18-49, no dependents'' is additionally 
exclusive of students and those reporting disability income. Program 
participants are those who reported receiving SNAP or Medicaid at any 
point between January 1, 2013, and December 31, 2014.
    We categorize each individual in each month into one of four 
categories: (1) employed and worked more than 20 hours a week on 
average, (2) employed and worked less than 20 hours a week on average, 
(3) unemployed and seeking employment, or (4) not in the labor force. 
If a worker was employed at variable weekly hours but maintained hours 
above the monthly threshold (80 hours for a 4 week month and 120 hours 
for a 5 week month), then we categorize them as ``employed and worked 
more than 20 hours a week for that month.'' Individuals are considered 
to have a stable employment status if they do not change categories 
over 2 years, and are considered to have made an employment status 
transition if they switched between any of these categories at least 
once. There is no employment status transition when a worker changes 
jobs but works more than 20 hours a week at each job.
Exposure To Proposed Work Requirements in SNAP
    Among working-age adults, SNAP and Medicaid serve a mix of the 
unemployed, low-income workers, and those who are not in the labor 
force (USDA 2017b). Figure 3 describes employment status by those 
groups who are currently exposed to work requirements and who would be 
newly subject to work requirements under the House proposal.
    During the Great Recession, waivers to work requirements were 
implemented nationwide. During the time period covered by the SIPP 
(2013-14), eight states stopped implementing these waivers fully, and 
ten states partially (Silberman 2013).\2\ For analytic purposes, we 
look at employment status transitions among 18 to 49 year olds without 
dependents as the demographic group currently exposed to work 
requirements, regardless of whether they lived in state in which 
waivers were implemented during 2013 and 2014. Those receiving SNAP 
benefits who are in the demographic group currently exposed to work 
requirements--adults aged 18-49 with no dependents--generally 
participate in the labor market, with just 25 percent consistently not 
in the labor force (discussed below). While 58 percent worked at least 
20 hours per week in at least 1 month over 2 years, 25 percent were 
over the threshold at some point but fell below the 20 hour threshold 
during at least 1 month over 2 years. Very few are always working less 
than 20 hours a week or always unemployed (less than two percent in 
either case), and 14 percent move across these categories.
Figure 3.
Employment Status over Two Years, SNAP Participants

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: Survey of Income and Program Participation (SIPP) 
        (U.S. Census Bureau 2013-14); authors' calculations.

    Those aged 18-49 who are not subject to the 3 month time limit 
because they have a dependent aged 6-17 but who would face it under the 
House proposal demonstrate a similar distribution of employment status 
as those without a dependent, but they are more likely to work. There 
are fewer individuals who are always out of the labor force (14 
percent) and more that consistently work 20 hours a week or more (46 
percent).\3\ There is also substantial month-to-month churn (16 
percent) between working above 20 hours per week and less than 20 hours 
per week and churn (12 percent) between working above 20 hours per week 
and being either unemployed or not in the labor force. This highlights 
the number who are actively in the workforce and meeting the 20 hour 
threshold in at least 1 month, but who might fail new work requirements 
from time to time.
    Older SNAP participants (aged 50-59 without dependents under age 6) 
who would also be newly exposed to work requirements and time limits 
have a distinct employment status pattern from those aged 18-49. Almost 
\1/2\ were permanently out of the labor force in large part due to 
their health. While 23 percent worked consistently above the threshold 
of 20 hours a week, nearly as many (18 percent) worked above the 
threshold at some point but also below the threshold at some point, 
meaning they would fail the work requirement despite having sometimes 
met the threshold.
    There is a meaningful portion of SNAP participants in the labor 
force and working, but not all are working above the monthly work 
requirement threshold consistently. Coglianese's (2018) finding that 
workers who are in and out of the labor force are not more likely to 
take another break later on suggests it is unclear how much more 
consistently work requirements would attach these people to the labor 
force.
    We next examine the reasons given for not working over the 2 year 
period, first for those aged 18-49 with a dependent between the ages of 
6 and 17, and second for those 50 to 59 without a dependent under age 6 
(figures 4a and 4b). The green crosshatch shows the share of the 
population that did not experience a gap in employment over the 2 year 
period, and thus were never asked why they were not working. Among 
those who were asked why they were not working for pay during at least 
1 week, we report the reason for not working in months they were not 
working. Those in solid shades of green were in the labor force but 
experienced at least one spell of unemployment or labor force 
nonparticipation. Those in the blue were out of the labor force for the 
entire 2 year period. Each person is assigned one reason--their most 
frequent reason--for not working.
Figure 4a.
Most-Frequent Reason for Not Working for Pay, SNAP Participants Aged 
        18-49 with Dependents Age 6-17
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: SIPP (U.S. Census Bureau 2013-14); authors' 
        calculations.
Figure 4b.
Most-Frequent Reason for Not Working for Pay, SNAP Participants Aged 
        50-59 with No Dependent under Age 6
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: SIPP (U.S. Census Bureau 2013-14); authors' 
        calculations.

    Among those aged 18-49 with dependents aged 6-17 who are newly 
exposed to work requirements (figure 4a), 86 percent were in the labor 
force at some point over 2 years but not all worked stably. Among those 
who did not work for pay for at least 1 week but were in the labor 
force, the overwhelming majority gave work-related reasons (68 
percent), such as temporary loss of job, temporary loss of hours (e.g., 
weather-related, not getting enough shifts, etc.), or a company 
shutting down a plant or location. Other large groups include those who 
are caregivers and those with health concerns. In a program with 
extensive good-cause waivers, it appears the bulk of these workers 
would not lose benefits if waivers were implemented with fidelity; but 
the administrative burden required to sort those with work-related 
problems from those who choose to not work could be quite high.
    Among those out of the labor force for the entire 2 year period, 
more than \1/2\ cite health reasons for being out of the labor force. 
In total, 0.3 percent of those aged 18-49 who would be newly exposed to 
work requirements and who were labor force non-participants said that 
they were not interested in working.
    Among individuals aged 50-59 (figure 4b), far more are out of the 
labor force consistently and far fewer have stable work. Overall, 
health (87 percent) and work-related (eight percent) issues dominate. 
The prevalence of health problems is striking considering we have 
already limited the sample to those not receiving disability payments. 
Fewer than one percent were retired or not interested in working.
    The share of older SNAP participants listing caregiving as a reason 
for being not in the labor force is notably smaller than the share of 
the younger SNAP participant population.
    Roughly 11 percent of SNAP participants aged 18-49 with a dependent 
6-17 that were out of the labor force for the entire 24 month period 
list caregiving as a reason for not being in the labor force. However, 
even 11 percent is smaller than many might expect. Many caregivers who 
are not in the labor force are in two-adult households where the other 
adult is working. In addition, many are in households with dependents 
aged 0-5, and those households are exempt from work requirements.
    In summary, based on 2013-14 data, 5.5 million adult SNAP 
participants would be newly exposed to work requirements with 3.8 
million who would have failed them at some point in this 2 year window. 
Notable among those who were asked about a spell of not working, 2.1 
million report health or disability issues and 1.5 million report work-
related issues. Only about 90,000 list a lack of interest or early 
retirement as their reason for not working.
Exposure to Proposed Work Requirements in Medicaid
    We study the work participation of Medicaid beneficiaries in a 
similar manner. Unlike SNAP, there is no current population of 
participants who face work requirements across the country to use as a 
comparison group. As noted above, previous Administrations and the 
courts have not viewed Medicaid work requirements as supporting core 
program goals; there are substantive doubts about whether work 
requirements for health insurance are appropriate. Nevertheless, we 
consider the employment status of Medicaid beneficiaries to illuminate 
how such requirements would function.
    Since Medicaid beneficiaries do not currently face work 
requirements, we do not separately examine the population aged 18-49 
without dependents. It is instructive to differentiate the work status 
transitions of younger (aged 18-49) and older (aged 50-64) Medicaid 
beneficiaries, restricted to those who either have a dependent 6-17 or 
no dependents, i.e., no dependents under the age of 6. We identify 
employment status transitions and the reasons given for not working 
among those targeted for work requirements over 24 consecutive months 
(January 2013-December 2014).
    Figure 5 shows that over 2 years (2013 and 2014), 80 percent of 
Medicaid beneficiaries aged 18-49 without a dependent child under age 6 
were in the labor force at some point. While about 40 percent 
consistently worked over the 20 hour threshold, 25 percent worked more 
than 20 hours at some point but would potentially lose benefits for 
falling below the 20 hour threshold for a month at another point.
    The picture is quite different for older Medicaid beneficiaries (50 
to 64) who would be exposed to work requirements. Of that population, 
44 percent were out of the labor force for all 24 months. About 29 
percent worked consistently more than 20 hours a week and about 17 
percent worked more than 20 hours at least once but failed to do so 
every month. The reasons given among working-age adult Medicaid 
beneficiaries not working for pay suggest that labor market reasons 
dominate among labor force participants and health reasons dominate 
among labor force non-participants (figures 6a and 6b). Once again, 
only a small number of labor force non-participants are not interested 
in work or are retired.
Figure 5.
Employment Status over Two Years, Medicaid Participants

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          SIPP (U.S. Census Bureau 2013-14); authors' calculations.
Figure 6a.
Most-Frequent Reason for Not Working for Pay, Medicaid Participants 
        Aged 18-49 with No Dependents under Age 6
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
Figure 6b.
Most-Frequent Reason for Not Working for Pay, Medicaid Participants 
        Aged 50-64 with No Dependents under Age 6
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: SIPP (U.S. Census Bureau 2013-14); authors' 
        calculations.

    Among older participants of Medicaid (aged 50-64 without a 
dependent under age 6, the population making up 37 percent of the 
sample population), 35 percent of those with Medicaid coverage are out 
of the labor force for health reasons; this group represents 79 percent 
of those who were not in the labor force for the full 2 years. It is 
worth noting that work requirements for this group would necessitate 
either lax requirements with a very large portion of the population 
getting waivers, or an administratively burdensome process to determine 
which individual's health concerns truly limit them from work.
Work Status in a Snapshot vs. Two Years
    In its report on work requirements, the Council of Economic 
Advisers (CEA 2018) looked at employment among adult program 
participants for the month of December 2013 using the SIPP and found 
that about three in five participants worked fewer than 20 hours per 
month. The CEA concludes that this level of work--or lack thereof--
``suggest[s] that legislative changes requiring them to work and 
supporting their transition into the labor market, similar to the 
approach in TANF, would affect a large share of adult beneficiaries and 
their children in these non-cash programs''.1-2
    A critical empirical takeaway from the analysis presented herein is 
that frequent movement between labor status categories over time 
increases the number of people exposed to losing benefits for failing 
to consistently meet a work requirement, and decreases the number of 
people who are entirely out of the labor market. We now examine how the 
analysis of work experiences differs when we compare a snapshot in 
time--one month--with analysis that includes transitions across status 
over 2 years. When we compare the 1 month of SIPP data cited in the CEA 
report (December 2013) against 24 months, we find that fewer program 
participants are labor force non-participants and fewer meet the work 
requirement threshold.
    Figure 7 demonstrates how observed employment status is different 
in 1 month versus 2 years. The first two bars show employment status 
categories for the full population aged 18-59 without dependents aged 
0-5, disability payments, or status as students. The second two bars 
show employment status categories in 1 month and 2 years for SNAP 
participants aged 18-59 with no dependents aged 0-5, disability 
payments, or status as students. An ``other'' transition during a 1 
month period are those who report being unemployed and a labor force 
non-participant during different weeks within December 2013.
Figure 7.
Employment Status in One Month vs. Two Years, SNAP

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          SIPP (U.S. Census Bureau 2013-14); authors' calculations.

    The first feature that jumps out of the data is that far fewer 
people are out of the labor force than is generally assumed. While a 1 
month snapshot shows that 20 percent of the overall population is not 
working (either out of the labor force or unemployed), over the course 
of 2 years more than 90 percent of the overall population is employed 
at some point. Many people are not truly on the sidelines as much as 
they are cycling in and out of the game. Furthermore, fewer people are 
solidly in the 20+ hours workforce. The share of the overall population 
that stably works more than 20 hours per week falls from 76 percent in 
the 1 month snapshot to 69 percent over 2 years.
    Looking only at those who participated in SNAP at any point during 
the 2 year period, the 1 month snapshot is also different from the 2 
year, both in terms of the number of participants out of the labor 
force and the number who would retain benefits under the work 
requirement proposal. Instead of 42 percent being out of the labor 
force and roughly 11 percent unemployed in the 1 month snapshot--
leading to more than \1/2\ of the group being labeled ``not working'' 
in the 1 month snapshot--roughly 29 percent are out of the labor force 
and just one percent are persistently unemployed over 2 years, meaning 
fewer than \1/3\ are not working consistently. Recall that the higher 
``not working'' rate among SNAP beneficiaries is largely driven by 
those aged 50-59. SNAP recipients aged 18-49 without dependents have a 
``not working'' rate of 25 percent over 2 years, and those with 
dependents aged 6-17 have a ``not working'' rate of just 14 percent. 
Almost a quarter of SNAP participants would fail the work requirements 
some months and pass them in others, with the majority giving work-
related reasons for their change in status.
    A similar pattern holds for Medicaid beneficiaries: the monthly 
snapshot overstates the number of labor force non-participants and 
understates those who would meet a work requirement. There is a ten 
percentage point-reduction in the share of those not working over 1 
month (39 percent) versus 2 years (29 percent). Forty-two percent would 
meet the work requirement in 1 month, but only 36 percent do over 2 
years. In addition, in the 2 year sample 22 percent of participants 
work over 20 hours in at least 1 month in the sample but fail to in 
other months (figure 8).
Figure 8.
Employment Status in One Month vs. Two Years, Medicaid

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          SIPP (U.S. Census Bureau 2013-14); authors' calculations.
Conclusion
    The combination of a strong labor market, work requirements to 
receive cash benefits through TANF, and work incentives generated by 
the EITC raised labor force participation rates among single mothers in 
the mid-1990s (Ziliak 2016), leading some to believe that further 
participation gains could be obtained by extending only the work 
requirement component to other programs (Haskins 2018; CEA 2018).
    Work requirements are intended to counter any work disincentives 
that come from a social safety net and to ensure that society is not 
unnecessarily supporting people who could otherwise support themselves. 
At the same time, such work requirements add administrative complexity 
to social programs and risk keeping benefits from parts of the 
population that should be receiving them. This economic analysis 
establishes a set of facts that are relevant when considering the 
expansion of work requirements.
    What types of populations will face these new work requirements? 
How many would fail to meet the requirements? Do program participants 
appear to already be in the labor force facing work-related constraints 
on hours or do they choose not to work? And how many would in theory be 
eligible for waivers relative to those individuals that society would 
like to push toward work?
    A large number of SNAP and Medicaid participants who would face new 
work requirements cycle in and out of the labor force and would thus 
lose benefits at certain times. Among those who are in the labor force, 
spells of unemployment are either due to job-related concerns or health 
issues. Very few reported that they were not working due to lack of 
interest.
    Among those out of the labor force for the entire 2 year period, 
health concerns are the overriding reason for not working, even after 
removing those who receive disability benefits from the sample. The 
older portion of the population newly exposed to work requirements is 
more likely to be out of the labor force for extended periods of time. 
Among this group, again, health reasons are the overriding factor in 
not working. Work requirements for this group might push more onto 
disability rolls, make the disability adjudication even more 
consequential, and require a separate health investigation to settle 
all the necessary waivers. Failure to receive a waiver would result in 
disenrollment; losing access to these programs would reduce resources 
available to purchase food and health insurance among otherwise 
eligible households.
    For those who qualify for exemptions, satisfy waiver requirements, 
or work enough to meet the requirements, there are still significant 
informational and administrative barriers to compliance. Program 
participants must understand how the work requirement policy relates to 
them, obtain and submit documentation, and do so at the frequency 
prescribed by the state (Wagner and Solomon 2018). Frequent exposure to 
verification processes, such as the monthly reporting periods 
prescribed in the Agricultural Act of 2014 (the farm bill) and many 
states' Medicaid proposals, increases the administrative burden on 
participants and enforcers, the likelihood of error, and cost (Bauer 
and Schanzenbach 2018b). These continuing roadblocks to participation, 
with attendant informational and transactional costs, are likely to 
result in lower take-up among the eligible population and disenrollment 
(Finkelstein and Notowidigdo 2018).
    Looking at snapshots of work experience, such as a single month, 
inflates both the number of SNAP and Medicaid participants who are out 
of the labor force and the number of people who work sufficient hours 
to satisfy work requirements. Over 24 consecutive months the number of 
SNAP and Medicaid program beneficiaries not working or seeking work as 
well as those working consistently above 20 hours fall substantially.
    There are safety net levers that can be used to pull those out of 
the labor force into work. Steps such as increasing the EITC might be a 
very effective way to increase work participation in this group without 
the same administrative burdens and negative spillovers to vulnerable 
populations. (See Hoynes, Rothstein, and Ruffini 2017 for a specific 
proposal along these lines.) That proposal is estimated to increase 
participation by 600,000 people. Raising the returns to work via the 
EITC or other measures, creating training or educational opportunities 
that can increase individuals' human capital, and providing child care 
or improved treatment and medical care to reduce health barriers to 
work could make full attachment to the labor force more viable for many 
individuals.
Endnotes
    1. See technical appendix tables 1 and 2 for additional work status 
transition statistics.
    2. The states not implementing able-bodied adult without dependents 
waivers at some point during 2013-14 are: Delaware, Guam, Iowa, Kansas, 
Nebraska, Oklahoma, Utah, Virginia, and Wyoming. States implementing a 
partial waiver (partial referring to different parts of the state or 
only part of the year): Colorado, Minnesota, New Hampshire, New York, 
North Dakota, Ohio, South Dakota, Texas, Vermont, Wisconsin.
    3. Those who meet the 20 hour threshold monthly hours variable 
include both those who meet the threshold every week and those whose 
hours varied each week but averaged to 20 hours per week each month. 
The volatility of their hours may suggest they are more likely to fail 
the work requirement threshold but they did not do so over the 2 year 
window.
References

 
 
 
    Abraham, Katharine G., and Melissa S. Kearney. 2018. ``Explaining
 the Decline in the U.S. Employment-to-Population Ratio: A Review of the
 Evidence.'' Working Paper 24333, National Bureau of Economic Research,
 Cambridge, MA.
    Agricultural Act of 2014 (farm bill), Pub. L. 113-79 (2014).
    Arkansas Department of Human Services. n.d. ``Access Arkansas.''
 (https://access.arkansas.gov.)
    Baicker, Katherine, Amy Finkelstein, Jae Song, and Sarah Taubman.
 2014. ``The Impact of Medicaid on Labor Market Activity and Program
 Participation: Evidence from the Oregon Health Insurance Experiment.''
 American Economic Review 104 (5): 322-28.
    Bauer, Lauren, and Diane Whitmore Schanzenbach. 2018a, July 27.
 ``Employment Status Changes Put Millions at Risk of Losing SNAP
 Benefits for Years.'' Blog. The Hamilton Project, Brookings
 Institution, Washington, D.C.
    ___. 2018b, August 9. ``Who Loses SNAP Benefits If Additional Work
 Requirements Are Imposed? Workers.'' Blog. The Hamilton Project,
 Brookings Institution, Washington, D.C.
    Black, Sandra E., Diane W. Schanzenbach, and Audrey Breitwieser.
 2017. ``The Recent Decline in Women's Labor Force Participation.'' In
 Driving Growth through Women's Economic Participation edited by Diane
 W. Schanzenbach and Ryan Nunn, 5-17. Washington, D.C.: The Hamilton
 Project.
    Blank, Rebecca. 2002.``Evaluating Welfare Reform in the United
 States.'' Journal of Economic Literature 40 (4): 1105-66.
    Blinder, Alan S., and Mark Zandi. 2015. ``The Financial Crisis:
 Lessons for the Next One.'' Policy Futures, Center on Budget and Policy
 Priorities, Washington, D.C.
    Bolen, Ed, Lexin Cai, Stacy Dean, Brynne Keith-Jennings, Catlin
 Nchako, Dottie Rosenbaum, and Elizabeth Wolkomir. 2018. ``House Farm
 Bill Would Increase Food Insecurity and Hardship.'' Center on Budget
 and Policy Priorities, Washington, D.C.
    Buchmueller, Thomas, John C. Ham, and Lara D. Shore-Sheppard. 2016.
 ``The Medicaid Program.'' In Economics of Means-Tested Transfer
 Programs in the United States, Vol. I, edited by Robert A. Moffitt, 21-
 136. Chicago, IL: University of Chicago Press.
    Bureau of Labor Statistics (BLS). 1977-2017. ``Current Population
 Survey Annual Social and Economic Supplement.'' Bureau of Labor
 Statistics, U.S. Department of Labor, Washington, D.C. Retrieved from
 IPUMS-CPS, University of Minnesota.
    ___. 2018. ``Current Population Survey Annual Social and Economic
 Supplement.'' Bureau of Labor Statistics, U.S. Department of Labor,
 Washington, D.C. Retrieved from IPUMS-CPS, University of Minnesota.
    Butcher, Kristin F., and Diane Whitmore Schanzenbach. 2018. ``Most
 Workers in Low-Wage Labor Market Work Substantial Hours, in Volatile
 Jobs.'' Policy Futures, Center on Budget and Policy Priorities,
 Washington, D.C.
    The Centers for Medicare & Medicaid Services. 2018. ``Remarks by
 Administrator Seema Verma at the 2018 Medicaid Managed Care Summit.''
 As prepared for delivery: September 27, 2018. The Centers for Medicare
 & Medicaid Services, Washington, D.C.
    Coglianese, John. 2018, February. ``The Rise of In-and-Outs:
 Declining Labor Force Participation of Prime Age Men.'' Working Paper,
 Harvard University, Cambridge, MA.
    Council of Economic Advisers (CEA). 2016. ``The Long-Term Decline in
 Prime-Age Male Labor Force Participation.'' Council of Economic
 Advisers, White House, Washington, D.C.
    ___. 2018. ``Expanding Work Requirements in Non-Cash Welfare
 Programs.'' Council of Economic Advisers, White House, Washington, D.C.
    Dague, Laura, Thomas DeLeire, and Lindsey Leininger. 2017. ``The
 Effect of Public Insurance Coverage for Childless Adults on Labor
 Supply.'' American Economic Journal: Economic Policy 9 (2): 124-54.
    Duggan, Mark, Gopi Shah Goda, and Emilie Jackson. 2017. ``The
 Effects of the Affordable Care Act on Health Insurance Coverage and
 Labor Market Outcomes.'' Working Paper 23607, National Bureau of
 Economic Research, Cambridge, MA.
    Fang, Hanming, and Michael P. Keane. 2004. ``Assessing the Impact of
 Welfare Reform on Single Mothers.'' Brookings Papers on Economic
 Activity (1): 1-95.
    Finkelstein, Amy, Nathaniel Hendren, and Erzo F.P. Luttmer. 2015.
 ``The Value of Medicaid: Interpreting Results from the Oregon Health
 Insurance Experiment.'' Working Paper 21308, National Bureau of
 Economic Research, Cambridge, MA.
    Finkelstein, Amy, and Matthew J. Notowidigdo. 2018. ``Take-up and
 Targeting: Experimental Evidence from SNAP.'' Working Paper 24652,
 National Bureau of Economic Research, Cambridge, MA.
    Gangopadhyaya, Anuj, Genevieve M. Kenney, Rachel A. Burton, and
 Jeremy Marks. 2018. ``Medicaid Work Requirements in Arkansas: Who Could
 Be Affected, and What Do We Know about Them?'' Urban Institute,
 Washington, D.C.
    Garfield, Rachel, Robin Rudowitz, and Anthony Damico. 2018.
 ``Understanding the Intersection of Medicaid and Work.'' Henry J.
 Kaiser Family Foundation, San Francisco, CA.
    Garfield, Rachel, Robin Rudowitz, and MaryBeth Musumeci. 2018.
 ``Implications of a Medicaid Work Requirement: National Estimates of
 Potential Coverage Losses.'' Henry J. Kaiser Family Foundation, San
 Francisco, CA.
    Garfield, Rachel, Robin Rudowitz, MaryBeth Musumeci, and Anthony
 Damico. 2018. ``Implications of Work Requirements in Medicaid: What
 Does the Data Say?'' Henry J. Kaiser Family Foundation, San Francisco,
 CA.
    Garthwaite, Craig, Tal Gross, and Matthew J. Notowidigdo. 2014.
 ``Public Health Insurance, Labor Supply, and Employment Lock.''
 Quarterly Journal of Economics 129 (2): 653-96.
    Goldman, Anna L., Steffie Woolhandler, David U. Himmelstein, David
 H. Bor, and Danny McCormick. 2018. ``Analysis of Work Requirement
 Exemptions and Medicaid Spending.'' JAMA Internal Medicine online,
 September 10.
    Gooptu, Angshuman, Asako S. Moriya, Kosali I. Simon, and Benjamin D.
 Sommers. 2016. ``Medicaid Expansion Did Not Result in Significant
 Employment Changes or Job Reductions in 2014.'' Heath Affairs 35 (1):
 111-18.
    Grogger, Jeffrey. 2004. ``Welfare Transitions in the 1990s: The
 Economy, Welfare Policy, and the EITC.'' Journal of Policy Analysis and
 Management 23 (4): 671-95.
    Hamilton, Gayle, Stephen Freedman, Lisa Gennetian, Charles
 Michalopoulos, Johanna Walter, Dianna Adams-Ciardullo, and Anna Gassman-
 Pines. ``National Evaluation of Welfare-to-Work Strategies.'' 2001.
 Prepared by MDRC, Washington, D.C. for the U.S. Department of Health
 and Human Services and U.S. Department of Education.
    Haskins, Ron. 2018, July 25. ``Trump's Work Requirements Have Been
 Tested Before. They Succeeded.'' The Washington Post.
    Hoynes, Hilary, Jesse Rothstein, and Krista Ruffini. 2017. ``Making
 Work Pay Better through an Expanded Earned Income Tax Credit.'' Policy
 Proposal 2017-09, The Hamilton Project, Brookings Institution,
 Washington, D.C.
    Hoynes, Hilary Williamson, and Schanzenbach, Diane Whitmore. 2012.
 ``Work Incentives and the Food Stamp Program.'' Journal of Public
 Economics 96 (1): 151-62.
    ___. 2016, January. ``The Safety Net as an Investment.'' Working
 Paper, University of California, Berkeley.
    Institute of Medicine. 2003. ``Spending on Health Care for Uninsured
 Americans: How Much, and Who Pays?'' In Hidden Costs, Values Lost:
 Uninsurance in America, edited by Institute of Medicine (U.S.)
 Committee on the Consequences of Uninsurance. Washington, D.C.:
 National Academies Press.
    Joint Economic Committee. 2018. ``Inactive, Disconnected, and
 Ailing: A Portrait of Prime-Age Men Out of the Labor Force.'' SCP
 Report 2-18, Joint Economic Committee, Washington, D.C.
    Juhn, Chinhui. 1992. ``Decline of Male Labor Market Participation:
 The Role of Declining Market Opportunities.'' The Quarterly Journal of
 Economics 107 (1): 79-121.
    Kaestner, Robert, Bowen Garrett, Jiajia Chen, Anuj Gangopadhyaya,
 and Caitlyn Fleming. 2017. ``Effects of ACA Medicaid Expansions on
 Health Insurance Coverage and Labor Supply.'' Journal of Policy
 Analysis and Management 36 (3): 608-42.
    Keith-Jennings, Brynne, and Dottie Rosenbaum. 2015. ``SNAP Benefit
 Boost in 2009 Recovery Act Provided Economic Stimulus and Reduced
 Hardship.'' Center on Budget and Policy Priorities, Washington, D.C.
    Meyer, Bruce D., and Dan T. Rosenbaum. 2001. ``Welfare, The Earned
 Income Tax Credit, and The Labor Supply of Single Mothers.'' The
 Quarterly Journal of Economics 116 (3): 1063-1114.
    Neale, Brian. 2018. ``RE: Opportunities to Promote Work and
 Community Engagement among Medicaid Beneficiaries.'' SMD Letter 18-002,
 Center for Medicare & Medicaid Services, U.S. Department of Health and
 Human Services, Washington, D.C.
    Patient Protection and Affordable Care Act (ACA), Pub. L. 111-148,
 124 Stat. 119 through 124 Stat. 1025 (2010).
    Pavetti, LaDonna, and Liz Schott. 2016. ``TANF at 20: Time to Create
 a Program that Supports Work and Helps Families Meet Their Basic
 Needs.'' Center on Budget and Policy Priorities, Washington, D.C.
    Price, Thomas E., and Seema Verma. 2017. ``Letter to the Nation's
 Governors.'' U.S. Department of Health and Human Services, Washington,
 D.C.
    Rosenbaum, Dorothy. 2013. ``The Relationship between SNAP and Work
 among Low-Income Households.'' Center on Budget and Policy Priorities,
 Washington, D.C.
    Rudowitz, Robin, and MaryBeth Musumeci. 2018. ``An Early Look at
 State Data for Medicaid Work Requirements in Arkansas.'' Henry J.
 Kaiser Family Foundation, San Francisco, CA.
    Schanzenbach, Diane W. 2018. ``Can Benefits and Incentives Promote
 Work?'' Journal of Policy Analysis and Management 37 (4): 903-11.
    Silberman, Lizbeth. 2013. ``Supplemental Nut[ri]tion Assistance
 Program (SNAP): Able Bodied Adults Without Dependents Waivers for
 Fiscal Year 2014.'' Policy Memorandum, [August] 2. U.S. Department of
 Agriculture, Food and Nutrition Service. (https://fns-
 prod.azureedge.net/sites/default/files/
 FY_2014_ABAWD_Waiver_Status.pdf.)
    Stewart v. Azar (Ronnie Maurice Stewart, et al., v. Alex M. Azar II,
 et al.). Civil Action No. 18-152 (JEB) (D.D.C. Apr. 10, 2018).
    Urban Institute. 2018, October. ``Work Requirements Tracker.'' Urban
 Institute, Washington, D.C.
    U.S. Census Bureau. 2013-14. ``Survey of Income and Program
 Participation.'' U.S. Census Bureau, U.S. Department of Commerce,
 Suitland, MD.
    U.S. Department of Agriculture (USDA). 2017a. ``Characteristics of
 USDA Supplemental Nutrition Assistance Program Households: Fiscal Year
 2016 (Summary).'' U.S. Department of Agriculture, Washington, D.C.
    ___. 2017b. ``Supplemental Nutrition Assistance Program (SNAP):
 Facts About SNAP.'' U.S. Department of Agriculture, Washington, D.C.
    Wagner, Jennifer, and Judith Solomon. 2018. ``States' Complex
 Medicaid Waivers Will Create Costly Bureaucracy and Harm Eligible
 Beneficiaries.'' Center on Budget and Policy Priorities, Washington,
 D.C.
    White House. 2018. ``Executive Order Reducing Poverty in America by
 Promoting Opportunity and Economic Mobility.'' White House, Washington,
 D.C.
    Wolkomir, Elizabeth, and Lexin Cai. 2018. ``The Supplemental
 Nutrition Assistance Program Includes Earnings Incentives.'' Center on
 Budget and Policy Priorities, Washington, D.C.
    Ziliak, James P. 2016. ``Temporary Assistance for Needy Families.''
 In Economics of Means-Tested Transfer Programs in the United States,
 Vol. I, edited by Robert A. Moffitt, 303-93. Chicago, IL: University of
 Chicago Press.
 

Appendix

                                                                    Appendix Table 1.
                                                          Employment Status, SNAP Participants
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                        Transitioned
                                                                                                    Transitioned      between 20+ hours
                    Stable (not in         Stable         Stable  (employed   Stable  (employed   between 20+ hours   and unemployment        Other
                     labor force)       (unemployed)         20+ hours)          <20 hours)         and <20 hours      or not in labor      transition
                                                                                                                            force
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                Age 18-49, no dependents
--------------------------------------------------------------------------------------------------------------------------------------------------------
          2013             34.3%                 5.5%               33.3%                4.1%                7.9%                4.9%            10.0%
          2014             32.6%                 5.5%               37.4%                3.5%                9.1%                7.2%             4.7%
       2013-14             24.6%                 1.7%               32.6%                1.7%               16.0%                9.3%            14.1%
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                Age 18-49, dependent 6-17
--------------------------------------------------------------------------------------------------------------------------------------------------------
          2013             20.4%                 4.9%               49.9%                2.4%                8.9%                6.0%             7.5%
          2014             21.0%                 4.2%               50.2%                2.4%                8.6%                9.9%             3.8%
       2013-14             14.0%                 0.7%               45.6%                0.4%               15.9%               12.3%            11.3%
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                             Age 50-59, no dependent under 6
--------------------------------------------------------------------------------------------------------------------------------------------------------
          2013             50.4%                 4.6%               25.8%                2.6%                5.7%                3.9%             7.0%
          2014             53.3%                 3.5%               26.1%                2.5%                5.9%                5.1%             3.6%
       2013-14             45.7%                 1.3%               23.0%                1.4%               10.1%                7.9%           10.7%
--------------------------------------------------------------------------------------------------------------------------------------------------------
Source: SIPP (U.S. Census Bureau 2013-14); authors' calculations.
Note: The sample is limited to U.S. citizens, non-active military, who reported receiving SNAP benefits at any point between January 2013 and December
  2014. Only respondents with 24 months of data were included. Those with children under age 6, full- or part-time students, and those who reported
  receiving disability benefits were excluded from the sample based on categorical work requirement exclusions. Those who were assigned to ``stable''
  categories were observed as not in the labor force, unemployed, above the 20 hour threshold, or below the 20 hour threshold per week. Those who were
  stable and employed more than 20 hours a week were assigned either by meeting the threshold every week or because the monthly hours total averaged to
  above 20 hours per week. Regardless of the number of transitions made, each person who was observed as switching between work statuses was assigned to
  one group in the following order: first, transitioned between more than and less than 80 hours per month; second, transitioned between more than 80
  hours per month and unemployment or labor force nonparticipation; third, other. ``Other'' includes those who transitioned between less than 80 hours
  per month and unemployment or labor force nonparticipation as well as those who transitioned between unemployment and labor force nonparticipation.


                                                                    Appendix Table 2.
                                                        Employment Status, Medicaid Participants
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                        Transitioned
                                                                                                    Transitioned      between 20+ hours
                    Stable (not in         Stable         Stable  (employed   Stable  (employed   between 20+ hours   and unemployment        Other
                     labor force)       (unemployed)         20+ hours)          <20 hours)         and <20 hours      or not in labor      transition
                                                                                                                            force
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                             Age 18-49, no dependent under 6
--------------------------------------------------------------------------------------------------------------------------------------------------------
          2013             27.7%                 3.8%               42.6%                3.6%                8.1%                4.1%            10.0%
          2014             26.4%                 4.2%               46.1%                3.3%                7.3%                7.6%             5.1%
       2013-14             19.6%                 1.1%               39.6%                1.1%               14.8%               10.9%            12.8%
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                             Age 50-64, no dependent under 6
--------------------------------------------------------------------------------------------------------------------------------------------------------
          2013             48.4%                 3.2%               32.9%                3.8%                5.5%                2.2%             4.0%
          2014             51.2%                 2.7%               29.9%                3.5%                5.0%                4.6%             3.1%
       2013-14             44.1%                 0.8%               28.5%                1.8%               11.7%                5.0%            8.2%
--------------------------------------------------------------------------------------------------------------------------------------------------------
Source: SIPP (U.S. Census Bureau 2013-14); authors' calculations.
Note: The sample is limited to U.S. citizens, non-active military, who reported receiving Medicaid benefits at any point between January 2013 and
  December 2014. Only respondents with 24 months of data were included. Those with children under age 6, full- or part-time students, those who reported
  receiving Medicare, and those who reported receiving disability benefits were excluded from the sample based on categorical work requirement
  exclusions. Those who were stable labor force non-participants are contrasted with those who were in the labor force (working or seeking work) at
  least once during the 2 year period. Those who were assigned to ``stable'' categories were observed as not in the labor force, unemployed, working
  above the 20 hour threshold, or working below the 20 hour threshold per week. Those who were stable and employed more than 80 hours per week were
  assigned either by meeting the 20 hours per week threshold every week or because the monthly hours total averaged above 20 hours per week. Regardless
  of the number of transitions made, each person who was observed as switching between work statuses was assigned to one group in the following order:
  first, transitioned between more than and less than 80 hours per month; second, transitioned between more than 80 hours per month and unemployment or
  labor force nonparticipation; third, other. ``Other'' includes those who transitioned between less than 80 hours per month and unemployment or labor
  force nonparticipation as well as those who transitioned between unemployment and labor force nonparticipation.

Technical Appendix
    Box Figure 1. Prime-Age Women's Labor Force Participation, by 
Marital Status and Presence of Children under Age 18

          Source: Current Population Survey Annual Social and Economic 
        Supplement (ASEC) (Bureau of Labor Statistics [BLS] 1977-2017); 
        authors' calculations.
          Note: ``Prime-age'' indicates ages 25 to 54, inclusive. 
        ``Married'' is defined by women who have a spouse in the 
        household or not in the household. ``Single'' is defined as all 
        other women, including divorced and widowed women. ``With 
        children'' is defined as having at least one child in the 
        household under the age of 18. ``No children'' is defined as 
        having no children in the household under the age of 18. 
        Population counts calculated using the Annual Social and 
        Economic Supplement weight.

    Figure 1. Exposure to Work Requirements among Adult SNAP 
Participants, 2017

          Source: ASEC (BLS 2018); authors' calculations.
          Notes: Those who would be exempt from work requirements if 
        the House bill work requirements were passed include those over 
        the age of 59, those with a dependent under the age of 6, full- 
        or part-time students, and those who receive disability 
        benefits. While in some states work requirements are waived for 
        those aged 18-49 with no dependents, state-level differences 
        are not accounted for in identifying those who are currently 
        exposed to work requirements. Population counts calculated 
        using the Annual Social and Economic Supplement weight among 
        U.S. citizens over the age of 18 who reported receiving SNAP 
        benefits at some point during 2017.

    Figure 2. Exposure to Work Requirements among Adult Medicaid 
Participants, 2017

          Source: ASEC (BLS 2018); authors' calculations.
          Note: States applying for waivers to add work requirements to 
        Medicaid have identified different categorical exemptions and 
        conditions for waivers. For this exercise, we identified the 
        most frequent categorical exemptions and applied those rules 
        nationally. Those who are over the age of 64 or who are dual 
        Medicare enrollees are exempt, those receiving disability 
        income are exempt, those with a dependent under the age of 6 
        are exempt, and full- or part-time students are exempt. 
        Population counts are calculated using the Annual Social and 
        Economic Supplement weight among U.S. citizens over the age of 
        18 who reported receiving Medicaid benefits at some point 
        during 2017.

    Figure 3. Employment Status over Two Years, SNAP Participants

          Source: Survey of Income and Program Participation (SIPP) 
        (U.S. Census Bureau 2013-14); authors' calculations.
          Note: The sample is limited to U.S. citizens, non-active 
        military, aged 18-59 who reported receiving SNAP benefits at 
        any point between January 2013 and December 2014. Only 
        respondents with 24 months of data were included. Those with 
        children under age 6, full- or part-time students, and those 
        who reported receiving disability benefits were excluded from 
        the sample based on categorical work requirement exclusions. 
        Those who were assigned to ``stable'' categories were observed 
        as not in the labor force, unemployed, working above the 20 
        hour threshold, or working below the 20 hour threshold per 
        week. Those who were stable and employed more than 20 hours a 
        week were assigned either by meeting the threshold every week 
        or because the monthly hours total averaged to above 20 hours 
        per week. Regardless of the number of transitions made, each 
        person who was observed as switching between work statuses was 
        assigned to one group in the following order: first, 
        transitioned between more than and less than 80 hours per 
        month; second, transitioned between more than 80 hours per 
        month and unemployment or labor force nonparticipation; third, 
        other. ``Other'' includes those who transitioned between less 
        than 80 hours per month and unemployment or labor force 
        nonparticipation as well as those who transitioned between 
        unemployment and labor force nonparticipation.

    Figures 4a and 4b. Most Frequent Reason for Not Working for Pay, 
SNAP Participants

          Source: SIPP (U.S. Census Bureau 2013-14); authors' 
        calculations.
          Notes: The sample is limited to U.S. citizens, non-active 
        military, aged 18-59 who reported receiving SNAP benefits at 
        any point between January 2013 and December 2014. Only 
        respondents with 24 months of data were included. Those with 
        children under age 6, full- or part-time students, and those 
        who reported receiving disability benefits were excluded from 
        the sample based on categorical work requirement exclusions. 
        Figure 4a is further restricted to those between the ages of 18 
        and 49 with a dependent between the ages of 6 and 17 while 
        figure 4b is limited to those between the ages of 50 and 59 
        with no dependents under the age of 6. Each person's most 
        frequent response for why they were not working was used to 
        calculate the distribution; ties were assigned in descending 
        order by work-related, health or disability, caregiving, 
        student, early retirement, not interested in working, and 
        other. The ``stable work, not asked'' group indicates that the 
        respondent was never asked this survey question because they 
        were working for pay every week. ``Work-related'' includes not 
        being able to find work, being laid off, or working for more 
        than 15 hours for no pay at a family business or farm. ``Health 
        or disability'' includes being unable to work because of an 
        injury, illness, or chronic health condition or disability. 
        ``Caregiving'' includes those not working due to pregnancy or 
        recent childbirth, or taking care of children or other persons. 
        Students included in the sample are those who did not report 
        that they were enrolled full- or part-time but reported not 
        working because they were going to school.

    Figure 5. Employment Status over Two Years, Medicaid Participants

          Source: SIPP (U.S. Census Bureau 2013-14); authors' 
        calculations.
          Note: The sample is limited to U.S. citizens, non-active 
        military, aged 18-64 who reported receiving Medicaid benefits 
        at any point between January 2013 and December 2014. Only 
        respondents with 24 months of data were included. Those with 
        children under age 6, full- or part-time students, those who 
        reported receiving Medicare, and those who reported receiving 
        disability benefits were excluded from the sample based on 
        categorical work requirement exclusions. See technical appendix 
        entry for figure 3 with regard to employment status assignment.

    Figures 6a and 6b. Most Frequent Reason for Not Working for Pay, 
Medicaid Participants

          Source: SIPP (U.S. Census Bureau 2013-14); authors' 
        calculations.
          Note: The sample is limited to U.S. citizens, non-active 
        military, aged 18-64 who reported receiving Medicaid benefits 
        at any point between January 2013 and December 2014. Only 
        respondents with 24 months of data were included. Those with 
        children under age 6, full- or part-time students, those who 
        reported receiving Medicare, and those who reported receiving 
        disability benefits were excluded from the sample based on 
        categorical work requirement exclusions. Those who were stable 
        labor force non-participants are contrasted with those who were 
        in the labor force (working or seeking work) at least once 
        during the 2 year period. Figure 6a is further restricted to 
        those between the ages of 18 and 49 with a dependent between 
        the ages of 6 and 17, whereas figure 6b is limited to those 
        between the ages of 50 and 64 with no dependents under the age 
        of 6. See technical appendix entry for figures 4a and 4b with 
        regard to reason assignment.

    Figure 7. Employment Status in One Month vs. Two Years, SNAP

          Source: SIPP (U.S. Census Bureau 2013-14); authors' 
        calculations.
          Note: The sample is limited to U.S. citizens, non-active 
        military, aged 18-59. Only respondents with 24 months of data 
        were included. Those currently exposed to work requirements, 
        those with children under age 6, full- or part-time students, 
        and those who reported receiving disability benefits were 
        excluded from the sample. The 1 month and 2 year samples differ 
        by reported SNAP benefit receipt. In the 1 month sample, 
        ``other'' refers to those who switched between labor force 
        nonparticipation and unemployment during the month of December 
        2013, the month chosen in the SIPP by CEA for its report on 
        work requirements.

    Figure 8. Employment Status in One Month vs. Two Years, Medicaid

          Source: SIPP (U.S. Census Bureau 2013-14); authors' 
        calculations.
          Note: The sample is limited to U.S. citizens, non-active 
        military, aged 18-64. Only respondents with 24 months of data 
        were included. Those with children under age 6, full- or part-
        time students, those who reported receiving Medicare, and those 
        who reported receiving disability benefits were excluded from 
        the sample based on categorical work requirement exclusions. 
        The 1 month and 2 year samples differ by reported Medicaid 
        benefit receipt. In the 1 month sample, ``other'' refers to 
        those who switched between labor force nonparticipation and 
        unemployment during the month of December 2013.
Advisory Council

 
 
 
George A. Akerlof, University Professor, Georgetown University
Roger C. Altman, Founder & Senior Chairman, Evercore
Karen L. Anderson, Senior Director of Policy & Communications, Becker
 Friedman Institute for Research in Economics, The University of Chicago
Alan S. Blinder, Gordon S. Rentschler Memorial Professor of Economics &
 Public Affairs, Princeton University; Nonresident Senior Fellow, The
 Brookings Institution
Robert Cumby, Professor of Economics, Georgetown University
Steven A. Denning, Chairman, General Atlantic
John M. Deutch, Institute Professor, Massachusetts Institute of
 Technology
Christopher Edley, Jr., Co-President & Co-Founder, The Opportunity
 Institute
Blair W. Effron, Partner, Centerview Partners LLC
Douglas W. Elmendorf, Dean & Don K. Price Professor of Public Policy,
 Harvard Kennedy School
Judy Feder, Professor & Former Dean, McCourt School of Public Policy,
 Georgetown University
Roland Fryer, Henry Lee Professor of Economics, Harvard University
Jason Furman, Professor of the Practice of Economic Policy, Harvard
 Kennedy School;Senior Counselor, The Hamilton Project
Mark T. Gallogly, Cofounder & Managing Principal, Centerbridge Partners
Ted Gayer, Executive Vice President Joseph A. Pechman Senior Fellow,
 Economic Studies, The Brookings Institution
Timothy F. Geithner, President, Warburg Pincus
Richard Gephardt, President & Chief Executive Officer, Gephardt Group
 Government Affairs
John Gray, President & Chief Operating Officer, Blackstone
Robert Greenstein, Founder & President, Center on Budget and Policy
 Priorities
Michael Greenstone, Milton Friedman Professor in Economics & the College
 Director of the Becker Friedman Institute for Research in Economics;
 Director of the Energy Policy Institute, University of Chicago
Glenn H. Hutchins, Co-founder, North Island
James A. Johnson, Chairman, Johnson Capital Partners
Lawrence F. Katz, Elisabeth Allison Professor of Economics, Harvard
 University
Melissa S. Kearney, Professor of Economics, University of Maryland;
 Nonresident Senior Fellow, The Brookings Institution
Lili Lynton, Founding Partner, Boulud Restaurant Group
Howard S. Marks, Co-Chairman, Oaktree Capital Management, L.P.
Mark McKinnon, Former Advisor to George W. Bush; Co-Founder, No Labels
Eric Mindich, Chief Executive Officer & Founder, Eton Park Capital
 Management
Alex Navab, Former Head of Americas Private Equity, KKR; Founder, Navab
 Holdings
Suzanne Nora Johnson, Former Vice Chairman, Goldman Sachs Group, Inc.
Peter Orszag, Vice Chairman & Global Co-Head of Healthcare, Lazard;
 Nonresident Senior Fellow, The Brookings Institution
Richard Perry, Managing Partner & Chief Executive Officer, Perry Capital
Penny Pritzker, Chairman, PSP Partners
Meeghan Prunty, Managing Director, Blue Meridian Partners Edna McConnell
 Clark Foundation
Robert D. Reischauer, Distinguished Institute Fellow & President
 Emeritus, Urban Institute
Alice M. Rivlin, Senior Fellow, Economic Studies, Center for Health
 Policy, The Brookings Institution
David M. Rubenstein, Co-Founder & Co-Executive Chairman, The Carlyle
 Group
Robert E. Rubin, Former U.S. Treasury Secretary, Co-Chair Emeritus,
 Council on Foreign Relations
Leslie B. Samuels, Senior Counsel, Cleary Gottlieb Steen & Hamilton LLP
Sheryl Sandberg, Chief Operating Officer, Facebook
Diane Whitmore Schanzenbach, Margaret Walker Alexander Professor,
 Director, The Institute for Policy Research, Northwestern University;
 Nonresident Senior Fellow, The Brookings Institution
Stephen Scherr, Chief Executive Officer, Goldman Sachs Bank USA
Ralph L. Schlosstein, President & Chief Executive Officer, Evercore
Eric Schmidt, Technical Advisor, Alphabet Inc.
Eric Schwartz, Chairman & CEO, 76 West Holdings
Thomas F. Steyer, Business Leader & Philanthropist
Lawrence H. Summers, Charles W. Eliot University Professor, Harvard
 University
Laura D'Andrea Tyson, Professor of Business Administration & Economics
 Director, Institute for Business & Social Impact, Berkeley-Haas School
 of Business
Jay Shambaugh, Director
 

Abstract
    Basic assistance programs such as the Supplemental Nutrition 
Assistance Program (SNAP, formerly the Food Stamps Program) and 
Medicaid ensure families have access to food and medical care when they 
are low-income. Some policymakers at the Federal and state levels 
intend to add new work requirements to SNAP and Medicaid. In this 
paper, we analyze those who would be impacted by an expansion of work 
requirements in SNAP and an introduction of work requirements into 
Medicaid. We characterize the types of individuals who would face work 
requirements, describe their labor force experience over 24 consecutive 
months, and identify the reasons why they are not working if they 
experience a period of unemployment or labor force nonparticipation. We 
find that the majority of SNAP and Medicaid participants who would be 
exposed to work requirements are attached to the labor force, but that 
a substantial share would fail to consistently meet a 20 hours per 
week-threshold. Among persistent labor force non-participants, health 
issues are the predominant reason given for not working. There may be 
some subset of SNAP and Medicaid participants who could work, are not 
working, and might work if they were threatened with the loss of 
benefits. This paper adds evidence to a growing body of research that 
shows that this group is very small relative to those who would be 
sanctioned under the proposed policies who are already working or are 
legitimately unable to work.
Figure 3.
Employment Status over Two Years, SNAP Participants

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: Survey of Income and Program Participation (SIPP) 
        (U.S. Census Bureau 2013-14); authors' calculations.

    The Chair. Thank you all for your testimony.
    We will now begin questioning. Members will be recognized 
for questioning in the order of seniority for Members who were 
here at the beginning of the hearing. After that, Members will 
be recognized in order of their arrival.
    I will now yield 5 minutes to Mr. McGovern.
    Mr. McGovern. Thank you very much, and thank you for your 
testimony.
    Last Congress, Republicans held 23 hearings on SNAP, and I 
disagree with Mr. Johnson. We didn't talk about this issue at 
all. In fact, they intentionally avoided a hearing on this. I 
requested a hearing and it never happened, and I now know why, 
because the bottom line is the ABAWD population is a 
complicated population. It doesn't fit into a nice, neat 
category that you can stigmatize, that you can demonize. This 
is a population that includes returning veterans, people with 
limited educational experiences, some who are aging out of 
foster care, people who have undiagnosed mental illnesses, 
people who live in rural areas who don't have access to 
transportation. I mean, there are lots and lots of issues 
involved in this population.
    And I should also point out for the record that the 
majority of able-bodied adults on SNAP right now actually work. 
They earn so little they still qualify for SNAP. And the notion 
that somehow this population is just lazy and just wants to 
benefit from this benefit, I will remind people that the 
average SNAP benefit is about $1.40 per person per meal, and 
so, it doesn't provide very much of anything.
    I asked Secretary Perdue when he was here in February to 
provide me the demographic data that the USDA used to justify 
this new rule. I have received nothing to date. Maybe it is 
lost in the mail, but hopefully we will get that at some point. 
But my frustration is we passed a farm bill and it was a 
bipartisan farm bill. It rejected all the cruel provisions that 
were contained in the House bill, but it passed overwhelmingly 
when it came back to the House, and yet, we have the 
Administration ignoring what Congress decided, which is 
frustrating.
    I want to also point out that in terms of the consequences 
of what the Administration is trying to do. I mean, Maine 
tightened up on the work requirements. There was an article in 
The Washington Post in May of 2017. Let me read the beginning 
of it. It said, ``For a period last year after he lost his food 
stamps, Tim Keefe, an out-of-work and homeless Navy veteran, 
used his military training to catch, skin and eat squirrels, 
roasting the animals over an open fire outside the tent he 
pitched in frigid Augusta, Maine. The new additions to Keefe's 
diet resulted from a decision by state authorities to tighten 
work requirements for recipients of the social safety net--
forcing the 49 year old who lost his job at a farm equipment 
factory because of an injury, off the food stamp rolls.'' I 
mean, this is the kind of stuff that can happen if we are not 
thoughtful about how we approach this issue. Yes, we want to 
help people get into the workforce. We ought to be investing in 
worker training, education, and transportation. There are a 
whole bunch of things we should be doing. Not cutting of their 
benefit because they find themselves in a difficult 
circumstance.
    I don't know how cutting off somebody's SNAP benefit is 
going to make it easier for them to get to a job where there is 
no transportation or somebody who, again, has an undiagnosed 
mental illness, how that is going to help them get into the 
workforce. This is a simple-minded approach to a complicated 
problem, I believe it is red meat to the right-wing base, who 
it seems, never tires of demonizing this population.
    Let me ask Ms. Hamler-Fugitt, Ms. Cunnyngham, and Dr. 
Shambaugh, I understand that a lot of your work has explored 
the complexities that arise from ABAWDs and low-income workers. 
Just for the record, again, do you believe that low-income 
persons who work less than 20 hours a week do so by choice, is 
this something they desperately want to do, or is it because of 
disadvantages?
    Ms. Hamler-Fugitt. To the Chair, to the Congressman, today 
in America, a job doesn't mean a living. There are people that 
work and they work hard; but, unfortunately in the current 
economy, jobs don't provide full time benefits. The folks who 
are part of our program, they work, they want a job.
    I would also point out that we have done longitudinal 
studies of the levers in our program. What we found is that 
when they do work, they generally work less than 30 hours a 
week for about $10 an hour with no benefits, and the average 
length of employment is 79 days, which is very interesting. 
Seventy-nine days would not trigger their eligibility for 
unemployment compensation. SNAP is a hunger lifeline for these 
individuals. A hungry worker is not a healthy worker, is not a 
productive worker.
    Dr. Shambaugh. I would just say, Congressman, not only do 
they want to work, but the evidence shows that most of them do 
work. And so, most of them are cycling in and out of jobs and 
the small portion who are not cycling in and out of jobs 
typically face significant barriers, health and otherwise, to 
work.
    Ms. Cunnyngham. I examine evidence and it is true that we 
know that, as you mentioned, that SNAP participants cycle in 
and out of work, that time limits can--well, to directly answer 
your question, yes, I believe people want to work and the 
evidence shows that most of them do work.
    Mr. McGovern. Thank you.
    The Chair. Thank you. The gentleman's time has expired.
    Ranking Member Johnson, you are now recognized.
    Mr. Johnson. Thank you, Madam Chair.
    Mr. Adolphsen, remind me. You had analyzed 600,000 ABAWDs 
in maybe three states. What were those three states?
    Mr. Adolphsen. Florida, Kansas, and Arkansas.
    Mr. Johnson. In Florida, Kansas, and Arkansas, did your 
research indicate that there were SNAP recipients who had 
learning disabilities?
    Mr. Adolphsen. Well by definition, an ABAWD is not 
disabled, but then the Department when they intake that person, 
particularly when the requirements are in place, will screen 
them for those types of barriers to make sure that they are 
directed to a place where they can get assistance for those.
    Mr. Johnson. In Florida, Kansas, and Arkansas, were there 
SNAP recipients who were not fully literate?
    Mr. Adolphsen. I think that is probably likely. With 
600,000 sample size, there were people that had issues that the 
Department would work with them on to help them get back to 
work.
    Mr. Johnson. In Florida, Kansas, and Arkansas, do you 
believe that there were SNAP recipients within your studied 
population who were caregivers for family members at home?
    Mr. Adolphsen. If they are actually responsible for the 
care of a child, they wouldn't be considered an ABAWD, so they 
wouldn't have been in that population.
    Mr. Johnson. In Florida, Kansas, and Arkansas, do you 
believe that there were returning veterans within the studied 
population?
    Mr. Adolphsen. Certainly.
    Mr. Johnson. Do you believe within Florida, Kansas, and 
Arkansas that within the studied population there would be 
people who lacked access to reliable transportation?
    Mr. Adolphsen. Yes, there certainly would be. For the folks 
that left the program, they would have had to earn more income 
and then come off the program. Because if they lacked 
transportation and therefore could not get to a job or 
training, they wouldn't have been in the group that we studied 
that left the program, because they would be exempt from the 
work requirement.
    Mr. Johnson. I guess I'm a little confused, sir. If the 
populations within those states--they sound an awful lot like 
the populations in my state and an awful lot like the 
populations in the states that the other witnesses described. I 
thought you said that there were successes for those 
populations that moved off the program? Did I misunderstand 
you?
    Mr. Adolphsen. No, absolutely, and I think that is what is 
concerning about the way the waivers operate right now is you 
have states doing very well with implementing the work 
requirement fully, in Florida, for example. And then you go to 
California and you have nearly one million people who have no 
work requirement. They are similar populations. They are 
similar income levels. They are the same age, same type of 
household situation. The waiver makes for a very uneven 
application of the program rules.
    Mr. Johnson. This has worked in some states, is that what 
your research indicates?
    Mr. Adolphsen. Absolutely. It has been very effective.
    Mr. Johnson. Give me a sense of the types of support that 
exist, things that states can do to help ABAWDs find meaningful 
employment?
    Mr. Adolphsen. Sure. There is a whole number of things. The 
Federal Government funds employment and training portions of it 
at 100 percent. States can get a 50/50 match for things like 
job search, education, job training. They can get funding for 
transportation, even buy equipment, things like boots if they 
need those for their job. States really can be very hands-on in 
helping people.
    The challenge is when you don't have the requirement in 
place, it also waives the requirement for the government to 
help them. They are simply sending them the benefit month after 
month, maybe checking in once a year for recertification. But 
they never really engage with them to find out where they can 
help.
    Mr. Johnson. Well, this is a population that clearly does 
have challenges, and no one should dispute that. I think you 
did a nice job, sir, of explaining that in the states you 
studied, there were people who had challenges. And I know my 
friends--Patty who works as a retail clerk in Mitchell who has 
some serious barriers. Mike and Paul, my friends who work in 
Mitchell, they have some challenges that I don't have. They 
have found meaningful employment. It is a meaningful part of 
their life. It is important that we remember that nobody is 
denigrating these folks. Nobody is suggesting that their path 
forward is easy. We are called to do an even better job than we 
are doing in helping them and work is a critically important 
part of that process.
    I want to thank you for your research, sir. I yield back, 
ma'am.
    The Chair. Thank you. Ms. Adams is recognized for 5 
minutes.
    Ms. Adams. Thank you, Madam Chair, and thank you to the 
Ranking Member as well, and to the individuals who are here 
testifying. Thank you.
    Before I begin, I just want to reiterate that USDA is 
unilaterally changing rules around requirements for ABAWDs, 
despite Congress' negotiating a farm bill at the end of 2018 
which explicitly avoids changing these requirements.
    While North Carolina does not currently have a waiver, 
lawmakers in my state are assessing the need to authorize 
waivers for some counties that meet current requirements. But 
you know, because states can do doesn't mean they will do or 
that they are able to.
    But I am deeply concerned that the proposed changes will 
take away needed flexibility for my state to help communities 
and individuals who are struggling with unemployment, opioid 
addiction, and other barriers to work.
    Ms. Hamler-Fugitt, Mecklenburg County is a part of my 
district. It is an area with a strong economic and population 
growth, and even in our county, we have more than 7,500 ABAWD 
individuals who are unable to find work, full-time work, and 
they are receiving SNAP benefits. If you can imagine the 
countless regions, especially in rural areas, that are seeing 
years of stagnant growth and continue to have high 
unemployment. The lack of access to work and the chronic 
barriers to work that many of these individuals face are some 
of the reasons that USDA estimates that 755,000 people will 
lose food assistance. And so, the rule would really force 
people who haven't been able to find work to enroll in E&T, or 
somehow find work when it has been impossible before.
    With your experience in this space, do you think that it 
makes sense to ask E&T Programs to do more than double their 
enrollment in programming with no funding, and what do you 
think the impacts will be to the quality of training?
    Ms. Hamler-Fugitt. To Congresswoman Adams, I can assure you 
that based on our firsthand experience in Ohio, slots just 
don't materialize. You have to have an infrastructure that is 
set up by the county or the state that does require funding as 
well to provide these services, and then the support services 
to get individuals into those work and training slots and 
ensuring that if they are available, that they are being 
trained for jobs that are currently available. It is a very 
expensive endeavor, and I have just done some numbers based on 
Ohio. If Ohio were to lose its current waivers in 35 counties, 
an estimated about 75,000 individuals would have to have a slot 
made available to them. The cost associated with that would be 
about $600 million. On average, the cost of a good Employment 
and Training Program varies from $4,500 on the low end to about 
$12,000 on the high end.
    Ms. Adams. Thank you very much. Let me just move onto ask 
Ms. Cunnyngham and Dr. Shambaugh, according to the National 
Education Association, more than 2.5 million children are being 
raised by their grandparents or other relatives because 
families are dealing with parental alcohol, substance abuse 
issues, and others. As a result, they face obstacles in 
securing an exemption from ABAWD time limits. How do you see 
this rule impacting those struggling with opioid and other 
forms of addiction, and do you expect that there will be 
unexpected consequences for the children? This is to Dr. 
Shambaugh and Ms. Cunnyngham.
    Dr. Shambaugh. Well, as has already been mentioned today, 
we know that there are people who are in the ABAWD population 
who sometimes are taking care of children, and so they are not 
considered a caregiver because they are not the primary 
caregiver in some cases, but they have some responsibility for 
children. Having them lose SNAP benefits would take resources 
away from a family with kids. I think that is something that is 
certainly is a concern.
    Ms. Adams. Okay. Ms. Cunnyngham?
    Ms. Cunnyngham. Sure. Well, another population that would 
be affected are non-custodial parents, so not only 
grandparents, but if you are looking for non-custodial parents 
to contribute to their children's well-being, it is important 
that they have a job. It is important that they are supported 
while looking for a job.
    Ms. Adams. Unexpected consequences. I have about 23 
seconds.
    Ms. Cunnyngham. Unexpected consequences, non-custodial 
parents are critical to the well-being of their children, and 
so we want to support them.
    Ms. Adams. Thank you. Madam Chair, I yield back.
    The Chair. Thank you very much. Mr. DesJarlais, you are 
recognized for 5 minutes.
    Mr. DesJarlais. Thank you, Madam Chair.
    We live in a country that takes care of people who can't 
take care of themselves, and that is the right thing to do. The 
SNAP Program is there to help people who are hungry, make sure 
that they have food in their mouths and that hunger is reduced, 
and it is our responsibility in Congress to make sure that we 
have the funding to take care of those in need.
    Thankfully, our economy is doing much better. Unemployment, 
as mentioned in your oral statement, Mr. Adolphsen, is at 50 
year lows. In fact, in my home State of Tennessee, about the 
only thing holding back economic growth is an adequate 
workforce. It is fortuitous that we are having this hearing 
today that Mr. McGovern had asked for.
    It seems like we probably agree on a lot of things that we 
are just not even seeing here. The fact that able-bodied people 
who can work, should work, is a pretty common concept. I asked 
Secretary Perdue when he was here last month if he had any idea 
why there is so much pushback to this idea, and he responded, 
``I have no clue.'' I can understand that when you look 
statistically that across all political spectrums, about 80 
percent of people believe that able-bodied people who can work, 
should work. We just have to figure out how to get that done.
    And I guess I would like to ask our witnesses, in concept, 
do you agree that able-bodied people who can work, should work? 
We will start at this end.
    Ms. Cunnyngham. Yes, I agree it is best for everyone, for 
the individuals.
    Mr. DesJarlais. Okay. Mr. Adolphsen?
    Mr. Adolphsen. Yes, I do.
    Ms. Hamler-Fugitt. Yes, sir.
    Dr. Shambaugh. Yes, I do.
    Mr. DesJarlais. That is a great place to work from. We all 
agree that that should happen, and the problem is, how do we do 
that? And we have heard of all kinds of barriers that stand 
before us, it is our job to solve that.
    Mr. Adolphsen, within the framework of the existing funding 
for employment and training, what changes have been made to 
increase the effectiveness of those funds, or how are we making 
it easier for people to find work and transition into the 
workforce?
    Mr. Adolphsen. Sure. Well, the funds in programs that you 
mentioned, Congressman, are really critical, and we can't leave 
that piece out. We put a lot of money into employment and 
training, not just in the SNAP Program, but also across the 
board. As was mentioned, we have adult ed, community colleges, 
Department of Labor career one-stops. There are a lot of 
resources designed to help these precise individuals, and for 
them to really be effective, the key is that people actually 
utilize them. We have seen problems where when these programs 
are purely voluntary, there is a type of requirement in place, 
there has been very little participation. And what we really 
want to see is these folks utilizing all those great resources 
that are being provided.
    That is one of the key values of the work requirement being 
in effect is it connects them to those resources, where 
otherwise, we are just loading the EBT card every month and we 
are not really working with them on getting back to work.
    Some of the places where it has been really effective 
recently, there has been a real urgency on connecting folks to 
a job and taking that first step. We saw in Florida, in 
particular, a lot of people took an initial first step and 
maybe landed at a temp agency or in fast food or retail, but 
that was just the first step. They quickly moved on to higher 
paying industries with more wages. The best way to make these 
programs effective is to get them in the door.
    Mr. DesJarlais. All right. It seems that at these hearings 
we always hear the best case scenarios from one side and the 
worst case scenarios from the other side, and I have heard that 
here today. You can be the group that wants to talk about 
successes. You can be the group that wants to talk about 
failures. I want to be the group that talks about successes, 
and we want more successes.
    Some of the other witnesses have talked extensively about 
all the barriers for able-bodied adults joining the workforce. 
Do you agree that these barriers are as bad as they say, or are 
you more optimistic?
    Mr. Adolphsen. Congressman, my experience at the agency, 
that was one of the things that was most disappointing is the 
whole system was revolved around this point of view of what 
people can't do. And it is really important that we flip that, 
and when someone walks through our door asking for help with 
the food benefit, right away we say, ``All right, here is your 
benefit. What can we do to get you moving forward? What are you 
able to do?'' Where can we help you maybe remediate some 
skills, things we talked about. But we really need to come at 
it from the point of view that these individuals are very 
capable. They can work. They can improve their skills, their 
education, and meet them at that point instead of starting 
right off the bat by having a long list of everything they 
can't do, moving forward.
    Mr. DesJarlais. I would agree. I think that we need to try 
to succeed, not just accept failing. Thank you for your 
testimony, all of you for being here.
    I yield back.
    The Chair. Thank you. Ms. Schrier, you are recognized for 5 
minutes.
    Ms. Schrier. Thank you to all of our witnesses. Thank you, 
Madam Chair.
    I wanted to first just repeat some of the really 
interesting things that I heard, because they really deserve 
emphasis.
    I believe it was you, Dr. Shambaugh, who said work 
requirements keep people out of SNAP, but have little to no 
effect on employment. And I thought that was a really profound 
statement. Another that I believe you said, but others said as 
well, was that the able-bodied workers without dependents is 
really a misnomer, that it misses a lot of people with 
undiagnosed mental illness, learning disabilities, and people 
on the autism spectrum who really are not truly able-bodied. I 
thought another really important point was that the average 
benefit is a $1.40 per meal, and I think about how we are 
nickel and diming over this benefit that is providing nutrition 
to people in need.
    I heard that only 1\1/2\ percent truly don't want to work, 
which is a small number, and in my opinion, not enough to throw 
the baby out with the bath water, and that the unemployment 
rates just don't paint a true picture, because the skill sets 
needed may not match where the job openings are.
    I wanted to talk a little bit about the State of 
Washington. The State of Washington is firmly committed to 
improving the lives of those on SNAP through work, by helping 
beneficiaries become self-sufficient through good paying jobs. 
In fact, in 2018 alone, the state spent $22 million of its own 
money on top of Federal funds on SNAP Employment and Training 
Programs, otherwise known as SNAP E&T. And making this 
investment among the top five states in the country, and in 
addition, we were granted one of ten SNAP E&T pilots that were 
funded in the 2014 Farm Bill. In 2016, our Assistant Secretary 
for Economic Services, David Stillman, testified before this 
Committee about our state's successful Education and Training 
Program, and under his leadership, the best practices learned 
are now being shared with others throughout the country.
    We have also engaged employers like Amazon, Microsoft, 
Providence Health, and others to be active partners in this 
training program, and the bottom line is that that part is 
working, and we know what we are doing.
    Now, the proposed rule will completely undermine all of 
that work. Our governor agrees, and in fact, I would like to 
enter into the record Governor Jay Inslee's letter of comment 
on the proposed rule, because it talks about the devastating 
impact that this will have on our state's 91,000 ABAWDs.
    [The letter referred to is located on p. 162.]
    Ms. Schrier. I had a question for Ms. Hamler-Fugitt, which 
is this: In Washington, we estimate that more than 43 percent 
of our state's ABAWD population currently experience 
homelessness, disproportionately higher than the broader SNAP 
population in the country at 11 percent. Nearly 60 percent of 
our ABAWD population is suffering from behavioral or physical 
health conditions, including substance use disorder. For these 
individuals, the USDA proposal would do nothing to help them 
find work, while adding yet another obstacle in their way, 
which is food insecurity and hunger. And the proposed rule 
would not achieve the goal of promoting self-sufficiency and 
jobs. It would make it more difficult to find employment.
    How does this description fit your experience in Ohio?
    Ms. Hamler-Fugitt. To the Congresswoman, very similar, 
except Washington State is to be commended for its commitment 
of $22 million of state general revenue funds to expand your 
program. Again, Washington residents are very fortunate to have 
that kind of leadership. Unfortunately, it is the luck of the 
draw. Other states or counties, in our situation, a county 
devolved system, that is up to local county commissioners. I 
would say it is very similar.
    One final remark. What is really missing from this is a 
standardized set of data. We need to be measuring the same 
thing. The assessments should be standardized across all states 
on the information, upon intake. These are not social workers 
that are doing this intake. These are clerks. They are not 
qualified to make determinations about one's mental or physical 
disabilities.
    Ms. Schrier. I appreciate that comment, especially as a 
medical professional myself, that these can be difficult 
diagnoses to make, and that we need to consider that the rate 
of undiagnosed everything, including learning disabilities in 
certain populations.
    Dr. Shambaugh, I had a question for you, and I would love 
to hear your thoughts on this excerpt from Governor Inslee's 
letter. ``While the unemployment rate does provide essential 
data, it does not take into account a community's 
individualized workforce needs or that its residents may not be 
well-suited to find and keep locally available jobs due to a 
lack of housing, hard skills, certifications, and employers in 
Washington.''
    Dr. Shambaugh. I think that is exactly right, and one thing 
that is important to recognize is even when the aggregate 
unemployment rate is low, it doesn't mean it is low for all 
groups or in all areas. As mentioned, in some places it is as 
high as ten percent. When it first crossed under seven percent 
nationally, it was still 10.7 for people with less than a high 
school degree. There are some people are going to be struggling 
a great deal, even when the overall rate is low.
    Ms. Schrier. Thank you. I appreciate that. Thank you to all 
of you.
    The Chair. Thank you very much. I now recognize Mr. 
Hagedorn, for 5 minutes.
    Mr. Hagedorn. Thank you, Madam Chair, Ranking Member. I 
appreciate the opportunity. Thanks for this hearing, and thanks 
to the witnesses.
    This is an issue that has been important to me for over 30 
years. I used to work for Congressman Arlan Stangeland, who was 
a Member of this Committee, and he and Congressman Stenholm of 
Texas at that time introduced bipartisan Work for Welfare 
legislation in the 1980s. We had broad support. Couldn't get a 
vote from the Majority party at that time, but a few years 
later, Newt Gingrich and some others picked up the provisions 
of that bill and passed it three times. It finally made it to 
the President's desk where he signed it, and the Clinton/
Gingrich--however you want to put it--welfare reform bill was 
highly successful. We drove down the cost of government. We 
empowered people and got them back into private-sector work. 
Most people would recognize, including even President Clinton, 
that that was quite a success.
    But over time, those work requirements went away, and we 
have had some issues with waivers and things of that nature and 
some loopholes that need to be closed.
    In my district we went around and we talked to all sorts of 
thought leaders, including mayors and social workers and 
others, and time after time, I was told there are people on the 
sidelines who could work, but for a number of reasons, are not 
working. I have never used the word lazy. That comes from the 
other side. Sometimes there are impediments. They lose benefits 
and so forth as they get in the workforce, and we need to look 
at that.
    But the Chairman himself, Chairman Peterson, who 
interestingly enough succeeded my old boss, Congressman 
Stangeland, in 1990. He has made comments to say that these 
waivers aren't working, particularly in the State of Minnesota. 
And he has talked not just here in Committee, but in press as 
well. And I would ask any of you, does anybody disagree with 
the Chairman's comments as it refers to the State of Minnesota 
and these waivers? Anyone? I guess not. Okay, we have some 
uniformity on that.
    What I would say is this: Work for Welfare is a concept 
that is empowering for people. It is a fairness issue for the 
taxpayers, because if people are able-bodied to work, they 
should do so, just like the taxpayers do to make it possible. 
It eliminates fraud. It drives down the cost of benefits.
    Now, if we want to talk in a serious way about helping 
folks, let's talk about the concept of transition wages. When 
people are moving from welfare into the workforce and they 
reach that cliff where they are going to lose medical benefits 
and other things, how do we transition them to keep them in 
private-sector work so they can continue to be upwardly mobile? 
That, to me, is important. I hope the Chair and others will 
work with me on those issues.
    Technical training: there are lots of jobs out there just 
begging, and we have to look at people's potential. That is the 
highest calling. What is the potential of each individual? Not 
just able-bodied folks, disabled people, people that want to be 
in the workforce and contribute. What is their potential? We 
have to have confidence in them and do whatever we can in order 
to promote that.
    And last, the Secretary was here a few weeks ago. He talked 
about this regulation, and I told him in my opinion--it was on 
the record--that it is God's work because it is moving people 
in the right direction. It is showing confidence in folks and 
it is not allowing states to cut them short and to do that.
    And I just say to my friends on the other side, if you 
don't like this regulation, you think the Executive Branch has 
gone too far, then join us in enacting the REINS Act. Let's 
make sure that every regulation coming out of the Executive 
Branch has to receive the vote of the House and the Senate and 
be affirmed before it goes live. That is only fair for all 
industries.
    With that, I would yield back.
    The Chair. Thank you very much. Mrs. Hayes, you are 
recognized for 5 minutes.
    Mrs. Hayes. Thank you, Madam Chair.
    In my home State of Connecticut, we have 107 towns, 
including 13 towns in my own district, that would lose the 2018 
waivers proposed under this rule. This would translate to about 
26,000 people in Connecticut automatically becoming vulnerable 
to losing the Federal help they need to simply put food on 
their tables.
    I will say to my friends on the other side that I was a 
SNAP beneficiary. I worked two jobs, was grossly underemployed, 
and was a full-time college student. It was temporary. It took 
much longer than 3 months, but it was temporary. While this 3 
month time limit for SNAP benefits for ABAWDs in theory should 
only impact adults who do not have children, in practice, it 
will inevitably and disproportionately impact children and 
young people.
    As a teacher, this damaging impact is personal for me. SNAP 
is the first line of defense amongst childhood hunger. It is 
also the first line of defense against economic instability and 
hunger for young people, especially for the 20,000 kids aging 
out of the foster care system every year.
    Madam Chair, I ask to submit two letters into the record, 
one from the NEA that talks about the effects on young children 
in the classroom, and the other one from Share Our Strength and 
No Kid Hungry.
    The Chair. Without objection.
    [The letters referred to are located on p. 158 and p. 159.]
    Mrs. Hayes. I have seen what a hungry 16 year old looks 
like, and it is not much different than a hungry 19 year old. 
Hunger is merely a symptom of poverty. The reality is that very 
few ABAWD recipients of SNAP are not interested in working. 
Rather, they are desperately underemployed, undereducated, or 
in low wage work that is highly unstable.
    The unemployment rate for young adults is nearly seven 
percent. According to a national survey of youth who aged out 
of the foster care system, only 12 percent were employed full 
time at age 19. Forty-four percent had not obtained a high 
school diploma or GED equivalent at age 19, leaving them at a 
significant disadvantage in seeking stable employment and 
livable wages.
    Instead of punishing people for being poor or for being in 
foster care, we need to further invest in job training and 
education as a way out of poverty so that people can, in fact, 
help themselves.
    By ripping away a lifeline of an already vulnerable 
population, this Administration is making yet another 
unconscionable attack on young people and poor people. This 
Administration is telling this population, one that has already 
struggled enough, that they don't deserve help the day they 
turn 18 and age out of the foster care system, that they don't 
deserve compassion from the Federal Government, and that they 
don't even deserve a hot meal.
    My question is for you, Dr. Shambaugh. What are the long-
term social and economic cost effects of ripping away this 
safety net for food security for young people aging out of the 
foster care system?
    Dr. Shambaugh. Thank you for the question. We have very 
good evidence, certainly for children in particular, that 
spending time in a household with SNAP relative to people in 
the same economic situation without SNAP benefits has 
substantial positive impacts on health, income, and earnings 
later in life. I think that would probably translate that type 
of result to that next population just a few years older, as 
you mentioned.
    We know that in many ways, you are making investments that 
make people more employable, better workers, healthier later in 
life, and in that sense, you wind up saving money in the long 
run.
    Mrs. Hayes. Well, I appreciate that because I have seen 
that. Because, in my experience, these young people do cycle 
back into the system, and it is not always with SNAP benefits. 
It is the criminal justice system. It is the unemployment 
system. It is all of these other programs that could have been 
prevented if the investment was made on the front end to help 
them to support themselves.
    Ms. Hamler-Fugitt, can you think of ways that we can 
increase outreach so that foster youth or young people who were 
formerly in foster care know about the benefits that are 
available to them and are better positioned to help themselves?
    Ms. Hamler-Fugitt. Congresswoman, we need to understand 
that as they are in the foster care system, that they are 
getting the kind of life skills and support necessary. They are 
being exposed to opportunities of higher education and 
training. We need to make sure that there are also transitional 
benefits.
    I know our state is continuing to do some more support 
around a bridges program that will provide that transition, but 
again, a lot of these children as they age out of the foster 
care system, they are dumped on their 18th birthday onto the 
street. That is wrong. I know for a fact that we have spent, in 
some cases, hundreds of thousands of dollars getting these 
young people to adulthood, only to turn our back on them. We 
can do better. I know we can.
    Mrs. Hayes. We can do better.
    Thank you, Madam Chair. I yield back.
    The Chair. Thank you. Mr. Bacon, you are recognized for 5 
minutes.
    Mr. Bacon. Thank you, Madam Chair, and I appreciate our 
four folks here today. I appreciate you sharing your 
perspective. It is a very important topic.
    I just want to take a moment to thank Chairman Peterson who 
has weighed in on this, in recent years, talking about some of 
the abuses of the state waiver process. I appreciate his 
candor, especially his candor coming from the other side of the 
aisle.
    The ABAWDs are just that. We are talking about able-bodied 
adults without young children, and in a time of record low 
unemployment, 2.7 percent in Nebraska, where we have more job 
openings than people seeking jobs right now, there is an 
expectation by most for folks to seek that work and there is an 
expectation of having some time limits within the SNAP Program, 
which there are. There is that expectation there that we should 
try to enforce that to the best of our ability.
    And so, my question for Mr. Adolphsen, if I may. In your 
studies, have you seen a contrast of those states who are 
enforcing these time limits versus those who are not or they 
are doing the waivers, and helping the ABAWDs get out of 
poverty? Is there some direct correlation between these 
individuals who are struggling, but once they get back in the 
workforce, how does that affect their prosperity or their 
getting out of poverty? Thank you.
    Mr. Adolphsen. Sure. Thank you, Congressman, for the 
question.
    What we looked at with those more than \1/2\ million adults 
that we tracked individually with their incomes, their incomes 
within a year doubled. In Arkansas, we had another year of data 
and their incomes tripled within the 2 year period. What you 
have is they are actually moving not just up in income, but out 
of poverty. The amount they were earning actually replaced the 
benefitted amount as well. In total, they were earning more 
than they were before, even when you count the benefit. We are 
seeing real upward economic mobility from that population.
    You contrast that with states where the waiver is in place, 
and there is no work requirement for this population. They are 
still on the program. By definition, they are still in poverty, 
right, they are in that income bracket that would keep them 
eligible. We know that three out of four of them aren't 
working. We are not seeing that same movement in those states.
    Mr. Bacon. Well, thank you. Really, full time work is the 
best way out of poverty, and often a year or 2 later, there are 
raises and promotions. It is the first step for getting people 
out of that poverty area.
    I just want to point out in Nebraska, we have such a 
shortage in some of our more technical career fields, whether 
it is welding, electrical work. There is a shortage of folks 
and they are being offered $40,000 a year jobs in training, 
while they are in training, with health insurance, just to get 
them started. We are having a hard time filling those 
positions.
    Thank you very much. I yield back, Madam Chair.
    The Chair. Thank you very much. Mr. Lawson, you are 
recognized for 5 minutes.
    Mr. Lawson. Thank you, Madam Chair, and welcome, witnesses, 
to the Committee.
    A perception of people who are receiving SNAP benefits that 
extend across America and dealing with the issue when I was in 
the state legislature to come into Congress was that people 
were lazy, they did not want to work, they just wanted 
government assistance.
    But when you go a little bit deeper into the situation, you 
find out that this is not true. In listening to your testimony 
this morning, one would think that because of all of the 
knowledge that you all have in dealing with people who are 
receiving SNAP benefits, that we could possibly learn a great 
deal as lawmakers about how the programs really should be 
established instead of some people saying what programs should 
be without any knowledge of it.
    I represent an area that has two major urban areas, and the 
rest of it is rural, and the rural community extends maybe for 
a distance of about 150 miles between from Gaston County--I 
would like to say from the Chattahoochee River to the St. 
John's River in Jacksonville. And in those areas, there is high 
unemployment and transportation issues where people have 
difficulty in trying to get into the city. And Florida in 
itself doesn't utilize the waivers.
    Now my question would be simply that food insecurity within 
college student population, which you know, I have a lot of 
student populations at universities and in the community 
colleges, is growing. And not only in the areas in Florida, but 
throughout the United States, especially at community colleges 
where many of these students are part-time and they are working 
and trying to make ends meet to really better themselves.
    Can you discuss any experience that you have had, Ms. 
Hamler-Fugitt, and am I saying that right, Fugitt? Okay. At 
Ohio Association of Foodbanks where part-time students have 
benefitted from their food stamps status while working hard to 
complete their degree?
    Ms. Hamler-Fugitt. Yes. To the Congressman, yes, that is 
one of the struggles. In fact, in my written testimony I talk 
about Mary, a young woman who is balancing both a pharmacy tech 
career track at a local community college, trying to work for a 
drugstore chain, trying to maintain that 20 hours of employment 
while also assisting her mother to care for her younger 
sisters. And unfortunately, she fell into a situation where 
because the sales weren't there in the store, she wasn't able 
to get her 20 hours that then put her SNAP benefits in peril, 
which then her lifelong dream of becoming a pharmacy tech, the 
first generation to graduate from high school or from college 
was left in the peril, where she was looking at having to drop 
out of college.
    I also want to say, as Ohio's largest charitable response 
to hunger, I am sure that your food banks are in a similar 
situation. The greatest demand we are currently having is for 
colleges and universities, both technical schools and 4 year 
institutions, as well as our K-12, to come on site and set up 
food pantries, not only to feed their students, but also to be 
able to feed the families of those students as well.
    Hunger is a problem in America. In my great state, it 
affects one in six. Hunger looks a whole lot like you and me, 
and it lives just six doors down.
    Mr. Lawson. That is incredible, and I noticed that many of 
the universities now are setting up food banks and working with 
some of the local grocery chains. One would think that once you 
are in college and you have this ambition to go to college, 
that there would be resources there with the financial aid so 
forth you get would help you through this. But most of those 
students who are also receiving financial aid and assistance--I 
know my time is running out--also have to rely heavily on food 
banks in the community as well as other people who are working 
in those communities.
    With that, Madam Chair, I yield back.
    The Chair. Thank you. Thank you very much.
    We will now recognize Mr. Davis.
    Mr. Davis. Thank you.
    The Chair. You are recognized for 5 minutes.
    Mr. Davis. Thank you, Madam Chair, and thanks to the panel.
    I am starting my fourth term here as a Member of the House 
Agriculture Committee, and we can probably all agree as 
Republicans and Democrats that the issue isn't about getting 
people off of SNAP benefits. The issue is about making sure 
that people who are on SNAP benefits have access to the jobs 
that we know are available in this country right now. And we 
can all agree, it may not be our Committee's jurisdiction, but 
there are some loopholes that still exist within our workforce 
training programs that perpetuate families staying on SNAP 
benefits, which is why we have right now nine million more 
people on SNAP benefits in this country today, when there are 
6.1 million jobs available, less than four percent 
unemployment, than when we had 9.5 percent unemployment.
    Today is a day that I certainly hope we can take your 
testimony and come together and try to find solutions. In the 
2018 Farm Bill, obviously a lot of us here in the House wanted 
to try and close what we saw as a loophole in our workforce 
investment programs by allowing for investment within SNAP 
education and training to allow families to go get training for 
jobs that we know are available in our communities. If we don't 
do it now when unemployment is at 3.9 percent, we are never 
going to do it when unemployment is at 9.5 percent. Help us 
come up with some solutions. Help us come up with some 
solutions to allow families to get money to go back and get 
training so they don't have to worry about what they can or 
cannot buy at the grocery store anymore, when they are doing 
everything they can to get a job in our communities that we 
know that are available.
    It is very frustrating to me that this debate becomes more 
about politics than clearly it does about policy. That is a 
very frustrating point for me as a legislator, and I certainly 
hope now that we have the other side in charge that we can come 
up with some solutions, because that loophole in our workforce 
investment system still exists. And if we do nothing, we are 
not helping those families who want to get off of SNAP 
benefits. That is my point.
    Being from Illinois, I also have an issue with the waiver 
process. Following the passage of the 2018 Farm Bill, I sent a 
letter to our then-Governor of Illinois, Bruce Rauner, asking 
what justification that the State of Illinois had to waive the 
ABAWD time limit in 101 of 102 counties. While individuals 
should not be penalized if jobs are unavailable, I inquired 
regarding what steps the state had taken to encourage the SNAP 
recipients to get training for employment. And the Governor's 
Administration at the time claimed a need for a waiver was due 
largely to administrative burdens, not out of any particular 
necessity in all of the 101 of 102 counties. Administrative 
governmental burdens.
    Again, it is cruel to do nothing to help a system recover. 
It is cruel not to help families get training for jobs that we 
know are available, even in the rural communities that I serve. 
That is what is cruel.
    Now, Mr. Adolphsen, I have a quick question for you. Why in 
the world a State like Illinois, where we have low 
unemployment, why in the world did our governor ask for a 
waiver for 101 of 102 counties?
    Mr. Adolphsen. Thank you, Congressman. You hit on it 
already. One reason is it is driven by the workload on their 
end, or perceived workload on their end. The other reason is 
simply to maximize the number of people that are waived. We 
have heard that explicitly from states. That is what they are 
doing. They are not looking strategically at where are people 
actually unable to get work or get into training. They are just 
looking at how do we maximize this waiver to cover as many 
people as possible so they don't have to be engaged, which is a 
problem. That is exactly what the rule tries to do, it is not 
getting rid of the waiver. The waiver exists. It is in Federal 
law. It is just making sure that it can only be used where it 
is actually targeted.
    On the administrative side, I heard the same thing in 
Maine. I have heard it in many other states. Well, that is a 
lot of work. Well first of all, that is the job of the 
government agency so yes, it is going to take some work. That 
is the job. On the other hand, it really hasn't proven to be an 
administrative burden in any way. The systems are all set up to 
do it because of the 1996 law, and it is work that they are 
already able to do.
    And I would just say, Congressman, quickly. I have heard a 
couple times the mention of college students. There are other 
ways to meet this requirement, other than working part-time. 
Individuals who are enrolled at least half time in any 
recognized school, training program, or institution of higher 
education are exempt from the requirement. That really doesn't 
come into play with this particular population or waiver.
    Mr. Davis. Great point. Thank you, and I yield back.
    The Chair. Thank you. Mr. Van Drew?
    Mr. Van Drew. I want to thank you, Madam Chair, and thank 
you all for being here today.
    This is obviously an issue of great concern all across the 
country in so many of our communities, and I find an 
interesting and intriguing part of the conversation is it seems 
as if we have almost gotten to a point we are saying we either 
are going to have programs to train individuals or we are going 
to give them SNAP benefits. And I don't really believe that is 
the issue. I believe the issue is that at certain points in 
people's lives, they need benefits in order to move on to the 
next point of their life, and that is really the hope and the 
desire of what we should be doing here.
    This proposed rule is going to have a major effect in many 
communities across the country. I come from an interesting 
district. I come from the State of New Jersey, and everybody 
always assumes because New Jersey is generally a wealthy state 
with a high per capita income that a lot of these issues don't 
exist. They exist in our urban areas up north, and they exist 
in my district, which is 40 percent of the state. I have 40 
percent of the state. I have eight counties and 92 towns, much 
of it rural, much of it seasonal, so we have a lot of shore 
communities. And what does that all mean? That means that a lot 
of folks don't have an easy opportunity to find access to full 
time good employment year-round. We certainly don't have the 
type of transportation that makes it easy, and we don't have 
some of the other amenities.
    I would point out in one of my deep south counties down in 
south Jersey, it was only a number of years ago that we got our 
first county college. It is different. It depends upon where 
you live. It depends on what the issues are. Unemployment is 
not unemployment everywhere. It is not the same. The numbers 
don't mean the same thing, and employment numbers don't mean 
the same thing everywhere.
    In northern New Jersey, if you are in the financial 
industry, that is a whole different thing if you are picking 
cranberries down in south Jersey. And everybody has to realize 
that and understand that.
    It isn't either/or. You can do both. We do have to train 
people. We do need more transportation. We do need more 
opportunity. Every single person up here wants everybody to 
work all the time. That is the goal. But in the process, people 
fall through the cracks and that is what we are talking about.
    Jobs just aren't always as easy to come by as some of the 
statistics show. One of the interesting parts of this is the 
Administration expects about \3/4\ of a million people to lose 
food assistance under this proposed rule, and probably would 
save, if my understanding is correct, about $15 billion in 
Federal savings from the cuts presented as a primary benefit of 
change. But what are the costs or what is the involvement going 
to be to those food pantries?
    We have food pantries and I deal with them and I visit 
them, and there are lots of good people who work real hard. 
They really are. They are hardworking people. They are just not 
making as much money, and they go to the food pantry and that 
helps them get through while they are trying to educate their 
kids more and they are trying to work harder, and they are 
working their two jobs, and God help them, sometimes three 
jobs. They are going to be getting hit more. They are going to 
have greater requirements upon them, and who is going to fill 
that? How are we going to take care of that? What is going to 
happen there? Or do you believe anything is going to happen at 
these food pantries and these types of facilities that exist 
out there? And I guess that is a question for Ms. Hamler-
Fugitt.
    Ms. Hamler-Fugitt. To the Chair, to the Congressman, food 
pantries can't do it. Our food banks and 3,540 member charities 
are already overwhelmed. Eighty percent of our member charities 
are faith-based organizations operating on budgets of less than 
$25,000 a year. They are overwhelmed with the demand, not only 
from working people who work every day and play by the rules 
but aren't earning enough to meet their basic needs, more 
senior citizens than we have ever seen, we are an aging state. 
They are the canary in the coal mine. More grandparents raising 
grandchildren, and now we place this additional burden on top 
of folks who have lost their SNAP benefits through no fault of 
their own because they can't find paid employment or work 
experience opportunity or SNAP Employment and Training Program. 
We can't do it. SNAP is the first line of defense against 
hunger in this country, not food banks.
    Mr. Van Drew. It is your considered opinion, then, that we 
are going to fall short? That they literally are not going to 
be able to keep up, and their shelves, at times, are going to 
be empty?
    Ms. Hamler-Fugitt. Yes, Congressman. It happens every day 
in your community and my community and across the U.S.
    Mr. Van Drew. And the last point--I know I am running out 
of time here, I am out of time. I might as well just admit it, 
right? Thank you, Chair. Oh, and Chair, also real quick, may I 
have unanimous consent to enter into the record Feeding 
America's comments on the proposed rule regarding able-bodied 
adults without dependents, and its impact on hunger and 
hardship?
    The Chair. Without objection.
    [The letter referred to is located on p. 346.]
    Mr. Van Drew. Thank you.
    The Chair. Thank you. Mr. Yoho?
    Mr. Yoho. Thank you, Madam Chair, and Dusty, I appreciate 
it. Thank you guys for being here. I know it has been a long 
day, and before I get started, Madam Chair, I would like to 
insert a February 19 letter signed by myself and 64 other 
Members in support of the Administration's proposed rule.
    The Chair. Without objection.
    [The letter referred to is located on p. 392.]
    Mr. Yoho. Thank you, ma'am.
    It shameful that politics gets involved in this, because it 
shouldn't. We are looking to reform programs that make America 
stronger. Our ultimate goal is we want a strong economy. We 
want strong job markets. We want people thriving and living the 
American dream, and I am not going to bore you with my story 
going from being broke as a church mouse to being on food 
stamps to where I am today.
    The programs are there for the people that truly need it, 
and we want to make sure the integrity of those programs are 
there. And Florida is one of the many states focused on work-
oriented reforms, and a new report shows the incredible impact 
they are having on our state, as you brought out, Mr. 
Adolphsen. Since the state implemented a food work requirement 
in 2016, nearly 94 percent of able-bodied childless adults have 
left Florida's food stamp program. Alabama saw 85 percent 
reduction. Maine saw more than 80 percent. And when we talk 
about able-bodied adults with no dependents, we are talking 
about a small group of people. We are looking for no physical 
disabilities, no mental disabilities, just a small section. And 
if we focus on that, what can we do? With these kinds of 
results, what can we do to implement these somewhere else in 
other states?
    And I guess the question for the panel is, was there any 
detriment that you know of people moving off in these states? I 
will take Florida, since that is where I am from. Can anybody 
say, ``Well, since 94 percent got off of that, there was this 
massive malnutrition or starvation''? Anybody? I will take that 
as a no. You were going to say something?
    Dr. Shambaugh. I would just say, we know what SNAP does to 
provide food security, and so, we know if not everyone is 
finding jobs, then there are people who are losing resources 
and are then left with food insecurity.
    Mr. Yoho. Is there any evidence that people that moved off 
of these programs fell into a bigger food insecurity? I mean, 
is there documented peer review articles?
    Dr. Shambaugh. There is also, just for the record, 
absolutely no documented evidence that the people moving to 
work were moving to work because of the work requirements. The 
only documented studies that have actually tried to study it 
carefully by looking at populations that face work requirements 
compared to those that didn't find literally no impact on the 
propensity to work, based on exposure to work requirements.
    Mr. Yoho. But if we have a reduction of 94 percent, 85 
percent in Alabama, 80 in Maine, we know that is a measurable--
--
    Dr. Shambaugh. We also know, though, that in the places 
that don't have work requirements, people cycle off SNAP all 
the time.
    Mr. Yoho. At what percent?
    Dr. Shambaugh. In 2008 to 2012, when the economy was 
terrible and work requirements were waived almost everywhere, 
the Department of Agriculture reports that 80 percent were off 
SNAP within 2 years. And that is with almost no----
    Mr. Yoho. Okay. I am glad you brought that up, because in 
the late 1990s, the share of Americans living in the country or 
city waived from SNAP work requirements was under 20 percent. 
It climbed a bit under the George Bush Administration to \1/3\. 
In 2009 a waiver program designed to accommodate exceptional 
circumstances became a national panacea. As part of the 
American Recovery and Reinvestment Act signed in by President 
Obama that February, Congress temporarily suspended the 
conditions on ABAWDs, SNAP, nationwide. The suspension was 
supposed to extend only though 2010, but no government 
initiative is temporary, and 8 years later, ABAWD time limit 
waivers are still in effect in at least part of 36 counties.
    The point that I want to get across is we are at full 
employment, pretty much, in this country. And we know prior to 
the recession, there was about 17, 18 million people on the 
SNAP Program. With the waivers, it went up to 41 to 42 million. 
It has come down to around 38 to 39 million. We should see, I 
would think, a ratio of decrease with the full employment. And 
we should take the politics out of this. Let's get people into 
higher paying jobs. As we have seen, there are 6.3 million 
unfulfilled jobs. In my district, we have people starting 
minimum wage, $15 an hour, through competition because the 
economy is so good. Let's maximize that and let's help people 
transition off and move from aid on a program, get them 
educated, and off a program. I am out of time, so I have to go.
    Thank you.
    The Chair. Thank you. Mr. Panetta, you are recognized for 5 
minutes.
    Mr. Panetta. Thank you, Madam Chair. It is good to be here. 
Ranking Member Johnson, good to see you, too.
    Thanks to all the witnesses for being here, as well as your 
preparation to be here. I know it takes a lot of work, so thank 
you very much.
    First of all, just some housekeeping. I would like to enter 
into the record this letter from MAZON, a national advocacy 
organization working to end hunger. It is a pretty extensive 
letter, almost a 20-page letter that talks about how this 
proposed rule would cause certain groups, like rural Americans, 
working poor, and veterans, to lose out on many of their 
benefits and the difficulties that they may undergo if this 
rule is in place. If I could enter that into the record?
    The Chair. Without objection.
    [The letter referred to is located on p. 379.]
    Mr. Panetta. Thank you, Madam Chair.
    I understand what Mr. Yoho and Mr. Davis are saying, and I 
agree with them partly. Not just because they are my friends, 
but because I agree that we have to start looking at policy 
when it comes to this issue. We know it can be very political 
on both sides, and I saw it last term as a freshman Member 
sitting right down there in the Agriculture Committee and 
dealing with the farm bill and the presentation of significant 
changes to the SNAP Program without any significant evidence 
supporting such changes. And when I sat there and I asked the 
Chairman about the evidence that he had in support of these 
changes that would actually work, the Chairman's response to me 
was well, we have 2 years to figure that out. And I believe 
that on this type of issue that is very sensitive, that is very 
important to my 74,000 recipients of SNAP in my district on the 
central coast of California, we don't have 2 years just to 
figure it out. We need to basically lay a foundation of 
evidence to do so.
    Now, part of that is my background. I am a prosecutor and I 
learned that I just couldn't go into court early on as a young 
misdemeanor deputy prosecutor and stand up and say he is guilty 
and sit back down. I had to prove my case with evidence, and 
then I could make my argument based on that evidence.
    That is something that obviously we need to do, not just on 
this Committee and the Agriculture Committee--not just on this 
Subcommittee. Obviously, it should be something done in 
Congress, to be frank. I think that is a common sense 
statement. But, this is the type of issue where you see the 
effects, where you see the politics at play, and I just hope, 
moving forward, we can continue to look at the evidence to 
support these types of programs, because we want to help 
people. That is wholehearted, that is important, and that is 
why we are here.
    We understand that the evidence is missing, and what we are 
seeing is that we can't just base one metric, a 20 hour work 
week--we can't just use that to tell us everything that we need 
to know about a recipient of SNAP benefits. We have to look at 
everything, and unfortunately, I do believe that this proposed 
rule does just that. It reduces benefits by singling out a 
group that USDA assumes is less deserving, those who are deemed 
able-bodied but are unable to work.
    We are learning today that there is more to this story 
about these recipients' stories than meets the eye, and that 
this proposed rule will harm those with vulnerabilities that we 
may not be able to see at first glance. And some of those 
deemed able-bodied may actually not be and others may face 
difficulties we would not otherwise anticipate.
    That is what this hearing is demonstrating, and why the 
USDA should rethink this proposal, gather more data, gather 
more evidence, and learn more about the challenges these 
targeted SNAP beneficiaries really face.
    Now obviously, one of those groups is veterans. In my 
district on the central coast, we have about 30,000 veterans 
for a number of reasons, but I just would like to throw out 
there to Ms. Hamler-Fugitt, basically with this proposed rule, 
what would be some of the obstacles that veterans would face in 
trying to find employment?
    Ms. Hamler-Fugitt. To Congressman Panetta, what we see in 
Franklin County in our vets population, we have a vets outreach 
worker that works specifically with this population. We are 
seeing servicemen and -women who have been on multiple tours of 
deployment returning to the community with a lot of issues, 
mental health issues, jobs that were promised that are not 
there for them to transition back into, and a desperate need 
for mental health treatment, as well as the transitional 
supports and housing supports. Again, just prioritizing that.
    If I could just reiterate what you said about the policy 
needs to be driven by empirical evidence. We have been having 
this conversation for more than 20 years, since this provision 
went into the 1996 Welfare Reform Act. And that is one thing 
that I urge all of you to do is to set the standard for data 
collection so we can measure this information, measure the 
participants, measure their outcomes across all states by using 
the same data sets.
    Mr. Panetta. Outstanding. Thank you. Madam Chair, I yield 
back.
    The Chair. Thank you very, very much. I thank all of you 
for being here today and your testimony.
    The Chair recognizes herself for 5 minutes.
    If I didn't know better, I would think that this was a 
hearing about waivers. It is not. If I didn't know better, I 
would think that there were no job requirements or training 
requirements for ABAWDs. There are. It is the law now.
    I listened to one of the witnesses talk about what happened 
in 2006. This is 2019. In 2006, we were a manufacturing society 
in most major cities in this country. Today, we are more 
service and we don't make anything in this country anymore. 
There was a time you could come out of high school and go into 
a factory and get a job. That doesn't exist today, because the 
same corporations we give big tax breaks to take all of their 
business and make everything offshore. Yes, there were jobs in 
2006. There aren't today for low-skilled and unskilled workers.
    Let's just talk about who really are ABAWDs. They are the 
people who clean these buildings that we work in every day and 
that some people sleep in every night. They are the people who 
serve us in the cafeteria, who fix our food. Those are ABAWDs. 
They work every single day, and even this government doesn't 
pay them enough to make a living. There are people who work in 
this building who qualify for SNAP every month, that $1.40 a 
meal. Let's talk about what it really is, and let's also talk 
about jobs.
    We know that over the next 20 years, 80 percent of all jobs 
will require some form of STEM education. Most of the people we 
are talking about, the poorest of the poor, don't have those 
skills, don't have that education. There may be jobs, but they 
don't qualify for them.
    If I had grown up maybe around a blueberry patch, I might 
have done that, too. There is not one in my neighborhood. My 
neighborhood is one where people just try to survive every day. 
I think that we have to be realistic about who we are talking 
about.
    I got into an elevator in this building. A person who 
cleans the building gets on with me, which is not really 
allowed for them to put their carts on with us. She wanted to 
tell me in tears how much it meant for her to get the SNAP 
benefits she gets every month.
    But no, we want to make this some big deal about being 
partisan, and it is not partisan. Hungry people are hungry 
people. People who work are people who work. If we are really 
honest with ourselves, and we started to talk to the people who 
are in these situations instead of believing that they are 
invisible and they are unworthy and undeserving, we might have 
a different outcome. Maybe we would sit down, as my colleague 
said here, and find a way to get them to the jobs that are 
available. Maybe we provide some transportation. Maybe we 
provide some training. Not just filling out an application, 
actually training them to do a job that exists.
    If I just didn't know better, I wouldn't even think I was 
in this country, if I didn't know better.
    And so, I just want to say to all of my colleagues, I know 
we all care about the people we represent, but maybe sometimes 
we need to come out of these buildings and talk to them. Maybe 
we need to go into a food bank and see who comes. That might be 
helpful, and not just assume who they are and what they are. 
And until the USDA can tell me who they are, then I am never 
going to support something like this, because not only does it 
not rely on any data when they could just wait a little while 
and get the data from the trials we have already done, but more 
importantly, because they don't know who they are talking 
about. They have no idea. And so, you just make up something 
for people.
    It is time that this Congress, the people's House, the 
Representatives of the people of this country, find out what 
our people want.
    And with that, I would close and ask my colleague if he has 
a closing statement. Oh, before that, Chairman Peterson asked 
that I enter into the record a letter of comment from 
Commissioner Tony Lourey with the Minnesota Department of Human 
Services. Without objection.
    [The letter referred to is located on p. 151.]
    Mr. Johnson. Yes, it is indeed true that the people who 
clean this building at night are worthy and deserving, and they 
are ABAWDs. And of course, it is absolutely true that the 
people who make the food in the cafeteria, they are worthy and 
they are deserving and they are ABAWDs.
    It is just as important to acknowledge that they are 
working and that their work is important, and that it is worthy 
of our respect. They are working and they are doing what they 
can to try to eke out a living and put themselves in a position 
where tomorrow can be better than yesterday. Work does that, 
and if a couple of things came out loud and clear that there is 
basis for agreement, it is first off that people who can work, 
should work. I want to thank the panelists for bringing that to 
the fore.
    Another thing that came out, particularly with Mr. Van 
Drew's comments and others, is the importance of data, the 
importance of evidence. Evidence is powerful and data can light 
our way forward, and that is why I am concerned that there was 
resistance on the part of some Members of this Committee to a 
robust data capture component championed by Chairman Conaway 
and others during the last farm bill discussion. I am hopeful 
that since we all acknowledge the importance of data, we can 
work together to have better data capture opportunities in days 
to come.
    I would close by saying this, Madam Chair. I have heard 
that you run a tight ship and a fair one. You do. Thank you for 
a good hearing.
    The Chair. Thank you very much, Mr. Johnson. I appreciate 
it and I appreciate your ability to work with me and 
willingness to do so.
    Thank you all very much for being here. I appreciate your 
testimony.
    Under the Rules of the Committee, the record of today's 
hearing will remain open for 10 calendar days to receive 
additional material and supplementary written responses from 
the witnesses to any questions posed by a Member.
    This hearing of the Subcommittee on Nutrition, Oversight, 
and Department Operations is adjourned.
    [Whereupon, at 10:50 a.m., the Subcommittee was adjourned.]
    [Material submitted for inclusion in the record follows:]
Submitted Letters by Hon. Marcia L. Fudge, a Representative in Congress 
                               from Ohio
February 1, 2019

  Hon. Sonny Perdue,
  Secretary,
  U.S. Department of Agriculture,
  Washington, D.C.

    Dear Secretary Perdue:

    As Chair of the Nutrition, Oversight, and Department Operations 
Subcommittee of the House Agriculture Committee, I write to request an 
extension of the 60 day comment period for the proposed rule regarding 
the treatment of Able-Bodied Adults Without Dependents (ABAWDs) in the 
Supplemental Nutrition Assistance Program (SNAP) entered into the 
Federal Register today. Given the complexity of and the significant 
interest in this topic, I request an immediate extension of the comment 
period from 60 to 120 days to allow for meaningful and robust comments.
    The proposed rule includes assumptions about ABAWDs and state 
administrative agencies that have been recently and thoroughly 
considered by Congress, and overwhelmingly rejected. In fact, 
legislative language similar to the proposed rule was included in the 
initial version of H.R. 2. This language was vetted in detail for 5 
months by Members of the 2018 Farm Bill Conference Committee before 
being struck from the final bill. As you know, the House and Senate 
ultimately passed a farm bill conference report by historic margins, 
and the President signed the bill without delay.
    During the farm bill signing ceremony, you said that USDA would do 
its ``best to implement that bill'' as Congress intended. This proposed 
rule does just the opposite. Therefore, I ask for an immediate 
extension of the comment period from 60 to 120 days to allow Members of 
Congress, and the countless other advocates in favor of protecting SNAP 
from unwarranted attacks, the opportunity to better inform USDA of the 
hardships that will result if the Department moves forward with this 
harmful and intolerable proposed rule.
            Respectfully,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Hon. Marcia L. Fudge,
Chair,
Subcommittee on Nutrition, Oversight, and Department Operations.

February 26, 2019

  Hon. Marcia L. Fudge,
  Chair,
  Subcommittee on Nutrition, Oversight, and Department Operations,
  U.S. House of Representatives,
  Washington, D.C.

    Dear Chair Fudge:

    Thank you for your letter dated February 1, 2019, requesting an 
extension of the public comment period for the recently proposed rule 
affecting Supplemental Nutrition Program (SNAP) work requirements and 
the participation time limit for able-bodied adults without dependents 
(ABAWDs).
    The proposed rule includes administrative actions within the 
authority delegated to the Secretary within the Food and Nutrition Act 
of 2008. It would encourage broader application of the statutory ABAWD 
work requirement, consistent with the Administration's focus on 
fostering self-sufficiency and promoting the dignity of work. I believe 
these proposed changes support our mutual goal of improving the lives 
of those participating in SNAP.
    I appreciate your interest in ensuring that the U.S. Department of 
Agriculture is able to receive meaningful and robust comments to this 
rule. Before the rule was published in the Federal Register on February 
1, 2019, and before the 60 day comment period began, the proposed rule 
was available on our website beginning December 20, 2018, thereby 
providing interested stakeholders additional time to review the 
proposal and begin formulating their comments. Given the additional 
amount of time that the rule has been on public display, I believe that 
a 60 day comment period is a sufficient amount of time to receive 
meaningful and robust comments.
    Thank you for your support. If you need further assistance, please 
have your staff contact Erin Wilson with the Office of Congressional 
Relations at (202) 720-7095 or [email protected].
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Hon. Sonny Perdue,
Secretary.
                                 ______
                                 
 Submitted Proposed Rule by Hon. Marcia L. Fudge, a Representative in 
                           Congress from Ohio
Federal Register
Vol. 84, No. 22
Friday, February 1, 2019
Proposed Rules
DEPARTMENT OF AGRICULTURE
Food and Nutrition Service
7 CFR Part 273
[FNS-2018-0004]
RIN 0584-AE57
Supplemental Nutrition Assistance Program: Requirements for Able-Bodied 
        Adults Without Dependents
    agency: Food and Nutrition Service (FNS), USDA.

    action: Proposed rule.

    summary: Federal law generally limits the amount of time an able-
bodied adult without dependents (ABAWD) can receive Supplemental 
Nutrition Assistance Program (SNAP) benefits to 3 months in a 36 month 
period, unless the individual meets certain work requirements. On the 
request of a state SNAP agency, the law also gives the Department of 
Agriculture (the Department) the authority to temporarily waive the 
time limit in areas that have an unemployment rate of over ten percent 
or a lack of sufficient jobs. The law also provides state agencies with 
a limited number of percentage exemptions that can be used by states to 
extend SNAP eligibility for ABAWDs subject to the time limit. The 
Department proposes to amend the regulatory standards by which the 
Department evaluates state SNAP agency requests to waive the time limit 
and to end the unlimited carryover of ABAWD percentage exemptions. The 
proposed rule would encourage broader application of the statutory 
ABAWD work requirement, consistent with the Administration's focus on 
fostering self-sufficiency. The Department seeks comments from the 
public on the proposed regulations.

    dates: Written comments must be received on or before April 2, 2019 
to be assured of consideration.

    addresses: The Food and Nutrition Service, USDA, invites interested 
persons to submit written comments on this proposed rule. Comments may 
be submitted in writing by one of the following methods:

   Preferred Method: Federal eRulemaking Portal: Go to http://
        www.regulations.gov. Follow the online instructions for 
        submitting comments.

   Mail: Send comments to Certification Policy Branch, Program 
        Development Division, FNS, 3101 Park Center Drive, Alexandria, 
        Virginia 22302.

   All written comments submitted in response to this proposed 
        rule will be included in the record and will be made available 
        to the public. Please be advised that the substance of the 
        comments and the identity of the individuals or entities 
        submitting the comments will be subject to public disclosure. 
        FNS will make the written comments publicly available on the 
        Internet via http://www.regulations.gov.

    for further information contact: Certification Policy Branch, 
Program Development Division, FNS, 3101 Park Center Drive, Alexandria, 
Virginia 22302. [email protected].

    supplementary information: 
Background
Acronyms or Abbreviations
[Phrase, Acronym or Abbreviation]
  Able-Bodied Adult without Dependent(s), ABAWD(s)
  Advanced Notice of Public Rulemaking, ANPRM
  Bureau of Labor Statistics, BLS
  Census Bureau's American Community Survey, ACS
  Code of Federal Regulations, CFR
  Department of Labor, DOL
  Employment and Training Administration, ETA
  Employment and Training, E&T
  Food and Nutrition Act of 2008, Act
  Food and Nutrition Service, FNS
  Labor Market Area(s), LMA(s)
  Labor Surplus Area(s), LSA(s)
  Supplemental Nutrition Assistance Program, SNAP
  The Personal Responsibility and Work Opportunity Reconciliation Act 
    of 1996, PRWORA
  U.S. Department of Agriculture, the Department or USDA
References
    The following references may be useful to help inform those wishing 
to provide comments.

  (1)  Section 6(d) and section 6(o) of the Food and Nutrition Act of 
            2008, as amended

  (2)  Title 7 of the Code of Federal Regulations, parts 273.7 and 
            273.24

  (3)  Food Stamp Program: Personal Responsibility Provisions of the 
            Personal Responsibility and Work Opportunity Reconciliation 
            Act of 1996, Proposed Rule, 64 FR 70920 (December 17, 
            1999). Available at: https://www.federalregister.gov/
            ?documents/?1999/?12/?17/?99-32527/?food-stamp-program-
            personalresponsibility-provisions-of-the-
            personalresponsibility-and-work

  (4)  Food Stamp Program: Personal Responsibility Provisions of the 
            Personal Responsibility and Work Opportunity Reconciliation 
            Act of 1996, Final Rule, 66 FR 4437 (January 17, 2001). 
            Available at: https://www.federalregister.gov/?documents/
            ?2001/?01/?17/?01-1025/?foodstamp-program-personal-responsi
            bilityprovisions-of-the-personal-responsibilityand-work

  (5)  Guide to Serving ABAWDs Subject to Time-limited Participation, 
            2015. Available at: https://fns-prod.azureedge.net/sites/
            default/files/Guide_to_
            Serving_ABAWDs_Subject_to_Time_Limit.pdf

  (6)  Guide to Supporting Requests to Waive the Time Limit for Able-
            Bodied Adults without Dependents, 2016. Available at: 
            https://fns-prod.azureedge.net/sites/default/files/snap/
            SNAP-Guide-to-Supporting-Requests-to-Waive-the-Time-Limit-
            for-ABAWDs.pdf

  (7)  Expiration of Statewide ABAWD Time Limit Waivers, 2015. 
            Available at: https://fns-prod.azureedge.net/sites/default/
            files/snap/SNAP-Expiration-of-Statewide-ABAWD-Time-Limit-
            Waivers.pdf

  (8)  ABAWD Time Limit Policy and Program Access, 2015. Available at: 
            https://fns-prod.azureedge.net/sites/default/files/snap/
            ABAWD-Time-Limit-Policy-and-Program-Access-Memo-Nov2015.pdf

  (9)  ABAWD Questions and Answers, 2015. Available at: https://fns-
            prod.azureedge.net/sites/default/files/snap/ABAWD-
            Questions-and-Answers-June%202015.pdf

  (10)  ABAWD Questions and Answers, 2013. Available at: https://fns-
            prod.azureedge.net/sites/default/files/snap/ABAWD-
            Questions-and-Answers-December-2013.pdf

  (11)  BLS Local Area Unemployment Statistics. Available at: https://
            www.bls.gov/lau/

  (12)  BLS Labor Surplus Area. Available at: https://www.doleta.gov/
            programs/lsa.cfm
The Rationale for Modifying Waiver Standards
    The President's Executive Order on Reducing Poverty in America by 
Promoting Opportunity and Economic Mobility (April 10, 2018) provided 
guiding principles for public assistance programs, one of which was to 
improve employment outcomes and economic independence by strengthening 
existing work requirements for work-capable individuals. The Executive 
Order directed Federal agencies to review regulations and guidance 
documents to determine whether such documents are consistent with the 
principles of increasing self-sufficiency, well-being, and economic 
mobility. Consistent with the Executive Order and the Administration's 
focus on fostering self-sufficiency, as well as the Department's 
extensive operational experience with ABAWD waivers, the Department has 
determined that the standards for waivers must be strengthened so that 
the ABAWD work requirement is applied to ABAWDs more broadly. The 
Department is confident that these changes would encourage more ABAWDs 
to engage in work or work activities if they wish to continue to 
receive SNAP benefits.
    The Department believes that the proposed changes reinforce the 
Act's intent to require these individuals to work or participate in 
work activities in order to receive SNAP benefits for more than 3 
months in a 36 month period. Section 6(o) of the Act, entitled, ``Work 
Requirements,'' allows these individuals to meet the ABAWD work 
requirement by working and/or participating in a qualifying work 
program at least 20 hours per week (averaged monthly to 80 hours per 
month) or by participating in and complying with workfare. For the 
purposes of meeting the ABAWD work requirement, working includes unpaid 
or volunteer work that is verified by the state agency. The Act 
specifically exempts individuals from the ABAWD time limit and 
corresponding work requirement for several reasons, including, but not 
limited to, age, unfitness for work, having a dependent child, or being 
pregnant.
    The Act authorizes waivers of the ABAWD time limit and work 
requirement in areas in which the unemployment rate is above ten 
percent, or where there is a lack of sufficient jobs. The Department 
believes waivers of the ABAWD time limit are meant to be used in a 
limited manner in situations in which jobs are truly unavailable to 
ensure enforcement of the ABAWD work requirements as much as possible 
to promote greater engagement in work or work activities.
    Immediately following the Great Recession, the vast majority of the 
states, including the District of Columbia, Guam, and the Virgin 
Islands, qualified for and implemented statewide ABAWD time limit 
waivers in response to a depressed labor market. In the years since the 
Great Recession, the national unemployment rate has dramatically 
declined. Despite the national unemployment rate's decline from 9.9 
percent in April 2010 to 3.9 percent in April 2018, a significant 
number of states continue to qualify for and use ABAWD waivers under 
the current waiver standards. Right now, nearly \1/2\ of ABAWDs live in 
areas that are covered by waivers despite a strong economy. The 
Department believes waiver criteria need to be strengthened to better 
align with economic reality. These changes would ensure that such a 
large percentage of the country can no longer be waived when the 
economy is booming and unemployment is low.
    The Department is committed to enforcing the work requirements 
established by Congress and is concerned about the current level of 
waiver use in light of the current economy. The regulations afforded 
states broad flexibility to develop approvable waiver requests. The 
Department's operational experience has shown that some states have 
used this flexibility to waive areas in such a way that was likely not 
foreseen by the Department.
    Some of the key concerns have stemmed from the combining of data 
from multiple individual areas to waive a larger geographic area (e.g., 
a group of contiguous counties) and the application of waivers in 
individual areas with low unemployment rates that do not demonstrate a 
lack of sufficient jobs. For example, some states have maximized the 
number of areas or people covered by waivers by combining data from 
areas with high unemployment with areas with low unemployment. This 
grouping has resulted in the combined area qualifying for a waiver when 
not all individual sub-areas would have qualified on their own. States 
have combined counties with unemployment rates under five percent with 
counties with significantly higher unemployment rates in order to waive 
larger areas. For example, current regulations required the Department 
to approve a state request to combine unemployment data for a populous 
county with a high unemployment rate of over ten percent with the 
unemployment data of several other less populous counties with very low 
unemployment rates that ranged between three and four percent. Other 
states have combined data from multiple areas that may only tenuously 
be considered an economic region. In some cases, states have grouped 
areas that are contiguous but left out certain low-unemployment areas 
that would otherwise logically be considered part of the region. In 
this manner, states have created questionable self-defined economic 
areas with gaping holes to leverage the flexibility of the regulations.
    The Department has also noted that, despite the improving economy, 
the lack of a minimum unemployment rate has allowed local areas to 
qualify for waivers based solely on having relatively high unemployment 
rates as compared to national average, regardless of how low local 
areas unemployment rates fall. Since the current waiver criteria have 
no floor, a certain percentage of states will continue to qualify for 
waivers even if unemployment continues to drop.
    It is the Department's understanding that the intent of Congress in 
passing the Personal Responsibility and Work Opportunity Reconciliation 
Act of 1996 was to provide SNAP to unemployed ABAWDs on a temporary 
basis (3 months in any 3 year period) with the expectation that they 
work and/or engage in a work program at least 20 hours per week, or 
participate in workfare, to receive SNAP on an ongoing basis. The 
Department is committed to implementing SNAP as Congress intended and 
believes that those who can work should work. The widespread use of 
waivers has allowed some ABAWDs to continue to receive SNAP benefits 
while not meeting the ABAWD work requirement for longer than 3 months. 
The proposed rule addresses these areas of concern and places 
safeguards to avoid approving waivers that were not foreseen by 
Congress and the Department, and to restrict states from receiving 
waivers in areas that do not clearly demonstrate a lack of sufficient 
jobs.
    As stated above, given the widespread use of ABAWD waivers during a 
period of historically low unemployment, the Department believes that 
the current regulatory standards should be reevaluated. Based on the 
Department's approximately 2 decades' experience with reviewing ABAWD 
waivers, the Department is proposing that the standards for approving 
these waivers be updated to ensure the waivers are applied on a more 
limited basis. The application of waivers on a more limited basis would 
encourage more ABAWDs to take steps towards self-sufficiency.
    The Department proposes stricter criteria for ABAWD waiver 
approvals that would establish stronger, updated standards for 
determining when and where a lack of sufficient jobs justifies 
temporarily waiving the ABAWD time limit. The proposed rule would also 
ensure the Department only issues waivers based on representative, 
accurate, and consistent economic data, where it is available. Limiting 
waivers would make more ABAWDs subject to the time limit and thereby 
encourage more ABAWDs to engage in meaningful work activities if they 
wish to continue to receive SNAP benefits. The Department recognizes 
that long-term, stable employment provides the best path to self-
sufficiency for those who are able to work. The Department believes it 
is appropriate and necessary to encourage greater ABAWD engagement with 
respect to job training and employment opportunities that would not 
only benefit ABAWDs, but would also save taxpayers' money. The 
Department and the states share a responsibility to help SNAP 
participants--especially ABAWDs--find a path to self-sufficiency. 
Through the stricter criteria for waiver approvals, the Department 
would encourage greater engagement in meaningful work activities and 
movement toward self-sufficiency among ABAWDs, thus reducing the need 
for nutrition assistance.
Waiver Standards Framework
    Current regulations at 7 CFR 273.24(f) set standards and 
requirements for the data and evidence that states must provide to FNS 
to support a waiver request. States enjoy considerable flexibility to 
make these waiver requests pursuant to the current regulations. For 
example, these regulatory standards give states broad flexibility to 
define the waiver's geographic scope. The discretion for states to 
define areas allows waivers based on data for combined areas that are 
not necessarily economically tied. An economically tied area is an area 
within which individuals can reside and find employment within a 
reasonable distance or can readily change employment without changing 
their place of residence. In addition, while the current regulations 
establish criteria for unemployment data that rely on standard Bureau 
of Labor Statistics (BLS) data or methods, the regulations also allow 
states to rely on alternative, less robust economic indicators, which 
include data other than unemployment data from BLS, to demonstrate a 
lack of sufficient jobs. Moreover, the waiver standards allow areas 
within states to qualify for waivers as a result of unemployment rates 
relative to the national average, without consideration for whether the 
national or local area unemployment rate is high or low. Put 
differently, under the current regulations, which do not include a 
local unemployment rate floor, even if the national unemployment rate 
falls, a particular area's unemployment rate may support a waiver if 
that area's unemployment rate is low but sufficiently higher than the 
national average. As a result of these and other shortcomings, the 
current regulations give states an opportunity to qualify for waivers 
and avoid the ABAWD time limit when economic conditions do not justify 
such relief. For these reasons, the Department believes that the waiver 
standards under this proposed rule will better identify areas that do 
not have a sufficient number of jobs to provide employment for ABAWDs.
    As of September 2018, the national unemployment rate is the lowest 
unemployment rate since 1969; however, states continue to request and 
qualify for ABAWD waivers based on the current waiver criteria, which 
define the lack of sufficient jobs in an area too broadly. In April 
2010, the national unemployment rate stood at 9.9 percent. From 2010 
through 2013, the vast majority of states qualified for and continued 
to implement statewide ABAWD time limit waivers. SNAP participation 
peaked at an average of 47.6 million recipients per month in FY 2013 
and has gradually declined since then. In July 2013, the national 
unemployment rate was 7.3 percent; 45 ABAWD time limit waivers covered 
the entire state,\1\ and six waivers covered specific areas within the 
state. In April 2018, SNAP participation totaled 39.6 million 
participants, and the national unemployment rate stood at 3.9 percent. 
In April 2018, eight waivers applied to an entire state, and 28 covered 
specific areas within a state. Although the national unemployment rate 
has dropped from 9.9 percent in April 2010 to 3.9 percent in April 
2018, many states continue to qualify for and use ABAWD time limit 
waivers under the current waiver standards, and nearly \1/2\ of all 
ABAWDs live in areas that are covered by waivers.
---------------------------------------------------------------------------
    \1\ The term ``state'' refers to any of the 50 states, the District 
of Columbia, and the U.S. territories[.]
---------------------------------------------------------------------------
    The Department is concerned that ABAWD time limit waivers continue 
to cover significant portions of the country and are out of step with a 
national unemployment rate hovering at less than four percent. Since 
the current waiver criteria have no floor, a certain percentage of 
states will continue to qualify for waivers even if unemployment 
continues to drop. In other words, regardless of how strong the economy 
is, the criteria are written in such a way that areas will continue to 
qualify even with objectively low unemployment rates. Many currently-
waived areas qualified based on 24 month local unemployment rates below 
six percent.
    The current criteria for waiver approval permit states to qualify 
for waivers without a sufficiently robust standard for a lack of 
sufficient jobs. The waiver criteria should be updated to ensure states 
submit data that is more representative of the economic conditions in 
the requested areas. Such reforms would make sure the Department issues 
waivers based on representative, accurate, and consistent economic 
data.
    This proposed rule would set clear, robust, and quantitative 
standards for waivers of the ABAWD time limit. The proposal would also: 
Eliminate waivers for areas that are not economically tied together; 
eliminate the ability of an area to qualify for a waiver based on its 
designation as a Labor Surplus Area (LSA) by the Department of Labor; 
limit the use of alternative economic indicators to areas for which 
standard data is limited or unavailable, such as Indian Reservations 
and U.S. Territories; and provide additional clarity for states 
regarding the waiver request process. The proposed changes would ensure 
the Department issues waivers only to provide targeted relief to areas 
that demonstrate a lack of sufficient jobs or have an unemployment rate 
above ten percent and that the ABAWD time limit encourages SNAP 
participants to find and keep work if they live in areas that do not 
lack sufficient jobs.
Background
Previous Action
    On February 23, 2018, the Department published an Advanced Notice 
of Public Rulemaking (ANPRM) entitled ``Supplemental Nutrition 
Assistance Program: Requirements and Services for Able-Bodied Adults 
Without Dependents'' (83 FR 8013) to seek public input to inform 
potential policy, program, and regulatory changes that could 
consistently encourage ABAWDs to obtain and maintain employment and 
thereby decrease food insecurity. The Department specifically asked 
whether changes should be made to: (1) The existing process by which 
state agencies request waivers of the ABAWD time limit; (2) the 
information and data states must provide to support the waiver request; 
(3) the Department's implementation of the waiver approval; and (4) the 
waiver's duration. The ANPRM generated nearly 39,000 comments from a 
range of stakeholders including private citizens, government agencies 
and officials, food banks, advocacy organizations, and professional 
associations.
    The comments addressed the broad scope of topics covered by the 
ANPRM. Comments about the ABAWD waiver included diverse perspectives, 
ranging from those who supported stricter waiver approval requirements 
to those who favored maintaining or expanding the criteria for waiver 
approval. Many commenters favored no change or expressed support for 
greater flexibility. Other commenters identified a number of areas of 
concern with current practices, including the use of waivers by states 
to waive the ABAWD work requirement and avoid promoting work, waiving 
areas with relatively low unemployment rates, and allowing the use of 
certain metrics for waiver approvals.
    The Department received more than 3,500 comments regarding 
potential reforms to the ABAWD time limit and waivers of the time limit 
through the Department's request for information (RFI) entitled, 
``Identifying Regulatory Reform Initiatives'' published July 17, 2017 
(82 FR 32649). This RFI requested ideas on how the Department can 
provide better customer service and remove unintended barriers to 
participation in the Department's programs in ways that least interfere 
with the Department's customers and allow the Department to accomplish 
its mission. The Department specifically requested ideas on 
regulations, guidance documents, or any other policy documents that 
require reform. While commenters disagreed with certain SNAP provisions 
outlined previously, specific changes to regulations and policies were 
not provided. The Department received a range of comments to the RFI in 
addition to the comments listed above that are not relevant to this 
proposed rule.
Summary of Proposed Changes
    The Department believes current regulations at 7 CFR 273.24(c) and 
7 CFR 273.24(f) should be updated and strengthened. The proposed rule 
focuses on updating the standards for ABAWD waivers. Current 
regulations at 7 CFR 273.24(f) set standards and requirements for the 
data and evidence that states must provide to FNS to support an ABAWD 
waiver request. States enjoy considerable flexibility to make these 
waiver requests pursuant to the current regulations. This flexibility 
has resulted in the widespread use of waivers during a period of low 
unemployment, which reduces the application of the work requirement.
    The Department proposes several changes. First, the proposed rule 
would limit the ability of areas to qualify for waivers as local 
economies and the overall national economy improve. Second, the 
proposed rule would no longer allow state agencies to combine 
unemployment data from areas with high unemployment with areas with 
lower unemployment and more plentiful employment opportunities in order 
to maximize the area waived. Instead, the proposed rule would ensure 
the Department issues waivers only to economically tied areas that meet 
the new criteria defining what is meant by a lack of sufficient jobs. 
The proposed rule would also limit the duration of waivers to 1 year, 
and curtail the use of less robust data to approve waivers. The 
subsequent sections provide details about the changes proposed in this 
rule.
Discussion of Proposed Changes
General
    The Department proposes that the rule, once finalized, would go 
into effect on October 1, 2019, which is the beginning of Federal 
Fiscal Year 2020. All waivers in effect on October 1, 2019, or 
thereafter, would need to be approvable according to the new rule at 
that time. Any approved waiver that does not meet the criteria 
established in the new rule would be terminated on October 1, 2019. 
States would be able to request new waivers if the state's waiver is 
expected to be terminated. The Department requests feedback from states 
regarding the implementation date. In addition, the Department proposes 
clarifying that any state agency's waiver request must have the 
Governor's endorsement to ensure that such a critical request is 
supported at the highest levels of state government.
Establishing Core Standards for Approval
    The Department proposes updating criteria for ABAWD time limit 
waivers to improve consistency across states and only allow approvals 
in areas where waivers are truly necessary. These revisions would 
include the establishment of core standards that would allow a state to 
reasonably anticipate whether it would receive approval from the 
Department. These core standards would serve as the basis for approval 
for the vast majority of waiver requests, save for areas with 
exceptional circumstances or areas with limited data or evidence, such 
as Indian Reservations and U.S. Territories. The proposed rule would 
continue to allow approvals for waivers based on data from BLS or a 
BLS-cooperating agency that show an area has a recent, 12 month average 
unemployment rate over ten percent.
    The proposed rule emphasizes that the basis for approval of waivers 
would be sound data and evidence that primarily relies on data from BLS 
or BLS-cooperating agencies. Any supporting unemployment data provided 
by the state would need to rely on standard BLS data or methods. BLS 
unemployment data is generally considered to be reliable and robust 
evidence for evaluating labor market conditions. BLS is an independent 
Federal statistical agency that is required to provide accurate and 
objective statistical information and is the principal fact-finding 
agency for the Federal Government in the broad field of labor economics 
and statistics. It collects, processes, analyzes, and disseminates 
essential statistical data for the public and Federal agencies.
    The proposed core standards for waiver approval would be codified 
in 7 CFR 273.24(f)(2).
Core Standards: Retaining Waivers Based On An Unemployment Rate Over 
        Ten Percent
    The Department does not propose changes to the regulations for 
waivers when an area has an unemployment rate over ten percent. The 
proposed rule would continue to allow approvals for waivers based on 
data from BLS or a BLS-cooperating agency that show an area has a 
recent, 12 month average unemployment rate over ten percent.
Core Standards: Establishing a Floor for Waivers Based On the 20 
        Percent Standard
    Current regulations at 7 CFR 273.24(f)(2) and (3) provide for 
waiver approvals for requested areas with an average unemployment rate 
at least 20 percent above the national average for a recent 24 month 
period, beginning no earlier than the same 24 month period that DOL 
uses to determine LSAs for the current fiscal year (otherwise known as 
the ``20 percent standard''). Under the current regulations, the 
Department adopted the 20 percent standard, in addition to LSA 
designation, to provide states with the flexibility to support waivers 
for areas in the country that are not considered by DOL for LSA 
designation and to allow states to use a more flexible 24 month 
reference period.
    There are key differences between the two standards. DOL's criteria 
for LSAs require an average unemployment rate that is at least 20 
percent above the national average and at least six percent for the 
preceding 2 calendar years (a 24 month period). DOL's local 
unemployment rate floor of six percent prevents areas with unemployment 
rates below that threshold from qualifying as LSAs. The 20 percent 
standard is the same, except that it allows for a flexible 24 month 
data reference period (no earlier than that which is used for LSAs) and 
it does not include any unemployment rate floor.
    Based upon operational experience, the Department has observed 
that, without an unemployment rate floor, local areas will continue to 
qualify for waivers under the Department's 20 percent standard based on 
high unemployment relative to the national average even as local 
unemployment rates fall to levels as low as five to six percent 
(depending upon the national rate). The Department believes that 
amending the waiver regulations to include an unemployment floor is a 
critical step in achieving more targeted criteria. While the 20 percent 
standard is similar to the calculation of an LSA, the Department 
believes it is appropriate to request public comment to explore a floor 
that is designed specifically for ABAWD waivers.
    The Department believes a floor should be set for the 20 percent 
standard so that areas do not qualify for waivers when their 
unemployment rates are generally considered to be normal or low. The 
``natural rate of unemployment'' is the rate of unemployment expected 
given normal churn in the labor market, with unemployment rates lower 
than the natural rate tending to result in inflationary pressure on 
prices. Thus, unemployment rates near or below the ``natural rate of 
unemployment'' are more indicative of the normal delay in unemployed 
workers filling the best existing job opening for them than a ``lack of 
sufficient jobs'' in an area. Generally, the ``natural rate of 
unemployment'' hovers around five percent. The Department believes that 
only areas with unemployment rates above the ``natural rate of 
unemployment'' should be considered for waivers. The Department seeks 
to establish a floor that is in line with the Administration's effort 
to encourage greater engagement in work and work activities. The 
Department believes that the seven percent floor for the 20 percent 
standard would strengthen the standards for waivers so that the ABAWD 
work requirement would be applied more broadly and fully consider the 
``lack of sufficient jobs'' criteria in the statute. Furthermore, this 
aligns with the proposal in the Agriculture and Nutrition Act of 2018, 
H.R. 2, 115th Cong.  4015 (as passed by House, June 21, 2018). As 
stated previously, the Department seeks to make the work requirements 
the norm rather than the exception to the rule because of excessive use 
of ABAWD time limit waivers to date. Using the proposed rule's seven 
percent floor for this criterion and eliminating waiver approvals based 
on an LSA designation (as well as utilizing the proposed limit on 
combining areas discussed below), an estimated 11 percent of ABAWDs 
would live in areas subject to a waiver. Currently, approximately 44 
percent of ABAWDs live in a waived area. The Department views the 
proposal as more suitable for achieving a more comprehensive 
application of work requirements so that ABAWDs in areas that have 
sufficient number of jobs have a greater level of engagement in work 
and work activities, including job training. In sum, the proposed rule 
modifies the current waiver criterion so that an area must have an 
average unemployment rate at least 20 percent above the national 
average and at least seven percent for a recent 24 month period, 
beginning no earlier than the same 24 month period that DOL uses to 
determine LSAs for the current fiscal year, to qualify for a waiver. 
The seven percent floor prevents a requested area with an unemployment 
rate 20 percent above the national average, but below seven percent, 
from qualifying for a waiver.
    Although the Department believes the local unemployment floor 
should be set at seven percent to best meet its goals of promoting 
self-sufficiency and ensuring areas with unemployment rates generally 
considered normal are not waived, it is requesting evidence-based and 
data-driven feedback on the appropriate threshold for the floor. 
Specifically, the Department requests feedback on which unemployment 
rate floor--six percent, seven percent, or ten percent--would be most 
effective at limiting waivers consistent with the Act's requirement 
that waivers be determined based on a lack of sufficient jobs.
    The Department is interested in public comments on establishing an 
unemployment floor of six percent, which would be consistent with DOL 
standards for LSAs. A six percent floor would require that an area 
demonstrate an unemployment rate of at least 20 percent above the 
national average for a recent 24 month period and at least a six 
percent unemployment rate for that same time period in order to receive 
waiver approval. The six percent floor also bears a relationship to the 
``natural rate of unemployment.'' in that it is approximately 20 
percent higher. As previously noted, the ``natural rate of 
unemployment'' generally hovers around five percent, meaning that 20 
percent above that rate is 6.0 percent. In combination with other 
changes in the proposed rule, the Department estimates that a six 
percent floor would reduce waivers to the extent that approximately 24 
percent of ABAWDs would live in waived areas. The Department is 
concerned that too many areas would qualify for a waiver of the ABAWD 
time limit with a six percent floor and that too few individuals would 
be subject to the ABAWD work requirements, which can be met through 
working or participating in a work program or workfare program, thereby 
moving fewer individuals towards self-sufficiency.
    The Department would also like to receive comments on establishing 
a floor of ten percent for the 20 percent standard. A ten percent floor 
would allow for even fewer waivers than the other options and would 
result in the work requirements being applied in almost all areas of 
the country. In combination with other changes in the proposed rule, 
the Department estimates that a ten percent floor would reduce waivers 
to the extent that approximately two percent of ABAWDs would live in 
waived areas.
    It is important to note that a ten percent floor would be distinct 
from the criteria for approval of an area with an unemployment rate of 
over ten percent. The ten percent unemployment floor would be attached 
to the 20 percent standard, which would mean an area would require an 
average unemployment rate 20 percent above the national average for a 
recent 24 month period and at least ten percent for the same period; 
the other similar, but separate standard requires an area to have an 
average unemployment rate of over ten percent for a 12 month period.
    Based on the Department's analysis, nearly 90 percent of ABAWDs 
would live in areas without waivers and would be encouraged to take 
steps towards self-sufficiency if a floor of seven percent was 
established. In comparison, a six percent floor would mean that 76 
percent of ABAWDs would live in areas without waivers and a ten percent 
floor would mean that 98 percent of ABAWDs would live in areas without 
waivers. A higher floor allows for the broader application of the time 
limit to encourage self-sufficiency.
    The Department is thus requesting comments on the various proposed 
options for setting a floor for the 20 percent standard. This will 
ensure that the Department fully considers the range of evidence 
available to establish a floor that meets the need of evaluating 
waivers.
Core Standards: Retaining the Extended Unemployment Benefits 
        Qualification Standard
    Under the proposed rule, the Department would continue to approve a 
state's waiver request that is based upon the requesting state's 
qualification for extended unemployment benefits, as determined by 
DOL's Unemployment Insurance Service. Extended unemployment benefits 
are available to workers who have exhausted regular unemployment 
insurance benefits during periods when certain economic conditions 
exist within the state. The extended benefit program is triggered when 
the state's unemployment rate reaches certain levels. Qualifying for 
extended benefits is an indicator, based on DOL data, that a state 
lacks sufficient jobs. Current regulations include this criterion as 
evidence of lack of sufficient jobs. The Department has consistently 
approved waivers based on qualification for extended unemployment 
benefits because it has been a clear indicator of lack of sufficient 
jobs and an especially responsive indicator of sudden economic 
downturns, such as the Great Recession. Therefore, the Department 
proposes to continue to include this criterion, reframed as a core 
standard for approval in this proposed regulation.
    The three provisions described above (the unemployment rate over 
ten percent standard, the 20 percent standard, and the qualification 
for extended unemployment benefits standard), would be considered the 
core standards for approval and, thus, the basis for most conventional 
waiver requests and approvals. The core standards would be codified in 
7 CFR 273.24(f)(2).
Criteria Excluded From Core Standards
    The proposed core standards would not include some of the current 
ABAWD time limit waiver criteria that are rarely used, sometimes 
subjective, and not appropriate when other more specific and robust 
data is available, such as unemployment rates from BLS. These excluded 
criteria include a low and declining employment-to-population ratio, a 
lack of jobs in declining occupations or industries, or an academic 
study or other publication(s) that describes an area's lack of jobs. 
These standards would no longer suffice for a waiver's approval if BLS 
data is available. These proposed changes would ensure that ABAWD time 
limit waiver requests are only approved in areas where waivers are 
truly necessary.
    The proposed rule would emphasize sound data and evidence that 
primarily relies on BLS and other DOL data for waiver approvals. Any 
supporting unemployment data that a state provides must, under the core 
standards, rely on standard data from BLS or a BLS-cooperating agency.
Other Data and Evidence in Exceptional Circumstances
    The proposed core standards would form the primary basis for 
determining waiver approval. However, the rule also proposes that the 
Department can approve waiver requests in exceptional circumstances 
based on other data and evidence. The Department proposes that other 
data and evidence still primarily rely on BLS unemployment data. Such 
alternative data would only be considered in exceptional circumstances 
or if BLS data is limited, unavailable, or if BLS develops a new method 
or data that may be applicable to the waiver review process. Given that 
economic conditions can change quickly, the Department believes it is 
appropriate to maintain a level of flexibility to approve waivers as 
needed in extreme, dynamic circumstances. Such waiver requests must 
demonstrate that an area faces an exceptional circumstance and provide 
data or evidence that the exceptional circumstance gives rise to an 
area not having a sufficient number of jobs to provide employment for 
the individuals in the area. For example, an exceptional circumstance 
may arise from the rapid disintegration of an economically and 
regionally important industry or the prolonged impact of a natural 
disaster. A short-term aberration, such as a temporary closure of a 
plant, would not fall within the scope of exceptional circumstances. 
For waiver requests in exceptional circumstances, the state agency may 
use additional data or evidence other than those listed in the core 
standards to support its need for a waiver under exceptional 
circumstances. In these instances, the state may provide data from the 
BLS or a BLS-cooperating agency showing an area has a most recent 3 
month average unemployment rate over ten percent. This provision to 
strengthen the standards for waivers would be codified in 7 CFR 
273.24(f)(3).
Restricting Statewide Waivers
    Current regulations at 7 CFR 273.24(f)(6) and the Department's 
policy guidance provide states with the discretion to define the areas 
to be covered by waivers. A state may request that a waiver apply to 
the entire state (statewide) or only to certain areas within the state 
(e.g., individual counties, cities, or towns), as long as the state 
provides data that corresponds to each requested area showing that the 
area meets one of the qualifying standards for approval.
    The proposed rule would eliminate statewide waiver approvals when 
sub-state data is available through BLS, except for those waivers based 
upon a state's qualification for extended unemployment benefits as 
determined by DOL's Unemployment Insurance Service. The Department 
proposes this change so that waivers of the ABAWD time limit are more 
appropriately targeted to those particular areas in which unemployment 
rates are high. Since statewide unemployment figures may include areas 
in which unemployment rates are relatively low, the Department believes 
that a more targeted approach would ensure that waivers exist only in 
areas that do not have a sufficient number of jobs to provide 
employment for the individuals living in that specific area. This 
proposed change further supports the Department's goal that more 
individuals are subject to the ABAWD time limit and work requirement, 
which can be met through working or participating in a work program or 
workfare program, consistent with the intent of the Act.
    The Department requests public comment specific to the proposed 
restriction on statewide waivers, especially with consideration to how 
the change may affect different states in different ways based upon 
geographic size, population, and other factors.
    These changes would be codified in 7 CFR 273.24(f)(4).
Restricting the Combining of Data to Group Sub-State Areas
    Current regulations at 7 CFR 273.24(f)(6) and the Department's 
policy guidance provide states considerable flexibility to define areas 
covered by ABAWD waivers. This flexibility allows states to combine 
data to group two or more sub-state areas, such as counties, together 
(otherwise referred to as ``grouped'' areas or ``grouping''). In order 
to meet the requirement for qualifying data or evidence that 
corresponds to the requested area, states use the unemployment and 
labor force data from the individual areas in the group to calculate an 
unemployment rate representative of the whole group. States can only 
group areas and support approval based on qualifying unemployment data. 
Under current regulations, states must demonstrate that the areas 
within any such group are contiguous and/or share the same Federal- or 
state-recognized economic region. For example, two or more contiguous 
counties could be grouped together, and the group's average 
unemployment rate could be calculated, by combining the unemployment 
and labor force data from each individual county.
    The Department's existing general conditions for the grouping of 
areas--that the areas must be either contiguous and/or share the same 
economic region--were intended to ensure that the areas grouped 
together are economically tied. However, in practice, the Department 
has learned that its standards for combining areas provide too much 
flexibility for state agencies and are often ineffective at ensuring 
that states are only grouping areas that are economically tied. For 
example, some states have grouped nearly all contiguous counties in the 
state together while omitting a few counties with relatively low 
unemployment in order to maximize the waived areas in the state. In 
other cases, states have grouped certain towns together that share the 
same economic region while omitting others with relatively low 
unemployment from the group, thereby maximizing the waived areas in the 
state.
    The proposed rule would prohibit states from grouping areas, except 
for areas that are designated a Labor Market Area (LMA) by the Federal 
Government.\2\ This change would ensure that only areas that are 
economically tied are grouped together. Moreover, the proposed rule 
would require states to include the unemployment data representative of 
all areas in the LMA in the state. As a result, states would be unable 
to omit certain areas within the LMA in the state for the purposes of 
achieving a qualifying unemployment rate for part of an LMA. These 
changes would be codified in 7 CFR 273.24(f)(5).
---------------------------------------------------------------------------
    \2\ An LMA is an economically integrated geographic area within 
which individuals can reside and find employment within a reasonable 
distance or can readily change employment without changing their place 
of residence. LMAs include Federally-designated statistical areas such 
as metropolitan statistical areas, micropolitan statistical areas, and 
other combined statistical areas. A nationwide list of every LMA is 
maintained by BLS.
---------------------------------------------------------------------------
    The Department requests public comments on whether it should 
include Labor Market Areas (LMAs) defined by the Federal Government as 
the basis for grouping areas or whether it should prohibit grouping 
entirely. If grouping were prohibited entirely, waived areas would be 
limited to individually qualifying jurisdictions with corresponding 
data (for example, counties and their equivalents, cities, and towns). 
The Department requests comments on the potential impacts of either 
policy. The Department believes that only allowing the use of Federally 
designated LMAs will limit the combination of areas that are not 
contiguous and economically integrated. The Department is interested in 
feedback on whether the LMA definition will target waivers to 
jurisdictions with a demonstrable lack of sufficient jobs without 
including jurisdictions that do not lack sufficient jobs.
Duration of Waiver Approvals and Timeliness of Data
    The proposed approach would limit the duration of waiver approvals. 
Under the current regulations, the Department typically approves 
waivers for 1 year. However, the current regulations allow the 
Department to approve shorter or longer waivers in certain 
circumstances. The Department proposes limiting a waiver's duration to 
1 year, but continuing to allow a waiver for a shorter period at a 
state's request. The Department believes that a 1 year waiver term 
allows sufficient predictability for states to plan and implement the 
waiver; at the same time, a 1 year waiver term ensures that the waiver 
request reflects current economic conditions.
    The proposed rule would also prioritize recent data by preventing 
states from requesting to implement waivers late in the Federal fiscal 
year, which broadens the available data reference period. Through 
operational experience, the Department has observed that several states 
that have historically requested 12 month waivers on a fiscal year 
basis (i.e., October 1 of 1 year through September 30 of the following 
year), have shifted their waiver request and implementation dates to 
later in the fiscal year (e.g., September 1 through August 31). The 
states that have made this shift have supported their waivers based on 
the 20 percent standard. In the current regulations, the 24 month data 
reference period for this waiver is tied to the fiscal year and only 
updates each year on October 1. The Department has noticed that as the 
unemployment rates have improved, states that shift the waiver 
operational period to later in the fiscal year have been able to 
capitalize on older data and qualify for waivers of the ABAWD time 
limit for additional time. States are able to take advantage of this 
loophole if their unemployment rates for the requested areas have been 
improving relative to the national average. As a result, these states 
are able to obtain a waiver and maximize the areas waived into the next 
fiscal year, using data that is no longer appropriate as of the October 
1 update.
    To curtail this practice, the Department proposes that waivers 
based on the 20 percent standard would not be approved beyond the 
fiscal year in which the waiver is implemented. In addition, these 
waivers must utilize data from a 24 month period no less recent than 
that DOL used in its current fiscal year LSA designation. Such an 
approach ensures waivers rely on sufficiently recent data for the 
current fiscal year and prevents states from using older data, which 
may not accurately reflect current economic conditions.
    This provision would streamline the implementation of the program 
and would be codified in 7 CFR 273.24(f)(6).
Areas With Limited Data or Evidence
    Current practices provide flexibility to state agencies to rely on 
alternative data sources regardless of whether the area has 
corresponding BLS unemployment data available. Currently, the 
Department may approve requests supported by an estimated unemployment 
rate of an area based on available data from BLS and Census Bureau's 
American Community Survey (ACS), a low and declining employment-to-
population ratio, a lack of jobs as a consequence of declining 
occupations or industries, or an academic study or other publication 
describing the area's lack of a sufficient number of jobs. At times, 
state agencies will use these alternative data sources to justify a 
waiver request even when the corresponding BLS data shows that the 
unemployment rate in the area is relatively low. As stated previously, 
the Department believes that waivers of the ABAWD time limit should be 
limited to only circumstances in which the area clearly does not have a 
sufficient number of jobs to provide employment for the individuals. By 
not restricting the use of these alternative to areas with limited data 
or evidence, the Department has permitted states to take advantage of 
these alternative data sources, when BLS employment data is readily 
available.
    Under the proposed rule, all of these criteria would only be 
applicable to areas for which BLS or a BLS-cooperating agency data is 
limited or unavailable, such as a reservation area or U.S. Territory. 
In these areas, the Department could approve requests supported by an 
estimated unemployment rate of an area based on available data from BLS 
and ACS, a low and declining employment-to-population ratio, a lack of 
jobs as a consequence of declining occupations or industries, or an 
academic study or other publication describing the area's lack of a 
sufficient number of jobs. Waiver requests for an area for which 
standard data from BLS or a BLS-cooperating agency is limited or 
unavailable would not be required to conform to the criteria for 
approval proposed under paragraphs (f)(2), (f)(3), (f)(4), (f)(5), and 
(f)(6). Additionally, the Department would consider other data in line 
with BLS methods or considered reliable. This allows for flexibility if 
new methods or data are developed for Indian Reservation or U.S. 
Territory regions currently with limited or no data.
    Using an estimated unemployment rate based on available data from 
BLS and ACS is part of current practice. The Department proposes 
codifying this criteria in the regulations only for areas with limited 
data or evidence, such as a reservation area or U.S. Territory. 
Currently, states often estimate unemployment rates for reservation 
areas by applying data from ACS to available BLS data. In addition, 
some Tribal governments generate their own labor force and/or 
unemployment data, which would remain acceptable to support a waiver.
    These changes would be codified in 7 CFR 273.24(f)(7).
Other Changes to Waivers
    The proposed rule would eliminate three provisions in current 
regulations: The designation as an LSA as a criterion for approval; the 
implementation of waivers before approval; and the historical seasonal 
unemployment as a criterion for approval. These provisions are 
eliminated to ensure that the ABAWD work requirement is applied in 
accordance with the Department's goal to strengthen work requirements.
    The proposed rule would no longer allow an area to qualify for a 
waiver based on DOL's Employment and Training Administration (ETA) 
designation of the area as an LSA for the current fiscal year. This 
change is central to the Department's efforts to raise the standards by 
which it determines whether an area is lacking a sufficient number of 
jobs to provide employment for ABAWDs in order to require more ABAWDs 
to engage in work, work training, or workfare if they wish to receive 
SNAP. As explained in a previous section, DOL's criteria for LSAs 
require an average unemployment rate that is at least 20 percent above 
the national average and at least six percent for the preceding 2 
calendar years (a 24 month period). The Department is eliminating LSA 
designation as a basis for waiver approval because LSAs are determined 
using a minimum unemployment rate floor of six percent, whereas the 
Department proposes using a minimum unemployment rate of seven percent 
for its similar, but more flexible, 20 percent standard. Continuing to 
allow LSA designation as a basis for waiver approval would be 
inconsistent. Moreover, LSAs are not designated for all different types 
of areas across the country, and having an LSA criteria separate from 
the 20 percent criteria could be seen as unnecessary moving forward.
    The proposed rule would bar states from implementing a waiver prior 
to its approval. Though rarely used, current regulations allow a state 
to implement an ABAWD waiver as soon as the state submits the waiver 
request based on certain criteria.\3\ By removing the current pertinent 
text in 273.24(f)(4), the proposed rule would require states to request 
and receive approval before implementing a waiver. This would allow the 
Department to have a more accurate understanding of the status of 
existing waivers and would provide better oversight in the waiver 
process. It would also prevent waivers from being implemented until the 
Department explicitly reviewed and approved the waiver.
---------------------------------------------------------------------------
    \3\ Under current regulations, the state must certify that data 
from the BLS or the BLS-cooperating agency show a most recent 12 month 
average unemployment rate over ten percent or that ETA designated the 
area as an LSA for the current fiscal year.
---------------------------------------------------------------------------
    The proposed rule would also remove the criterion of a historical 
seasonal unemployment rate over ten percent as a basis for approval. 
Historical seasonal unemployment does not demonstrate a prolonged lack 
of sufficient number of jobs to provide employment for the individuals. 
Historical seasonal unemployment rates, by definition, are limited to a 
relatively short period of time each year. Nor does a historical 
seasonal unemployment rate indicate early signs of a declining labor 
market. Historical seasonal unemployment rates are cyclical rather than 
indicative of declining conditions. Based on operational experience, 
the Department has not typically seen the use of this criterion by 
states. The Department has not approved a waiver under this criterion 
in more than 2 decades. For these reasons, the Department proposes 
removing a historical seasonal average unemployment rate as a way to 
qualify for a waiver.
    In addition, as stated previously, the proposed rule would no 
longer provide for statewide waivers except for those waivers approved 
based upon a state's qualification for extended unemployment benefits.
Ending the ``Carryover'' of ABAWD Exemptions
    The proposed rule would end the unlimited carryover and 
accumulation of ABAWD percentage exemptions, previously referred to as 
15 percent exemptions before the enactment of the Agriculture 
Improvement Act of 2018. Upon enactment, Section 6(o)(6) of the Act 
provides that each state agency be allotted exemptions equal to an 
estimated 12 percent of ``covered individuals,'' which are the ABAWDs 
who are subject to the ABAWD time limit in the state in Fiscal Year 
2020 and each subsequent fiscal year. States can use these exemptions 
available to them to extend SNAP eligibility for a limited number of 
ABAWDs subject to the time limit. When one of these exemptions is 
provided to an ABAWD, that one ABAWD is able to receive 1 additional 
month of SNAP benefits. The Act and current regulations give states 
discretion whether to use these exemptions, and, as a result, some 
states use the exemptions that are available to them and others do not.
    Each fiscal year, the Act requires the Department to estimate the 
number of exemptions that each state be allotted and to adjust the 
number of exemptions available to each state. Based on the Act's 
instructions, the regulations provide the specific formulas that the 
Department must use to estimate the number of exemptions, which are 
referred to as ``earned'' exemptions, and to adjust the exemptions 
available to the state each year. The proposed rule would not change 
any part of the calculation that the Department follows to estimate 
earned exemptions, or any other part of 273.24(g). The proposed rule 
would only change the calculation that the Department uses to adjust 
the number of exemptions available for each fiscal year at 7 CFR 
273.24(h).
    The regulation's current interpretation of Section 6(o)(6)(G) of 
the Act, which requires the adjustment of exemptions, causes unused 
exemptions to carry over and accumulate from 1 year to the next, unless 
the state uses all of its available exemptions in a given year. For FY 
2018, states earned approximately 1.2 million exemptions, but had about 
an additional 7.4 million exemptions available for use due to the 
carryover of unused exemptions from previous fiscal years. The 
Department views the carryover of significant amounts of unused 
exemptions to be an unintended outcome of the current regulations. The 
Department is concerned that such an outcome is inconsistent with 
Congressional intent to limit the number of exemptions available to 
states each year. Concerns about the carryover of exemptions were also 
expressed by the September 2016, USDA Office of the Inspector General 
(OIG) audit report ``FNS Controls Over SNAP Benefits for Able-Bodied 
Adults Without Dependents.'' Therefore, the Department proposes 
revising 7 CFR 273.24(h) to end the unlimited carryover of unused 
percentage exemptions. The Department proposes this change to implement 
the Act more effectively and to advance further the Department's goal 
to promote self-sufficiency.
    In order to address the carryover issue, the proposed rule would 
change the adjustment calculation that the Department uses to increase 
or decrease the number of exemptions available to each state for the 
fiscal year based on usage during the preceding fiscal year. The 
proposed rule would no longer allow for unlimited carryover from all 
preceding years. Instead, each state agency's adjustment would be based 
on the number of exemptions earned in the preceding fiscal year minus 
the number of exemptions used in the preceding fiscal year. The 
resulting difference would be used to adjust (by increasing or 
decreasing) the earned exemption amount. In addition, the adjustment 
will apply only to the fiscal year in which the adjustment is made.
    The three examples below show how the proposed rule's adjustment 
calculation would work in practice based on no exemption use, varied 
exemption use, and exemption overuse. These examples assume that a 
state earns five new exemptions every year over a 4 year period.
Example 1, No Exemption Use
    Example 1 shows how the proposed adjustment calculation would work 
for a state that uses zero exemptions, and how it would end the 
carryover and accumulation of unused exemptions. The state earned five 
exemptions for the current fiscal year (FY) of 2021 in this example 
(row A). The state's adjustment for FY 2021 is based on the number of 
exemptions earned in the previous year (FY 2020) minus the number of 
exemptions used for the previous year (FY 2020). In this example, we 
assume the state earned five exemptions in FY 2020 and used no 
exemptions in FY 2020, so the adjustment for FY 2021 is five (row B). 
The adjustment of five (row B) is then added to the five earned for FY 
2021 (row A) to obtain the state's total of ten exemptions after 
adjustment for FY 2021 (row C). In FY 2021, the state uses zero 
exemptions (row D), so it does not have any overuse liability for that 
year because row E results in a positive number. In FY 2022, FY 2023, 
and FY 2024, the calculation is the same and results are the same each 
year. The number of exemptions available to the state is increased 
based on the number earned for and used in the preceding fiscal year, 
but the state does not carryover accumulated exemptions indefinitely. 
Whereas the state would have 25 total exemptions after adjustment for 
FY 2024 under the current regulations, the state would have ten total 
exemptions after adjustment for FY 2024 under the proposed regulation.

                                                    Example 1
----------------------------------------------------------------------------------------------------------------
                                                  Fiscal year (FY)          2021      2022      2023      2024
----------------------------------------------------------------------------------------------------------------
A.........................................  Earned for current FY.......         5         5         5         5
B.........................................  (+) Adjustment for current           5         5         5         5
                                             FY (earned minus used for
                                             previous FY).
C.........................................  (=) Total after adjustment          10        10        10        10
                                             for current FY.
D.........................................  (^) Used in current FY......         0         0         0         0
E.........................................  (=) Liability for overuse?     10 (No)   10 (No)   10 (No)   10 (No)
                                             (Yes or No).
----------------------------------------------------------------------------------------------------------------

Example 2, Varied Exemption Use
    Example 2 shows how the proposed adjustment calculation would work 
for a state that uses different amounts of exemptions each fiscal year 
and therefore receives an increase or decrease in the exemptions 
available to it each subsequent fiscal year. In other words, the number 
of exemptions available to the state is adjusted for an increased total 
exemptions 1 year, then a decreased total exemptions the next. The 
state earned five exemptions for the current FY of 2021 (row A). The 
state's adjustment for FY 2021 is based on the number of exemptions 
earned in the previous year (FY 2020) minus the number of exemptions 
used for the previous year (FY 2020). We assume the state earned five 
exemptions in FY 2020 but used zero exemptions in FY 2020, so the 
state's total after adjustment for FY 2021 is ten (row C). In FY 2021, 
the state uses eight exemptions (row D), so it does not have any over-
usage liability for that year (row E). That is, though the state only 
earned five exemptions for FY 2021, the adjustment allowed the state to 
avoid any over usage liability for FY 2021. However, for the purposes 
of adjustment in FY 2022, the eight used exemptions are subtracted from 
the five earned exemptions for FY 2021, not from the ten adjusted 
exemption amount available in FY 2021. Therefore, the adjustment amount 
for FY 2022 is negative three. In FY 2022, the state again earns five 
exemptions but the adjustment is negative three (the result of 
subtracting row D, FY 2021 from row A, FY 2022). The state then has a 
total of two exemptions for FY 2022. The state chooses to use two 
exemptions for FY 2022, therefore it has no overuse in FY 2022. This 
example shows how the proposed regulation increases or decreases the 
number of exemptions available to states while also limiting the 
average number of exemptions in effect to 12 percent over time. As 
shown in row D, the state can use no more than ten exemptions over the 
course of any 2 year period, which is equal to the ten exemptions 
earned over every 2 year period.

                                                    Example 2
----------------------------------------------------------------------------------------------------------------
                                                  Fiscal year (FY)          2021      2022      2023      2024
----------------------------------------------------------------------------------------------------------------
A.........................................  Earned for current FY.......         5         5         5         5
B.........................................  (+) Adjustment for current           5        ^3         3        ^3
                                             FY (earned minus used for
                                             previous FY).
C.........................................  (=) Total after adjustment          10         2         8         2
                                             for current FY.
D.........................................  (^) Used in current FY......         8         2         8         2
E.........................................  (=) Liability for overuse?      2 (No)    0 (No)    0 (No)    0 (No)
                                             (Yes or No).
----------------------------------------------------------------------------------------------------------------

Example 3, Exemption Overuse
    Example 3 shows how the proposed adjustment calculation would work 
for a state that overuses exemptions. In this example, we again assume 
the state earned five exemptions in FY 2020 but used zero exemptions in 
FY 2020, so the state's total after adjustment for FY 2021 is ten (row 
C). In FY 2021, the state uses six exemptions (row D); once again, it 
does not have any over-usage liability for that year (row E), but the 
adjustment for FY 2022 will be negative one (the result of subtracting 
row D, FY 2021 from row A, FY 2022). Put differently, the five 
exemptions earned for FY 2022 offset the adjustment of negative one. 
The state then has a total of four exemptions for FY 2022 (row C). 
However, the state uses six exemptions in FY 2022. Because the state 
used more exemptions in FY 2022 than its total after adjustment for FY 
2022, it has an overuse liability of two for FY 2022. The Department 
would consider the exemption overuse an over-issuance and would hold 
the state liable for the total dollar value of the exemptions, as 
estimated by the Department.

                                                    Example 3
----------------------------------------------------------------------------------------------------------------
                                                 Fiscal year (FY)          2021       2022      2023      2024
----------------------------------------------------------------------------------------------------------------
A........................................  Earned for current FY.......         5          5         5         5
B........................................  (+) Adjustment for current           5         ^1        ^1         1
                                            FY (earned minus used for
                                            previous FY).
C........................................  (=) Total after adjustment          10          4         4         6
                                            for current FY.
D........................................  (^) Used for current FY.....         6          6         4         4
E........................................  (=) Liability for overuse?      4 (No)   ^2 (Yes)    0 (No)    2 (No)
                                            (Yes or No).
----------------------------------------------------------------------------------------------------------------

    Under the proposed rule, the Department would continue to provide 
states with its estimated number of exemptions earned for each upcoming 
fiscal year as data becomes available, typically in September. The 
Department would also continue to provide states with the exemption 
adjustments as soon as updated caseload data is available and states 
have provided final data on the number of exemptions used in the 
preceding fiscal year, typically in January.
    The Department also seeks comments from states on how to treat 
state agencies' existing total number of percentage exemptions, which 
in some cases have carried over and accumulated over many years, and on 
when the proposed change should be implemented. Under the proposed 
rule, these accumulated percentage exemptions would not be available to 
states once the change is implemented. Additionally, because the 
adjusted number of exemptions is based on the preceding fiscal year, 
the change in regulatory text will impact state's ability to use 
exemptions in the fiscal year preceding the fiscal year that the 
provision goes into effect. Therefore, the Department seeks comment on 
how to best handle these issues.
    The proposed rule would not change or affect the ``caseload 
adjustments'' at 273.24(h)(1), which apply to any state that has a 
change of over ten percent in its caseload amount. However, the 
Department is taking this opportunity to correct the cross-reference 
that this paragraph makes to 273.24(g)(2) for accuracy. The proposed 
regulation cross-references 273.24(g)(3), instead of (g)(2). The 
Department is making this change because it is more accurate and 
precise to cross-reference to 273.24(g)(3), given that the caseload 
adjustments apply to the number of exemptions estimated as earned for 
each state for each fiscal year.
Procedural Matters
Executive Order 12866 and 13563
    Executive Orders 12866 and 13563 direct agencies to assess all 
costs and benefits of available regulatory alternatives and, if 
regulation is necessary, to select regulatory approaches that maximize 
net benefits (including potential economic, environmental, public 
health and safety effects, distributive impacts, and equity). Executive 
Order 13563 emphasizes the importance of quantifying both costs and 
benefits, of reducing costs, of harmonizing rules, and of promoting 
flexibility. This proposed rule has been determined to be economically 
significant and was reviewed by the Office of Management and Budget 
(OMB) in conformance with Executive Order 12866.
Regulatory Impact Analysis
    As required for rules that have been designated as economically 
significant by the Office of Management and Budget, a Regulatory Impact 
Analysis (RIA) was developed for this proposed rule. It follows this 
rule as an Appendix.* The following summarizes the conclusions of the 
regulatory impact analysis:
---------------------------------------------------------------------------
    * Editor's note: the document referred to was not published in the 
Federal Register; and therefore, is not published in this hearing.
---------------------------------------------------------------------------
    The Department has estimated the net reduction in Federal spending 
associated with the proposed transfer rule to be approximately $1.1 
billion in fiscal year (FY) 2020 and $7.9 billion over the 5 years 
2020-2024. This is a reduction in Federal transfers (SNAP benefit 
payments); the reduction in transfers represents a 2.5 percent decrease 
in projected SNAP benefit spending over this time period.
    Under current authority, the Department estimates that about 60 
percent of ABAWDs live in areas that are not subject to a waiver and 
thus face the ABAWD time limit. Under the revised waiver criteria the 
Department estimates that nearly 90 percent of ABAWDs would live in 
such an area. Of those newly subject to the time limit, the Department 
estimates that approximately \2/3\ (755,000 individuals in FY 2020) 
would not meet the requirements for failure to engage meaningfully in 
work or work training.
Regulatory Flexibility Act
    The Regulatory Flexibility Act (5 U.S.C. 601-612) requires Agencies 
to analyze the impact of rulemaking on small entities and consider 
alternatives that would minimize any significant impacts on a 
substantial number of small entities. Pursuant to that review, it has 
been certified that this rule would not have a significant impact on a 
substantial number of small entities.
    This proposed rule would not have an impact on small entities 
because the proposed rule primarily impacts state agencies. As part of 
the requirements, state agencies would have to update their procedures 
to incorporate the new criteria for approval associated with requesting 
waivers of ABAWD time limit. Small entities, such as smaller retailers, 
would not be subject to any new requirements. However, all retailers 
would likely see a drop in the amount of SNAP benefits redeemed at 
stores if these provisions were finalized, but impacts on small 
retailers are not expected to be disproportionate to impact on large 
entities. As of FY 2017, approximately 76 percent of authorized SNAP 
retailers (nearly 200,000 retailers) were small groceries, convenience 
stores, combination grocery stores, and specialty stores, store types 
that are likely to fall under the Small Business Administration gross 
sales threshold to qualify as a small business for Federal Government 
programs. While these stores make up the majority of authorized 
retailers, collectively they redeem less than 15 percent of all SNAP 
benefits. The proposed rule is expected to reduce SNAP benefit payments 
by about $1.7 billion per year. This would equate to about a $100 loss 
of revenue per small store on average per month ($1.7 billion  15%/
200,000 stores/12 months). In 2017, the average small store redeemed 
more than $3,800 in SNAP each month; the potential loss of benefits 
represents less than three percent of their SNAP redemptions and only a 
small portion of their gross sales. Based on 2017 redemption data, a 
2.7 percent reduction in SNAP redemptions represented between 0.01 and 
0.5 percent of these stores gross sales.
Executive Order 13771
    Executive Order 13771 directs agencies to reduce regulation and 
control regulatory costs and provides that the cost of planned 
regulations be prudently managed and controlled through a budgeting 
process.
    This proposed rule is expected to be an Executive Order 13771 
deregulatory action. The rule does not include any new costs. FNS is 
proposing a reduction in burden hours since state agencies are no 
longer able to group areas together for waiver approval. The reduction 
would result in an estimated collective savings of $12,092 for state 
agencies.
Unfunded Mandates Reform Act
    Title II of the Unfunded Mandates Reform Act of 1995 (UMRA), Public 
Law 104-4, establishes requirements for Federal agencies to assess the 
effects of their regulatory actions on state, local and Tribal 
governments and the private sector. Under section 202 of the UMRA, the 
Department generally must prepare a written statement, including a cost 
benefit analysis, for proposed and final rules with ``Federal 
mandates'' that may result in expenditures by state, local or Tribal 
governments, in the aggregate, or the private sector, of $100 million 
or more in any 1 year. When such a statement is needed for a rule, 
Section 205 of the UMRA generally requires the Department to identify 
and consider a reasonable number of regulatory alternatives and adopt 
the most cost effective or least burdensome alternative that achieves 
the objectives of the rule.
    This proposed rule does not contain Federal mandates (under the 
regulatory provisions of Title II of the UMRA) for state, local and 
Tribal governments or the private sector of $100 million or more in any 
1 year. Thus, the rule is not subject to the requirements of sections 
202 and 205 of the UMRA.
Executive Order 12372
    SNAP is listed in the Catalog of Federal Domestic Assistance under 
No. 10.551. For the reasons set forth in the Final Rule codified in 7 
CFR part 3015, subpart V and related Notice (48 FR 29115), this Program 
is excluded from the scope of Executive Order 12372, which requires 
intergovernmental consultation with state and local officials.
Federalism Summary Impact Statement
    Executive Order 13132 requires Federal agencies to consider the 
impact of their regulatory actions on state and local governments. 
Where such actions have Federalism implications, agencies are directed 
to provide a statement for inclusion in the preamble to the regulations 
describing the agency's considerations in terms of the three categories 
called for under Section 6(b)(2)(B) of Executive Order 13132.
    The Department has determined that this rule does not have 
Federalism implications. Therefore, under Section 6(b) of the Executive 
Order, a Federalism summary impact statement is not required.
Executive Order 12988, Civil Justice Reform
    This proposed rule has been reviewed under Executive Order 12988, 
Civil Justice Reform. This rule is not intended to have preemptive 
effect with respect to any state or local laws, regulations or policies 
which conflict with its provisions or which would otherwise impede its 
full and timely implementation. This rule is not intended to have 
retroactive effect unless so specified in the Effective Dates section 
of the final rule. Prior to any judicial challenge to the provisions of 
the final rule, all applicable administrative procedures must be 
exhausted.
Civil Rights Impact Analysis
    FNS has reviewed the proposed rule, in accordance with the 
Department Regulation 4300-4, ``Civil Rights Impact Analysis'' to 
identify and address any major civil rights impacts the proposed rule 
might have on minorities, women, and persons with disabilities. While 
we believe that a reduction in the number of ABAWD waivers granted to 
state agencies will adversely affect potential program participants in 
all groups who are unable to meet the employment requirements, and have 
the potential for disparately impacting certain protected groups due to 
factors affecting rates of employment of members of these groups, we 
find that the implementation of mitigation strategies and monitoring by 
the Civil Rights Division of FNS will lessen these impacts.
Executive Order 13175
    This rule has been reviewed in accordance with the requirements of 
Executive Order 13175, ``Consultation and Coordination with Indian 
Tribal Governments.'' Executive Order 13175 requires Federal agencies 
to consult and coordinate with Tribes on a government-to-government 
basis on policies that have Tribal implications, including regulations, 
legislative comments or proposed legislation, and other policy 
statements or actions that have substantial direct effects on one or 
more Indian Tribes, on the relationship between the Federal Government 
and Indian Tribes or on the distribution of power and responsibilities 
between the Federal Government and Indian Tribes.
    The USDA's Office of Tribal Relations (OTR) has assessed the impact 
of this rule on Indian Tribes and determined that this rule has Tribal 
implications that require Tribal consultation under E.O. 13175. FNS 
invited Tribal leaders to a consultation held on March 14, 2018. Tribal 
leaders did not provide any statement or feedback to the Department on 
the rule. FNS and OTR will determine if a future consultation is 
needed. If a Tribe requests consultation, FNS will work with the Office 
of Tribal Relations to ensure meaningful consultation is provided where 
changes, additions, and modifications identified herein are not 
expressly mandated by Congress
Paperwork Reduction Act
    The Paperwork Reduction Act of 1995 (44 U.S.C. Chap. 35; 5 CFR 
1320) requires the Office of Management and Budget (OMB) approve all 
collections of information by a Federal agency before they can be 
implemented. Respondents are not required to respond to any collection 
of information unless it displays a current valid OMB control number. 
In accordance with the Paperwork Reduction Act of 1995, this proposed 
rule will contain information collections that are subject to review 
and approval by the Office of Management and Budget; therefore, FNS is 
submitting for public comment the changes in the information collection 
burden that would result from adoption of the proposals in the rule.
    Comments on this proposed rule must be received by April 2, 2019. 
Comments are invited on: (a) Whether the proposed collection of 
information is necessary for the proper performance of the functions of 
the agency, including whether the information shall have practical 
utility; (b) the accuracy of the agency's estimate of the burden of the 
proposed collection of information, including the validity of the 
methodology and assumptions used; (c) ways to enhance the quality, 
utility, and clarity of the information to be collected; and (d) ways 
to minimize the burden of the collection of information on those who 
are to respond, including use of appropriate automated, electronic, 
mechanical, or other technological collection techniques or other forms 
of information technology.
    All responses to this notice will be summarized and included in the 
request for OMB approval. All comments will also become a matter of 
public record.
    Title: Supplemental Nutrition Assistance Program Waivers of Section 
6(o) of the Food and Nutrition Act.
    OMB Number: 0584-0479.
    Expiration Date: [July 31, 2021].
    Type of Request: Revision of a currently approved collection.
    Abstract: Section 6(o) of the Food and Nutrition Act of 2008, (the 
Act, as amended through Pub. L. 113-xxx), limits the amount of time an 
able-bodied adult without dependents (ABAWD) can receive Supplemental 
Nutrition Assistance Program (SNAP) benefits to 3 months in a 36 month 
period, unless the individual is working and/or participating in a work 
program half-time or more, or participating in workfare. The Act 
exempts individuals from the time limit for several reasons, including 
age, unfitness for work, or having a dependent child. The ABAWD time 
limit and work requirement currently apply to people ages 18 through 
49, unless they are already exempt from the general work requirements, 
medically certified as physically or mentally unfit for employment, 
responsible for a child under 18, or pregnant. ABAWDs are also work 
registrants and must meet the general work requirements. In addition, 
ABAWDs subject to the time limit must work and/or participate in a work 
program 80 hours per month or more, or participate in and comply with 
workfare to receive SNAP for more than 3 months in a 36 month period. 
Participation in SNAP E&T, which is a type of work program, is one way 
a person can meet the 80 hour per month ABAWD work requirement, but 
other work programs are acceptable as well.
    The Act also provides state agencies with flexibility to request a 
waiver of this time limit if unemployment is high or the area does not 
have a sufficient number of jobs to provide employment. State agencies 
can request to waive the ABAWD time limit if an area has an 
unemployment rate of over ten percent or the state can meet one of the 
regulatory options to show it does not have a sufficient number of jobs 
to provide employment. If the time limit is waived, individuals are not 
required to meet the ABAWD work requirement to receive SNAP for more 
than 3 months in a 36 month period. This collection of information is 
necessary for FNS to perform its statutory obligation to review waivers 
of the SNAP ABAWD time limit.
    This is a revision of a currently approved information collection 
request associated with this rulemaking. In the previous submission, 
the Food and Nutrition Service (FNS) estimated 35 hours for each waiver 
request for a total of 1,198 hours. Based on the experience of FNS 
during calendar year 2018, FNS projects that 36 out of 53 state 
agencies would submit requests for a waiver of the time limit for ABAWD 
recipients based on a high unemployment rate or lack of sufficient 
number of jobs. FNS estimates a response time of 28 hours for each 
waiver request based on labor market data, which require detailed 
analysis of labor markets within the state. FNS projects a total of 
1,008 hours, which would be a reduction of 190 hours compared to the 
1,198 hours estimated provided in the pending approval.
    FNS is proposing a reduction in burden hours since state agencies 
are no longer able to group areas together for waiver approval. The 
reduction will burden hours would result in an estimated collective 
savings of $12,092 for state agencies. This rule does not require any 
recordkeeping burden. Reporting detail burden details are provided 
below.
    Respondents: State agencies.
    Estimated Number of Respondents: 36.
    Estimated Number of Responses per Respondent: 1.
    Estimated Total Annual Burden on Respondents: 1,008.

--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                 Response                                                       Differences
                                 Requirement (7    Estimated     annually      Total      Hours per      Annual      Previous      due to    Differences
       OMB No. 0584-0479          CFR 273.24(f)    number of       per         annual      response      burden     submission    program       due to
                                                  respondents   respondent   responses                   hours     total hours    changes     adjustment
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                             Affected Public: State Agencies
--------------------------------------------------------------------------------------------------------------------------------------------------------
Reporting burden..............  Submissions of             36            1           36           28        1,008        1,190         ^182            0
                                 waiver request
                                 based on labor
                                 market data..
                                7 CFR 273.24(f)--           0            0            0            0            0            8           ^8            0
                                 Submission of
                                 waiver request
                                 based on Labor
                                 Surplus Area
                                 designation..
                               -------------------------------------------------------------------------------------------------------------------------
  Reporting totals..............................           36                                               1,008                      ^190
                                                 -------------------------------------------------------------------------------------------------------
  Total Reporting Burden due to Rulemaking......                                                            1,008
--------------------------------------------------------------------------------------------------------------------------------------------------------

E-Government Act Compliance
    The Department is committed to complying with the E-Government Act 
of 2002, to promote the use of the Internet and other information 
technologies to provide increased opportunities for citizen access to 
government information and services, and for other purposes.
List of Subjects in 7 CFR Part 273
    Able-bodied adults without dependents, Administrative practice and 
procedures, Employment, Indian reservations, Time limit, U.S. 
territories, Waivers, Work requirements.
    Accordingly, FNS proposes to amend 7 CFR part 273 to read as 
follows:
part 273--certification of eligible households
     1. The authority citation for part 273 continues to read as 
follows:

          Authority: 7 U.S.C. 2011-2036.

     2. In  273.24, revise paragraph (f) to read as follows:
 273.24  Time Limit for able-bodied adults.
          * * * * *
    (f) Waivers--.(1) General. The state agency may request FNS 
approval to temporarily waive the time limit for a group of individuals 
in the state in the area in which the individuals reside. To be 
considered for approval, the request must be endorsed by the state's 
governor and supported with corresponding data or evidence 
demonstrating that the requested area:

          (i) Has an unemployment rate of over ten percent; or
          (ii) Does not have a sufficient number of jobs to provide 
        employment for the individuals.

    (2) Core standards. FNS will approve waiver requests under (1)(i) 
and (ii) that are supported by any one of the following:

          (i) Data from the Bureau of Labor Statistics (BLS) or a BLS-
        cooperating agency that shows an area has a recent 12 month 
        average unemployment rate over ten percent;
          (ii) Data from the BLS or a BLS-cooperating agency that shows 
        an area has a 24 month average unemployment rate 20 percent or 
        more above the national rate for a recent 24 month period, but 
        in no case may the 24 month average unemployment rate of the 
        requested area be less than seven percent. The 24 month period 
        must be no earlier than the same 24 month period used by the 
        Department of Labor's Employment and Training Administration to 
        designate Labor Surplus Areas for the current fiscal year; or
          (iii) Evidence that an area qualifies for extended 
        unemployment benefits as determined by the Department of Labor 
        (DOL).

    (3) Other data and evidence. FNS may approve waiver requests that 
are supported by data or evidence other than that listed under 
paragraph (f)(2) of this section if the request demonstrates an 
exceptional circumstance in an area. In addition, the request must 
demonstrate that the exceptional circumstance has caused a lack of 
sufficient number of jobs, such as data from the BLS or a BLS-
cooperating agency that shows an area has a most recent 3 month average 
unemployment rate over ten percent. Supporting unemployment data 
provided by the state must rely on standard BLS data or methods.
    (4) Restriction on statewide waivers. FNS will not approve 
statewide waiver requests if data for the requesting state at the sub-
state level is available from BLS, except for waivers under paragraph 
(f)(2)(iii) of this section.
    (5) Restricting the combining of data to group sub-state areas. The 
state agency may only combine data from individual areas that are 
collectively considered to be a Labor Market Area by DOL.
    (6) Duration of waiver approvals. In general, FNS will approve 
waivers for 1 year. FNS may approve waivers for a shorter period at the 
state agency's request and waivers under paragraph (f)(2)(ii) of this 
section will not be approved for a period beyond the fiscal year in 
which the waiver is implemented.
    (7) Areas with limited data or evidence. Waiver requests for an 
area for which standard BLS data or a BLS-cooperating agency data is 
limited or unavailable, such as a reservation area or U.S. Territory, 
are not required to conform to the criteria for approval under 
paragraphs (f)(2), (f)(3), (f)(4), (f)(5) and (f)(6) of this section. 
The supporting data or evidence provided by the state must correspond 
to the requested area.

          (i) FNS may approve waivers for these areas if the requests 
        are supported by sufficient data or evidence, such as:

                  (A) Estimated unemployment rate based on available 
                data from BLS and Census Bureau's American Community 
                Survey;
                  (B) A low and declining employment-to-population 
                ratio;
                  (C) A lack of jobs in declining occupations or 
                industries; or
                  (D) An academic study or other publication describing 
                the area as lacking a sufficient number of jobs to 
                provide employment for its residents.

          (ii) In areas with limited data or evidence, such as 
        reservation areas or U.S. Territories, FNS may allow the state 
        agency to combine data from individual areas to waive a group 
        of areas if the state agency demonstrates that the areas are 
        economically integrated.
          * * * * *
     3. In  273.24, revise paragraph (h) to read as follows:
          * * * * *
    (h) Adjustments. FNS will make adjustments as follows:

          (1) Caseload adjustments. FNS will adjust the number of 
        exemptions estimated for a state agency under paragraph (g)(3) 
        of this section during a fiscal year if the number of SNAP 
        recipients in the state varies from the state's caseload by 
        more than ten percent, as estimated by FNS.
          (2) Exemption adjustments. During each fiscal year, FNS will 
        increase or decrease the number of exemptions allocated to a 
        state agency based on the difference between the number of 
        exemptions used by the state for the preceding fiscal year and 
        the number of exemptions estimated for the state for the 
        preceding fiscal year under paragraphs (g)(3) and (h)(1) of 
        this section. The increase or decrease will only apply for the 
        fiscal year in which the adjustment is made. For example:

                  (i) If the state agency uses fewer exemptions in the 
                preceding fiscal year than were estimated for the state 
                agency by FNS for the preceding fiscal year under 
                paragraphs (g)(3) and (h)(1) of this section, FNS will 
                increase the number of exemptions allocated to the 
                state agency for the current fiscal year by the 
                difference to determine the adjusted exemption amount.
                  (ii) If the state agency uses more exemptions in the 
                preceding fiscal year than were estimated for the state 
                agency by FNS for the preceding fiscal year under 
                paragraphs (g)(3) and (h)(1) of this section, FNS will 
                decrease the number of exemptions allocated to the 
                state agency for the current fiscal year by the 
                difference to determine the adjusted exemption amount.
          * * * * *
    Dated: December 20, 2018.

Brandon Lipps,
Acting Deputy Under Secretary, Food, Nutrition, and Consumer Services.

[FR Doc. 2018-28059 Filed 1-31-19; 8:45 a.m.]

billing code 3410-30-p
                                 ______
                                 
 Submitted Comment Letter by Hon. Collin C. Peterson, a Representative 
  in Congress from Minnesota; Authored by Tony Lourey, Commissioner, 
                 Minnesota Department of Human Services
March 29, 2019

  Brandon Lipps,
  Administrator, Food and Nutrition Service
  U.S. Department of Agriculture;

  Certification Policy Branch,
  Program Development Division,
  United States Department of Agriculture--Food and Nutrition Service,
  Alexandria, Virginia 22302

  Re: Docket No. FNS-2018-0004, RIN 0584-AE57, Comments in Response to 
            Proposed Rulemaking: Supplemental Nutrition Assistance 
            Program: Requirements and Services for Able-Bodied Adults 
            without Dependents

    Dear Mr. Lipps:

    The Minnesota Department of Human Services (MN DHS) oversees the 
state's Supplemental Nutrition Assistance Program (SNAP) to provide 
critical food assistance to low-income families. As Commissioner of the 
department, I have serious concerns about the proposed rule regarding 
SNAP waivers that the Food and Nutrition Service (FNS) published in the 
Federal Register on February 1, 2019. This rule will likely increase 
hunger and deprivation among thousands of people in Greater Minnesota 
by causing them to lose their benefits.
    Under current law, working-age adults who do not have dependent 
children must either have a job or be enrolled in officially-recognized 
employment training for 20 hours per week in order to receive more than 
3 months of SNAP benefits in a 3 year time period. States can waive the 
time limit for this population in geographic areas that have an 
unemployment rate that is 20 percent above the national average. In 
Minnesota, 30 counties and 11 American Indian reservations and Tribal 
areas, all of which are in rural areas, currently receive these SNAP 
waivers.\1\ The proposed rule would limit the existing criteria for 
granting SNAP waivers in a way that would cause much of the population 
in these areas to lose SNAP benefits.
---------------------------------------------------------------------------
    \1\ The following counties are currently eligible for a waiver from 
the 3 month time limit: Aitkin, Becker, Beltrami, Carlton, Cass, 
Clearwater, Cook, Cottonwood, Crow Wing, Hubbard, Isanti, Itasca, 
Kanabec, Kittson, Koochiching, Lake, Lake of the Woods, Mahnomen, 
Marshall, Mille Lacs, Morrison, Murray, Norman, Pennington, Pine, Red 
Lake, Roseau, St. Louis, Todd, Wadena.
---------------------------------------------------------------------------
    Understanding the low-wage labor market is critical to 
understanding the role that SNAP plays in helping workers mitigate the 
instability of low-wage work. SNAP is a critical support for workers 
who earn wages that are so low that they live in poverty despite 
working. It also helps these workers when they experience a spell of 
unemployment. The vast majority of working-age SNAP recipients in 
Minnesota work in low-wage jobs that offer little employment security, 
erratic and unpredictable schedules, and few benefits. These industries 
include hotels and restaurants, retail, temporary placement agencies, 
and health care's low-wage occupations. The jobs in these industries 
are much more likely than other sectors to be part-time and have high 
worker turnover. Many of the adults subject to SNAP time limits lack 
basic skills in reading, math, and writing and face other barriers to 
employment which can limit their job prospects. This group of SNAP 
recipients is also more likely than the larger SNAP population and the 
overall statewide population to be homeless, lack transportation, have 
an addiction, or experience domestic violence.\2\ SNAP helps mitigate 
the effects of low pay and job unpredictability to help workers weather 
the inevitable unemployment spells that come with low-wage jobs.
---------------------------------------------------------------------------
    \2\ U.S. Government Accountability Office (2003). Food Stamp 
Employment and Training Program Better Data Needed to Understand Who Is 
Served and What the Program Achieves: https://www.gao.gov/assets/240/
237571.pdf.
---------------------------------------------------------------------------
    The concerns outlined below highlight changes proposed in the rule 
that would further undermine the well-being of low-wage workers 
receiving SNAP in Minnesota:

  (1)  The rule proposes to eliminate statewide waivers, which would 
            leave Minnesota vulnerable during severe economic crises. 
            In addition to providing a nutrition safety net during 
            periods of economic volatility, the use of SNAP benefits 
            also boosts local economies by providing economic stimulus 
            to grocers, farmers, and others in the food pipeline. The 
            Great Recession which began in 2008 eliminated 160,000 jobs 
            in Minnesota. When people lose their jobs, the wider 
            economy is vulnerable because those individuals can no 
            longer make purchases or pay bills. SNAP not only ensures 
            that people who are unemployed can purchase groceries, but 
            also that local food retailers still have customers and can 
            keep their staff employed during difficult economic times.

        A USDA Economic Research Service analysis estimated that each 
            $1 in Federal SNAP benefits generates $1.79 in economic 
            activity. Those dollars help food retailers (many of which 
            are operating on thin margins) improve food access for all 
            residents. Historically, Minnesota has had a relatively 
            strong economy and only had a statewide waiver during the 
            2008 recession. That is exactly the sort of scenario in 
            which programs like SNAP must respond quickly and 
            effectively to diminish the impact of the crisis on 
            individuals and slow a widening economic crisis.

  (2)  The proposed rule changes the criteria used to qualify a region 
            for a SNAP waiver based on high unemployment. The current 
            standard for ``insufficient jobs'' that can qualify an area 
            for a waiver is an unemployment rate of at least 20% above 
            the national average. This rule would create an additional 
            standard by requiring waivered areas to also have a minimum 
            unemployment rate of either 6%, 7%, or 10% (the proposed 
            rule asks for public comment on the impact of each of these 
            unemployment rates).

        The unemployment rate is not a complete measure of economic 
            stress and establishing a minimum unemployment rate in this 
            arbitrary manner lacks the evidence-based rigor needed when 
            making a major policy change. Minnesota has very distinct 
            regions, some of which rely primarily on agriculture, 
            mining, food processing, health care, or mixed sectors 
            which each follow distinct economic cycles. Some regions 
            can be flourishing in our state while others are struggling 
            economically. If FNS were to apply a minimum unemployment 
            rate of 7%, only four of the 30 counties \3\ that are 
            included in the waiver would continue to qualify. All 
            American Indian reservations and Tribal areas would 
            continue to qualify. Under such a change, 2,650 Minnesotans 
            would be subject to the 3 month time limit.
---------------------------------------------------------------------------
    \3\ The counties that would still qualify under an unemployment 
rate floor of seven percent are: Clearwater, Itasca, Koochiching, and 
Marshall.

  (3)  The proposed rule would limit local control and state 
            flexibility in defining areas of high unemployment by 
            forcing states to make the determinations using only small 
            Labor Market Areas recognized by the Bureau of Labor 
            Statistics (BLS). This approach fails to recognize the 
            economic reality in rural areas of Minnesota. The Bureau of 
            Labor Statistics designated a small Labor Market Area by 
            measuring whether at least 25% of a county's residents or 
            employees are associated with a neighboring county. 
            Applying that narrow methodology to SNAP waivers misses 
            that fact that in some counties, workers may have to travel 
            in all directions and often beyond a contiguous county for 
            their job. States have the best understanding of the 
            regional patterns in their labor markets and can best 
            account for that when applying for waivers. Using the BLS 
            small Labor Market Area for such determinations is 
---------------------------------------------------------------------------
            misguided.

  (4)  While the 2018 Farm Bill modified the number of exemptions from 
            SNAP time limits that states can receive each year from 15% 
            to 12%, it did not change their ability to carry over 
            unused exemptions. The proposed rule would no longer allow 
            states to carry over all unused exemptions from 1 year to 
            another. This change restricts states' ability to use the 
            program's policies to respond to shifts in the labor market 
            and the economy. Minnesota would naturally use fewer 
            exemptions when the labor markets across the state are 
            relatively strong and would increase the use of exemptions 
            when the labor markets weaken. That ability to respond 
            should not be restricted.

        The proposal would also allow FNS to apply this aspect of the 
            rule change retroactively, which would also be harmful to 
            Minnesota. States that have earned exemptions and were 
            allowed to carry them over across Federal fiscal years 
            should be able to continue to do so. Our current 
            accumulations from previous years should not be dismissed. 
            States know their residents and their geographic regions 
            best, and should be allowed to determine how these 
            exemptions could be used to address continued challenges 
            for some of their low-wage workers.

  (5)  Implementing the proposed rule changes by October 1, 2019 would 
            undoubtedly lead to errors and confusion. Major changes in 
            complex systems need to be well-planned so they can be 
            well-implemented. If any of the provisions of the proposed 
            rule are enacted, they should not be implemented any sooner 
            than October 1, 2020.

    If the changes outlined in this proposed rule go into effect, they 
would force many workers in areas with unemployment rates at least 20% 
more than the national rate to lose their SNAP benefits. They would be 
forced to find jobs that are not available or to enroll in employment 
services that do not exist. There is not enough funding in the SNAP 
Employment and Training program to serve the people currently subject 
to time limits, much less thousands of new workers subject to the time 
limit. If Minnesota were to apply the small increase in funding for the 
SNAP Employment and Training program from the 2018 Farm Bill to all the 
individuals affected by this rule change, we estimate that we would 
only have $35 per person to spend on employment and training services 
for people that face multiple barriers to work.
    Congress had the opportunity to include these policy changes in the 
recently passed farm bill but chose to not do so. To make these changes 
through executive action, without providing the resources to help low-
wage workers improve their odds of getting jobs, only increases 
hardship for people who are already struggling to afford the basics. 
The rules governing eligibility for waivers and individual exemptions 
have been in place for nearly 20 years. In that time, they have proven 
to be reasonable, transparent, and manageable for states to 
operationalize.
    Although this rule may be meant to increase the number of people 
engaged in work, these changes would actually undermine low-wage 
workers' ability to reach stability. Minnesota's economic well-being 
depends on all workers being able to meet their basic needs and provide 
local businesses with customers, even when the economy weakens. I urge 
you, for the benefit of working people in Greater Minnesota, to reject 
the changes proposed in this rule.
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Tony Lourey,
Commissioner.
                                 ______
                                 
    Submitted Letter by Hon. James P. McGovern, a Representative in 
                      Congress from Massachusetts
February 27, 2018

  Hon. Glen Thompson,
  Chairman,
  Subcommittee on Nutrition,
  House Committee on Agriculture,
  Washington, D.C.

    Dear Chairman Thompson:

    It has been a pleasure serving with you on the Nutrition 
Subcommittee, and I have appreciated the Majority's diligence in 
conducting a thorough review of the Supplemental Nutrition Assistance 
Program (SNAP) over the past several years.
    During the 23 hearings our Committee has held on SNAP, we've heard 
from experts--conservative and liberal--that SNAP works. We've learned 
that benefits should not be cut, and that current benefits are 
inadequate. We also learned that SNAP does not discourage work, and 
that eliminating work waivers will hamper state flexibility and 
increase hunger.
    Despite all of these hearings and findings, I'm concerned by 
reports that the Committee is drafting a bill, behind closed doors, 
that will seek to dramatically undermine access to SNAP benefits for 
the population of very vulnerable able-bodied adults without 
dependents, known as ABAWDs. My concern has only grown in the past 
several weeks as the Administration has proposed drastic changes to 
this population through its budget proposal and solicited feedback on 
advancing its goal of moving ABAWDs out of the SNAP program.
    I am now respectfully requesting that the Nutrition Subcommittee 
hold a hearing on the ABAWD population before making any changes to 
current SNAP law impacting this group of vulnerable adults.
    Members of this Committee deserve the opportunity to learn more 
about the ABAWD population from expert witnesses before voting on any 
legislation that could limit their access to modest food benefits.
    Thank you for your consideration of this request, and I look 
forward to hearing from you soon.
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Hon. James P. McGovern,
Ranking Minority Member,
Subcommittee on Nutrition.
                                 ______
                                 
   Submitted Article by Hon. James P. McGovern, a Representative in 
                      Congress from Massachusetts
Trump to poor Americans: Get to work or lose your benefits
The Washington Post
Wonkblog/Analysis
By Caitlin Dewey and Tracy Jan
May 22, 2017 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          A group of homeless men and women receive meals from 
        volunteers on May 18 in Morgantown, West Virginia. West 
        Virginia is one of the nation's poorest states where nearly one 
        in five struggled to afford basic necessities in 2015. (Spencer 
        Platt/Getty Images)

    For a period last year after he lost his food stamps, Tim Keefe, an 
out-of-work and homeless Navy veteran, used his military training to 
catch, skin and eat squirrels, roasting the animals over an open fire 
outside the tent he pitched in frigid Augusta, Maine.
    The new additions to Keefe's diet resulted from a decision by state 
authorities to tighten work requirements for recipients of the social 
safety net--forcing the 49 year old, who lost his job at a farm 
equipment factory because of an injury, off the food stamp rolls.
    ``I was eating what I could find, and borrowed from friends and 
strangers,'' Keefe said in testimony to the Maine legislature. ``There 
were many times . . . when I would go 2 or even 3 days without food. If 
one was inclined to lose a lot of weight, I could recommend this diet 
wholeheartedly.''
    Now the Trump Administration in its first major budget proposal has 
proposed more stringent work requirements--similar to those in effect 
in Maine and other states--to limit eligibility for food stamps and a 
host of other benefits as part of sweeping cuts to anti-poverty 
programs.
    The White House budget proposal, due to be unveiled on Tuesday, 
would reduce spending on anti-poverty programs from food stamps to tax 
credits and welfare payments by $274 billion over a decade, largely by 
tightening eligibility for these programs, according to Administration 
officials. With additional reforms on Medicaid and disability 
insurance, total safety net cuts would top $1 trillion over 10 years.
    Making low-income Americans work to qualify for so-called welfare 
programs is a key theme of the budget. ``If you are on food stamps and 
you are able-bodied, we need you to go to work,'' said budget director 
Mick Mulvaney during a White House briefing on Monday.
    He said the strengthened requirements in the budget focuses on 
putting the 6.8 million unemployed or underemployed Americans back to 
work. ``There is a dignity to work,'' he said, ``and there's a 
necessity to work to help the country succeed.''
    The White House did not offer details Monday on how the work 
requirements would be implemented, other than saying it would be 
``phased in'' for able-bodied adults without dependent children.
    The White House estimated the combined reforms to the Supplemental 
Nutrition Assistance Program, better known as food stamps, would 
generate nearly $193 billion in savings over a decade.
    In addition to SNAP reforms, Trump will propose taking the earned 
income and child tax credits away from undocumented immigrants working 
in the United States, many of whom pay taxes or have American born-
children. That reform alone would save $40 billion over a decade, 
according to the White House.
    Anti-poverty advocates say the White House could implement its 
desired reforms to SNAP in two ways: require recipients to work more 
than the current minimum of 20 hours a week, or cut the unemployment 
waivers in areas with high joblessness rates.
    The influential Heritage Foundation, as well as a number of House 
conservatives have championed a crackdown on waivers, leading many 
anti-poverty advocates to conclude that is the most likely way the 
White House would implement its proposed reforms.
    Robert Rector, a senior research fellow at the Heritage Foundation 
who has asked the White House to prioritize work requirements, said the 
Trump Administration needs to ``go after'' the four million able-bodied 
adults without dependents in the food stamp program.
    ``You say to them, `We will give you assistance, but come to the 
office 1 day a week to do job search or community service,' '' Rector 
said. ``When Maine did that, they found almost immediately that their 
caseload dropped 85 percent.''
    Critics say such a change could endanger people like Keefe, a 
veteran who has been unable to find a job after injuring his wrist on 
the job at a plow factory in Rockland, Maine. As a result, Keefe now is 
medically unable to lift more than 25 pounds--which disqualifies him 
from other work in manufacturing.
    The Navy veteran was one of several thousand former food stamp 
recipients who lost benefits when Maine, in 2015, declined to renew its 
waiver and reinstated statewide work requirements. He has spent much of 
the last year living in a tent.
    ``I don't wanna worry no one,'' said Keefe, who recently testified 
to Maine's Committee on Health and Human Services about the impact the 
work requirement had on him. But, he added: ``I hope they understand 
that people fall through the cracks.''
    The Trump Administration is considering other changes to SNAP. 
While details remain sparse, Mulvaney said the Federal Government would 
be asking states to share in the costs for the food stamps program, 
through a phased-in ``state match'' so they have a ``little more skin 
in the game.''
    ``We believe in the social safety net. We absolutely do,'' Mulvaney 
said. ``What we've done is not to try and remove the safety net for 
folks who need it, but to try and figure out if there's folks who don't 
need it that need to be back in the workforce.''
    Suspending employment waivers would hit hard in areas with high 
unemployment such as southern and central California, where the 
unemployment rate can spike as high as 19 percent, as well as cities 
such as Detroit and Scranton, Pa., where joblessness remains rampant. 
The change would also hit hard in large portions of New Mexico, Oregon, 
Washington, Georgia, Kentucky, Tennessee, West Virginia, Idaho and 
Michigan.
    ``It's unconscionable, cruel and ineffective,'' said Josh Protas, 
the Vice President of Public Policy at MAZON, a national anti-hunger 
organization. ``I'm honestly not sure what their goal is.''
    Critics say the changes in unemployment waivers would be 
devastating for Native American families living on reservations in 
North and South Dakota, Arizona and Montana where there is chronic 
poverty and high unemployment.
    ``The President's budget proposal will force kids in rural America 
to go hungry while wasting billions of taxpayer dollars on misplaced 
priorities like a wall that won't keep us safe,'' said Senator Jon 
Tester (D-MT), in a statement to the Post. ``Parents in Montana and 
across Indian Country should not have to choose between food for their 
tables, gas for their cars, and shoes for their kids.''
    The number of Americans on SNAP remains high, however. In 2016, 44 
million Americans receive the benefits, compared to just 28 million 
people in 2008.
    ``They have not come down like we would expect them to do,'' 
Mulvaney said. ``That raises a very valid question: Are there folks on 
SNAP who shouldn't be?''
    Anti-hunger advocates argue that, generally speaking, there are 
not. Because SNAP benefits decrease gradually with increased income, 
there is no incentive for people to avoid work to get benefits--a 
phenomenon economists call the ``welfare cliff.'' And benefits are too 
small for people to subsist on them without working: The average food 
stamp benefit was $465 a month for a family of four in 2015. Most 
people are on the program for between 7 and 9 months on average.
    ``The notion that people would prefer not to work to get that 
benefit, give me a break,'' said U.S. Representative Jim McGovern, (D-
Mass.) a longtime anti-hunger advocate. ``This is a lousy and rotten 
thing to do to poor people. They look at SNAP as an ATM to pay for 
their other priorities.''
    Additionally, \3/4\ of households using SNAP contain children, 
seniors, or people with disabilities, said Elaine Waxman, a senior 
fellow in the Income and Benefits Policy Center at the Urban Institute. 
Without SNAP, the country would have had three to 4.5 million more 
people in poverty during the recession, she said.
    More than \1/4\ of able-bodied adults without dependents on SNAP do 
not have a high school diploma, Waxman said; another 57 percent don't 
have college degrees--putting them at a disadvantage when it comes to 
finding work.
    A number are also veterans, young adults aging out of the foster 
care system, and felons recently released from jail. SNAP recipients 
who cannot find work, for these or other reasons, are supposed to 
attend job training programs--but they're not widely available because 
of lack of funding.
    ``This is the trick. On the one hand, you want people to do 
something, when in fact a lot of folks may not realistically be able to 
find a job,'' Waxman said. ``Most states don't want to put the money 
in. This is a dilemma that we're in.''
    The evidence that stricter work requirements actually cause people 
to get jobs is mixed, at best. In Kansas, which reinstated the 
requirements in October 2014, 40 percent of unemployed adults were 
still unemployed a year after being kicked off SNAP. Among former SNAP 
participants who lost benefits, the average annual income was only 
$5,562, according to the Foundation for Government Accountability, a 
right-wing think tank based in Florida.
    Progress has also been hotly debated in Maine, a state that 
conservatives regularly hold up as evidence that stricter work-
requirements are effective. When the state dropped its waiver in 2015, 
the number of unemployed adults in the program immediately fell by 
nearly 80 percent.
    But a May 2016 report by the state found that nearly 60 percent of 
those affected individuals did not report any income in the year after 
they left the program--suggesting they were still unemployed or 
underemployed a year later.
    On the national level, Michael Tanner, a senior fellow who focuses 
on social welfare issues at the Cato Institute, a libertarian think 
tank, said he doesn't think similar mandates will have a huge impact on 
moving large numbers of recipients into employment or result in 
significant budget savings. Most SNAP recipients who can work are 
already working, and many of those who are not meet one of the various 
exemptions such as being disabled.
    ``It's making a statement that Republicans think people who are on 
public assistance should be doing all they can to get off,'' Tanner 
said, ``and that means working whenever possible.''
    McGovern, who sits on the House Agriculture Committee, said he was 
surprised to learn about the White House proposal given Agriculture 
Secretary Sonny Perdue's testimony before the Committee last week 
saying he did not favor any major changes to the food stamps program.
    ``It's been a very important, effective program,'' Perdue said, 
according to a recording of the hearing. ``As far as I'm concerned we 
have no proposed changes. You don't try to fix things that aren't 
broken.''
    The Trump Administration is advocating other ``fixes'' to the 
safety net, as well. The budget will also propose requiring people to 
have a Social Security [N]umber to collect tax credits. Mulvaney said 
it is unfair that taxpayers support immigrants working illegally in 
this country.
    ``How do I go to somebody who pays their taxes and say, `Look, I 
want you to give this earned income tax credit to somebody who is 
working here illegally? That's not defensible,'' Mulvaney said.
    Rector, of the Heritage Foundation, said he also hopes Trump will 
prioritize work requirements for those receiving housing subsidies. 
Mulvaney did not address that on Monday.
    Diane Yentel, President of the National Low Income Housing 
Coalition, said the majority of Americans receiving housing subsidies 
are elderly, disabled or already include someone who works. Of the 
remaining households, nearly \1/2\ include a preschool child or an 
older child or adult with a disability who needs the supervision of a 
caregiver.
    Establishing work requirements for the remaining six percent of 
households who are `work able' but not employed would require state and 
local housing agencies already facing funding shortfalls to establish 
cumbersome monitoring and enforcement systems for a very narrow segment 
of rental assistance recipients, she said.
    ``This is neither cost effective nor a solution to the very real 
issue of poverty impacting millions of families living in subsidized 
housing or in need,'' Yentel said in a statement to the Post.

          Correction: This story incorrectly stated the average annual 
        income for SNAP participants in Kansas who had lost and then 
        found jobs was $5,562. That figure applied to all SNAP 
        participants who had lost the benefit.

          Caitlin Dewey is the food policy writer for Wonkblog. 
        Subscribe to her daily newsletter: tinyletter.com/cdewey, 
        @caitlindewey.
          Tracy Jan covers the intersection of race and the economy for 
        The Post. She previously was a national political reporter at 
        The Boston Globe, @TracyJan.
                                 ______
                                 
  Submitted Comment Letter by Hon. Jahana Hayes, a Representative in 
     Congress from Connecticut; Authored by Marc Egan, Director of 
          Government Relations, National Education Association
March 19, 2019

  Certification Policy Branch,
  SNAP Program Development Division,
  Food and Nutrition Service, USDA,
  Alexandria, Virginia

  RE: Proposed Rule: Supplemental Nutrition Assistance Program (SNAP): 
            Requirements for Able-Bodied Adults without Dependents RIN 
            0584-AE57

    Dear Certification Policy Branch:

    Thank you for the opportunity to comment in opposition to USDA's 
Proposed Rule on Requirements for Able-Bodied Adults without Dependents 
(ABAWDs).
    In theory, the 3 month time limit for Supplemental Nutrition 
Assistance Program (SNAP) benefits for ABAWDs impacts only adults who 
do not have children. In practice, it also harms children living in 
low-income, food-insecure households. Making it more difficult for 
states to waive the 3 month time limit for low-income individuals 
facing barriers to employment, as the proposed rule would do, makes it 
more likely that vulnerable children will go hungry or be poorly 
nourished.
First line of defense against childhood hunger
    SNAP, our nation's largest Federal food assistance program, is the 
first line of defense against childhood hunger. The program provides 
low-income households with monthly funds specifically designated for 
food purchases. Research links participation in SNAP for 6 months with 
an 8.5 percentage point decrease in food insecurity in households with 
children, according to USDA itself (Measuring the Effect of 
Supplemental Nutrition Assistance Program (SNAP) Participation on Food 
Security, (https://fns-prod.azureedge.net/sites/default/files/
Measuring2013.pdf) Aug. 2013).
    Food insecurity is a major threat to the health and well-being of 
the 12.5 million children in America--one in six--living in food-
insecure households. The consequences are devastating. Every day, 
educators like the three million members of the National Education 
Association (NEA) see firsthand how hungry children struggle to learn. 
Access to enough healthy food is essential to academic success.
    In 2015, 19.2 million children relied on SNAP for consistent access 
to food--44 percent of the program's participants. In addition to 
fighting food insecurity, SNAP significantly reduces child poverty and 
helps struggling families make ends meet: the program lifted 1.5 
million children out of poverty in 2017 alone.
Overly tight requirements are cruel and counterproductive
    Federal law limits SNAP eligibility for childless, unemployed or 
underemployed adults age 18-50 (except those who are exempt) to just 3 
months out of every 3 years unless they obtain and maintain an average 
of 20 hours a week of employment--and can prove it. These requirements 
are often already untenable for individuals who face structural 
barriers to employment and/or sufficient regular work hours. Data from 
2013 and 2014 show that the overwhelming majority of SNAP participants 
struggling to work 20 hours a work are not uninterested in working--
they are experiencing the consequences of volatile low-wage labor 
markets, caregiving duties, or personal health issues.
    The proposed rule would limit states' flexibility and tighten 
requirements for waiving this 3 month time limit for ABAWDs, causing an 
estimated 750,000 individuals to lose access to SNAP--an approach that 
is counterproductive as well as cruel. Denying people critical food 
assistance harms their health and productivity, hindering their ability 
to find and keep employment and achieve economic self-sufficiency.
Proposed changes do not reflect today's realities
    Technically, children under age 18 and the adults who live with 
them are exempt from the 3 month time limit for SNAP. This approach 
does not fully reflect the complex arrangements necessary for low-
income families to put food on the table. Specifically:

   Children with non-custodial parents (NCPs). Some 4.5 million 
        poor and low-income custodial parents rely on child support 
        payments from NCPs and use SNAP to put food on the table for 
        their children. NCPs are often low-income themselves: 2.1 
        million were below the poverty line in 2015 and 1.5 million 
        accessed SNAP to supplement their resources. Since NCPs are not 
        exempt from the 3 month time limit for ABAWDs, the proposed 
        rule threatens them as well as their children. An NCP who loses 
        SNAP benefits may no longer be able to make child support 
        payments.

   Children whose extended family members provide financial 
        support. Some low-income children receive food, financial 
        assistance, or care from extended family members, family 
        friends, or a parent's significant other who is receiving SNAP 
        benefits--people who are often struggling financially 
        themselves. The most economically precarious households are the 
        most likely to rely on such networks. So-called ABAWDs who lose 
        their SNAP benefits may have to stop providing support for 
        children they previously helped.

   Children impacted by the opioid crisis: Today, more than 2.5 
        million children are being raised by their grandparents or 
        other relatives, in part because families are dealing with 
        parental alcohol and substance abuse issues, which are growing 
        rapidly due to the opioid epidemic. The adults who provide 
        informal kinship care for children impacted by substance abuse 
        issues may not do so on a consistent schedule, however. As a 
        result, they may face obstacles in securing an exemption from 
        ABAWD time-limits. If they lose access to SNAP in the face of 
        tightened waiver requirements, the children they care for could 
        experience increased poverty and food insecurity as a result.

   Youth aging out of foster care and unaccompanied homeless 
        youth: SNAP plays a significant role in the health and well-
        being of youth in foster care and unaccompanied homeless youth 
        who often lack support systems. They disproportionately 
        experience significant barriers to obtaining a high school 
        diploma, entering college, obtaining a driver's license, 
        accessing health insurance, maintaining housing stability, 
        obtaining steady employment, and accessing sufficient food. 
        SNAP can help address their food insecurity, but because former 
        foster youth and unaccompanied homeless youth often meet the 
        definition of an Able-Bodied Adult Without Dependents, they 
        face obstacles accessing this critical assistance and would 
        likely disproportionately suffer under tightened state waiver 
        requirements. This is of particular concern after recent 
        changes made by the Agriculture Improvement Act of 2018 (P.L. 
        115-334) that reduced states' automatic exemption threshold 
        from 15 percent to 12 percent.
Conclusion
    SNAP time limits for ABAWDs adversely affect children and 
vulnerable youth, even though they are not the policy's intended 
targets. The proposed rule would exacerbate this problem. Furthermore, 
it flies in the face of Congressional intent. Congress just concluded a 
review and reauthorization of SNAP in the Agriculture Improvement Act 
of 2018, and explicitly rejected the proposed changes. This proposed 
rule is executive overreach that clearly disregards Congressional 
intent. The National Education Association represents educators who 
will see in their classrooms every day how vulnerable children, as a 
result of this rule, will experience a reduction in important resources 
that help meet their basic needs. NEA strongly opposes the proposed 
rule because it would limit SNAP benefits for more low-income adults, 
as well as children who may rely on them to help meet basic needs.
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Marc Egan,
Director of Government Relations.
                                 ______
                                 
  Submitted Comment Letter by Hon. Jahana Hayes, a Representative in 
    Congress from Connecticut; Authored by Lisa Davis, Senior Vice 
         President, No Kid Hungry Campaign, Share Our Strength
March 29, 2019

  Certification Policy Branch,
  SNAP Program Development Division,
  Food and Nutrition Service, USDA,
  Alexandria, Virginia

  Re: Proposed Rulemaking: Supplemental Nutrition Assistance Program 
            (SNAP): Requirements for Able-Bodied Adults Without 
            Dependents; RIN 0584-AE57, Docket ID: FNS-2018-0004

    Dear Certification Policy Branch:

    Thank you for the opportunity to comment on USDA's Proposed 
Rulemaking on Supplemental Nutrition Assistance Program (SNAP) 
Requirements for Able-Bodied Adults Without Dependents (ABAWDs).
    Share Our Strength is a national anti-hunger and anti-poverty 
organization. Through our No Kid Hungry campaign, we work to end 
childhood hunger in the United States by ensuring children have access 
to healthy food, every day all year round.
    While we support the stated goal of fostering self sufficiency, we 
are deeply concerned that the proposed changes to further restrict 
ABAWD's ability to receive SNAP benefits would cause significant 
hardship to very low-income individuals, restrict state flexibility and 
do nothing to help those struggling to find employment and secure jobs. 
To the contrary, the loss of food assistance will likely create 
additional financial and emotional stress making it harder to achieve 
this goal. The proposed rule also circumvents the will of Congress by 
attempting to implement, through executive action, policy changes 
Congress rejected in the bipartisan Agriculture Improvement Act of 2018 
(the farm bill) which was recently enacted by an overwhelming majority.
    Current law limits individuals between the ages of 18 through 49, 
who have not received a disability certification or are raising minor 
children, to just 3 months of SNAP benefits out of every 3 years unless 
they can document they are working or participating in a job training 
program at least 20 hours per week. However, states aren't required to 
offer work or training options to those impacted and most states do 
not. When several states began re-instating time limits that had been 
waived during the recession, at least 500,000 ABAWDs lost SNAP.\1\ And, 
mostly recently, reinstatement of the time-limit for ABAWDs in Kentucky 
led to an estimated 13,000 individuals to lose their SNAP benefits, not 
because they found employment, but because they reached their benefit 
time-limit.\2\ This represented a 20 to 22 percent decline in ABAWDs 
caseload in the state between January 2017 and September 2018.
---------------------------------------------------------------------------
    \1\ Bolen, Ed, et al., 2016. More than 500,000 Adults Will Lose 
SNAP Benefits in 2016 as Waiver Expire (https://www.cbpp.org/research/
food-assistance/more-than-500000-adults-will-lose-snap-benefits-in-
2016-as-waivers-expire). Center on Budget and Policy Priorities.
    \2\ Waxman, Elaine and Nathan Joo. 2019. Reinstating SNAP Work-
Related Time Limits: A Case Study of Able-Bodied Adults Without 
Dependents in Kentucky (https://www.urban.org/sites/default/files/
publication/100027/reinstating_snap_time_limits_1.pdf). Urban 
Institute.
---------------------------------------------------------------------------
    Recognizing that communities across the United States often face 
specific local challenges around employment and that state leaders are 
better equipped than their Federal counterparts to evaluate local 
economic conditions, states have long had the ability to seek waivers 
from the strict 3 month limit in areas where jobs are lacking and to 
waive the requirements for portions of their caseload who face 
particular challenges meeting the work requirement. This flexibility 
allows states to be responsive to local labor market variables and to 
protect individuals who live in areas of high unemployment, areas where 
economic conditions are lagging, and/or areas impacted by catastrophic 
events such as a natural disaster. The proposed rule would undermine 
states' flexibility, implementing a one-size-fits-all approach that 
eliminates some waiver grounds and restricts others.
    We agree that the best pathway from poverty to self-sufficiency is 
through adequate and stable employment. However, even though national 
unemployment has dropped to about four percent, millions of people in 
communities across the country continue to struggle to make ends meet 
due to difficulty finding a job, low wages and inadequate hours, 
limited skills, poor health or inadequate transportation. This rule 
would do nothing to help those impacted obtain employment. To the 
contrary, it would increase hunger and economic hardship by eliminating 
SNAP benefits for more than 750,000 \3\ unemployed and underemployed 
Americans according to USDA's own calculations. Other studies estimate 
the impact to be higher--with 1.2 million individuals loosing food 
access.\4\
---------------------------------------------------------------------------
    \3\ United States Department of Agriculture. 2018. Proposed 
Rulemaking: Supplemental Nutrition Assistance Program: Requirements for 
Able-Bodied Adults Without Dependents (https://s3.amazonaws.com/public-
inspection.federalregister.gov/2018-28059.pdf). FNS-2018-0004, RIN 
0584-AE57. PP40.
    \4\ Cunnyangham, Karen. 2019. Proposed Changes to the Supplemental 
Nutrition Assistance Program: Waivers to Work-Related Time Limits 
(https://www.mathematica-mpr.com/our-publications-and-findings/
publications/proposed-changes-to-the-supplemental-nutrition-assistance-
program-waivers-to-work-related-time). Mathematica Policy Research; 
Federal Poverty Level for a single individual is $12,490 for 2019.
---------------------------------------------------------------------------
    Those hit hardest would be those facing the greatest challenges in 
the labor market, including people of color, young adults aging out of 
foster care, veterans, homeless individuals, and those with limited 
education or skills or under-diagnosed physical or mental health 
issues. Research shows that only \1/2\ of ABAWDs nationally have a high 
school diploma or the equivalent,\5\ making it difficult to find and 
maintain stable employment in today's knowledge-based economy. Children 
aging out of foster care are particularly vulnerable. By age 24, only 
\1/2\ of these youths will obtain employment and only three to four 
percent will have earned a college degree by age 26, making them 
especially vulnerable to hunger and poverty.\6\
---------------------------------------------------------------------------
    \5\ Bolen, Ed. 2015. Approximately 1 Million People Would Lose Food 
Assistance Benefits in 2016 As State Waivers Expire: Affected 
Individuals Are Very Poor: Few Qualify for Other Help. Center On Budget 
and Policy Priorities.
    \6\ Shared Justice. 2017. Aging Out of Foster Care: 18 and On Your 
Own (http://www.sharedjustice.org/most-recent/2017/3/30/aging-out-of-
foster-care-18-and-on-your-own).
---------------------------------------------------------------------------
    Those impacted by SNAP time limits are often living in extreme 
poverty. According to latest research, 88 percent of ABAWDs that would 
be impacted by the proposed rule are making less than $6,245 per year 
per individual.\7\ They constitute a relatively small portion of all 
SNAP recipients--representing 12 percent or seven million individuals 
nationwide--and their numbers do not appear to be increasing despite 
claims to the contrary.\8\
---------------------------------------------------------------------------
    \7\ Supra note at 9.
    \8\ Center on Poverty and Social Policy. 2018. Understanding Recent 
Trends In Food Stamp Usage and implications for Increased Work 
Requirements (https://static1.squarespace.com/static/
5743308460b5e922a25a6dc7/t/5b69b61970a6adeee8860dc8/1533654555824/
Poverty+and+So
cial+Policy+Brief_2_5.pdf). Columbia University
---------------------------------------------------------------------------
    Further restricting benefits for ABAWDs is poor public policy and 
counterproductive, particularly in light of the growing body of 
research demonstrating SNAP's effectiveness and short and longer-term 
impact on health and economic security. In 2015 alone, SNAP lifted 8.4 
million people out of poverty.\9\ SNAP does this by freeing up 
resources that participants can spend on other critical needs such as 
housing, childcare, health care costs and transportation. In addition, 
studies found that SNAP participation was tied to an annual reduction 
of $1,400 in health care costs among low-income adults.\10\
---------------------------------------------------------------------------
    \9\ Wheaton, Laura and Victoria Tran. 2018. The Anti-Poverty 
Effects of the Supplemental Nutrition Assistance Program (https://
www.urban.org/sites/default/files/publication/96521/
the_antipoverty_effects_of_the_supplemental_nutrition_assistance_program
_3.pdf). Urban Institute.
    \10\ Berkowitz, Seth, et al., 2017. Supplemental Nutrition 
Assistance Program (SNAP) Participation and Health Care Expenditure 
Among Low-Income Adults (https://jamanetwork.com/journals/
jamainternalmedicine/article-abstract/2653910?redirect=true). JAMA 
Internal Medicine.
---------------------------------------------------------------------------
    SNAP already functions as an effective work support program. Most 
SNAP participants who can work are working or have worked in the past 
year, often for limited hours or in seasonal employment. This is 
particularly true for those considered ABAWDs: 25 percent are working 
while receiving benefits and 75 percent worked the year before or after 
receiving benefits. The experience of Franklin County, Ohio, 
demonstrates the challenges ABAWDs face in meeting the 20 hours per 
week work requirement due to unpredictable work schedules and lack of 
stable jobs.\11\
---------------------------------------------------------------------------
    \11\ Ohio Association of Food Banks. 2015. Franklin County: Work 
Experience Program, Able-Boded Adults Without Dependents (http://
admin.ohiofoodbanks.org/uploads/news/ABAWD_Report_2014-2015-v3.pdf).
---------------------------------------------------------------------------
    There is no evidence to suggest that restricting state waiver 
authority and thus eliminating benefits for hundreds of thousands of 
current SNAP beneficiaries would serve to increase employment and 
earnings among ABAWDs. Instead, it would punish those who were unable 
to find stable employment of at least 20 hours per week by denying them 
food benefits at a time when they most need it. The effect would be 
increased hunger and hardship. In fact, SNAP is one of the only 
supports available to individuals who fall under the ABAWD definition, 
as childless adults are not eligible for most other safety-net 
programs.
    Rather than reducing state flexibility and further restricting SNAP 
benefits for ABAWDs, policy change should be focused on addressing the 
underlying barriers to employment among those impacted such as limited 
education and skills, physical and mental health issues, unstable 
housing and lack of access to transportation. Investments in effective 
employment and training programs that are based on an individualized 
assessment of the beneficiary and tailored to their skills and 
challenges would be a much more effective way to help SNAP ABAWDs move 
from poverty to self-sufficiency. Research shows that SNAP Employment 
and Training (E&T) programs remain limited in their capacity to meet 
current needs, serving only a small percentage of those who are subject 
to work requirements,\12\ reinforcing the challenges facing ABAWDs who 
would be impacted by the proposed rule.
---------------------------------------------------------------------------
    \12\ Waxman, Elaine, et al., 2019. Poverty, Vulnerability, and the 
Safety Net (https://www.urban.org/urban-wire/social-safety-net-2019-
four-trends-watch-snap). Urban Institute.
---------------------------------------------------------------------------
    While the SNAP Employment and Training pilots authorized and funded 
through the 2014 Farm Bill will offer important learnings and best 
practices, work requirements should not be expanded unless adequate and 
effective job training programs and supports are in place to ensure 
meaningful pathways to self-sufficiency.
    We encourage strong coordination between SNAP Employment and 
Training with other federally funded job training and placement 
programs, as well as adequate funding for programs and services that 
support work, such as child-care, transportation, mental health 
counseling and casework management.
    Work requirements or benefit time limits that are not accompanied 
by the resources to ensure those impacted can find and sustain 
employment run counter to the objective of achieving economic self-
sufficiency and serve only to restrict benefits, thus increasing hunger 
and poverty rather than increasing employment and wages.
    We urge you to maintain states' flexibility to both request time 
limit waivers when jobs and employment supports are not available and 
to waive the work requirements for portions of their caseload who face 
particular challenges in meeting the work requirement. The rules 
governing areas eligibility for waivers were enacted with bipartisan 
support, have been in place for nearly 20 years and every state except 
Delaware has availed themselves of waivers at some point since the time 
limit became law. The waiver rules are reasonable, transparent, and 
manageable for states to operationalize. Thus, any change that would 
restrict, impede, or add uncertainty to states' current ability to 
waive areas with high unemployment should be avoided.
    Therefore, we respectfully request USDA to withdraw this harmful 
proposal. Congress has deliberated on these issues and rejected the 
restrictions included in the proposed rule in the 2018 Farm Bill, 
opting instead to including provisions to strength, encourage, and 
prioritize effective job training and employment-related activities.
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Lisa Davis,
Senior Vice President, No Kid Hungry Campaign,
Share Our Strength.
                                 ______
                                 
   Submitted Comment Letter by Hon. Kim Schrier, a Representative in 
Congress from Washington; Authored by Hon. Jay Inslee, Governor, State 
                             of Washington
March 29, 2019

  The Honorable Sonny Perdue,
  Secretary,
  U.S. Department of Agriculture
  Washington, D.C.

    Dear Secretary Perdue:

    On behalf of the State of Washington, I write to express my grave 
concerns with the Food and Nutrition Service's (FNS) proposed rule, 
``Supplemental Nutrition Assistance Program (SNAP): Requirements for 
Able-Bodied Adults Without Dependents (ABAWDs).'' This misguided and 
harmful policy would severely restrict access to food assistance for 
those who need it most, exacerbating hunger and making it even more 
difficult for people in poverty to find work. It removes state 
flexibility, rips away food assistance from 755,000 vulnerable 
Americans, worsens our homelessness crisis, and fails to achieve the 
Administration's stated goal of improving self-sufficiency. I strongly 
urge that it be withdrawn.
    Evidence shows that SNAP is one of the most important lifelines for 
families and communities facing economic hardship, lifting millions of 
Americans out of poverty and food insecurity every year. More than 42 
million people across the country rely on SNAP for food assistance, 
including more than 920,000 in Washington alone.\1\ It is a 
particularly significant safety net for our most vulnerable, as 75 
percent of SNAP households include a child, an elderly person, or a 
person with disabilities.\2\ The program is also a key economic driver 
that supports food producers, farmers' markets, and retailers. Every 
dollar spent on nutrition assistance expands the economy by 
approximately $1.70, boosting local economies and supporting 260,000 
individual retailers nationwide.\3\
---------------------------------------------------------------------------
    \1\ Center for Budget and Policy Priorities (CBPP), March 2018.
    \2\ Washington State Department of Agriculture (WSDA), January 
2018.
    \3\ CBPP, April 2018.
---------------------------------------------------------------------------
    This Administration's proposal would radically alter the SNAP 
program for certain populations and take away needed flexibility from 
states, imposing a top-down, one-size-fits-all approach that prevents 
Washington from addressing the unique and individualized needs of our 
local communities. It would directly harm our people and our economy, 
threatening to rip away food assistance from more than 91,000 
individuals who currently receive an average monthly benefit of 
$210.40, while reducing annual total revenue for Washington by over 
$32.6 million. Nationally, the proposed changes would result in a loss 
of $85 billion in economic activity for grocery stores, farmers, and 
other local food retail suppliers. It is a cruel and mean-spirited 
policy that damages people and businesses alike.
    Congress rejected these exact changes on a bipartisan basis last 
year. In considering the 2018 Farm Bill (P.L. 115-[334]), which was 
approved by large majorities in both chambers and signed by the 
President on December 20, 2018, Congress debated and subsequently 
excluded these changes to the SNAP program that would strip state 
flexibility and impose harsh, inflexible requirements on beneficiaries. 
To any objective observer, it is clear that these changes were not 
intended to be made and that USDA's proposal runs counter to 
Congressional intent. I encourage USDA to heed the advice of Congress 
in withdrawing this deeply harmful policy.
    I appreciate the opportunity to share our state's concerns and hope 
you give them the attention and consideration they deserve. Below, 
please find additional feedback from our state on specific questions 
raised by USDA in the proposed rule.
Labor Market Areas for Grouping
    In USDA's proposal, the Department specifically requested comments 
on the use of Labor Market Areas (LMAs) for grouping areas. We believe 
LMAs defined by the Federal Government should be included as the basis 
for grouping areas, and that grouping should not be prohibited 
entirely. States are currently given discretion to define groups or 
areas to be combined, provided the areas are contiguous or considered 
part of the same economic region. Availability of jobs is examined when 
counties are in close proximity to counties where individuals often 
commute. Washington uses this discretion for LMA groupings because we 
understand our residents are disadvantaged when they are required to 
travel unreasonable distances for employment. People should be able to 
readily change jobs without being forced to change their place of 
residence, particularly as most ABAWDs have limited resources and 
cannot easily commute or change residences to obtain employment.
    If LMAs are not a basis for grouping, participants may not be able 
to reside and find employment within a reasonable distance or change 
jobs without also having to change their residence. Denying states the 
ability to group counties would negatively impact an estimated 91,203 
individuals in Washington identified as ABAWDs. The loss of waivers for 
these counties would also cause a negative impact on our local 
economies.
Setting a Floor for the 20 Percent Standard
    Washington does not support USDA's proposal to establish a floor 
for the 20 percent standard, which would further limit state 
flexibility and restrict necessary waivers to appropriately serve SNAP 
beneficiaries. We do not believe that a floor of six percent, seven 
percent or ten percent is needed or advisable. (See Table 2 for 
additional data on how these changes would adversely affect our state.) 
We believe the current floor setting that has been established at 20 
percent above the average national unemployment rate is appropriate and 
necessary.
    The current standard is essential to allow flexibility in 
requesting necessary waivers. This flexibility is granted with the 
knowledge that state and local leaders are best equipped to develop 
solutions for their specific labor markets and industries. While the 
unemployment rate does provide essential data, it does not take into 
account a community's individualized workforce needs or that its 
residents may not be well-suited to find and keep locally available 
jobs due to lack of housing, skills, training, or other barriers. To 
illustrate this, Table 1 highlights the top ten occupations, hard 
skills, certifications, and employers in Washington according to our 
Employment Security Department (ESD):

              Table 1: Employmer Demand in Washington State
------------------------------------------------------------------------
    Occupations      Hard Skills      Certifications        Employers
------------------------------------------------------------------------
Software            Microsoft      Driver's License      Amazon
 Developers          Office
Registered Nurses   Quality        Commercial Driver's   Providence
                     Assurance      License               Health &
                                                          Services
Retail              Microsoft      Class A Commercial    State of
 Salespersons        PowerPoint     Driver's License      Washington
Computer            Freight+                             Peace Health
 Occupations
First-Line          Software       Basic Life Support    University of
 Supervisors of      Development   Certified Registered   Washington
 Retail Sales       Java            Nurse                Microsoft
 Workers
Marketing Managers  Structured     Certification in      Catholic Health
Stock Clerks and     Query          Cardiopulmo-nary      Initiatives
 Order Filers        Language       Resuscitation        MultiCare
                    Python                                Health System
Customer Service    Bilingual      Security Clearance    Schweitzer
 Representatives    Forklifts      Continuing Education   Engineering
                                                          Laboratories
Heavy and Tractor-                 First Aid             Kaiser
 Trailer Drivers                    Certification         Permanente
                                   HAZMAT
First-Line
 Supervisors of
 Food Preparation
 and Serving
 Workers
------------------------------------------------------------------------

    Job readiness in these fields can be an insurmountable goal for 
individuals who must navigate numerous and repetitive barriers on a 
daily basis. From homelessness and housing instability to domestic 
violence, mental health, and substance use disorder, there are myriad 
and significant barriers facing ABAWDs that prevent them from 
effectively seeking and obtaining employment. In many cases, these 
barriers must be addressed first for an individual to be ready for job 
training and the workforce. A person experiencing homelessness must 
primarily focus on where they are going to sleep and eat, for example, 
not where are they going to find work.
    In Washington, we estimate that more than 43 percent of our state's 
ABAWD population is currently experiencing homelessness--
disproportionately higher than the broader SNAP population, of which 
only 11 percent are experiencing homelessness. Nearly 60 percent of the 
ABAWD population is suffering from behavioral or physical health 
conditions, including substance use disorder.\4\ For these individuals, 
USDA's proposal would do nothing to help them find work, while adding 
yet another obstacle in their way--food insecurity. It would not 
achieve USDA's stated goal of promoting self-sufficiency and in fact 
would make it more difficult for ABAWDs to find employment.
---------------------------------------------------------------------------
    \4\ Department of Social and Health Services (DSHS) Economic 
Services Administration (ESA), January 2019.
---------------------------------------------------------------------------
    Large percentages of SNAP recipients also experience labor market 
fluctuations due to seasonal employment, part-time work, or 
underemployment, and would be directly harmed by USDA's proposal 
despite their participation in the workforce. The vast majority of 
those who transition between working more than 20 hours a week and a 
different employment status--less than 20 hours a week, seeking 
employment, or not in the labor force--are working on a monthly basis 
but still may not meet USDA's one-size-fits-all work requirement. Under 
the proposed rule, a large number of individuals would lose food 
assistance as a result of volatility in the labor market and through no 
fault of their own.
    We support current Federal regulations that allow states to waive 
the 3 month time limit in geographic areas with high unemployment or 
insufficient jobs. Creating an unemployment rate floor would negatively 
impact a large number of counties across our state, including wide 
swaths of rural and economically disadvantaged communities. The loss of 
waivers would affect SNAP eligibility for tens of thousands of 
Washington citizens who may otherwise not qualify for food assistance.
    Table 2 below illustrates how the proposed changes would impact 
SNAP recipients in Washington under USDA's proposed changes. A review 
of data shows that there is no difference in the number of counties and 
SNAP recipients adversely affected at seven or ten percent.

        Table 2: Impact of Proposed Changes to Washington State 5
------------------------------------------------------------------------
  Proposed Change    Loss of Grouping       6% Floor     7% or 10% Floor
------------------------------------------------------------------------
SNAP Recipients     15,321              75,407           91,203
 Adversely
 Affected
\5\ DSHS ESA,
 January 2019.
Counties Adversely  Asotin, King, San   Adams, Asotin,   All counties
 Affected            Juan, Snohomish,    Benton,          except Ferry
                     Walla Walla,        Chelan, Clark,   (38 counties)
                     Whitman (6          Columbia,
                     counties)           Cowlitz,
                                         Douglas,
                                         Franklin,
                                         Island,
                                         Jefferson,
                                         King, Kitsap,
                                         Kittitas,
                                         Klickitat,
                                         Lincoln,
                                         Pierce, San
                                         Juan, Skagit,
                                         Skamania,
                                         Snohomish,
                                         Spokane,
                                         Thurston,
                                         Walla Walla,
                                         Whatcom, and
                                         Whitman (26
                                         counties)
------------------------------------------------------------------------

Eliminating the Carryover Exemption Provision
    Washington strongly disagrees with USDA's interpretation of the 
ABAWD exemption provision of the 2018 Farm Bill, which decreases ABAWD 
exemptions granted to states from 15 percent to 12 percent It is our 
interpretation that the law did not intend for USDA to limit the 
carryover of exemptions for ``covered individuals,'' and only lowered 
the percentage of exemptions granted to each state. We believe this 
proposal is contrary to Congressional intent and should be withdrawn.
    The 2018 Farm Bill and current regulations give states flexibility 
over whether and when to use and carryover these exemptions. Washington 
depends on this flexibility to effectively operate our program. In 
2015, Washington was one of ten states awarded a SNAP Employment and 
Training (E&T) pilot, which tests innovative approaches to employment 
for work registrants. Participants were randomly assigned to a control 
and treatment group. Washington was able to use our 15 percent 
exemptions to ensure participants assigned to the control group 
remained engaged and eligible for food assistance to ensure accuracy of 
our pilot. The elimination of carryover exemptions would significantly 
impact our state's ability to carry out the E&T pilot and effectively 
operate our SNAP program.
Conclusion
    Washington strongly opposes USDA's proposal threatening food 
assistance for more than 91,000 individuals in our state and 755,000 
Americans nationwide. We understand that obtaining employment can be 
difficult for many ABAWDs working to reach their full potential, many 
of whom face significant barriers--including homelessness and substance 
use disorder--with little or no resources. We also understand that 
state flexibility is necessary to meet the unique needs of the ABAWD 
population and our local economies. The current rules, which have been 
in place for 20 years, are reasonable, transparent, manageable, and 
effective. We see no rational justification for this Administration's 
sweeping changes that would undermine our state's success in reducing 
hunger and moving people to employment. I urge that it be withdrawn.\6\
---------------------------------------------------------------------------
    \6\ Centers for Disease Control and Prevention (CDC), July 2017.
---------------------------------------------------------------------------
    We appreciate your consideration of our state's perspective. If you 
have any questions, please contact the Director of my Washington, D.C. 
Office, Casey Katims, at [email protected]. Thank you.
            Very truly yours,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Hon. Jay Inslee,
Governor.

CC:

Washington Congressional Delegation;
Cheryl Strange, Secretary, Department of Social and Health Services 
(DSHS);
David Stillman, Assistant Secretary, DSHS Economic Services 
Administration;
Babette Roberts, Director, DSHS Community Services Division.
                                 ______
                                 
   Submitted Comment Letter by Hon. Kim Schrier, a Representative in 
Congress from Washington; Authored by Stacy Dean, Vice President, Food 
       Assistance Policy, Center on Budget and Policy Priorities
April 1, 2019

  Ms. Sasha Gersten-Paal,
  Chief,
  Certification Policy Branch,
  Program Development Division,
  Food and Nutrition Service,
  Alexandria, VA

  Re: Proposed Rule: Supplemental Nutrition Assistance Program: 
            Requirements and Services for Able-Bodied Adults Without 
            Dependents RIN 0584-AE57

    Dear Ms. Gersten-Paal:

    We are writing to provide comments on USDA's Notice of Proposed 
Rule Making (NPRM) regarding the Supplemental Nutrition Assistance 
Program's (SNAP) Requirements and Services for Able-Bodied Adults 
Without Dependents. The proposed rule would restrict longstanding state 
flexibility to waive areas from SNAP's 3 month time limit as well as 
limit states' ability to exempt certain individuals from the time 
limit. As a result, USDA estimates that when fully implemented in a 
typical month some 755,000 individuals would lose food assistance 
benefits because they could not document an average of 80 hours per 
month of employment or that they qualify for an exemption. USDA does 
not provide any evidence to support its assertion that the policy would 
result in greater employment or earnings. This is likely because such 
evidence does not exist. Instead, there is an extensive body of 
research that suggests the very likely outcome of the proposed policy 
is that more individuals will experience hardship and poverty, 
including a risk of hunger. Moreover, given available research on work 
requirements and the labor market, the proposed policy is very likely 
to have even worse outcomes for African Americans, Native Americans, 
Latinos, and individuals with disabilities.
    The Center on Budget and Policy Priorities is a nonpartisan 
research and policy institute. We pursue Federal and state policies 
designed both to reduce poverty and inequality and to restore fiscal 
responsibility in equitable and effective ways. We apply our deep 
expertise in programs and policies that help low-income people in order 
to help inform debates and achieve better policy outcomes. We work to 
protect and strengthen programs that reduce poverty and inequality and 
increase opportunity for people trying to gain a foothold on the 
economic ladder. Our work on Federal nutrition programs, including 
SNAP, is a core component of our organization's work. Our food 
assistance analyst team includes nine people, including eight analysts 
and researchers who work on SNAP policy and operations. We have deep 
expertise on SNAP time limit policy including waivers and individual 
exemptions. Three members of our team, as well as our organization's 
President, have worked on SNAP for more than 2 decades, including 
during the time period when the law governing the time limit was 
enacted and the current regulations were proposed and codified.
    We have deep concerns with the proposed policy and offer extensive 
comments to support our strong recommendation that USDA withdraw the 
NPRM and maintain current policy. In addition to causing harm to 
vulnerable individuals who are in between jobs or underemployed, the 
proposed policy runs counter to Congressional intent. When legislating 
the time limit policy, Congress established a waiver authority that 
allows for states to waive the rule for areas with insufficient jobs 
for individuals subject to the rule. Given that individuals who fall 
into the group subject to the time limit face extreme difficulty in the 
labor market, a fact validated by extensive research, the proposed rule 
would undercut Congressional intent by setting arbitrary limits 
unrelated to the purpose of the waiver.
    The proposed rule is also poorly argued, internally inconsistent, 
and wildly out of sync with extensive research findings. It offers 
little, and in some cases, no reasoning or evidence to support such a 
dramatic change in a longstanding Federal policy that would have 
significant consequences on participants, states and other key 
stakeholders such as retailers and small business. The Department also 
provided flawed and contradictory analysis in the NPRM and did not 
include information available to the agency that would have informed 
the rulemaking process. USDA's rationale for such a sweeping and 
harmful change was cursory at best making it almost impossible to 
comment in a way that is responsive to its thinking. Because USDA did 
not make its reasoning transparent or provide evidence to support its 
position, we feel obligated to review and provide years of well-known 
research and data (some of which USDA funded) that provides evidence 
counter to USDA's proposed policy. We strongly encourage USDA to review 
these materials as we are concerned the Department is unaware of the 
overwhelming evidence that undermines their assertions and poorly 
formed conclusions in the proposed rule. This has resulted in lengthy 
comments in which we conclude that the best course of action for the 
proposed policy and under the rulemaking process would be for USDA to 
withdraw the NPRM. We strongly urge that course of action.
    In this proposed rule, USDA proposed many damaging and ill-advised 
changes to waivers and individual exemptions from the 3 month time 
limit. The major changes include:

   Mandating that areas must have a minimum of a seven percent 
        average unemployment rate over a 2 year period in order to 
        qualify for a waiver from the time limit;

   Restricting states' flexibility to define the area they wish 
        to waive;

   Eliminating several waiver criteria that have been part of 
        program rules for over 20 years, including a low and declining 
        employment-to-population ratio;

   No longer allowing states to implement waivers that meet 
        USDA's criteria while not requiring that USDA approve waivers 
        in a timely manner;

   Requiring states to seek their governor's written consent; 
        and

   Restricting states' ability to accumulate unused individual 
        exemptions.

    Our comments on the proposed regulation fall into several major 
categories:
Proposed Changes to Waiver Criteria
   Chapter 1: Overview of Waivers from the Three-Month Time 
        Limit--Their Purpose and History

   Chapter 2: FNS Waiver Policy Has Been Consistent for the 
        Last 22 Years

   Chapter 3: Setting a Floor for Waivers for Areas With 20% 
        Above National Unemployment Is Inconsistent with Congressional 
        Intent and Would Be Harmful to Vulnerable Individuals

   Chapter 4: Dropping Several Key Criteria from the 
        Insufficient Jobs Criteria Is Inconsistent with the Statute

   Chapter 5: Restricting State Flexibility on Grouping Areas 
        Is Counter to Evidence

   Chapter 6: Taking Away Food Benefits from Individuals Who 
        Cannot Document 20 Hours a Week of Work Will Not Increase Labor 
        Force Participation for This Population

   Chapter 7: Proposed Rule's Requirement That State Waiver 
        Requests Have the Governor's ``Endorsement'' Violates 
        Congressional Intent

   Chapter 8: Proposed Rule Would Make Implementing Time Limit 
        Harder by Removing Provisions That Give States Certainty Around 
        Approval
Proposed Changes to Individual Exemptions
   Chapter 9: Eliminating the Carryover of Unused Individual 
        Exemptions Would Cause Hardship and Exceeds Agency Authority
Problems with the Proposed Rule Process
   Chapter 10: The Proposed Rule Fails to Provide Sufficient 
        Rationale or Supporting Evidence for the Proposed Policy

   Chapter 11: The Proposed Rule's ``Regulatory Impact 
        Analysis'' Highlights FNS' Faulty Justification and Includes 
        Numerous Unclear or Flawed Assumptions

   Chapter 12: The Proposed Rule Would Disproportionately 
        Impact Individuals Protected by Civil Rights Laws, Violating 
        the Food and Nutrition Act's Civil Rights Protections

   Chapter 13: The Proposed Rule Fails to Adequately Estimate 
        the Impact on Small Entities
Appendix that includes all cited studies and references
   Appendix A: CBPP Bios

   Appendix B: Materials Cited in Comments

    We strongly urge USDA to withdraw the rule and maintain current 
policy. If you have any questions regarding our comments, please do not 
hesitate to contact us.
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Stacy Dean,
Vice President, Food Assistance Policy.
   center on budget and policy priorities comments on rin 0584-ae57: 
 supplemental nutrition assistance program: requirements and services 
               for able-bodied adults without dependents
Table of Contents
Proposed Changes to Waiver Criteria
   Chapter 1: Overview of Waivers from the Three-Month Time 
        Limit--Their Purpose and History

   Chapter 2: FNS Waiver Policy Has Been Consistent for the 
        Last 22 Years

   Chapter 3: Setting a Floor for Waivers for Areas With 
        Percent Above National Unemployment Is Inconsistent with 
        Congressional Intent and Would Be Harmful to Vulnerable 
        Individuals

   Chapter 4: Dropping Several Key Criteria From Waiver 
        Criteria Is Inconsistent With the Statute

   Chapter 5: Restricting State Flexibility on Grouping Areas 
        Is Counter to Evidence

   Chapter 6: Taking Away Food Benefits from Individuals Who 
        Cannot Document 20 Hours a Week of Work Will Not Increase Labor 
        Force Participation for This Population

   Chapter 7: Proposed Rule's Requirement That State Waiver 
        Requests Have the Governor's ``Endorsement'' Violates 
        Congressional Intent

   Chapter 8: Proposed Rule Would Make Implementing The Time 
        Limit Harder by Removing Provisions That Give States Certainty 
        Around Approval
Proposed Changes to Individual Exemptions
   Chapter 9: Eliminating the Carryover of Unused Individual 
        Exemptions Would Cause Hardship and Exceeds Agency Authority
Problems with the Proposed Rule Process
   Chapter 10: The Proposed Rule Fails to Provide Sufficient 
        Rationale or Supporting Evidence for the Proposed Policy Change

   Chapter 11: The Proposed Rule's ``Regulatory Impact 
        Analysis'' Highlights FNS' Faulty Justification and Includes 
        Numerous Unclear or Flawed Assumptions

   Chapter 12: The Proposed Rule Would Disproportionately 
        Impact Individuals Protected by Civil Rights Laws, Violating 
        the Food and Nutrition Act's Civil Rights Protections

   Chapter 13: The Proposed Rule Fails to Adequately Estimate 
        the Impact on Small Entities
Appendix that includes all cited studies and references
   Appendix A: CBPP Bios

   Appendix B: Materials Cited in Comments

    Note, throughout these comments, we use the terms: Food and 
Nutrition Service (FNS), U.S. Department of Agriculture (USDA), and 
``the Department'' somewhat interchangeably. We are not aware of a 
particular convention and it is not our intent to suggest difference 
when we use one term vs. the other. In addition, when we refer to 
``state'' or ``states'' we intend to include counties in their role 
administering the program in county-administered states.
Chapter 1: Overview of Waivers from the Three-Month Time Limit--Their 
        Purpose and History
    The time limit is one of the harshest rules in the Supplemental 
Nutrition Assistance Program (SNAP, formerly known as the Food Stamp 
Program). Childless adults on SNAP are extremely poor. Like adults with 
children, childless adults often turn to SNAP for assistance when they 
are no longer able to make ends meet, especially as they lose jobs, 
their hours are cut, or their wages hover at the Federal minimum. While 
participating in SNAP, their income averages 29 percent of the poverty 
line, the equivalent of about $3,400 per year for a single person in 
2016.\1\ The U.S. Department of Agriculture (USDA), which administers 
SNAP, has established standards that have remained consistent over the 
last 20 years under which states can request a waiver of the time limit 
for areas with consistently high unemployment. States request waivers 
for multiple reasons, including to ease administrative burden, 
implement more effective work programs, and exempt vulnerable 
individuals who likely will struggle to find work. The proposed rule 
would severely weaken this flexibility, increasing administrative 
burden for states and hardship for SNAP participants who struggle to 
find work. This chapter describes the history of these waivers, 
Congressional intent and early implementation of waiver rules, and the 
reasons why states choose to waive areas in their state.
---------------------------------------------------------------------------
    \1\ Steven Carlson, et al., ``Who Are the Low-Income Childless 
Adults Facing the Loss of SNAP in 2016?'' Center on Budget and Policy 
Priorities, February 8, 2016, http://www.cbpp.org/research/food-
assistance/who-are-the-low-income-childless-adults-facing-the-loss-of-
snap-in-2016.
---------------------------------------------------------------------------
    One of SNAP's harshest rules limits unemployed individuals aged 18 
to 50 not living with children to 3 months of SNAP benefits in any 36 
month period when they aren't employed or in a work or training program 
for at least 20 hours a week.\2\ Under the rule, implemented as part of 
the 1996 welfare law, states are not obligated to offer affected 
individuals a work or training program slot, and most do not. SNAP 
recipients' benefits are generally cut off after 3 months irrespective 
of whether they are searching diligently for a job or willing to 
participate in a qualifying work or job training program. As a result, 
this rule is, in reality, a time limit on benefits and not a work 
requirement, as it is sometimes described.
---------------------------------------------------------------------------
    \2\ For a more comprehensive discussion of the time limit rule, 
see: Ed Bolen, et al., ``More Than 500,000 Adults Will Lose SNAP 
Benefits in 2016 as Waivers Expire,'' Center on Budget and Policy 
Priorities, updated March 18, 2016, http://www.cbpp.org/research/food-
assistance/more-than-500000-adults-will-lose-snap-benefits-in-2016-as-
waivers-expire.
---------------------------------------------------------------------------
    In addition to being harsh policy that punishes individuals who are 
willing to work but can't find a job, the rule is one of the most 
administratively complex and error-prone aspects of SNAP law. Many 
states also believe that the rule undermines their efforts to design 
meaningful work requirements, as the time limit imposes unrealistic 
dictates on the types of job training that can qualify. For these 
reasons, many states and organizations that represent SNAP participants 
have long sought the rule's repeal.
    The time limit law does provide states with the ability to seek 
waivers from USDA to temporarily suspend the 3 month limit for 
individuals in areas with insufficient jobs. These waivers are the 
primary subject of the proposed rulemaking along with states' authority 
to use flexible individual exemptions to exempt individuals of their 
choosing from the time limit. Since passage of the welfare law, many 
states have sought waivers for counties, cities, or reservations with 
relatively high and sustained unemployment. Every state except Delaware 
has sought a waiver at some point since the time limit's enactment.\3\
---------------------------------------------------------------------------
    \3\ ``FNS Controls Over SNAP Benefits for Able-Bodied Adults 
Without Dependents,'' USDA Office of Inspector General, Audit Report 
27601-0002-31, September 2016, https://www.usda.gov/oig/webdocs/27601-
0002-31.pdf.
---------------------------------------------------------------------------
    States can choose (or choose not) to request a waiver. In some 
cases, states with areas that have a persistently struggling labor 
market, such as the Central Valley in California or rural West 
Virginia, have sought waivers to avoid penalizing those who cannot find 
a 20 hour per week job within 3 months. In other cases, governors have 
sought waivers because extraordinary events have hurt their local labor 
markets, such as the 2010 Gulf of Mexico oil spill, Hurricane Katrina, 
or layoffs from a major local employer.
    Many states also seek waivers from the time limit because they 
would prefer to devote the resources needed to implement the 
administratively complex time limit to implementing a more rational and 
appropriate work requirement tailored to their local economy and to 
available job training programs.
Figure 1
Estimated Impact of USDA Proposed Rule
Share of U.S. Waived from SNAP's 3 Month Time Limit

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Note: Represents share of U.S. population living in a waived 
        area, i.e., county or city.
          * Estimated [share] of U.S. population living in a waived 
        area under USDA proposed rule, if rule were in effect in 2018.
          Source: CBPP analysis of state waivers; U.S. Census Bureau 
        population estimates.

    USDA's guidelines regarding waiver criteria, articulated in 
guidance and regulations, have set clear, consistent standards for 
waivers since soon after the statute adopted the time limit and waiver 
provisions in 1996. A review of waivers over the last 20 years shows 
that just over \1/3\ of the country (as measured by the share of the 
total population living in waived counties) is waived in a typical 
year.\4\ (See Figure 1.)
---------------------------------------------------------------------------
    \4\ During the recession and its aftermath, Congress made a large 
portion of the country temporarily eligible for a waiver in recognition 
of widespread elevated unemployment. Some have misinterpreted this 
temporary expansion of waivers as a permanent expansion of the policy 
or an Obama Administration-led effort to eliminate the time limit.
---------------------------------------------------------------------------
    In the NPRM, USDA states that the current rate of waivers was 
unforeseen, which is inconsistent with the historical record that 
demonstrates that USDA's original estimate of the extent of waiver 
coverage under its rules was in line with current actual coverage. In 
the NPRM preamble, the Department states: ``The proposed rule addresses 
these areas of concern and places safeguards to avoid approving waivers 
that were not foreseen by Congress and the Department, and to restrict 
states from receiving waivers in areas that do not clearly demonstrate 
a lack of sufficient jobs.'' \5\ This statement stands in contrast to 
USDA's own documents. USDA was fully cognizant that its original 
proposed waiver policy, which it later codified into final regulations, 
could result in more than \1/3\ of the country being waived. In an 
internal summary of waivers from April 23, 1997 entitled, ``Time Limit 
Waivers for Able-bodied Food Stamp Participants,'' FNS staff wrote to 
Office of Management and Budget staff that ``Thirty percent to 45 
percent of the able-bodied caseload may be waived. However, USDA's best 
estimate is that the areas that have been waived represent 
approximately 35 percent of the able-bodied caseload in the nation as a 
whole.'' \6\ This was written at a time of relatively low unemployment 
and early in the implementation of waivers when take up of waivers was 
relatively low. This would suggest that current policy, which has 
resulted in 36 percent of the general population living in waived areas 
except during the Great Recession and its aftermath, is consistent with 
what USDA originally intended rather than something that has exceeded 
its vision. Moreover, the memo does not suggest any concern with the 
share of the country waived. And, these criteria were nearly exact to 
those codified in final rules.
---------------------------------------------------------------------------
    \5\ Supplemental Nutrition Assistance Program: Requirements and 
Services for Able-Bodied Adults without Dependents, 84 Fed. Reg.  980 
(proposed rule February 1, 2019) found at https://
www.federalregister.gov/documents/2019/02/01/2018-28059/supplemental-
nutrition-assistance-program-requirements-for-able-bodied-adults-
without-dependents#p-45, hereafter we will refer to this as the 
``NPRM.''
    \6\ FNS White Paper, ``Time Limit Waivers for Able-Bodied Food 
Stamp Recipients,'' April 23, 1997. Faxed from FNS to OMB analyst 
Lester Cash on April 25, 1997.
---------------------------------------------------------------------------
    Under USDA's proposed rule, however, areas eligible for waivers 
would be dramatically reduced. Our organization applied the proposed 
rule to the areas waived in 2018 and determined that:

   Of the 985 counties (or county equivalents) waived in 2018, 
        639 counties (65 percent of all waived counties) in 28 states 
        would have lost their waivers.

   Of the 309 towns located outside of waived counties in 2018, 
        285 towns (92 percent of all waived towns) would have lost 
        their waivers, including 259 New England towns.

   170 out of the 273 reservations (62 percent of all waived 
        reservations) waived in 2018 would have lost their waivers.\7\
---------------------------------------------------------------------------
    \7\ Based on CBPP internal analysis of unemployment data from the 
U.S. Bureau of Labor Statistics and the U.S. Census Bureau. The list of 
areas is included in Appendix B as ``CBPP Summary of Areas That Would 
Have Lost Their Waivers form the SNAP Three-Month Time Limit in 2018 if 
the Proposed Rule Were Implemented in 2018.''

    Under the proposed policy, we estimate that the share of the U.S. 
population living in waived areas would have declined by over 80 
percent in 2018, from 36 percent to 6.1 percent of the U.S. population. 
The proposed rule would therefore result in a dramatic reduction in 
states' ability to waive areas from the time limit. Unfortunately, that 
appears to be USDA's goal rather than designing and implementing a 
policy consistent with the statute, i.e., setting waiver criteria and 
policy that would allow states to waive areas with insufficient job for 
individuals subject to the time limit.
A. Current Rules Governing Waivers for Areas With Insufficient Jobs for 
        Individuals Subject to the Time Limit
    The SNAP time limit provision is based in substantial part on an 
amendment successfully offered on the House floor on July 18, 1996, by 
Reps. Robert Ney and John Kasich. When considering the appropriateness 
of some of the proposals in the proposed rule, it is illuminating to 
example the floor debate to see what Congress did--and did not--think 
it was requiring.
    The floor debate indicates that the amendment's cosponsors believed 
that then food stamp workfare (participation in which would have 
exempted an individual from benefit termination) to be widespread and 
assumed that large numbers of those who cannot find a private-sector 
job would be offered a workfare slot. For example, Rep. Kasich stated 
on the floor: `` . . . let me be clear what the amendment does so that 
there is no confusion. If you are [able]-bodied, single, between the 
ages of 18 and 5-, and you get food stamps, we are saying you have to 
work . . . If you cannot get a job, you go to a workfare program; 45 
out of 50 states have a workfare program.'' \8\
---------------------------------------------------------------------------
    \8\ 142 Cong. Rec. H7905 (daily ed. July 18, 1996). In fact, only 
about ten states had food stamp workfare programs at that time, and 
most such programs were very small. Many of them operated in only a few 
counties in these states, an some were only open to families with 
children. Even today, SNAP workfare is unavailable to a great many 
people subject to the time limits.
---------------------------------------------------------------------------
    The sponsors heatedly disputed the statements by opponents of the 
amendment that the amendment would cause substantial hardship by 
denying assistance to people who want to work but cannot find a job or 
a workfare slot. And, they emphasized that the amendment contains 
waivers and other means to avert such situations. For example:

   Rep. Ney stated: `` . . . if we read the text, there are 
        hardship exemptions. It can be waived. There are safeguards in 
        this.'' \9\ Mr. Ney also noted: `` . . . it is an amendment 
        that provides some safety, it provides a course of a safety net 
        [sic], it has the ability to have waivers from the state 
        department of human services.'' \10\
---------------------------------------------------------------------------
    \9\ 142 Cong. Rec. H7905 (daily ed. July 18, 1996).
    \10\ 142 Cong. Rec. H7905 (daily ed. July 18, 1996).

   Rep. Kasich also addressed this issue. ``It is only if you 
        are able-bodied, if you are childless, and you live in an area 
        where you are getting food stamps and there are jobs available, 
        then it applies. So, if you are able-bodied, you go and you 
        have to work 20 hours to get your food stamps. The of course if 
        you cannot find a job then you do workfare. That is what it is. 
        But there are a number of exemptions in here for people who 
        find themselves in particularly difficult circumstances . . .'' 
        \11\
---------------------------------------------------------------------------
    \11\ 142 Cong. Rec. H7905 (daily ed. July 18, 1996) (emphasis 
added).

    As their statements indicate, the amendment's sponsors visualized 
the amendment largely as one under which people were prodded to look 
for work, were generally provided a workfare slot if a private sector 
jobs was not available and would be protected by a waiver if there were 
insufficient jobs and workfare slots for them. The sponsors did not see 
their amendment as one under which large numbers of individuals who 
want to work but cannot fund a job end up with neither work nor food 
stamps. It should be noted that the sponsors were not cognizant of the 
extremely limited number of food stamp workfare slots throughout the 
country.
    In the final legislation Congress established that states could 
waive areas lacking jobs. USDA has established criteria to implement 
that authority that have been consistent for 2 decades. The rule was 
designed to permit states to seek waivers in areas where jobs aren't 
available. To qualify for a waiver, states must provide detailed 
evidence of high unemployment in local areas, in accordance with 
rigorous requirements set by USDA. USDA has consistently used the same 
criteria to define high unemployment since the late 1990s.
    The Federal law gives states the option to request a waiver of the 
time limit if they can document that a given geographic area has an 
insufficient number of jobs (or has an unemployment rate over ten 
percent). The standards that define how a state may document 
``insufficient jobs'' were first outlined in FNS Guidance issued in 
December 1996.\12\ In the guidance, USDA offered several reflections on 
its understanding of Congressional intent at the time. First, USDA 
shared its belief that Congress understood that this group of 
individuals could find it especially challenging to find permanent 
employment and that waivers are intended recognize this problem. ``USDA 
believes that the law provided authority to waive these provisions in 
recognition of the challenges that low-skilled workers may face in 
finding and keeping permanent employment. In some areas, including 
parts of rural America, the number of employed persons and the number 
of job seekers may be far larger than the number of vacant jobs. This 
may be especially so for person with limited skills and minimal work 
history.''
---------------------------------------------------------------------------
    \12\ U.S. Department of Agriculture, Food and Nutrition Service 
(FNS) ``Guidance for States Seeking Waivers for Food Stamp Limits,'' 
FNS guidance to states, December 3, 1996.
---------------------------------------------------------------------------
    In addition, the guidance provided key background on some of the 
policy that USDA seeks to restrict in the NPRM. With respect to how 
states can set or define the area within the state that it seeks to 
waive, USDA said, ``USDA will give states broad discretion in defining 
areas that best reflect the labor market prospects of program 
participants and administrative needs.'' \13\ The guidance also 
recognized that the statute seeks to identify whether or not there are 
sufficient jobs for individuals subject to the time limit. ``The 
guidance that follows offers some examples of the types and sources of 
data available to states as the consider waiver requests for areas with 
insufficient jobs. Because there are not standard data or methods to 
make the determination of the sufficiency of jobs, the list that 
follows is not exhaustive. States may use these data sources as 
appropriate, or other data as available, to provide evidence that the 
necessary conditions exist in the area for which they intend the waiver 
to apply. The absence of a particular data source or approach (for 
example, data or statistics compiled by a university is not meant to 
imply that it would not be considered by USDA if requested by a 
state.'' \14\
---------------------------------------------------------------------------
    \13\ U.S. Department of Agriculture, Food and Nutrition Service 
(FNS) ``Guidance for States Seeking Waivers for Food Stamp Limits,'' 
FNS guidance to states, December 3, 1996. Copy included in the 
appendix.
    \14\ Ibid.
---------------------------------------------------------------------------
    In its original NPRM that covered how USDA would regulate the 
waiver authority, FNS included the conceptual framework of the criteria 
detailed guidance but did not include all of the specifics in the 
actual regulation language.\15\ Commenters, including the Center on 
Budget and Policy Priorities comments that USDA should include and 
codify the details of the guidance into rule in order to prevent 
changes in how waiver policy was interpreted and applied, allowing for 
consistency.\16\ Other commenters expressed appreciation for the 
substance of the waiver criteria as articulated in the guidance and 
provided for in the NPRM.\17\ USDA adopted the suggestion and included 
the guidance almost verbatim in the final rule. These criteria were 
modified only slightly in USDA's final regulation waivers based on the 
experience learned during the waiver application and approval process 
(for example, states were allowed to apply to more recent time periods 
the criteria the Labor Department uses to identify Labor Surplus Areas 
in order to determine if an area qualifies for a waiver). The 
regulations were proposed by the Clinton Administration and fully 
codified in regulations under the Bush Administration in 2001. In 
setting the waiver criteria, USDA adhered to longtime Labor Department 
standards to identify areas with labor-market weakness. To qualify for 
the insufficient jobs standard, a state must demonstrate that a 
geographic area (as defined by the state) meets specified criteria.
---------------------------------------------------------------------------
    \15\ 64 Fed. Reg. No. 242, page 70920, RIN: 0584-AC39 (proposed 
rule December 17, 1999.)
    \16\ CBPP Comments on 64 Fed. Reg. No. 242, page 70920, RIN: 0584-
AC39, February 17, 2000.
    \17\ Comments submitted by the Greater Upstate Law Project, Center 
for Civil Justice and the American Public Human Services Administration 
on 64 Fed. Reg. No. 242, page 70920, RIN: 0584-AC39, February 17, 2000.
---------------------------------------------------------------------------
    Federal regulations deem waiver requests that are based on certain 
criteria as ``readily approvable''--meaning USDA approves them once it 
confirms that the data are correct--because the data clearly establish 
high unemployment in the area. (In other words, USDA cannot arbitrarily 
deny a state that provides adequate documentation showing that the 
area's unemployment rate would qualify it for a waiver.) These criteria 
are:

   Designation as a Labor Surplus Area--a criterion that 
        several Federal agencies use to prioritize government contracts 
        or assistance.\18\
---------------------------------------------------------------------------
    \18\ U.S. Department of Labor, ``Labor Surplus Area: Frequently 
Asked Questions,'' updated August 21, 2015, https://www.doleta.gov/
programs/lsa_faq.cfm.

   An average unemployment rate at least 20 percent above the 
        national average over a recent 24 month time period. This 
        standard tracks the Labor Department's definition of a Labor 
---------------------------------------------------------------------------
        Surplus Area but can use more recent data.

   An average 12 month unemployment rate over ten percent.

    In addition, waivers based on unemployment rates that meet the 
criteria to qualify for additional weeks of Extended Benefits (EB) 
under the Unemployment Insurance (UI) system may also be approved by 
USDA.\19\ States may also make the case for a waiver for a given area 
based on certain other criteria; approval of these waivers is left to 
the discretion of the Secretary of Agriculture. One example is a low 
and declining employment-to-population ratio,\20\ a measure that labor 
economists use to capture weak labor markets in areas where there is a 
notable lack of jobs relative to the size of the working-age 
population. States have used this criterion sparingly, and USDA 
requires states to demonstrate additional evidence of weak labor 
markets for approval, such as a spike in unemployment or a significant 
company layoff that affects local labor markets.\21\ Typically, only a 
handful of rural counties and Indian reservations receive waivers under 
this criterion.
---------------------------------------------------------------------------
    \19\ The EB program has criteria in law under which unemployed 
workers in a state are eligible to receive extended unemployment 
benefits, and states can opt to offer EB benefits under certain 
additional criteria. (For more information, see ``Conformity 
Requirements for State UI Laws,'' Department of Labor, https://
workforcesecurity.doleta.gov/unemploy/pdf/uilaws_
extended.pdf.) Because these unemployment criteria (known as 
``triggers'') establish high unemployment, a state is eligible for a 
waiver if it meets the criteria under the triggers, even if the state 
does not elect to provide EB benefits under that trigger.
    \20\ The employment-to-population ratio is the share of the non-
institutional, civilian adult population (over age 16) that is 
employed. The employment-to-population ratio provides useful 
information in assessing labor market conditions over the business 
cycle because it takes into account changes in labor market ``slack'' 
(insufficient jobs) due to changes in both unemployment and labor-force 
participation. For more information, see Sarah Donovan, ``An Overview 
of the Employment-Population Ratio,'' Congressional Research Service, 
May 27, 2015, https://fas.org/sgp/crs/misc/R44055.pdf.
    \21\ U.S. Department of Agriculture, Food and Nutrition Service 
(FNS), ``Supplemental Nutrition Assistance Program--Guide to Supporting 
Requests to Waive the Time Limit for Able-Bodied Adults without 
Dependents (ABAWD),'' December 2, 2016, https://www.fns.usda.gov/sites/
default/files/snap/SNAP-Guide-to-Supporting-Requests-to-Waive-the-Time-
Limit-for-ABAWDs.pdf.
---------------------------------------------------------------------------
    USDA has not issued major policy changes since the criteria were 
initially published via guidance in 1996, and state waiver requests 
have consistently been evaluated according to these criteria. The 
agency has provided guidance to states on the specifics of how to do 
the required calculations and what information to attach.\22\
---------------------------------------------------------------------------
    \22\ For example, see: ``SNAP--Guide to Supporting Requests to 
Waive the Time Limit for Able-Bodied Adults without Dependents 
(ABAWD)'': https://fns-prod.azureedge.net/sites/default/files/snap/
SNAP-Guide-to-Supporting-Requests-to-Waive-the-Time-Limit-for-
ABAWDs.pdf.
---------------------------------------------------------------------------
B. Congressional Action to Expand Waivers During the Great Recession
    Waiver criteria have been consistent since 1996, with the exception 
of temporary expansions in response to the Great Recession. In response 
to the 2007 recession, Congress took action that had the effect of 
temporarily expanding the circumstances under which an area could 
qualify for a waiver. Some have mistakenly portrayed these temporary 
expansions as a permanent expansion of waiver authority. These 
temporary policies were the only two expansions in waiver criteria 
since the time limit took effect in 1996--and both have ended.

   In recognition of the Great Recession's impact on job loss 
        and increased hardship for unemployed workers, Congress enacted 
        the Federal Emergency Unemployment Benefits (EUC) program in 
        2008. EUC, like the Federal emergency unemployment insurance 
        programs enacted in every major recession since 1958, was a 
        temporary program that provided additional weeks of UI to 
        qualifying jobless workers during periods when jobs were hard 
        to find.\23\ EUC established several ``tiers,'' with each tier 
        making a specified number of additional weeks of UI benefits 
        available to jobless workers in the state, depending on the 
        state's unemployment rate. Workers in states with higher 
        unemployment rates would be in higher tiers and hence could 
        receive more weeks of UI benefits. Because qualifying for 
        higher tiers of benefits under EUC signified higher 
        unemployment and a lack of jobs, the Bush Administration 
        allowed states to qualify for a waiver based on qualifying for 
        at least the second tier of EUC.\24\
---------------------------------------------------------------------------
    \23\ Chad Stone, ``Congress Should Renew Emergency Unemployment 
Compensation Before the End of the Year,'' Center on Budget and Policy 
Priorities, November 21, 2013, http://www.cbpp.org/research/congress-
should-renew-emergency-unemployment-compensation-before-the-end-of-the-
year.
    \24\ USDA Memo, ``SNAP--ABAWD Statewide Waivers--New Criteria for 
Unemployment Insurance Extended Benefits Trigger,'' January 8, 2009, 
https://fns-prod.azureedge.net/sites/default/files/snap/
ABAWD%20Statewide%20Waivers.pdf. When all states were eligible for both 
the first and second tiers of EUC, USDA required states to be eligible 
for at least the third tier to qualify for a waiver.

      Congress extended and modified the EUC program several times, 
        allowing it to operate through January 1, 2014.\25\ Many states 
        qualified for at least the second tier of EUC through December 
        2013. As a result, they qualified for waivers from the time 
        limit into 2015 (since USDA approved waivers for up to 1 year 
        from the date a state qualified for EUC).
---------------------------------------------------------------------------
    \25\ U.S. Department of Labor, ``Emergency Unemployment 
Compensation Expired on January 1, 2014,'' updated July 1, 2015, http:/
/ows.doleta.gov/unemploy/supp_act.asp.

   Meanwhile, the 2009 Recovery Act suspended the time limit 
        nationwide for part of 2009 and all of Fiscal Year 2010. States 
        had the option to retain the time limit if they offered work 
        opportunities, such as job training and workfare, to all 
        individuals subject to the rule. During this time, states 
        didn't have to request a waiver (though almost every state 
        qualified for a statewide waiver due to the exceptionally high 
        levels of unemployment across the country). The suspension of 
        the time limit ended in September 2010. After that, most states 
        continued to qualify for statewide waivers for a few years 
---------------------------------------------------------------------------
        under EUC-related and other, longstanding USDA waiver criteria.

    The requirement that states demonstrate to USDA that an area 
exceeds a high threshold of persistent unemployment in order to qualify 
for a waiver has limited the waivers' scope. A review of waivers over 
the last 20 years shows that just over \1/3\ of the country (as 
measured by share of the total population living in waived counties) 
has been waived in a typical year.\26\ Only during the recession and 
its aftermath was more than \1/2\ the county temporarily waived from 
the time limit, and that was due to widespread elevated unemployment. 
Some have mistakenly interpreted the temporary suspension of the time 
limit in 2009-2010, or the temporary expansion of waivers during the 
aftermath of the recession when job growth remained sluggish for some 
time, as a permanent expansion of the policy or an Obama 
Administration-led effort to eliminate the time limit.
---------------------------------------------------------------------------
    \26\ ``SNAP Time Limits: Waivers from the Time Limit Are Back to 
Historic Norms,'' Center on Budget and Policy Priorities, March 24, 
2017, http://www.cbpp.org/sites/default/files/atoms/files/3-24-
17fa1.pdf.
---------------------------------------------------------------------------
C. Why Do States Seek Waivers?
    Individual state decisions to seek a time-limit waiver have varied 
over time depending on states' leadership and the economic 
circumstances at the time of their request. Nevertheless, the reasons 
remain consistent with those put forward by USDA in their early 
guidance. USDA's Office of Inspector General documented states' 
motivation in a recent audit of this policy.\27\ Because states waive 
the time limit to exempt individuals in areas lacking jobs and to ease 
administrative burden, the proposed rule would significantly increase 
the burden on states and make the time limit less reflective of areas 
lacking jobs, as we explain in greater depth later.
---------------------------------------------------------------------------
    \27\ ``FNS Controls Over SNAP Benefits for Able-Bodied Adults 
Without Dependents.''

   The time limit provision is very complicated and difficult 
        to administer. State administrators have expressed strong 
        concern with the complexity of the time-limit provision since 
        its passage in 1996. The rule requires them to track 
        individuals with a level of specificity that is inconsistent 
        with how they otherwise operate SNAP and other low-income 
        assistance programs. States find the rule to be error-prone and 
        believe that it can increase their payment error rate. Some 
        states seek waivers, in part, to ease the administrative burden 
---------------------------------------------------------------------------
        associated with the rule.

   Waiving the time limit allows states to set a genuine work 
        requirement. Under the time limit, states are not required to 
        offer a job or training program to every individual (or, for 
        that matter, to any affected individuals), and they do not 
        receive sufficient funds through the SNAP Employment and 
        Training (E&T) program to do so. In addition, the law limits 
        the types of slots a state can provide, making them expensive 
        and out of sync with the needs of much of this population. As a 
        result, very few states commit to offering work opportunities 
        to all individuals subject to the time limit.

      Waivers, by contrast, can make meaningful work requirements a 
        reality. A state requesting a waiver of the 3 month time limit 
        can still require individuals to engage in work-related 
        activities as a condition of receiving benefits through the 
        SNAP E&T program. Every state operates a SNAP E&T program, 
        through which the state can provide a wide range of employment-
        related activities to a broad range of individuals who are able 
        to work. While there is little evidence that SNAP E&T 
        requirements lead to long-term sustainable jobs, they do allow 
        a state to require a SNAP participant to engage in work 
        activities in order to remain eligible.
      Some states require SNAP participants to participate in a job 
        search program, as a way of testing an individual's willingness 
        to work, to remain eligible. These job search programs are 
        relatively inexpensive to operate. But stand-alone job search 
        is explicitly prohibited from being a qualifying E&T activity 
        for childless adults subject to the time limit. The only 
        activities states are allowed to offer to individuals subject 
        to the time limit are job training, education, and workfare 
        programs, which typically are too expensive to offer to all 
        such individuals.\28\ Moreover, this population often isn't a 
        state's priority for such investments.
---------------------------------------------------------------------------
    \28\ Hours spent in job search can count toward an individual's 
required 20 hours per week, so long as they constitute less than \1/2\ 
of the total number of hours spent in E&T activities.
---------------------------------------------------------------------------
      In short, if a childless adult searches diligently for work but 
        is unable to find a job or a slot in a work or training 
        program, he or she loses benefits after 3 months, despite 
        showing effort and willingness to work. Waivers, by contrast, 
        allow states to ensure that they are denying benefits based 
        only on bad conduct, not bad luck.

   States wish to protect individuals living in relatively high 
        unemployment areas. Even in states with relatively low 
        statewide unemployment rates, parts of the state may have 
        significantly weaker labor markets, with few jobs available. 
        The flexibility that allows states to apply for area waivers 
        recognizes that parts of a state may have insufficient jobs for 
        low-income workers. For example, some states may seek waivers 
        for areas where a dominant industry is struggling.

      States frequently use waiver authority for rural areas, where 
        about \3/4\ of adults say good jobs are hard to come by where 
        they live.\29\ Urban areas as a whole have fully recovered the 
        jobs lost in the recession, while the number of jobs in rural 
        areas continued to remain below pre-recession levels in 
        2017.\30\
---------------------------------------------------------------------------
    \29\ ``The State of American Jobs,'' Pew Research Center, October 
6, 2016, http://www.ledevoir.com/documents/pdf/
etude_travail_pewresearch.pdf.
    \30\ U.S. Department of Agriculture, ``Rural America at a Glance: 
2017 Edition,'' November 2017, https://www.ers.usda.gov/webdocs/
publications/85740/eib-182.pdf?v=43054.
---------------------------------------------------------------------------
D. Current Waiver Authority Is Insufficient to Address Needs of 
        Unemployed Workers
    While a waiver offers a necessary, temporary reprieve from the time 
limit for individuals living in areas with high unemployment, both the 
waiver authority and the underlying time limit are not responsive to 
the immediate employment challenges that many people subject to the 
rule face, even in areas of more modest unemployment. That, in part, is 
why USDA's proposed rule to restrict states' ability to seek waivers is 
so surprising and ill-informed. Geographic waivers provide needed but 
inadequate protection for individuals subject to the time limit. While 
the underlying rule exempts some individuals from the time limit (such 
as people with physical or mental conditions and those caring for 
incapacitated individuals) and states can exempt a limited number of 
additional individuals in unique circumstances, the waiver flexibility 
allows states the option to fully exempt all individuals who face 
insufficient job opportunities for reasons other than area 
unemployment.\31\ As noted above, USDA indicated in their early 
guidance on waivers that the unemployment rate can mask the labor 
market realities for individuals subject to this rule.
---------------------------------------------------------------------------
    \31\ Federal regulations identify certain individuals as exempt 
(see 7 CFR  273.24(c)) and states receive a limited number of 
individual exemptions they can use to exempt any individual subject to 
the rule, though these are underutilized in most states (see 7 CFR  
273.24(g)).
---------------------------------------------------------------------------
    Many of the individuals subject to the time limit struggle to find 
employment even in normal economic times. States utilize waivers in 
recognition of this fact, which also demonstrates why the proposed rule 
is so harsh. Those subject to this rule are extremely poor, tend to 
have limited education, and sometimes face barriers to work such as a 
criminal justice history or racial discrimination. While participating 
in SNAP, childless adults have average incomes of 33 percent of the 
poverty line--the equivalent of about $4,000 per year for a single 
person in 2019. About a quarter have less than a high school education, 
and \1/2\ have at most a high school diploma or GED.\32\ SNAP 
participants subject to the 3 month cutoff are more likely than other 
SNAP participants to lack basic job skills like reading, writing, and 
basic mathematics, according to the Government Accountability 
Office.\33\ As we will discuss in much greater depth, an extensive body 
of research shows why these adults likely face much higher unemployment 
rates than their area's unemployment rate and why the proposed rule 
would severely curtail waivers in areas where these individuals do not 
have access to adequate job opportunities.
---------------------------------------------------------------------------
    \32\ Steven Carlson, Dorothy Rosenbaum, and Brynne Keith-Jennings, 
``Who Are the Low-Income Childless Adults Facing the Loss of SNAP in 
2016?'' Center on Budget and Policy Priorities, February 8, 2016, 
http://www.cbpp.org/research/food-assistance/who-are-the-low-income-
childless-adults-facing-the-loss-of-snap-in-2016.
    \33\ ``Food Stamp Employment and Training Program,'' United States 
General Accounting Office (GAO-3-388), March 2003, p. 17.
---------------------------------------------------------------------------
    A much preferable alternative to the USDA's proposed rule would 
have been an effort to make it more possible for states to waive the 
time limit for more individuals who live in areas with insufficient 
jobs for those subject to its eligibility restriction. Restricting this 
flexibility would be counter to the intent of the law, inconsistent 
with more than 2 decades of practice, and would not produce the stated 
outcomes USDA claims its proposal would achieve.
Chapter 2: FNS Waiver Policy Has Been Consistent for the Last 22 Years
A. Current Rules Governing Waivers for Areas with Insufficient Jobs for 
        Individuals Subject to the Time Limit
    Congress established that states could waive areas lacking jobs, 
and U.S. Department of Agriculture (USDA) has established criteria that 
have been consistent for 2 decades. When the time limit was being 
debated in Congress as part of the 1996 welfare law, its proponents 
claimed that the proposed rule was not intended to take effect in areas 
where jobs weren't available. Then-Congressman and co-author of the 
provision John Kasich said, ``It is only if you are able-bodied, if you 
are childless, and if you live in an area where you are getting food 
stamps and there are jobs available, then it applies.'' \34\ The rule 
was designed to permit states to seek waivers in areas where jobs 
aren't available. To qualify for a waiver, states must provide detailed 
evidence of high unemployment in local areas, in accordance with 
rigorous requirements set by USDA. USDA has consistently used the same 
criteria to define high unemployment since the late 1990s.
---------------------------------------------------------------------------
    \34\ Congressional Record, 104th Congress, Welfare and Medicaid 
Reform Act of 1996 (House of Representatives--July 18, 1996), page 
H7905, https://www.congress.gov/crec/1996/07/18/CREC-1996-07-18.pdf.
---------------------------------------------------------------------------
    The Federal law gives states the option to request a waiver of the 
time limit if they can document that a given geographic area has an 
insufficient number of jobs (or has an unemployment rate over ten 
percent). The standards that define how a state may document 
``insufficient jobs'' for individuals subject to the time limit were 
first outlined in FNS guidance issued in December 1996.\35\ In the 
guidance, USDA offered several reflections on its understanding of 
Congressional intent at the time. First, USDA shared its belief that 
Congress understood that this group of individuals could find it 
especially challenging to find permanent employment and that waivers 
are intended to recognize this problem:
---------------------------------------------------------------------------
    \35\ U.S. Department of Agriculture, Food and Nutrition Service 
(FNS) ``Guidance for States Seeking Waivers for Food Stamp Limits,'' 
FNS guidance to states, December 3, 1996.

          USDA believes that the law provided authority to waive these 
        provisions in recognition of the challenges that low-skilled 
        workers may face in finding and keeping permanent employment. 
        In some areas, including parts of rural America, the number of 
        employed persons and the number of job seekers may be far 
        larger than the number of vacant jobs. This may be especially 
---------------------------------------------------------------------------
        so for persons with limited skills and minimal work history.

    In addition, the guidance provided key background on some of the 
policy that USDA seeks to restrict in the NPRM. With respect to how 
states can set or define the area within the state that it seeks to 
waive, USDA said, ``USDA will give states broad discretion in defining 
areas that best reflect the labor market prospects of program 
participants and administrative needs.'' \36\ The guidance also 
recognized that the statute seeks to identify whether or not there are 
sufficient jobs for individuals subject to the time limit:
---------------------------------------------------------------------------
    \36\ Ibid.

          ``The guidance that follows offers some examples of the types 
        and sources of data available to states as they consider waiver 
        requests for areas with insufficient jobs. Because there are 
        not standard data or methods to make the determination of the 
        sufficiency of jobs, the list that follows is not exhaustive. 
        States may use these data sources as appropriate, or other data 
        as available, to provide evidence that the necessary conditions 
        exist in the area for which they intend the waiver to apply. 
        The absence of a particular data source or approach (for 
        example, data or statistics compiled by a university) is not 
        meant to imply that it would not be considered by USDA if 
        requested by a state.'' \37\
---------------------------------------------------------------------------
    \37\ Ibid.

    These criteria were modified only slightly in USDA's final 
regulation on waivers based on the experience learned during the waiver 
application and approval process (for example states were allowed to 
apply to more recent time periods the criteria the Labor Department 
uses to identify Labor Surplus Areas in order to determine if an area 
qualifies for a waiver.) The regulations were proposed by the Clinton 
Administration and fully codified in regulations under the Bush 
Administration in 2001. In setting the waiver criteria, USDA adhered to 
long-time Labor Department standards to identify areas with labor-
market weakness. To qualify for the insufficient jobs standard, a state 
must demonstrate that a geographic area (as defined by the state) meets 
specified criteria.
    Federal regulations deem waiver requests that are based on certain 
criteria as ``readily approvable''--meaning USDA approves them once it 
confirms that the data are correct--because the data clearly establish 
high unemployment in the area. (In other words, USDA cannot arbitrarily 
deny a state that provides adequate documentation showing that the 
area's unemployment rate would qualify it for a waiver.) These criteria 
are:

   Designation as a Labor Surplus Area--a criterion that 
        several Federal agencies use to prioritize government contracts 
        or assistance.\38\
---------------------------------------------------------------------------
    \38\ U.S. Department of Labor, ``Labor Surplus Area: Frequently 
Asked Questions,'' updated August 21, 2015, https://www.doleta.gov/
programs/lsa_faq.cfm.

   An average unemployment rate at least 20 percent above the 
        national average over a recent 24 month time period. This 
        standard tracks the Labor Department's definition of a Labor 
---------------------------------------------------------------------------
        Surplus Area but can use more recent data.

   An average 12 month unemployment rate over ten percent.

    In addition, waivers based on unemployment rates that meet the 
criteria to qualify for additional weeks of Extended Benefits (EB) 
under the Unemployment Insurance (UI) system may also be approved by 
USDA.\39\ States may also make the case for a waiver for a given area 
based on certain other criteria; approval of these waivers is left to 
the discretion of the Secretary of Agriculture. One example is a low 
and declining employment-to-population ratio,\40\ a measure that labor 
economists use to capture weak labor markets in areas where there is a 
notable lack of jobs relative to the size of the working-age 
population. States have used this criterion sparingly, and USDA 
requires states to demonstrate additional evidence of weak labor 
markets for approval, such as a spike in unemployment or a significant 
company layoff that affects local labor markets.\41\ Typically, only a 
handful of rural counties and Indian reservations receive waivers under 
this criterion.
---------------------------------------------------------------------------
    \39\ The EB program has criteria in law under which unemployed 
workers in a state are eligible to receive extended unemployment 
benefits, and states can opt to offer EB benefits under certain 
additional criteria. (For more information, see ``Conformity 
Requirements for State UI Laws,'' Department of Labor, https://
workforcesecurity.doleta.gov/unemploy/pdf/uilaws_
extended.pdf.) Because these unemployment criteria (known as 
``triggers'') establish high unemployment, a state is eligible for a 
waiver if it meets the criteria under the triggers, even if the state 
does not elect to provide EB benefits under that trigger.
    \40\ The employment-to-population ratio is the share of the non-
institutional, civilian adult population (over age 16) that is 
employed. The employment-to-population ratio provides useful 
information in assessing labor market conditions over the business 
cycle because it takes into account changes in labor market ``slack'' 
(insufficient jobs) due to changes in both unemployment and labor-force 
participation. For more information, see Sarah Donovan, ``An Overview 
of the Employment-Population Ratio,'' Congressional Research Service, 
May 27, 2015, https://fas.org/sgp/crs/misc/R44055.pdf.
    \41\ U.S. Department of Agriculture, Food and Nutrition Service 
(FNS), ``Supplemental Nutrition Assistance Program--Guide to Supporting 
Requests to Waive the Time Limit for Able-Bodied Adults without 
Dependents (ABAWD),'' December 2, 2016, https://www.fns.usda.gov/sites/
default/files/snap/SNAP-Guide-to-Supporting-Requests-to-Waive-the-Time-
Limit-for-ABAWDs.pdf.
---------------------------------------------------------------------------
    USDA has not issued major policy changes since the criteria were 
initially published via guidance in 1996, and state waiver requests 
have consistently been evaluated according to these criteria. The 
agency has provided guidance to states on the specifics of how to do 
the required calculations and what information to attach.\42\
---------------------------------------------------------------------------
    \42\ For example, see: U.S. Department of Agriculture, Food and 
Nutrition Service (FNS), ``Supplemental Nutrition Assistance Program--
Guide to Supporting Requests to Waive the Time Limit for Able-Bodied 
Adults without Dependents (ABAWD),'' December 2, 2016, https://
www.fns.usda.gov/sites/default/files/snap/SNAP-Guide-to-Supporting-
Requests-to-Waive-the-Time-Limit-for-ABAWDs.pdf.
---------------------------------------------------------------------------
B. Department Claims to Return to Original Policy Intent and That 
        Current Waiver Standards are Inconsistent
    In the 2019 NPRM, the Department declared its commitment to 
``implement SNAP as Congress intended,'' \43\ implying that waiver 
policy has diverged significantly from the original policy set in the 
1996 welfare reform law. It also claims that the rule will ``improve 
consistency across states,'' \44\ but fails to define what the current 
inconsistency is, why the current standards are causing such 
inconsistency, does not provide any evidence to support its claim of 
inconsistency, or explain why and how it is a problem. Two possible 
interpretations of the ``inconsistency'' claim are that current waiver 
standards do not apply consistently to all states, or that the current 
standards produce inconsistent waived areas across states. Neither of 
these claims holds up to scrutiny.
---------------------------------------------------------------------------
    \43\ 2019 NPRM, p. 8.
    \44\ 2019 NPRM, p. 16.
---------------------------------------------------------------------------
FNS Waiver Criteria Have Not Changed Significantly Since 1996
    The Department's suggestion that waiver policy has deviated from 
Congressional intent suggests that either the Department now knows 
something that it did not 22 years ago when it put forward guidance to 
implement the law or that the final regulations deviated from the 
original guidance set in December 1996.
    On the first count, the Department provided no information or 
evidence from legislative history that would suggest that its knowledge 
or understanding of Congressional intent has improved since it issued 
its first guidance on waiver policy just a few short months after the 
welfare law passed. In fact, the NPRM does not provide any reference to 
legislative history to help reviewers understand why current policy is 
out of sync with the goal of the statute. It is impossible to respond 
to the Department's reasoning other than to provide the available 
legislative history as we have in Chapter 1 (Overview of Waivers From 
the Three-Month Time Limit--Their Purpose and History) which explains 
how legislative history runs counter to the Department's assertions.
    Similarly, we observe no significant policy shift in the waiver 
policy that the Department originally set forth in its December 1996 
guidance from current policy. In fact, comparing waiver standards from 
1996 to the current standards can provide insight into how much waiver 
policy has significantly changed over the past 2 decades. The best 
evidence for this comes from FNS' 1996 guidance, which describes in 
detail the waiver criteria that were available to states at the time. 
Table 2.1 below compares the key waiver criteria included in FNS' 
December 1996 guidance to the current criteria described in FNS' 
December 2016 guidance (which is the most recent articulation of the 
rules set forth in the 2001 Federal regulations).

                                Table 2.1
            FNS Waiver Policy Has Been Consistent Since 1996
------------------------------------------------------------------------
                                          January 2001    December 2016
                      December 1996 FNS    Final Rule     FNS Guidance
                        Guidance \45\         \46\            \47\
------------------------------------------------------------------------
                                  Waiver Eligibility Criteria
------------------------------------------------------------------------
Labor Surplus Area    Yes                Yes            Yes
 Designation (LSA)
\45\ U.S. Department
 of Agriculture,
 Food and Nutrition
 Service (FNS)
 ``Guidance for
 States Seeking
 Waivers for Food
 Stamp Limits,'' FNS
 guidance to states,
 December 3, 1996.
LSA-Like: 24 month    No                 Yes            Yes
 average
 unemployment rate
 20 percent above
 the national
 average using more
 current data than
 LSA
\46\ 66 Fed. Reg.,
 No. 11, 4438
 (January 17, 2001).
Qualification for     Yes                Yes            Yes
 Extended
 Unemployment
 Benefits
\47\ ``SNAP--Guide
 to Supporting
 Requests to Waive
 the Time Limit for
 Able-Bodied Adults
 without Dependents
 (ABAWD),'' FNS
 guidance issued
 December 2, 2016,
 https://fns-
 prod.azureedge.net/
 sites/default/files/
 snap/SNAP-Guide-to-
 Supporting-Requests-
 to-Waive-the-Time-
 Limit-for-
 ABAWDs.pdf.
12 month average      Yes                Yes            Yes
 unemployment rate
 over ten percent
3 month average       Yes                Yes            Yes
 unemployment rate
 over ten percent
Historical seasonal   Yes                Yes            Yes
 unemployment rate
 over ten percent
Employment-to-        Yes                Yes            Yes
 Population Ratios
Demonstration of      Yes                Yes            Yes
 lack of jobs in
 declining
 occupations or
 industries
Demonstration of      Yes                Yes            Yes
 lack of jobs in an
 area
------------------------------------------------------------------------
                                Other Waiver-Eligibility Policy
------------------------------------------------------------------------
Combining data for    Yes                Yes            Yes
 geographic and
 economic regions
Estimating            Yes                Yes            Yes
 unemployment rates
 for Tribal lands
Requesting 2 year     No                 No             Yes
 waivers
------------------------------------------------------------------------

    Table 2.1 demonstrates that the waiver criteria set in the 1996 
welfare reform law have remained remarkably consistent over the past 2 
decades. For example, FNS' 1996 guidance indicated that high 
unemployment areas can be waived by being designated as Labor Surplus 
Areas (LSA), qualifying for extended unemployment benefits, or having 
average unemployment rates of over ten percent. These are the same 
criteria described in current FNS guidance. Moreover, criteria that are 
seldom used by states, such as demonstrating historical seasonal 
unemployment or a lack of jobs in declining occupations are described 
in the 1996 guidance and remain the same today. The meaningful change 
was to allow states to use more recent unemployment data when 
considering whether an area met the LSA criteria of having average 
unemployment rates at least 20 percent above the national average for a 
recent 24 month period. This variation of the LSA criteria also permits 
areas to qualify with 24 month average unemployment rates below six 
percent. This criterion is informally known as ``LSA-like.'' Using more 
recent unemployment data allows for a more current assessment of the 
unemployment situation of an area and is an enhancement of the LSA 
criteria, not a significant change. This was added in the early 2000s 
and is codified in current Federal regulations.\48\
---------------------------------------------------------------------------
    \48\ 66 Fed. Reg., No. 11, 4438 (January 17, 2001).
---------------------------------------------------------------------------
    Similarly, the 1996 guidance included other waiver policies such as 
the ability to combine data and estimating unemployment rates for 
Tribal lands, urging states to ``consider areas within, or combinations 
of counties, cities and towns'' and to ``consider the particular needs 
of rural areas and Indian reservations.'' \49\ These policies remain in 
place in current guidance, with small changes made over the years.
---------------------------------------------------------------------------
    \49\ U.S. Department of Agriculture, Food and Nutrition Service 
(FNS) ``Guidance for States Seeking Waivers for Food Stamp Limits,'' 
FNS guidance to states, December 3, 1996.
---------------------------------------------------------------------------
    The small changes that have occurred are largely refinements of the 
original criteria, not major additions to waiver policy. For example, 
FNS guidance issued in December 2004 revised the method for calculating 
average unemployment rates over 24 month periods.\50\ Current FNS 
guidance also provides specific instructions on how to round 24 month 
average unemployment rates, and a standard methodology for estimating 
unemployment rates for Native American reservations.\51\ FNS also 
offered states ``the option of 2 year waiver approvals'' in a February 
2006 memorandum; while this was an addition to waive policy at the 
time, it was not a major revision of waiver standards--the criteria for 
2 year waivers are more restrictive than those for shorter waivers.\52\ 
(See Chapter 8 for more.)
---------------------------------------------------------------------------
    \50\ ``ABAWD Waivers--New Method for Calculating Average 
Unemployment Rates,'' FNS Northeastern Region Food Stamp Regional 
Letter, April 28, 2004.
    \51\ U.S. Department of Agriculture, Food and Nutrition Service 
(FNS), ``Supplemental Nutrition Assistance Program--Guide to Supporting 
Requests to Waive the Time Limit for Able-Bodied Adults without 
Dependents (ABAWD),'' December 2, 2016, https://www.fns.usda.gov/sites/
default/files/snap/SNAP-Guide-to-Supporting-Requests-to-Waive-the-Time-
Limit-for-ABAWDs.pdf.
    \52\ ``FSP--2-Year Approval of Waivers of the Work Requirements for 
ABAWDs under 7 CFR 273.24,'' FNS memorandum issued February 3, 2006.
---------------------------------------------------------------------------
    The final rule published in January 2001 offers clear evidence that 
the Department at the time intended to codify the waiver policies from 
its 1996 guidance, so that they would become a consistent set of rules 
that states use to determine their waiver eligibility in the future. In 
the final rule, the Department discussed the comments issued in 
response to its NPRM on the waiver policy, and why this influenced its 
codification of waiver criteria. It acknowledged that it did not 
include the 1996 guidance in its initial regulations, not because it 
deviated from the Department's intent, but because ``[the guidance] was 
extensive and detailed.'' \53\ The Department also explained that it 
``received several comments suggesting [the Department] include all or 
some of the guidance in the regulations. Commenters argued that unless 
the guidance is incorporated into the regulations, a subsequent 
Administration could abolish it without public comment. Based on these 
comments, [it] decided to incorporate some of the more pertinent 
aspects of the guidance into the regulation. More specifically, [it] 
modified the regulations at 7 CFR 273.24(f) to include a non-exhaustive 
list of the kinds of information a state agency may submit to support a 
claim of ten percent unemployment or `lack of sufficient jobs.' '' \54\ 
The final rule goes on to list the same waiver eligibility criteria 
described in Table 2.1 as part of the December 2016 guidance, and shows 
that Department recognized at the time that a consistent and 
predictable waiver policy would be an essential asset to states in the 
future.
---------------------------------------------------------------------------
    \53\ 66 Fed. Reg., No. 11, 4438 (January 17, 2001).
    \54\ Ibid.
---------------------------------------------------------------------------
    This evidence demonstrates that current waiver criteria are not 
wildly out of step with the original intent of waiver policy at its 
inception. The original guidance set the flexibility that states 
currently have in waiving areas, contrary to the Department's claim in 
the proposed rule that they use their flexibility ``in a way that was 
not likely foreseen.'' \55\
---------------------------------------------------------------------------
    \55\ NPRM, p. 7.
---------------------------------------------------------------------------
    Furthermore, the consistency in waiver policy over the decades has 
been important for states, which have relied on it for 20 years. The 
Department's claim that its proposed rule will allow ``States to 
reasonably anticipate whether it would receive approval'' ignores the 
reality that current waiver policy already accomplishes this goal. In 
reality, the rule would make it harder for states to obtain waivers and 
would disrupt their long-standing waiver implementation procedures.
The Proposed Rule Does Not Provide Evidence of Inconsistency in Current 
        Waiver Standards
    As noted earlier, the Department does not explain or justify in the 
rule its implication that current waiver standards are inconsistent, 
and reasonable interpretations of what it meant do not hold up to 
scrutiny. For example, the Department may have meant that there is not 
a consistent set of waiver standards that apply to all states. This is 
not the case, as waiver standards apply uniformly to all states. States 
might use different criteria to show their eligibility for waivers; for 
example, a state with unemployment well above ten percent might request 
a waiver based on the ten percent threshold, whereas another state with 
rapidly rising unemployment might request a waiver based on qualifying 
for extended unemployment benefits. The fact that states use different 
criteria reflects differences in their demographic composition and 
economies, among other factors. It does not mean, as the Department 
might be implying, that states do not have the option of using any of 
the criteria to show their waiver eligibility, particularly as their 
local economic conditions change over time.
    Over the past 2 decades, FNS has regularly updated its guidance to 
states to inform them of their options as the economy changed. One of 
the strengths of the current rules and USDA's application of them is 
the extraordinary consistency with which USDA applied the rules across 
the years and states. Until 2017, states could predict with extreme 
accuracy whether the Department would approve a waiver based on the 
listed criteria and guidance. It was only after the current 
Administration took office that USDA began denying waivers that it had 
long approved--such as no longer approving 2 year waivers for areas 
that met the standards set in guidance or for areas eligible under the 
Employment-to-Population ratio.\56\ While there were not a lot of these 
types of requests historically, it was the new Administration that 
introduced uncertainty into the process. Similarly, waiver requests 
that would typically be approved in 1 to 3 months can now take upwards 
of 6 months to approve. This has resulted in USDA sometimes not 
approving waivers until after the requested start date. FNS regional 
and national office staff have not known what would and would not be 
approved or when. The political leadership at USDA has introduced 
uncertainty and inconsistency in the review and approval process. 
Moreover, they have been inconsiderate of states' need for certainty 
and predictability in order to implement waivers after approval.
---------------------------------------------------------------------------
    \56\ U.S. Department of Agriculture, Food and Nutrition Service 
(FNS), ``Supplemental Nutrition Assistance Program--Guide to Supporting 
Requests to Waive the Time Limit for Able-Bodied Adults without 
Dependents (ABAWD),'' December 2, 2016, https://www.fns.usda.gov/sites/
default/files/snap/SNAP-Guide-to-Supporting-Requests-to-Waive-the-Time-
Limit-for-ABAWDs.pdf.
---------------------------------------------------------------------------
    If the Department meant instead that the current waiver standards 
do not produce consistent waived areas across states, then it is making 
an unreasonable argument. The only inconsistency across states is the 
Department's own application of the flexibility afforded to it, not in 
USDA's application of the rules (until recently). It is incumbent upon 
USDA to define and demonstrate the inconsistency it observes given that 
this argument is a core element of its reason to re-regulate these 
long-standing rules.
    The evidence shows how little of the Department's proposed rule is 
based on a clear knowledge of the waiver policy's history and an 
intimate understanding of the waiver standards' application to states. 
This clearly demonstrates the brittle nature of the Department's 
justifications of the changes to current waiver policy contained in the 
proposed rule.
Chapter 3: Setting a Floor for Waivers For Areas With 20 Percent Above 
        National Unemployment Is Inconsistent With Congressional Intent 
        and Would Be Harmful to Vulnerable Individuals
    The most significant change of the proposed rule would drastically 
roll back waivers of the time limit by requiring states to show that 
areas meet an unemployment rate threshold of 20 percent above the 
national average (which the Department of Agriculture, or the 
Department, and we will refer to as the ``20 percent standard'') and, 
if the 20 percent standard is below a specific threshold, meet this 
specific threshold, referred to as the ``unemployment rate floor'' to 
qualify for a waiver. We believe this proposal is out of sync with the 
goal and purpose of the underlying legislation. Furthermore, the 
Department did not discuss whether it considered a substantial body of 
relevant research that contradicts the claims it made in support of 
this change and provided little to no evidence to back up its proposal, 
making it difficult for us to comment on the process the Food and 
Nutrition Service (FNS) used to develop this regulation. Below we 
discuss each of the following reasons in detail that explain the flaws 
in this process:

   This proposal is contrary to Congressional intent, which 
        clearly was to allow states flexibility to use a variety of 
        metrics to demonstrate that the population subject to the time 
        limit does not have access to enough jobs. Congress has 
        rejected past proposals to impose an unemployment rate floor 
        and otherwise restrict the current waiver criteria.

   Considerable evidence shows that the adults without 
        dependent children potentially subject to the rule face 
        overlapping labor market disadvantages, and therefore 
        experience significantly higher unemployment rates than the 
        general unemployment rate for their area. Because an area's 
        overall unemployment rate overstates job availability for the 
        individuals subject to the time limit, imposing an unemployment 
        rate floor would disqualify many areas from eligibility for a 
        waiver where childless SNAP participants have very few job 
        opportunities. The statute clearly gives states that want to 
        the ability to waive the time limit for some or all individuals 
        in areas where there aren't enough jobs to employ these 
        individuals.

   The Department misleadingly cites the unemployment rate 
        floor used by the Department of Labor in establishing Labor 
        Surplus Areas (LSAs) to support the proposal, without 
        recognizing that LSAs are meant for different purposes, and 
        that LSAs also include an unemployment rate ceiling.

   The Department uses the concept of a ``natural rate of 
        unemployment'' to support the proposed unemployment rate floor 
        of seven percent, which is a misinterpretation of a 
        macroeconomic concept that is not a fixed or precisely 
        identifiable unemployment rate. Furthermore, the Department 
        then suggests a significantly higher unemployment rate floor 
        than what it states the natural rate is without explaining how 
        the natural rate relates to the proposed unemployment rate 
        floor of seven percent. This lack of explanation for choosing 
        the substantially higher rate of seven percent demonstrates how 
        this specific unemployment rate floor was chosen arbitrarily.

   While no specific rate of unemployment would properly 
        reflect these individuals' circumstances, evidence shows that 
        seven percent unemployment specifically is too high, given that 
        many of these individuals are often in groups that experience 
        unemployment rates significantly above that level and they 
        often face barriers to employment.

   The proposal would fail to adequately provide states with 
        waiver coverage during times of rising unemployment, as the 
        combination of the high unemployment rate with the lengthy 24 
        month lookback would preclude many states with rising 
        unemployment from eligibility. The Department lacked 
        transparency in not referencing whether they examined the 
        potential impact of this proposal at other times in the 
        business cycle besides the current moment.

   The Department attempts to support its proposed unemployment 
        rate floor by explaining that such a floor would decrease the 
        share of who it refers to as ``ABAWDs'' living in a waived 
        area. This justification ignores Congressional intent and lacks 
        transparency in the underlying assumptions and methodology used 
        to estimate this metric.

   The Department also sought feedback on alternative 
        unemployment rate floors of six and ten percent, which are both 
        unworkable and an inappropriate reading of the statute. Its 
        proposal of these alternate floors demonstrates the 
        arbitrariness of the proposed seven percent floor, but also 
        shows that it is impossible to designate a specific 
        unemployment rate floor that would adequately interpret the law 
        by accurately reflecting jobs available to childless adults.

    In proposing an unemployment rate floor for waivers based on the 20 
percent standard, the Department ignores the intent of Congress and 
uses misleading justifications with no transparent evidence to support 
its claims. While the current 20 percent standard may not perfectly 
represent areas that lack jobs for childless adults because the overall 
unemployment rate masks divergent labor market opportunities for sub-
groups such as these individuals, the proposed rule would only 
exacerbate the shortcomings of current policy.
A. Unemployment Rate Floor Proposal Inconsistent With Congressional 
        Intent
    When Congress established the 3 month time limit in the Personal 
Responsibility and Work Opportunity Reconciliation Act of 1996 
(PRWORA), Public Law 104-193, it established that a state may seek a 
waiver for a geographic area. Congress gave states this authority in 
recognition that individuals may not have success in finding a job if 
there are limited job opportunities. When the House Committee on Budget 
reported the original bill, the report stated:

          The Committee understands that there may be instances in 
        which high unemployment rates in all or part of a state or 
        other specified circumstances may limit the jobs available for 
        able-bodied food stamp participants between 18 and 50 years 
        with no dependents. Therefore the Secretary, upon request from 
        a state, is provided with the authority to waive job 
        requirements in these circumstances or if unemployment rates 
        are above ten percent.\57\
---------------------------------------------------------------------------
    \57\ H.R. Report 104-651, Welfare and Medicaid Reform Act of 1996, 
https://www.congress.gov/congressional-report/104th-congress/house-
report/651.

    Congress created waiver authority to enable states to waive areas 
with ``high unemployment rates'' or ``otherwise specific 
circumstances,'' indicating that a range of circumstances may be 
indicative of depressed labor market conditions. The welfare reform law 
established that a state could seek a waiver for an area if it: ``(i) 
has an unemployment rate of over ten percent; or (ii) does not have a 
sufficient number of jobs to provide employment for the individuals.'' 
\58\ (Herein, as with the current regulations, we will use ``area'' to 
refer to geographic areas, which generally refers to areas for which 
states generally seek waivers, such as counties, cities, towns, Tribal 
areas, or metropolitan areas.)
---------------------------------------------------------------------------
    \58\ Food and Nutrition Act, 7 U.S.C.  2015(o)(4). This language 
is identical to the language in P.L. 104-193, PRWORA.
---------------------------------------------------------------------------
    Congress therefore created two distinct categories to establish the 
circumstances under which a state can request a waiver:

   The first criterion establishes that an area with an 
        unemployment rate of ten percent may qualify for a waiver. The 
        unemployment rate measures the share of the labor force that is 
        actively looking for work. Historically, a ten percent 
        unemployment rate is an indicator of severe labor market 
        distress, such as during an economic downturn. Since the Bureau 
        of Labor Statistics (BLS) began publishing monthly unemployment 
        rates in 1948, the national unemployment rate has equaled or 
        exceeded ten percent only during the 1981-1982 recession and 
        during the Great Recession of 2007-2009. Congress recognized 
        that a local area with such a high unemployment rate likely 
        would not offer adequate job opportunities so that people who 
        are subject to the time limit could find work. With such high 
        unemployment, even the most readily employable jobseekers will 
        likely struggle to find work, and those who are more 
        disadvantaged will face even more challenges. States that 
        prefer to waive only areas with extremely high unemployment 
        rates can also request waivers based on this criterion.

   The second criterion is focused on measuring employment 
        opportunities for the specific individuals affected by the time 
        limit. Congress recognized that while useful for measuring the 
        health of a local labor market, the unemployment rate may not 
        give a complete picture of job availability for all workers in 
        that area, particularly for individuals facing labor market 
        disadvantages. An area may not have a sufficient number of jobs 
        because the share of jobseekers who are out of work is 
        relatively high, as indicated by the employment rate. Even with 
        a low unemployment rate, however, there can be instances where 
        there aren't enough jobs to provide employment for specific 
        individuals or groups. Even if there are enough jobs in number 
        to match the number of jobseekers, the individuals' skills 
        might not match the requirements of the available jobs, the 
        jobs may be inaccessible due to geographic or transit 
        limitations, or employers may discriminate against some 
        jobseekers based on their race, work history, disability, or 
        other characteristics, for example.

    In its original interpretation, the Department recognized that 
Congress intended for the ``insufficient jobs'' criterion to include a 
range of metrics that are targeted towards the individuals subject to 
the time limit. The Department published guidance on December 3, 1996, 
which stated:

          The statute recognizes that the unemployment rate alone is an 
        imperfect measure of the employment prospects of individuals 
        with little work history and diminished opportunities. It 
        provides states with the option to seek waivers for areas in 
        which there are not enough jobs for groups of individuals who 
        may be affected by the new time limits in the Food Stamp 
        Program.
          To some extent, the decision to approve waivers based on an 
        insufficient number of jobs must be made on an area-by-area 
        basis. Examples of such situations include areas where an 
        important employer has either relocated or gone out of 
        business. In other areas there may be a shortage of jobs that 
        can be filled by persons with limited skills and work 
        experience relative to the number of persons seeking such 
        jobs.\59\
---------------------------------------------------------------------------
    \59\ USDA, ``Guidance for States Seeking Waivers for Food Stamp 
Limits,'' December 3, 1996.

    The Department therefore originally (in 1996) interpreted the 
intent of Congress in creating the second category for waiver authority 
as a recognition of the shortcomings of the unemployment rate for 
measuring job opportunities for the individuals subject to the time 
limit, and established that it could use flexibility in determining 
whether a state demonstrates a lack of jobs. In response to comments, 
when preparing the original final rule, the Department balanced the 
need to provide specific guidance that would be codified in regulation 
so that it would remain consistent across subsequent Administrations 
with the need to retain the flexibility that the Department recognized 
that Congress had created in its original lawmaking. The final rule 
---------------------------------------------------------------------------
stated:

          Based on these comments, we have decided to incorporate some 
        of the more pertinent aspects of the guidance into the 
        regulation. More specifically, we have modified the regulations 
        at 7 CFR 273.24(f) to include a non-exhaustive list of the 
        kinds of information a state agency may submit to support a 
        claim of ten percent unemployment or `lack of sufficient jobs.' 
        \60\
---------------------------------------------------------------------------
    \60\ 66 Fed. Reg., No. 11, 4438 (January 17, 2001), p. 4462. 
https://www.federalregister.gov/d/01-1025/p-205.

    FNS' original (2001) interpretation therefore was clear that in 
providing guidance about specific methods states can use to demonstrate 
a lack of jobs in an area, it was not precluding states from using 
other data or metrics to demonstrate insufficient jobs, given that it 
is a concept not easily shown by any one numeric quantity or metric.
    By proposing an unemployment rate floor, the Department is 
proposing to restate the waiver criteria in a manner that is 
inconsistent with the intent of the statute. Currently, states can 
waive areas with insufficient jobs to employ a specific, more 
disadvantaged, population. The current 20 percent standard already has 
limitations in its ability to reflect jobs available for individuals 
subject to the time limit, who likely experience much higher 
unemployment rates than the overall unemployment rate in their area. As 
we discuss in detail below, areas with unemployment rates that are 20 
percent above the national average may still lack jobs for those with 
barriers to unemployment. As we will explain, there are several reasons 
why current aspects of the 20 percent standard in the context of the 
current regulations allow for a greater ability to demonstrate a lack 
of sufficient jobs than the proposed regulation would allow. The 
proposed regulations would therefore significantly worsen the problem 
with the current 20 percent standard as a measure of ``insufficient 
jobs.''
    First, under the current regulation, an area with elevated 
unemployment compared to national unemployment can qualify for a 
waiver, without meeting a specific unemployment rate standard. Defining 
high unemployment at a relative level rather than a specific 
unemployment rate threshold allows for greater consideration of trends 
such as those in labor force participation, which may affect low 
unemployment rates, especially relevant for disadvantaged groups. If 
workers who are not employed stop looking for work and therefore exit 
the labor force, measures of labor force participation will decline. 
Because the unemployment rate measures the share of the labor force 
that is not employed but is actively seeking work, lower labor force 
participation may be a signal of weak labor markets that is not 
reflected in the unemployment rate (for example, if discouraged workers 
stop looking for work).
    Overall labor force participation has fallen over the last 2 
decades, including particularly sharply during the Great Recession, and 
only began rebounding in about 2015. Labor force participation fell 
sharply among prime-age workers (thus less affected by population aging 
and retirement) with lower educational attainment from 2000 to 2015 and 
in 2018 were still below 2000 levels.\61\
---------------------------------------------------------------------------
    \61\ Audrey Breitwieser, Ryan Nunn, and Jay Shambaugh. ``The recent 
rebound in prime-age labor force participation,'' Brookings 
Institution, August 2, 2018. https://www.brookings.edu/blog/up-front/
2018/08/02/the-recent-rebound-in-prime-age-labor-force-participation/.
---------------------------------------------------------------------------
    Lower unemployment rates are thus less indicative of strong labor 
markets in recent years than in the past, and particularly so for a 
group that tends to fare worse in the labor market, such as those with 
lower levels of education. The 20 percent standard, which currently 
does not have a floor, relies on unemployment rates, which are an 
imperfect proxy of jobs available for this population. Because the 
current unemployment rate threshold needed to qualify for a waiver 
varies along with national trends, however, the current standard gives 
more flexibility to capture those trends. Not having a specific 
unemployment rate floor therefore allows for the 20 percent standard to 
better capture insufficient jobs than it would with a specific floor. 
In addition, currently states have the ability to group together 
counties to better represent local labor market opportunities, which 
the proposed rule would also restrict. (We discuss these changes in 
more detail in Chapter 5.) This flexibility also helps mitigate some of 
the shortcomings in the current 20 percent standard.
    Second, the Department is also proposing to eliminate other 
criteria existing in current regulations that can serve as an 
alternative to measuring ``insufficient jobs'' in cases where the 20 
percent standard does not adequately reflect job opportunities. In the 
context of these changes, the 20 percent standard takes on increasing 
importance as one of the sole methods to demonstrate a lack of 
sufficient jobs. The effect of these proposed changes largely results 
in a requirement that states demonstrate a specific unemployment rate 
threshold to qualify for a waiver under the ``insufficient jobs'' 
criterion, when Congress expressly intended for this criterion to 
encompass a broader range of metrics.
    The Department proposes to eliminate most of the remaining 
alternatives to metrics based on the unemployment rate that current 
regulations at 7 CFR  273.24(f)(2)(ii) allow, such as the elimination 
of the option to demonstrate a ``low and declining employment-
population ratio'' or to demonstrate declining industries. The 
Department would also sharply reduce the ability of states to request 
waivers for groups of neighboring counties, which may be useful in 
cases where the unemployment rate is a particularly poor proxy for 
labor market opportunities for individuals subject to the time limit. 
(We discuss the changes to employment-population ratio and other means 
of showing a lack of sufficient jobs in Chapter 4, and changes to 
grouping in Chapter 5.) With these changes, for the most part an area 
could only qualify for a waiver by demonstrating that it has a 12 month 
unemployment rate average of at least ten percent, a 2 year 
unemployment rate of at least seven percent, or that it qualifies for 
extended unemployment insurance benefits, the eligibility for which is 
based on a recent 3 month insured or total unemployment rate.
    The proposal does allow for states to demonstrate ``exceptional 
circumstances,'', but even then suggests that it must support this 
claim with evidence, such as of a ten percent unemployment rate: ``the 
request must demonstrate that the exceptional circumstance has caused a 
lack of sufficient number of jobs, such as data from the BLS or a BLS-
cooperating agency that shows an area has a most recent 3 month average 
unemployment rate over ten percent.'' \62\ Under the proposed rule, 
states will largely be limited to demonstrating that an area meets a 
specific unemployment rate threshold to qualify for a waiver under the 
``insufficient jobs'' category of waivers, which does not align with 
the intent of Congress to provide for multiple metrics under this 
category.
---------------------------------------------------------------------------
    \62\ NPRM, p. 992.
---------------------------------------------------------------------------
    Congress regularly includes specific unemployment rate thresholds 
for policy purposes when that is its intent. Congress included ten 
percent unemployment as one of the criteria to qualify for a waiver of 
SNAP's 3 month time limit, as stated above. Similarly, in the same 
legislation, Public Law 104-193, Congress created a specific definition 
of a ``needy state'' under the TANF program, which allows states 
additional weeks of job search and readiness. One of the qualifications 
for a ``needy state'' was a 3 month unemployment rate of at least 6.5 
percent that exceeds 110 percent of the unemployment rate for the same 
period in either of the last 2 years.\63\ Congress clearly understood 
that unemployment rates may be an appropriate threshold in some 
instances, but chose to include a criterion that was more loosely 
defined and allowed for alternative economic measures to demonstrate a 
lack of jobs. Congress also chose to allow waivers based on economic 
circumstances that reflect jobs available for a targeted population, 
the individuals subject to the time limit. Had Congress intended to 
allow states to qualify for waivers only based on unemployment rates, 
it would have only included waiver criteria with those unemployment 
rate parameters, rather than including the second criteria targeted 
towards childless adult SNAP participants.
---------------------------------------------------------------------------
    \63\ Personal Responsibility and Work Opportunity Reconciliation 
Act of 1996, P.L. 104-193,  403(b)(6), https://www.congress.gov/104/
plaws/publ193/PLAW-104publ193.pdf.
---------------------------------------------------------------------------
    In the original final rule, published in 2001, the Department made 
clear that they interpreted the ``lack of sufficient jobs'' as 
encompassing a broad range of metrics and not exclusively tied to 
demonstrating a high unemployment rate. By proposing a specific 
unemployment rate threshold for the 20 percent standard, reducing the 
ability of states to group together areas, and eliminating most of the 
alternative criteria that would let states use alternative information, 
the Department has substantially changed its interpretation of how 
states can demonstrate that an area lacks jobs for the individuals 
subject to the time limit. In practice, except during times when states 
qualify for extended benefits, under the proposed regulation, states 
would largely be limited to showing that an area has a seven percent 
unemployment rate over 2 years to show it lacks enough jobs to employ 
people subject to the time limit. The Department did not attempt to 
demonstrate that a specific unemployment rate threshold shows an area 
lacks jobs for these individuals, instead discussing the unrelated fact 
that the proposal would have the effect of narrowing the number of 
waived areas, which we explain below. The Department therefore provides 
no evidence that the changes in the rulemaking are aligned with the 
intent of the statute to allow waivers in areas lacking sufficient 
jobs, a broader concept than areas meeting specific unemployment rate 
thresholds.
Congress Recently Rejected Proposals to Limit Current Waiver-Approval 
        Standards
    Congress also has rejected attempts to narrow waiver-approval 
criteria to impose an unemployment rate floor for the ``20 percent 
standard.'' H.R. 2, the House Agriculture Improvement Act of 2018, as 
passed by the House on June 21, 2018, included a restriction similar to 
the Administration's proposal, requiring an area to have an 
unemployment rate of at least seven percent to qualify based on having 
a 2 year unemployment average greater than the national average. The 
Senate did not include such a restriction on waivers. The Conference 
Committee adapted the Senate's approach, which then passed and was 
signed into law. As Rep. Marcia Fudge, a conferee, noted in the 
Congressional Record:

          The Conference Committee also rejected House provisions that 
        would shorten SNAP's 3 month time limit to 1 month and expand 
        the population subject to the rule to a broader group of 
        recipients. We also rejected the House's proposal to limit 
        states' flexibility to waive high-unemployment areas from the 3 
        month limit.\64\
---------------------------------------------------------------------------
    \64\ See the floor statement by Congresswoman Fudge, 164 Cong. Rec. 
H10149 (daily ed. December 12, 2018), https://www.congress.gov/
congressional-record/2018/12/12/house-section/article/H10142-3.

    Similarly, the Conference Report noted that Congress chose not to 
---------------------------------------------------------------------------
change the underlying statute:

          The Managers also acknowledge that waivers from the ABAWD 
        time limit are necessary in times of recession and in areas 
        with labor surpluses or higher rates of unemployment. The 
        Managers intend to maintain the practice that bestows authority 
        on the state agency responsible for administering SNAP to 
        determine when and how waiver requests for ABAWDs are 
        submitted.\65\
---------------------------------------------------------------------------
    \65\ H.R. Rept. 115-1072, Agriculture Improvement Act of 2018, 
Title IV (3), https://www.congress.gov/congressional-report/115th-
congress/house-report/1072.

    Congress therefore chose not to change the criteria by which states 
could request area waivers. While the Administration cited the House-
passed version of the H.R. 2 to support the proposed seven percent 
unemployment floor, Congress ultimately rejected this proposal in favor 
of the Senate approach, demonstrating intent to keep the current 
interpretation of the ``insufficient jobs'' criterion intact.
B. Unemployment Rates Overstate Jobs Available to Childless Adult SNAP 
        Participants
    By proposing an unemployment rate floor for the ``20 percent 
standard,'' the Department argues that areas with unemployment rates 
below this threshold offer enough jobs so that those individuals can 
find work. For example, when describing its support for its proposed 
unemployment rate floor, the Department states, ``The Department views 
the proposal as more suitable for achieving a more comprehensive 
application of work requirements so that ABAWDs in areas that have 
sufficient number of jobs have a greater level of engagement in work 
and work activities, including job training.'' \66\ The Department 
therefore states that areas with unemployment rates below its proposed 
floor of seven percent over 2 years offer a sufficient number of jobs 
to the individuals subject to the time limit. This interpretation that 
areas with lower unemployment have enough jobs to employ adults without 
dependent children ignores the reality that overall unemployment rates 
overstate jobs available to disadvantaged individuals.
---------------------------------------------------------------------------
    \66\ NPRM, p. 984.
---------------------------------------------------------------------------
    The Department states that the unemployment rate floor proposal 
would prevent areas with low unemployment from qualifying for a waiver 
but ignores evidence that the individuals subject to the time limit are 
in demographic groups that experience higher unemployment rates than 
their area's average. In explaining why it chose to propose an 
unemployment rate floor, the Department noted:

          Based upon operational experience, the Department has 
        observed that, without an unemployment rate floor, local areas 
        will continue to qualify for waivers under the Department's 20 
        percent standard based on high unemployment relative to the 
        national average even as local unemployment rates fall to 
        levels as low as five to six percent (depending upon the 
        national rate).

    The Department is therefore stating that the floor is necessary to 
prevent areas with unemployment rates it considers ``low'' from 
qualifying for a waiver. Adult SNAP participants without dependent 
children, however, are likely to face barriers to employment that 
result in fewer jobs available for those individuals than for the 
general population. It is unrealistic to set a specific threshold that 
guarantees that the labor market creates a sufficient number of jobs to 
provide employment to this group, and any such threshold based on the 
overall unemployment rate in an area would guarantee that many areas 
where childless adult SNAP participants could not find work were 
ineligible. When it explained its position that it does not believe 
areas with low unemployment rates should qualify for waivers, the 
Department did not provide any research to support its position that 
areas with low unemployment rates provide enough jobs so that the 
individuals subject to the time limit can find work, nor did it address 
the extensive research that demonstrates that these individuals 
struggle to find work even when unemployment rates are low. Because the 
Department did not provide this information, it is difficult for 
commenters to understand how they are interpreting a specific 
unemployment rate as measuring job availability for this population and 
to respond to this reasoning.
    The unemployment rate is a broad labor market metric that masks 
differences in the labor market outcomes experienced by different 
groups. Some groups, such as African American workers, have 
historically and consistently higher unemployment rates. The recent 
Great Recession also demonstrated how less-advantaged groups fared more 
poorly in the recession, losing more jobs and recovering more slowly.
    Evidence shows that the adults targeted by the time limit often 
face barriers to work. While these low-income adults without dependents 
are a diverse group and there has been limited research on this 
specific population, the available evidence demonstrates that many face 
greater struggles to find work than the overall population. This group, 
while diverse, has many characteristics that, as we will explain below, 
are associated with worse labor market outcomes:

   Over \3/4\ of this group have a high school diploma or less, 
        and studies show that many lack skills sought by employers.

   This group is demographically diverse. Of adult SNAP 
        participants aged 18 through 49 who do not receive disability 
        income or have children in the household, about 53 percent are 
        male, and 47 percent are female. About \2/5\ are aged 18 
        through 29, \1/4\ are aged 30 to 39, and \1/3\ are aged 40 to 
        49. About \2/5\ are white, over \1/4\ are African American, and 
        approximately 20 percent are Latino.\67\ They live in a range 
        of areas: about \2/5\ live in urban areas, \2/5\ live in 
        suburban areas, and about 15 percent live in rural areas.\68\
---------------------------------------------------------------------------
    \67\ We looked at U.S. Agriculture Department's Fiscal Year 2017 
SNAP Households Characteristic data (QC), the 2017 American Community 
Survey (ACS) 1 year estimates, and the March 2018 Community Population 
Survey (CPS). Reporting of race/ethnicity is voluntary and is missing 
for 13 percent of ABAWDs in QC. About 12 percent of ABAWDs self-
identified or were coded by an eligibility worker as ``Latino or 
Hispanic'', but the share increased to 17 percent in high-reporting 
states (missing for less than ten of SNAP participants). CPS and ACS 
capture more Hispanics than QC. Hispanics account for 22 percent of 
ABAWDs in CPS and 20 percent in ACS. Compared to ACS, the disability 
income questions are much more detailed and comprehensive in CPS.
    \68\ CBPP analysis of the March 2018 Current Population Survey. 
Some 12 percent are unknown.

   Like most SNAP participants, this group largely works, but 
        in low-wage jobs that provide little stability, and as a 
        result, many move in and out of work and experience periods 
---------------------------------------------------------------------------
        when they are out of work.

   Research indicates that many of these individuals face 
        barriers to employment, including low skills, inconsistent work 
        history, health conditions that limit their ability to work, 
        inadequate access to transportation, criminal justice history, 
        or unstable access to housing.

    Because this population is distinct from the United States 
population, and faces greater disadvantages with regards to accessing 
employment, an overall unemployment rate or other overall labor force 
metric will largely overstate the jobs available to this group. The 
section below explains the research documenting the unique barriers to 
employment that childless adult SNAP participants face, and the higher 
unemployment rates associated with many of these characteristics.
Childless and Non-Custodial Parent Adult SNAP Participants Are Likely 
        to Have Lower Levels of Educational Attainment, Which Is 
        Associated With Higher Unemployment Rates and More Sensitivity 
        to Labor Market Shocks
    The majority of adult SNAP participants without dependents have a 
high school education or less. According to 2017 USDA Household 
Characteristics data, about \1/4\ (24 percent) of non-disabled 
individuals aged 18 through 49 in households without children report 
having less than a high school education, and about 54 percent report a 
high school diploma or a GED. (Some eight percent do not report 
educational attainment.) \69\ They are more likely than other SNAP 
participants to lack basic job skills like reading, writing, and basic 
mathematics, according to a 2003 Government Accountability Office (GAO) 
study.\70\ A more recent study of SNAP employment and training (E&T) 
participants, which includes many childless adults ages 18 through 49, 
but did not separately report results for that population, found that 
\3/4\ of employment and training providers surveyed found that at least 
some of the E&T participants they serve lack basic skills when they 
enter the program, over \1/2\ said some participants have low literacy 
levels or were high school dropouts, and over \2/5\ cited that 
participants' skills were mismatched to industry needs or were out of 
date. Over \1/4\ of E&T participants surveyed identified limited 
education as a barrier to employment.\71\ Caseworkers in a work 
experience program in Ohio found signs of functional illiteracy even 
among those with a high school degree.\72\
---------------------------------------------------------------------------
    \69\ CBPP analysis of FY 2017 USDA Household Characteristics data.
    \70\ ``Food Stamp Employment and Training Program,'' United States 
General Accounting Office, revised March 2003, https://www.gao.gov/
assets/240/237571.pdf.
    \71\ Gretchen Rowe, Elizabeth Brown, and Brian Estes, ``SNAP 
Employment and Training (E&T) Characteristics Study: Final Report,'' 
United States Department of Agriculture, Food and Nutrition Services, 
revised October 2017, https://fns-prod.azureedge.net/sites/default/
files/ops/SNAPEandTCharacteristics.pdf.
    \72\ ``A Comprehensive Assessment of Able-Bodied Adults Without 
Dependents and Their Participation in the Work Experience Program in 
Franklin County, Ohio,'' Ohio Association of Foodbanks, revised 2014, 
http://admin.ohiofoodbanks.org/uploads/news/WEP-2013-2014-report.pdf.
---------------------------------------------------------------------------
    Research shows that adults with lower educational attainment have 
higher unemployment rates than those with more education. (Figure 3.1.) 
For example, in 2018, while the unemployment rate for workers with a 
bachelor's degree or more was 2.1 percent, the unemployment rate for 
high school graduates was 4.1 percent, and for those with less than a 
high school education, 5.6 percent. African Americans with less than a 
high school diploma had an unemployment rate of 10.4 percent.\73\
---------------------------------------------------------------------------
    \73\ ``Employment Status of the Civilian Population 25 Years and 
Over by Educational Attainment,'' Bureau of Labor Statistics, revised 
February 1, 2019, https://www.bls.gov/news.release/empsit.t04.htm.
---------------------------------------------------------------------------
Figure 3.1
Unemployment Higher Among Those With Less Education
Monthly unemployment rate

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: U.S. Bureau of Labor Statistics.

    Workers with less education are more likely to lose jobs during an 
economic downturn and will recover more slowly in the aftermath of a 
recession. Researchers have found that an increase of one percentage 
point in the state unemployment rate leads to almost a two-percentage-
point increase in unemployment for workers with less than a high school 
degree compared to less than 0.5-percentage-point increase for those 
with a college degree.\74\ Workers with a high school diploma had lower 
employment rates in 2007 than college graduates: 55 percent for those 
with only high school education, compared to 72.5 percent for those 
with a bachelor's degree. Employment rates, or the share of the 
population with a job, fell more sharply for the group with lower 
levels of educational attainment, and in 2018 had yet to return to pre-
recession levels.\75\ Counties with large shares of workers with less 
than a high school degree also saw greater employment losses during the 
Great Recession.\76\
---------------------------------------------------------------------------
    \74\ Hilary Hoynes, Douglas L. Miller, and Jessamyn Schaller, ``Who 
Suffers During Recessions?'' Journal of Economic Perspectives (Summer 
2012), pp. 27-48. https://pubs.aeaweb.org/doi/pdfplus/10.1257/
jep.26.3.27.
    \75\ Lauren Bauer and Jay Shambaugh, ``Workers with Low Levels of 
Education Still Haven't Recovered From the Recession,'' The Hamilton 
Project (September 2018), pp. 1-4, http://www.hamiltonproject.org/blog/
employment_rate_gap_workers_with_low_levels_of_education_
still_havent_recov.
    \76\ Brian Thiede and Shannon Monnat, ``The Great Recession and 
America's Geography of Unemployment'' Demographic Research (September 
2016), pp. 891-928. https://www.ncbi.nlm.nih.gov/pmc/articles/
PMC5486972/.
---------------------------------------------------------------------------
    Workers with less education may be hit harder by recessions in part 
because when unemployment rises, employers may raise the skill 
requirements for positions: one study found that a one-percentage-point 
increase in the local unemployment rate raises the fraction of jobs 
requiring a bachelor's degree by about 0.4 percentage points and the 
fraction of jobs requiring 2 or more years of experience by about 0.8 
percentage points.\77\ Evidence also suggests that for workers entering 
the labor market during a recession, the effects can be long-lasting: 
those workers had reduced earnings that persisted up to 10 years into 
workers' careers, and the effect was most pronounced for those with 
less than a high school education, driven by greater losses in 
employment.\78\
---------------------------------------------------------------------------
    \77\ Alicia Modestino, Daniel Shoag, and Joshua Balance, 
``Upskilling: Do Employers Demand Greater Skill When Skilled Workers 
are Plentiful?'' Harvard Kennedy School Taubman Center for State and 
Local Government (May 2015), pp. 1-4. https://www.hks.harvard.edu/
sites/default/files/centers/taubman/files/Upskilling.pdf.
    \78\ Hannes Schwandt and Till von Wachter, ``Unlucky Cohorts: 
Estimating the Long-Term Effects of Entering the Labor Market in a 
Recession in Large Cross-Sectional Data Sets,'' Journal of Labor 
Economics, Vol. 37, No. 51 (January 2019), S161-S198, https://
www.journals.uchicago.edu/doi/10.1086/701046.
---------------------------------------------------------------------------
    The majority of adult SNAP participants without dependents have a 
high school diploma or lower educational attainment. Evidence shows 
that workers with a high school diploma or less have higher 
unemployment rates, lower employment rates, experience greater 
employment losses during economic downturns, and recover more slowly. 
The overall unemployment rate therefore will significantly overstate 
the employment opportunities available to less-educated workers, 
particularly during a recession and the aftermath. FNS does not appear 
to have considered any of this research in developing this proposal. We 
urge FNS to carefully review this literature, which demonstrates that 
because adults with less education, typically have higher unemployment 
rates than the overall average in their area, the proposed unemployment 
rate floor would be a much higher rate for adults with less education, 
the majority of childless adults.
Over Two-Fifths of Childless Adult SNAP Participants Aged 18-49 Are 
        African American or Latino, Groups That Experience Higher 
        Unemployment Rates and More Employment Discrimination
    Over \1/4\ of childless adult SNAP participants targeted by the 
time limit are African American and approximately 20 percent are 
Latino.\79\ These groups, particularly African Americans, also have 
higher unemployment rates than white Americans and are more affected by 
recessions.
---------------------------------------------------------------------------
    \79\ CBPP analysis of FY 2017 USDA Household Characteristics data, 
the March 2018 Current Population Survey, and 2017 American Community 
Survey (ACS) 1 year estimates.
---------------------------------------------------------------------------
    Black and Latino workers generally have higher unemployment rates 
than white Americans. According to data published by the BLS, in the 
fourth quarter of 2018, for example, the overall unemployment rate was 
3.6 percent and 3.2 percent for white workers, but Latinos had an 
unemployment rate of 4.3 percent, and the unemployment rate for African 
Americans was 6.1 percent.\80\ In fact, for about the past 4 decades, 
unemployment rates among black workers have been about double those of 
white workers.\81\ This relationship is true even when comparing 
unemployment rates for those with similar education levels. The 
unemployment rate among African American workers with less than a high 
school education in 2018 was 10.4 percent, more than double the 
unemployment rate of whites with the same education level, which was 
5.1 percent. Black high school graduates had unemployment rates of 6.7 
percent in 2018, close to double the unemployment rate for white high 
school graduates in 2018, of 3.5 percent.\82\
---------------------------------------------------------------------------
    \80\ ``Table E-16. Unemployment Rates by age, sex, race, and 
Hispanic or Latino ethnicity,'' Bureau of Labor Statistics, revised 
January 4, 2019, https://www.bls.gov/web/empsit/cpsee_e16.htm.
    \81\ Valerie Wilson, ``Before the State of the Union, a fact check 
on black unemployment,'' Economic Policy Institute, February 2019, pp. 
1-4. https://www.epi.org/blog/before-the-state-of-the-union-a-fact-
check-on-black-unemployment/.
    \82\ ``Labor Force Statistics from the Current Population Survey, 
Table 7. Employment status of the civilian noninstitutional population 
25 years and over by educational attainment, sex, race, and Hispanic or 
Latino ethnicity,'' Bureau of Labor Statistics, revised January 18, 
2019, https://www.bls.gov/cps/cpsaat07.htm.
---------------------------------------------------------------------------
    These disparities are also found at the local level. Researchers 
have found significant racial disparities in labor force statistics 
within the same city, which may be explained in part by complex and 
deeply rooted factors such as industry concentration, investments in 
housing and infrastructure, and demographic trends. Chicago, San 
Francisco, Washington, and the borough of Manhattan all had relatively 
low black employment rates in 2015 (56, 53, 64, and 62 percent, 
respectively), and white employment rates that were at least 20 
percentage points higher (83, 84, 88, and 85 percent, 
respectively).\83\ It is unclear if the Department considered the 
consistently high unemployment rates among African American and Latino 
workers when proposing a minimum unemployment rate floor of seven 
percent, which would essentially be an unemployment rate that is close 
to 14 percent for African Americans.
---------------------------------------------------------------------------
    \83\ Martha Ross and Natalie Holmes, ``Employment by Race and 
Place: Snapshots of America,'' Brookings Institution, February 2017, 
pp. 1-16. https://www.brookings.edu/blog/the-avenue/2017/02/27/
employment-by-race-and-place-snapshots-of-america/.
---------------------------------------------------------------------------
    Employment outcomes for African Americans are also more affected by 
the business cycle than white Americans. One study found that over the 
period of 1990 through 2004, as the unemployment rate increased by one 
percentage point, men were 0.16 percentage points more likely to become 
unemployed, but this rate rose to 0.27 percentage points for African 
American men. Black men were also less likely to transition from 
unemployment to employment than white men, though the researchers found 
that this relationship didn't change significantly during the business 
cycle, the same study found. These results control for differences in 
education and other characteristics.\84\ Another study found that black 
and Latino workers are more likely to work part-time for economic 
reasons than white workers, even after controlling for other 
demographic and economic differences between the groups. This analysis 
found that this involuntary part-time work rose for all groups during 
the Great Recession, but recovered much more quickly for white men than 
for black men, with black men much less likely to transition from part-
time to full-time work in the years following the recession than white 
men.\85\
---------------------------------------------------------------------------
    \84\ Kenneth Couch and Robert Fairlie, ``Last Hired First Fired? 
Black-White Unemployment and the Business Cycle,'' Demography (February 
2010), pp. 227-247. https://www.ncbi.nlm.nih.gov/pmc/articles/
PMC3000014/.
    \85\ Tomaz Cajner, et al., ``Racial Gaps in Labor Market Outcomes 
in the Last Four Decades and over the Business Cycle,'' Federal Reserve 
Board, June 2017, pp. 1-33, https://papers.ssrn.com/sol3/
papers.cfm?abstract_id=2996084.
---------------------------------------------------------------------------
    Multiple deep-rooted factors contribute to these employment 
disparities. For example, decades of discriminatory housing policies 
have contributed to unequal access to quality education for black 
children, which may affect employment opportunities later in life.\86\ 
In addition to these complex causes, a large body of research also 
demonstrates that employer discrimination contributes to higher 
unemployment rates among African Americans, especially compared to 
white Americans.
---------------------------------------------------------------------------
    \86\ Richard Rothstein, ``The Racial Achievement Gap, Segregated 
Schools, and Segregated Neighborhoods--A Constitutional Insult,'' 
Economic Policy Institute, November 12, 2014, https://www.epi.org/
publication/the-racial-achievement-gap-segregated-schools-and-
segregated-neighborhoods-a-constitutional-insult/.
---------------------------------------------------------------------------
    Researchers have conducted dozens of field studies over the past 3 
decades in which they have compared outcomes for otherwise identical 
job applications that differ only by racial or ethnic markers (such as 
identical resumes with distinct names). One meta-analysis of such 
studies found that white applicants receive 36 percent more callbacks 
than African Americans with the same qualifications, and 24 percent 
more callbacks than Latinos. They found there was little change in the 
callback disparities between white and black Americans over the 25 
years studied, from 1990 to 2015, and a slight reduction in the 
disparities between Latino and white applicants, though barely 
statistically significant.\87\ We strongly urge FNS to review all of 
these studies, as they help explain why an unemployment rate is an 
especially poor predictor of job availability for African American 
workers, who may not be hired for available jobs due to discrimination. 
For example:
---------------------------------------------------------------------------
    \87\ Lincoln Quillian, et al., ``Meta-Analysis of Field Experiments 
Shows No Change in Racial Discrimination in Hiring Over Time,'' 
Proceedings of the National Academy of Sciences of the United States of 
America (April 2017), pp. 1-6, https://www.pnas.org/content/early/2017/
09/11/1706255114.

   Two field studies, in Milwaukee and New York City, found 
        consistently higher callbacks for white applicants compared to 
        African American applicants. Both studies had young men (ages 
        21 to 24) play the role of job applicants. They were matched 
        with applicants with similar appearance and verbal and social 
        skills, and presented with similar resumes demonstrating 
        similar levels of education and job experience, and they 
        received job interview training to be similarly prepared. In 
        both Milwaukee and New York, white applicants received 
        callbacks or job offers at roughly double the rate of African 
        American applicants.\88\
---------------------------------------------------------------------------
    \88\ Devah Pager and Bruce Western, ``Identifying Discrimination at 
Work: The Use of Field Experiments,'' Journal of Social Issues (2012) 
pp. 221-237, http://scholar.harvard.edu/files/pager/files/
identifying_discrimination_pager_western.pdf?m=1462807104.

   Another field study found that black applicants were about 
        \1/2\ as likely to receive a callback as white applicants. This 
        study also found that white applicants who were recently 
        released from prison had similar levels of callbacks as black 
        and Latino applicants: whites with criminal records obtained 
        positive responses in 17.2 percent of job applications, 
        compared to 15.4 percent for Latinos and 13.0 percent for 
        blacks.\89\
---------------------------------------------------------------------------
    \89\ Devah Pager, Bruce Western, and Bart Bonikowski, 
``Discrimination in a Low-Wage Labor Market: A Field Experiment,'' 
American Sociological Review (October 2009), pp. 777-799, http://
scholar.harvard.edu/files/bonikowski/files/pager-western-bonikowski-
discrimination-in-a-low-wage-labor-market.pdf.

   One field experiment found that when comparing outcomes of 
        identical resumes with names that were typically associated 
        with white or black identities, white applicants had a 50 
        percent higher chance of being called back.\90\
---------------------------------------------------------------------------
    \90\ Marianne Bertrand and Sendhil Mullainathan, ``Are Emily and 
Greg More Employable than Lakisha and Jamal? A Field Experiment on 
Labor Market Discrimination,'' National Bureau of Economic Research 
(July 2003), pp. 1-27 https://www.nber.org/papers/w9873.

    While they make up a small share of childless adults subject to the 
time limit, Native Americans are likely to be disproportionately 
affected by this proposed rule given the estimate that many Tribal 
reservations may lose waiver eligibility, as outlined in Chapter 1. 
Native Americans also traditionally have higher unemployment rates and 
worse labor force outcomes than white Americans, in part due to sparse 
job opportunities on or near Tribal and other rural areas and the 
legacy of historical factors contributing to lower educational 
attainment and other barriers to employment. (Figure 3.2.) \91\
---------------------------------------------------------------------------
    \91\ Ed Bolen and Stacy Dean, ``Waivers Add Key State Flexibility 
to SNAP's Three-Month Time Limit,'' Center on Budget and Policy 
Priorities, updated February 6, 2018, https://www.cbpp.org/research/
food-assistance/waivers-add-key-state-flexibility-to-snaps-three-month-
time-limit.
---------------------------------------------------------------------------
Figure 3.2
Native Americans Face Higher Unemployment
2006-2018 annual averages

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Note: ``Native American'' refers to individuals identifying 
        as American Indian or Alaska Native alone or in combination 
        with some other racial category. ``White'' refers to 
        individuals identifying as white alone. Both Native American 
        and white data includes Hispanics.
          Source: U.S. Bureau of Labor Statistics, Current Population 
        Survey.

    This evidence shows that black and Latino workers as well as Native 
American workers, have historically and consistently higher 
unemployment rates than white workers, and that these outcomes cannot 
be explained solely by differences in education or other 
characteristics. Significant numbers of individuals subject to the time 
limit therefore are in groups that experience unemployment rates that 
are significantly higher than that of their state or local area. An 
unemployment rate floor would therefore disallow states from requesting 
waivers from areas where black and Latino workers have few job 
opportunities, even if the general unemployment rate for their area is 
relatively low. If implemented, this proposal would therefore 
disproportionately harm black, Latino and Native American adults 
subject to the time limit. This disparate impact of this policy is 
therefore in conflict with 7 U.S.C.  2020(c)(2), which establishes 
that program administration of SNAP must be consistent with existing 
civil rights law.
Childless Adult SNAP Participants Are More Likely to Work in Jobs With 
        High Rates of Un- And Underemployment and Instability
    The individuals who are targeted by the time limit do work, but in 
occupations where workers experience instability, including 
underemployment, gaps in employment, and higher unemployment rates. The 
general unemployment rate for the area therefore does not reflect the 
unemployment rates for workers such as those in service occupations, 
who are more likely to be unemployed at any given time than other 
workers.
    SNAP participants who work generally work in service or sales 
occupations, such as cashiers, cooks, home health aides, janitors, or 
drivers.\92\ A recent study of SNAP E&T participants, which includes 
many childless adult SNAP participants ages 18-49, found that sales and 
service occupations, such as cashiers and food preparation workers, 
were among the most common reported by participants.\93\
---------------------------------------------------------------------------
    \92\ ``SNAP Helps 1 in 10 Workers in the United States Put Food on 
the Table,'' Center on Budget and Policy Priorities, revised November 
2018, https://www.cbpp.org/sites/default/files/atoms/files/
factsheets_11-27-18fa_us.pdf.
    \93\ Gretchen Rowe, Elizabeth Brown, and Brian Estes, ``SNAP 
Employment and Training (E&T) Characteristics Study: Final Report,'' 
United States Department of Agriculture, Food and Nutrition Services, 
revised October 2017, https://fns-prod.azureedge.net/sites/default/
files/ops/SNAPEandTCharacteristics.pdf.
---------------------------------------------------------------------------
    There are higher unemployment rates among workers in many of these 
occupations. People who report their occupation as a service occupation 
had unemployment rates about 23 percent higher than the general 
unemployment rate in 2018, with food preparation and serving workers 
reporting unemployment rates about 56 percent higher than the overall 
average.\94\ One analysis that looked at working-age workers who did 
not receive disability income and did not have young children in the 
household found that unemployment rates among those individuals were 
especially high for cashiers, housekeepers, and laborers in 2017.\95\
---------------------------------------------------------------------------
    \94\ ``Labor Force Statistics from the Current Population Survey, 
Table 25, Annual Averages, Unemployed persons by occupation and sex,'' 
Bureau of Labor Statistics, revised January 18, 2019, https://
www.bls.gov/cps/cpsaat25.htm.
    \95\ Kristin Butcher and Diane Whitmore Schanzenbach, ``Most 
Workers in Low-Wage Labor Market Work Substantial Hours, in Volatile 
Jobs,'' Center on Budget and Policy Priorities, revised July 24, 2018, 
https://www.cbpp.org/sites/default/files/atoms/files/7-24-18pov.pdf.
---------------------------------------------------------------------------
    Low-skill and low-wage workers are also more likely to be working 
part time for economic reasons, and to cycle in and out of the labor 
force. For example, one study found that the share of workers in low-
skill jobs (classified by the types of tasks required, which are manual 
and routine, as opposed to cognitive and non-routine) who were working 
part time involuntarily was about three times that of workers in the 
highest-skill occupations (11 percent versus 3 percent), and another 
found that involuntary part-time workers were concentrated in the 
retail trade and hospitality industries.\96\ Another study observed a 
broader trend of workers cycling in and out of the labor force.\97\
---------------------------------------------------------------------------
    \96\ Maria E. Canon, Marianna Kudlyak, Guannan Luo, and Marisa 
Reed, ``Flows To and From Working Part Time for Economic Reasons and 
the Labor Market Aggregates During and After the 2007-09 Recession,'' 
Economic Quarterly, Vol. 100, No. 2, Second Quarter 2014, Pp.87-111. 
https://www.richmondfed.org/-/media/richmondfedorg/publications/
research/economic_
quarterly/2014/q2/kudlyak.pdf; Lonnie Golden, ``Still falling short on 
hours and pay,'' Economic Policy Institute, December 5, 2016, https://
www.epi.org/publication/still-falling-short-on-hours-and-pay-part-time-
work-becoming-new-normal/#epi-toc-8.
    \97\ John Coglianese ``The Rise of In-and-Outs: Declining Labor 
Force Participation of Prime Age Men,'' Working Paper, February 28, 
2018, https://scholar.harvard.edu/coglianese/publications/rise-of-in-
and-outs.
---------------------------------------------------------------------------
    One of the reasons why these workers might have higher rates of un- 
and underemployment and non-labor force participation is higher 
turnover: because these occupations lack stability, workers are 
likelier to move in and out of jobs, and likelier to be unemployed or 
out of the labor force at a given time or take a part-time job when 
they would desire a full-time job. Low-paying jobs often have irregular 
schedules that change from week to week. Workers in low-wage jobs are 
sometimes given little notice of schedule changes or are expected to be 
on call, and are more likely to work part-time hours when they would 
prefer a full-time schedule.\98\ Low-wage jobs are also more likely to 
lack paid sick leave or other paid leave. For example, only 46 percent 
of workers in jobs with average hourly wages in the bottom 25 percent 
of the wage distribution had paid sick leave in 2016, compared to 91 
percent of workers in the highest-paid jobs (and 72 percent 
overall).\99\
---------------------------------------------------------------------------
    \98\ For more information, see Brynne Keith-Jennings and Vincent 
Palacios, ``SNAP Helps Millions of Workers,'' Center on Budget and 
Policy Priorities, May 10, 2017, https://www.cbpp.org/research/food-
assistance/snap-helps-millions-of-low-wage-workers.
    \99\ Bureau of Labor Statistics, Employee Benefits Survey, https://
www.bls.gov/ncs/ebs/benefits/2017/ownership/civilian/table32a.htm. This 
survey constructs hourly wage percentiles by looking at average 
reported hourly wages for workers by occupation. (Because these 
averages are for occupations, workers may fall into a different 
percentile than their occupation as a whole if they earn more or less 
than the average for their occupation.)
---------------------------------------------------------------------------
    Workers in jobs with lower wages, more volatility, and fewer 
benefits are more likely to experience turnover, research shows. For 
example, a study found that workers with access to paid sick leave or 
paid vacation were more likely to stay in their current job. This study 
found these effects even when controlling for other characteristics of 
workers, such as education level or income, or characteristics of jobs 
(such as the size of the firm and other benefits provided) that are 
associated with more job separations.\100\ Another study that examined 
data from a large chain of retailers found that workers who earned 
lower wages and had more schedule volatility (which was driven by 
changes in consumer demand, not by employee choice) were more likely to 
leave their jobs; this study found that these effects were not due to 
worker ability.\101\
---------------------------------------------------------------------------
    \100\ Heather Hill, ``Paid Sick Leave and Job Stability,'' Work and 
Occupations, Vol. 40, Issue 2, 2013, https://www.ncbi.nlm.nih.gov/pmc/
articles/PMC3825168/.
    \101\ Saravanan Kesavan and Camelia Kuhnen, ``Demand fluctuations, 
precarious incomes, and employee turnover,'' working paper, May 2017, 
http://public.kenan-flagler.unc.edu/faculty/kuhnenc/research/
kesavan_kuhnen.pdf.
---------------------------------------------------------------------------
    In addition to job conditions, the lack of key supports such as 
stable housing may also contribute to volatility or periods of 
joblessness among low-income workers. For example, recent research 
finds that low-income renters who experience a forced move (such as 
following an eviction) are more likely to be laid off from their jobs, 
compared to similar renters who did not experience a forced move.\102\
---------------------------------------------------------------------------
    \102\ Matthew Desmond and Carl Gershenson, ``Housing and Employment 
Insecurity Among the Working Poor,'' Social Problems, Vol. 63, Issue 1, 
February 2016, https://academic.oup.com/socpro/article-abstract/63/1/
46/1844105.
---------------------------------------------------------------------------
    At least in part because of these conditions, workers in low-wage 
jobs are more likely to be employed at jobs for shorter periods and to 
experience periods of non-work. Workers with lower levels of education 
(who are more likely to work in low-wage jobs) spend more weeks 
unemployed than those with more education and they experience less wage 
growth over the course of their lifetimes, according to studies by the 
Bureau of Labor Statistics that follow workers over time. These studies 
also find that younger workers with less education are more likely to 
have short-term jobs of 6 months or less, compared to workers of a 
similar age with more education.\103\ Workers in jobs that tend to have 
low wages, such as in the leisure and hospitality industry and service 
occupations, also tend to stay at jobs for shorter lengths than workers 
in other jobs.\104\ We strongly urge FNS to carefully review this 
research, which helps explain why unemployment rates may not capture 
the dynamic nature of work for low-wage workers, who comprise the vast 
majority of SNAP participants.
---------------------------------------------------------------------------
    \103\ Bureau of Labor Statistics, ``America's Young Adults at 29: 
Labor Market Activity, Education and Partner Status: Results from a 
Longitudinal Survey,'' April 8, 2016, [https://www.bls.gov/
news.release/archives/nlsyth_04082016.pdf]; Bureau of Labor Statistics, 
``Number of Jobs Held, Labor Market Activity, and Earnings Growth Among 
the Youngest Baby Boomers: Results from a Longitudinal Survey 
Summary,'' August 24, 2017, https://www.bls.gov/news.release/
nlsoy.nr0.htm.
    \104\ Bureau of Labor Statistics, ``Employee Tenure in 2016,'' 
https://www.bls.gov/news.release/pdf/tenure.pdf.
---------------------------------------------------------------------------
    Working SNAP participants often work in occupations and industries 
with low wages and more volatility. Compared to all workers, a greater 
share of workers who participate in SNAP are employed in service 
occupations and in industries such as retail and hospitality, where 
jobs are more likely to pay low wages and have other features of low 
quality.\105\ Childless adults are also likely to experience gaps in 
employment, despite being employed regularly. For example, one study 
that compared a snapshot, December 2013, with the 24 month period 
surrounding that month (January 2013 through December 2014), found that 
while \3/4\ of childless adult SNAP participants were in the labor 
force at some point during this period, only about \1/3\ consistently 
worked at least 20 hours per week throughout the entire period.\106\ 
Because childless adult SNAP participants work in jobs that contribute 
to periods of non-employment, unemployment rates in their area likely 
do not capture their labor trends. The Department did not say whether 
it considered the unemployment rate floor in the context of the types 
of jobs that childless adults are likely to work in. Its lack of 
transparency makes it difficult to assess the claim that individuals 
subject to the time limit have access to a sufficient number of jobs in 
an area with seven percent unemployment over a 2 year period.
---------------------------------------------------------------------------
    \105\ Brynne Keith-Jennings and Vincent Palacios, ``SNAP Helps 
Millions of Workers,'' Center on Budget and Policy Priorities, May 10, 
2017, https://www.cbpp.org/research/food-assistance/snap-helps-
millions-of-low-wage-workers.
    \106\ Lauren Bauer, ``Workers Could Lose SNAP Benefits Under 
Trump's Proposed Rule,'' The Hamilton Project, December 2018, http://
www.hamiltonproject.org/blog/workers_could_lose_
snap_benefits_under_trumps_proposed_rule.
---------------------------------------------------------------------------
Many Childless Adult SNAP Participants Have Health Conditions That 
        Limit Their Ability to Work
    While adults are exempt from the time limit if they are ``medically 
certified as physically or mentally unfit for employment,'' \107\ 
evidence suggests that many childless adult non-elderly SNAP 
participants have health conditions that serve as a barrier to 
employment. These adults may not fit the state's definition of ``unfit 
for work'' or may struggle to understand the rules or document their 
condition in order to obtain an exemption. Research shows that adults 
with disabilities and other health issues tend to have higher 
unemployment rates and fewer employment opportunities than individuals 
without such conditions.
---------------------------------------------------------------------------
    \107\ 7 CFR  273.24(c)(2).
---------------------------------------------------------------------------
    Various sources illustrate the health conditions that many 
childless adult SNAP participants have. Survey data indicate that among 
individuals aged 18 through 49 who do not receive disability benefits 
or identify as disabled, nor have children in their household, about 
\1/5\ report a health problem or disability that prevents them from 
working or limits the type of work they can do, report leaving the job 
or the labor force due to disability, or report not having worked in 
the last year due to disability.\108\ Research funded by FNS reports 
that state caseworkers found multiple barriers to employment among 
individuals subject to the time limit as they worked to implement 
welfare reform's time limits and work requirements. The most frequently 
cited barriers included medical or mental health issues, or substance 
use disorders.\109\ A detailed study of childless adults who were 
referred to a work experience program in Franklin County (Columbus), 
Ohio found that \1/3\ have a mental or physical limitation, including 
depression, post-traumatic stress disorder, mental or learning 
disabilities, or physical injuries.\110\ A more recent study of 
characteristics of employment and training participants, which includes 
many childless adults ages 18-49, found that about 30 percent of E&T 
participants identified health issues as a barrier to employment.\111\
---------------------------------------------------------------------------
    \108\ CBPP analysis of the March 2018 Current Population Survey.
    \109\ John L. Czajka, et al., ``Imposing a Time Limit on Food Stamp 
Receipt: Implementation of the Provisions and Effects on Food Stamp 
Program Participation, Volume I,'' U.S. Department of Agriculture, Food 
and Nutrition Service, September 2001, https://fns-prod.azureedge.net/
sites/default/files/abawd.pdf.
    \110\ ``A Comprehensive Assessment of Able-Bodied Adults Without 
Dependents and Their Participation in the Work Experience Program in 
Franklin County, Ohio,'' Ohio Association of Foodbanks, revised 2015, 
http://ohiofoodbanks.org/wep/WEP-2013-2015-report.pdf.
    \111\ Gretchen Rowe, Elizabeth Brown, and Brian Estes, ``SNAP 
Employment and Training (E&T) Characteristics Study: Final Report,'' 
United States Department of Agriculture, Food and Nutrition Services, 
revised October 2017, https://fns-prod.azureedge.net/sites/default/
files/ops/SNAPEandTCharacteristics.pdf.
---------------------------------------------------------------------------
    Research finds that people with disabilities tend to face greater 
barriers to employment and have worse labor force outcomes than 
individuals without disabilities. The Bureau of Labor Statistics, for 
example, finds that the unemployment rate among working-age individuals 
with a disability (ages 16 to 64), was ten percent in 2017, more than 
double the unemployment rate for working-age individuals without a 
disability, which was 4.2 percent.\112\ Earlier BLS research describes 
some of the barriers to employment that individuals with disabilities 
identified. About \1/2\ of the individuals who were out of the labor 
force or unemployed in May 2012 identified experiencing these 
challenges, with the most common being that their own disability 
limited their work ability, and with smaller shares identifying a lack 
of training, lack of transportation, or the need for 
accommodations.\113\ Other research has identified challenges 
jobseekers with disabilities face, such as difficulty finding 
appropriate jobs (which could include difficulty finding jobs that 
provide appropriate accommodations), lack of social networks to 
facilitate job connection, or lack of accessible transportation for 
job-seeking.\114\ Research has also documented that employers' 
attitudes may also harm jobseekers with disabilities, as employers may 
inadvertently underestimate the capacity of individuals with 
disabilities.\115\
---------------------------------------------------------------------------
    \112\ ``Economic News Release: Table 1, Employment status of the 
civilian noninstitutional population by disability status and selected 
characteristics, 2018 annual averages,'' Bureau of Labor Statistics, 
revised June 21, 2018, https://www.bls.gov/news.release/disabl.t01.htm.
    \113\ ``Persons With a Disability: Barriers to Employment, Types of 
Assistance, and Other Labor-Related Issues,'' Bureau of Labor 
Statistics, revised April 24, 2013, https://www.bls.gov/news.release/
pdf/dissup.pdf.
    \114\ Pamela Loprest and Elaine Maag, ``Barriers to and Supports 
for Work Among Adults with Disabilities: Results from the NHIS-D,'' The 
Urban Institute, January 2001, pp. 1-22, https://www.bls.gov/
news.release/pdf/dissup.pdf.
    \115\ Katrina Vornholt, et al., ``Disability and Employment--
Overview and Highlights,'' European Journal of Work and Organizational 
Psychology (2018), pp. 40-55, https://www.tandfonline.com/doi/pdf/
10.1080/1359432X.2017.1387536.
---------------------------------------------------------------------------
    Many childless adults report physical or mental health conditions 
that limit their ability to find a job and to work, and research shows 
that adults with disabilities face higher unemployment rates. While 
some of these individuals may be exempt from the time limit based on 
disability, others face difficulty documenting their health conditions, 
or have conditions that fall short of the ``unfit for work'' standard. 
Imposing an unemployment rate floor for 20 percent standard waivers 
would therefore cause many areas to lose waivers where individuals with 
health conditions are unable to find jobs. The Department did not 
address the disparate impact of the proposed rule on people with 
disabilities and other health conditions.
Childless Non-Elderly Adults Seeking Work May Face Geographic or 
        Transportation Limitations With Respect to Available Jobs
    As stated above, about \2/5\ each of childless adult SNAP 
participants without disabilities live in urban or suburban areas, and 
about 15 percent live in rural areas. Many rural areas have stalled 
economic development, which may result in relatively few job 
opportunities available. Even in urban or suburban areas where jobs may 
be more plentiful, many workers face transportation limitations to 
access those jobs. Some childless adults may face obstacles to finding 
work based on geographic factors.
    Individuals living in rural areas are less likely to be employed 
than other areas. Factors such as out-migration of younger workers and 
the aging of the remaining workforce, and declining infrastructure and 
investment have contributed to this trend. Beginning in the 1970s, the 
share of men with less than a high school education who are employed 
has declined more in rural areas than in urban areas. By 2016, only 
about 50 percent of these men with lower educational attainment were 
employed, about 15 percentage points lower than men without a high 
school diploma in urban areas.\116\ While both rural and urban counties 
saw steep employment losses that began recovering in 2010, the recovery 
stalled for rural counties, contributing to a significant employment 
gap between rural and urban counties, according to a 2014 USDA study. 
Counties with large share of African Americans saw greater impacts from 
the recession, which can only partially be explained by factors such as 
industry mix or educational composition of those counties, this study 
found.\117\
---------------------------------------------------------------------------
    \116\ James Ziliak, ``Restoring Economic Opportunity for `The 
People Left Behind': Employment Strategies for Rural America,'' Aspen 
Institute, revised 2019. https://www.aspeninstitute.org/longform/
expanding-economic-opportunity-for-more-americans/restoring-economic-
opportunity-for-the-people-left-behind-employment-strategies-for-rural-
america/.
    \117\ Tom Hertz, et al., ``Rural Employment Trends in Recession and 
Recovery,'' U.S. Department of Agriculture, Economic Research Service, 
revised August 2014, https://www.ers.usda.gov/webdocs/publications/
45258/48731_err172.pdf?v=0.
---------------------------------------------------------------------------
    In addition to individuals living in rural areas may facing 
significant challenges to finding work, many individuals in all types 
of areas do not have access to available jobs, either because their 
skills do not align with the job requirements, or because 
transportation options to those jobs are inadequate. In these areas, 
the area unemployment rate is an especially poor proxy for jobs 
available to those individuals, if such jobs are inaccessible. We 
strongly urge FNS to carefully review all of the below studies, which 
are key to understanding why an unemployment rate for an area does not 
accurately portray the number of jobs available to the individuals 
subject to the time limit.

   The number of jobs within a typical commuting distance in 
        major metro areas fell by seven percent between 2000 and 2012, 
        with steeper losses for Latino and African American residents, 
        which fell by 17 and 14 percent respectively, and for residents 
        with income below the poverty line, which fell by 17 percent, 
        compared to six percent for non-poor residents. The majority of 
        Census tracts with high poverty rates or a majority of 
        residents of color experienced losses in accessible jobs.\118\
---------------------------------------------------------------------------
    \118\ Elizabeth Kneebone and Natalie Holmes, ``The Growing Distance 
Between People and Jobs in Metropolitan America,'' Metropolitan Policy 
Program at Brookings, March 2015, pp. 1-24, http://kedc.com/wp-content/
uploads/2015/04/Brookings_JobCommuteDistance2015.pdf.

   In a number of metropolitan areas, low-income workers live 
        far from available jobs, one recent study found by comparing 
        the distance between the residence of low-wage jobseekers and 
        job postings based on data from an online marketplace for 
        hourly jobs. This study found, for example, that in 12 major 
        metropolitan areas, within at least nine percent of [ZIP C]odes 
        in each area, job postings far exceeded jobseekers in those 
        [ZIP C]odes.\119\ In some of these cities such as San 
        Francisco, jobs may be clustered in areas of the city where 
        housing costs are high, and low-wage job-seekers live farther 
        away and have limited transit options. For example, in Boston, 
        low-wage job postings far exceed the number of applicants in 41 
        percent of [ZIP C]odes, and in New York, San Francisco, 
        Chicago, Minneapolis, and Denver, available jobs far outnumber 
        job seekers in about \1/4\ or more of [ZIP C]odes. There are 
        also many pockets of metropolitan areas with the opposite 
        issue, where many job seekers are clustered, but available jobs 
        are far from where they live. This is the case for over \1/2\ 
        of the [ZIP C]odes in Atlanta and Miami, where job applicants 
        far outweigh open jobs (measured as [ZIP C]odes in the bottom 
        quintile of job seekers minus job postings within 6.3 miles of 
        the [ZIP C]ode's center). In cities such as Columbus, Detroit, 
        Austin, and Nashville, there are far more job applicants then 
        there are available jobs in over \1/4\ of [ZIP C]odes. In many 
        cities in the study such as Columbus, Nashville, Dallas, and 
        Washington, D.C., sizable shares of [ZIP C]odes have both 
        problems, demonstrating the mismatch between the distribution 
        of available jobs and workers.\120\
---------------------------------------------------------------------------
    \119\ The researchers identified [ZIP C]odes where job postings 
``far exceeded'' jobseekers as [ZIP C]odes that are in the top quintile 
of job seekers minus job postings within 6.3 miles of the [ZIP C]ode's 
population-weighted centroid, which is the average distance between job 
seekers and applicants in their dataset.
    \120\ Christina Stacy, Brady Meixell, and Serena Lei. ``Too Far 
from Jobs: Spatial Mismatch and Hourly Workers,'' Urban Institute, 
February 21, 2019, https://www.urban.org/features/too-far-jobs-spatial-
mismatch-and-hourly-workers.

   Increasing job accessibility, a measure of proximity to 
        employment opportunities relative to other nearby jobseekers, 
        significantly increases the chance of finding a job for African 
        Americans and Latinos, a 2006 study that looked at jobseekers 
        in three major metropolitan areas found. This shows how 
        disparities in access to jobs contribute to disparities in 
        labor market outcomes. The authors identified several factors 
        as contributing to this phenomenon, including that African 
        Americans were more likely to live in central cities farther 
        from suburban areas with more jobs and were less able to move 
        to a new neighborhood due to housing segregation and 
        residential discrimination; jobs were located in cheaper 
        suburban areas due to land use policy; and that there was a 
        lack of public transit options and lower car ownership rates 
---------------------------------------------------------------------------
        among African Americans.

      This study found that increasing accessibility by one standard 
        deviation above the mean value would increase the probability 
        of completing a job search within 6 months by 61 percent for 
        black non-college graduates, while not increasing this effect 
        for white workers with similar levels of education. Having 
        access to a car, searching in a job-rich area, being able to 
        accept a longer commute, having a higher-quality social 
        network, and having more education were all associated with an 
        increase in the probability of finding a job within 6 months, 
        while being black or having child care concerns were associated 
        with a decrease in this probability. The cumulative effects of 
        spatial job search variables such as job accessibility or car 
        ownership rates accounted for about 40 percent of the gap 
        between the time it takes black and white jobseekers to find a 
        job.\121\
---------------------------------------------------------------------------
    \121\ Rucker Johnson, ``Landing a job in urban space: The extent 
and effects of spatial mismatch,'' Regional Science and Urban Economics 
(February 2006), pp. 331-372, https://www.ssc.wisc.edu/gwallace/
Papers/Johnson%20(2006).pdf.

   Increased job accessibility reduced the length of time it 
        took recently laid-off workers in nine metropolitan areas to 
        find a job. This study looked specifically at jobseekers who 
        had been employed but laid off to ensure that these individuals 
        were searching for reasons unrelated to characteristics 
        associated with higher unemployment. It found that ``an 
        increase in one unit in job accessibility (from ^0.5 to 0.5) is 
        approximately equal to an increase from the 20th to the 80th 
        percentile of job accessibility. Such an increase is associated 
        with a 5.0 percent reduction in search duration for finding any 
        job, and a 6.6 and 8.3 percent reduction for accessions to a 
        new job with 75 and 90 percent of prior job earnings, 
        respectively.'' Black and Hispanic workers were more sensitive 
        to job accessibility than were white workers.\122\
---------------------------------------------------------------------------
    \122\ Fredrik Andersson, et al., ``Job Displacement and the 
Duration of Joblessness: The Role of Spatial Mismatch,'' National 
Bureau of Economic Research (April 2014), pp. 1-50. https://
www.nber.org/papers/w20066.pdf.

   An analysis of job accessibility in the Chicago metropolitan 
        area found that increasing job accessibility is linked with 
        lower unemployment. At the mean, an increase in job 
        accessibility of one standard deviation was associated with a 
        0.43-point reduction in the unemployment rate overall, a 0.57-
        point reduction in the African American unemployment rate, and 
        a 0.47-reduction in the unemployment rate for low-income 
        households.\123\
---------------------------------------------------------------------------
    \123\ Jangik Jin and Kurt Paulsen, ``Does Accessibility Matter? 
Understanding the Effect of Job Accessibility on Labour Market 
Outcomes,'' Urban Studies (2018), pp. 92-115. https://
journals.sagepub.com/doi/abs/10.1177/0042098016684099.

    Research shows how geographic factors can influence labor market 
outcomes such as employment for individuals in ways that are not 
readily apparent based on unemployment rates. An individual living in 
an area with a relatively low unemployment rate may not have access to 
jobs for which they are qualified due to transportation limitations. 
Some rural areas may have relatively low unemployment rates due in part 
to low labor force participation, and individuals living there may have 
relatively few job opportunities. We are concerned that when 
considering an unemployment rate floor for the purposes of time limit 
waivers, the Department did not appear to consider whether job 
accessibility may limit the potential for childless SNAP participants 
to obtain a job, even if the area in which they live has a relatively 
low unemployment rate.
Childless Adult SNAP Participants Report Other Barriers That Are 
        Associated With Higher Unemployment Rates
    The unemployment rate for the area does not reflect the 
availability of jobs for adult SNAP participants because these 
individuals face many disadvantages compared to the overall labor 
force. In addition to some of the characteristics already discussed 
that are associated with higher unemployment rates and other worse 
labor force outcomes, many individuals face barriers to work that may 
make it more difficult to find available jobs, complete a job search, 
be selected by employers, or maintain a job once employed. Here again, 
FNS' proposed rule seemed to either ignore or dismiss without 
explanation the considerable research that finds that unemployment 
rates do not reflect job availability for the individuals subject to 
the time limit. We encourage FNS to carefully review these research, 
which demonstrates how barriers to employment limit job availability 
for people subject to the time limit even if the unemployment rate is 
low.

   Housing instability and homelessness. Several studies have 
        reported that some childless adult SNAP participants lack 
        access to stable housing and some experience homelessness. A 
        USDA research report looking at individuals first subject to 
        the time limit found that homelessness was among the barriers 
        that case managers reported.\124\ A GAO study that looked at 
        employment and training programs for childless adults also 
        found that some case managers reported housing difficulties as 
        a barrier to work; for example, Colorado officials estimated 
        that about 40 percent of their employment and training 
        participants experienced homelessness.\125\ Similarly, a more 
        recent USDA study of employment and training providers found 
        that over \2/5\ of these providers identified a lack of stable 
        housing as a barrier for at least a quarter of participants in 
        these programs, which include many adults targeted by the time 
        limit.\126\
---------------------------------------------------------------------------
    \124\ John L. Czajka, et al., ``Imposing a Time Limit on Food Stamp 
Receipt: Implementation of the Provisions and Effects on Food Stamp 
Program Participation, Volume I,'' U.S. Department of Agriculture, Food 
and Nutrition Service, September 2001. https://fns-prod.azureedge.net/
sites/default/files/abawd.pdf.
    \125\ ``Food Stamp Employment and Training Program,'' United States 
General Accounting Office, revised March 2003, https://www.gao.gov/
assets/240/237571.pdf.
    \126\ Gretchen Rowe, Elizabeth Brown, and Brian Estes, ``SNAP 
Employment and Training (E&T) Characteristics Study: Final Report,'' 
United States Department of Agriculture, Food and Nutrition Services, 
revised October 2017, https://fns-prod.azureedge.net/sites/default/
files/ops/SNAPEandTCharacteristics.pdf.

      Barriers to work among individuals experiencing homelessness are 
        well-documented, including limited skills and inconsistent work 
        histories, lack of transportation, or physical or mental health 
        conditions.\127\ Those who are homeless may lack consistent 
        access to resources needed to maintain personal hygiene and 
        meet dress codes, and the sleep deprivation and stress of 
        lacking housing may also affect these workers. People 
        experiencing homelessness also lack access to a reliable 
        mailing address and may not have consistent access to a phone 
        or computer for job application and communication needs.\128\ 
        Individuals who experience evictions are also more likely to be 
        laid off, research finds.\129\
---------------------------------------------------------------------------
    \127\ David Long, John Rio, and Jeremy Rosen. ``Employment and 
Income Supports for Homeless People,'' 2007 National Symposium on 
Homelessness Research, https://aspe.hhs.gov/system/files/pdf/180356/
report.pdf.
    \128\ ``Taking Away Medicaid for Not Meeting Work Requirements 
Harms People Experiencing Homelessness,'' Center on Budget and Policy 
Priorities, revised December 14, 2018, https://www.cbpp.org/sites/
default/files/atoms/files/4-18-18health.pdf.
    \129\ Matthew Desmond and Carl Gershenson, ``Housing and Employment 
Insecurity among the Working Poor,'' Social Problems (January 2016), 
pp. 1-22. https://scholar.harvard.edu/files/mdesmond/files/
desmondgershenson.sp2016.pdf?m=1452638824.

   Criminal records. Some childless adult SNAP participants may 
        face additional barriers to work due to having a criminal 
        record. For example, a study of childless adults referred to a 
        work experience program in Franklin County, Ohio found that 
        about \1/3\ reported having a criminal record.\130\ Over \1/2\ 
        of SNAP E&T providers reported that a significant share of 
        participants reported a criminal record as a barrier to 
        work.\131\ An in-depth interview study of SNAP participants who 
        experienced periods of time with no other income, approximately 
        \1/2\ of whom were ages 18 through 49 and who did not have 
        dependent children, found that about \1/5\ of study 
        participants had a criminal record that served as a barrier to 
        employment.\132\ While many of these studies have relatively 
        small study populations and some are focused on populations who 
        are more likely to have criminal justice records and are 
        therefore not always representative samples, it is clear that 
        there are individuals potentially subject to the time limit who 
        have experience with the criminal justice system.
---------------------------------------------------------------------------
    \130\ ``A Comprehensive Assessment of Able-Bodied Adults Without 
Dependents and Their Participation in the Work Experience Program in 
Franklin County, Ohio,'' Ohio Association of Foodbanks, 2014, http://
admin.ohiofoodbanks.org/uploads/news/WEP-2013-2014-report.pdf.
    \131\ Gretchen Rowe, Elizabeth Brown, and Brian Estes, ``SNAP 
Employment and Training (E&T) Characteristics Study: Final Report,'' 
United States Department of Agriculture, Food and Nutrition Services, 
revised October 2017, https://fns-prod.azureedge.net/sites/default/
files/ops/SNAPEandTCharacteristics.pdf.
    \132\ Claire Wilson and Brian Estes, ``Examining the Growth of the 
Zero-Income SNAP Caseload: Characteristics, Circumstances, and Dynamics 
of Zero-Income SNAP Participants,'' U.S. Department of Agriculture, 
revised October 2014. https://fns-prod.azureedge.net/sites/default/
files/ops/ZeroIncome-Vol2.pdf.

      Formerly incarcerated individuals face steep barriers to work, 
        and as a result, on average face periods of unemployment 
        following release. Longitudinal studies that have tracked 
        prisoners upon their release, for example, find that up to \1/
        2\ remain without a job 12 months after release.\133\ A recent 
        CBPP paper summarizes some of this research: \134\
---------------------------------------------------------------------------
    \133\ National Research Council, The Growth of Incarceration in the 
United States: Exploring Causes and Consequences, (Washington, D.C.: 
The National Academies Press, 2014), pp. 233-259. https://www.nap.edu/
catalog/18613/the-growth-of-incarceration-in-the-united-states-
exploring-causes.
    \134\ Elizabeth Wolkomir, ``How SNAP Can Better Serve the Formerly 
Incarcerated,'' Center on Budget and Policy Priorities, revised March 
16, 2018. https://www.cbpp.org/sites/default/files/atoms/files/3-6-
18fa.pdf.

     Studies document that the majority of individuals 
            returning from incarceration face health conditions. For 
            example, one study found that \1/2\ of men and \2/3\ of 
            women had been diagnosed with chronic physical ailments 
            such as asthma, diabetes, hepatitis, or HIV/AIDS.\135\ 
            People leaving jail and prison are three to six times 
            likelier than others to suffer from mental illness, another 
            study found.\136\
---------------------------------------------------------------------------
    \135\ Kamala Mallik-Kane and Christy A. Visher, ``Health and 
Prisoner Reentry: How Physical, Mental, and Substance Abuse Conditions 
Shape the Process of Reintegration,'' Urban Institute Justice Policy 
Center, revised February 2008. https://www.urban.org/sites/default/
files/publication/31491/411617-Health-and-Prisoner-Reentry.PDF.
    \136\ Henry J. Steadman, et al., ``Prevalence of Serious Mental 
Illness Among Jail Inmates,'' Psychiatric Services (June 2009), pp. 1-
5. https://ps.psychiatryonline.org/doi/pdf/10.1176/ps.2009.60.6.761.

     Formerly incarcerated individuals also tend to lack 
            education and training sought by employers. They have an 
            average of fewer than 12 years of education and in some 
            cases limited cognitive capacity, a history of behavioral 
            problems, or a low level of functional literacy.\137\ 
            Furthermore, they miss out on opportunities to gain work 
            experience while in prison, and often do not have access to 
            training programs.\138\
---------------------------------------------------------------------------
    \137\ National Research Council, The Growth of Incarceration in the 
United States: Exploring Causes and Consequences, (Washington, D.C.: 
The National Academies Press, 2014), pp. 233-259. https://www.nap.edu/
catalog/18613/the-growth-of-incarceration-in-the-united-states-
exploring-causes.
    \138\ Bruce Western, ``Collateral Costs: Incarceration's Effect on 
Economic Mobility,'' The Pew Charitable Trusts, revised 2010. https://
www.pewtrusts.org//media/legacy/uploadedfiles/pcs_assets/2010/
collateralcosts1pdf.pdf.

     Evidence also suggests that employers are more averse 
            to hiring those with criminal convictions than any other 
            disadvantaged group, and formerly incarcerated individuals 
            can also face occupational licensing and other 
            restrictions.\139\
---------------------------------------------------------------------------
    \139\ Elizabeth Wolkomir, ``How SNAP Can Better Serve the Formerly 
Incarcerated,'' Center on Budget and Policy Priorities, revised March 
16, 2018. https://www.cbpp.org/sites/default/files/atoms/files/3-6-
18fa.pdf.
---------------------------------------------------------------------------
Unemployment Rates Significantly Overstates Jobs Available to Childless 
        Adult SNAP Participants, Evidence Suggests
    The preamble to the proposed rule suggests that adding an 
unemployment rate floor to qualify for a waiver is necessary because 
``the Department believes a floor should be set for the 20 percent 
standard so that areas do not qualify for waivers when their 
unemployment rates are generally considered to be normal or low.'' 
\140\ The Department proposes this unemployment floor to interpret the 
statute which provides that states can waive areas that lack a 
``sufficient number of jobs to provide employment for the 
individuals,'' therefore suggesting that areas with unemployment rates 
below the proposed threshold do have enough jobs to provide employment 
for the individuals who are subject to the time limit.
---------------------------------------------------------------------------
    \140\ NPRM, p. 984.
---------------------------------------------------------------------------
    A significant body of research, provided above, demonstrates why 
FNS' reasoning is flawed and lacks transparency. The area unemployment 
rate is a poor proxy for employment opportunities available to adult 
SNAP participants without dependent children. These individuals on 
average are more likely than other workers to have limited education 
and skills, experience discrimination, lack geographic access to jobs, 
face housing instability, and experience other barriers to employment. 
Many of these individuals likely experience multiple barriers that 
affect their ability to find a job. For example, an African American 
worker with less than a high school education living far from available 
jobs with no reliable transportation options will likely have access to 
far fewer jobs than their area unemployment rate suggests. A rural area 
may have a low unemployment rate in part because of low labor force 
participation, where many have given up searching for work due to few 
job opportunities. Not only did FNS not provide any evidence on whether 
it considered research showing how the unemployment rate overstates 
jobs available based on different demographic characteristics, it also 
did not provide research that shows how these economic conditions may 
interact with each other and affect the opportunities to find work for 
the very disadvantaged population that is subject to the time limit.
    Because these adults are in many groups that have significantly 
higher unemployment rates than the overall unemployment rate for their 
area, the proposed unemployment rate floor would likely disqualify many 
areas where these individuals have few opportunities. The unemployment 
rate among the group of individuals subject to the time limit in a 
county with a seven percent unemployment rate is likely much higher 
than seven percent. Under the proposed regulation, states would be much 
less effective at identifying areas where there are not enough jobs for 
the individuals subject to the time limit, as many of these areas would 
have overall unemployment rates below the proposed threshold. The 
Department did not discuss how the unemployment rate relates to job 
availability for individuals subject to the time limit. It appears FNS 
ignored the considerable research that shows how the individuals 
subject to the time limit belong to demographic groups with much higher 
unemployment rates than the average or face barriers to accessing jobs. 
Without this research, it is difficult to understand how it came to the 
conclusion that a seven percent 2 year unemployment rate (or a 1 year 
unemployment rate of ten percent) accurately captures the number of 
jobs available to this population.
C. Citation of Labor Surplus Area Unemployment Floor Inappropriate for 
        this Population
    The Department cites the fact that the Department of Labor (DOL) 
has an unemployment floor in its classification of Labor Surplus Areas 
(LSAs) as support for its proposal to impose a similar floor for the 
purposes of waiver criteria, implying its approach is consistent with 
DOL's. The Department, however, proposes a higher floor than DOL uses 
without providing evidence, when research suggests if anything the 
floor for this population would be substantially lower than DOL's. The 
Department also fails to acknowledge that DOL uses a ten percent 
ceiling in its identification of LSAs, demonstrating that its citation 
of LSAs is either misleading or based on incomplete information.
    In the preamble for the rule, the Department explained how the DOL 
has an unemployment rate floor for Labor Surplus Areas, implying that 
implementing an unemployment rate floor for an area to qualify for a 
waiver based on having unemployment rates 20 percent above the national 
average would be appropriate to be consistent with DOL's approach.\141\ 
Labor Surplus Areas are areas that DOL identifies as having a ``surplus 
of labor'' based on having an unemployment rate of 20 percent higher 
than the national average for a designated 24 month period. Federal, 
state, and local agencies use LSA designations for multiple purposes. 
Executive Order 12073 required executive agencies to ``emphasize 
procurement set asides in order to strengthen our nation's economy.'' 
\142\ DOL lists several other agencies that use Labor Surplus Areas, 
such as ``The Small Business Administration uses the LSA list for bid 
selections for small business awards in Historically Underutilized 
Business Zones (HUBZones).'' \143\
---------------------------------------------------------------------------
    \141\ NPRM, p. 983.
    \142\ Executive Order 12073, Federal procurement in labor surplus 
areas, 43 FR 36873, 3 CFR, 1978 Comp., p. 216, https://
www.archives.gov/federal-register/codification/executive-order/
12073.html.
    \143\ U.S. Department of Labor, ``Labor Surplus Areas: Frequently 
Asked Questions,'' https://www.doleta.gov/programs/lsa_faq.cfm.
---------------------------------------------------------------------------
    DOL has an unemployment rate floor so that the minimum unemployment 
rate used for identification of LSAs is at least six percent. DOL also 
has an unemployment rate ceiling; when national unemployment is high 
enough that 20 percent above the national average exceeds ten percent 
unemployment over 2 years, DOL will designate LSAs that have 
unemployment rates above ten percent.\144\ DOL also allows states to 
demonstrate that areas meet alternative criteria to demonstrate an 
exceptional circumstance, such as a recent 3 month unemployment rate at 
the threshold required for LSA designation.
---------------------------------------------------------------------------
    \144\ Ibid.
---------------------------------------------------------------------------
    The Department established identification of Labor Surplus Areas as 
one of a non-exhaustive list of methods of demonstrating ``insufficient 
jobs'' in its original 1996 guidance, and it was codified as an example 
in the final 2001 regulation.\145\ The guidance and final rule also 
allowed states to use similar data demonstrating unemployment rates 
that are 20 percent above the national average for a 24 month period.
---------------------------------------------------------------------------
    \145\ USDA, ``Guidance for States Seeking Waivers for Food Stamp 
Limits,'' December 3, 1996 and Federal Register, Vol. 66, No. 11, 
January 17, 2001. P. 4462. https://www.federalregister.gov/d/01-1025/p-
205.
---------------------------------------------------------------------------
    FNS decided, in its 1996 guidance and 2001 regulation, to include 
LSA designation as one of many ways that states can demonstrate that an 
area lacks sufficient jobs. This decision likely reflects the fact that 
unemployment rates are readily available on a monthly basis and are 
statistically reliable for sub-state areas. There are few alternative 
measures available at the state and local level that states can use to 
demonstrate a lack of jobs. For example, at the national and state 
level there are measures such as ``alternative measures of labor 
underutilization,'' a broader set of metrics that include ``discouraged 
workers,'' who are workers who want a job but have not recently 
searched for work because they believe no jobs are available to them, 
as well as others such as workers who would like to work full time but 
can only find a part-time job.\146\ Those metrics cannot be reliably 
calculated at the sub-state level. It is reasonable to use metrics 
developed by DOL for the express purposes of identifying areas with 
excess labor compared to jobs, given that these metrics can facilitate 
the process for states. It is also reasonable, however, to adapt these 
criteria to more accurately capture the intent of the law, which is to 
capture jobs available for a sub-population. Current regulations make 
such an adaptation by adopting the ``20 percent standard'' without a 
floor.
---------------------------------------------------------------------------
    \146\ Bureau of Labor Statistics, ``Alternative Measures of Labor 
Underutilization for States, 2018 Annual Averages.'' https://
www.bls.gov/lau/stalt.htm.
---------------------------------------------------------------------------
    In proposing to use a similar unemployment floor to that used by 
DOL in the LSA designation, the Department does not consider that while 
a specific floor may be appropriate for DOL's purposes in establishing 
LSAs, the waiver criteria are meant to represent a fully distinct 
concept from LSAs and therefore adapting them to the different purpose 
of waiver criteria is necessary and appropriate. The Department 
correctly notes that waiver criteria do not currently require an area 
to meet a specific threshold if the area's unemployment rate is above 
the national average, unlike the LSA criteria.
    These measures are meant to serve different purposes, however:

   The suggested criteria for waivers of the time limit are 
        meant to establish a lack of ``sufficient jobs'' for a specific 
        sub-population, childless adult SNAP participants, which faces 
        labor market disadvantages. As noted in section B above, 
        childless adult SNAP participants belong to groups that have 
        higher unemployment rates than their local area, which makes 
        defining a specific unemployment rate at which those 
        individuals have access to enough jobs difficult, if not 
        impossible. For example, since this group may experience 
        unemployment rates that are double the rates of their area, a 
        six percent or seven percent unemployment rate ``floor'' would 
        mean a 12 or 14 percent unemployment rate for this group of 
        individuals. At any unemployment rate threshold, it is likely 
        that large groups of childless adult SNAP participants would 
        not have access to available jobs given their serious barriers 
        to work. Given the uncertainty and difficulty in establishing 
        whether there are jobs available for this population, not 
        requiring a specific unemployment rate threshold appropriately 
        allows for greater flexibility in determining areas with 
        insufficient jobs.

   LSAs, on the other hand, establish the economic condition of 
        an area to enable prioritization of procurement contracts and 
        economic development purposes. In creating the threshold for 
        LSAs, DOL does not need to consider whether jobs are available 
        for specific types of individuals, and instead is focused on 
        understanding the macroeconomic conditions of an area in order 
        to direct economic stimulus. It therefore may be reasonable for 
        the Department of Labor to establish specific thresholds that 
        meet those criteria for those purposes, given that the criteria 
        attempt to establish levels at which economic stimulus is 
        needed.

    In citing Labor Surplus Areas as a reason to implement an 
unemployment rate floor, the Department also fails to acknowledge the 
unemployment rate ceiling that DOL uses in LSA designation. Not only 
does DOL have an unemployment rate floor of six percent unemployment in 
its criteria for LSA designation, it also has an unemployment rate 
ceiling of ten percent unemployment. This means that when designating 
LSAs, the unemployment rate threshold DOL uses is 20 percent above the 
national average but not more than ten percent. The Department not only 
does not acknowledge this fact, it suggests this unemployment rate as a 
possible floor for the time limit:

          The Department would also like to receive comments on 
        establishing a floor of ten percent for the 20 percent 
        standard. A ten percent floor would allow for even fewer 
        waivers than the other options and would result in the work 
        requirements being applied in almost all areas of the 
        country.\147\
---------------------------------------------------------------------------
    \147\ NPRM, p. 984.

    DOL establishes this unemployment rate ceiling for LSA designation 
in recognition that a sustained level of unemployment at ten percent is 
adequately high to demonstrate that an area has severe labor market 
weakness. These criteria ensure that during times of widespread 
elevated unemployment, areas can qualify without having exceptionally 
high unemployment rates. For example, the LSA list in Fiscal Year 2012 
was based on unemployment rates between January 2009 and December 2010, 
during the height of the Great Recession when national unemployment was 
9.5 percent, and 20 percent above that would have been 11.4 
percent.\148\ Areas were also eligible in Fiscal Year 2013 and Fiscal 
Year 2014 with ten percent unemployment for the same reason.\149\ 
Without a ceiling, many areas that were struggling during the height of 
the recession and recovery would have been ineligible for LSA 
designation.
---------------------------------------------------------------------------
    \148\ Department of Labor, ``Labor Surplus Area Classification 
under Executive Orders 12073 and 10582 2012,'' https://www.doleta.gov/
lsa/Archived/2011-2012/Federal_Register_
2012_Final.pdf.
    \149\ Department of Labor, ``Labor Surplus Area Classification 
under Executive Orders 12073 and 10582 2013'' https://www.doleta.gov/
lsa/Archived/2012-2013/Federal_Register_
2013_Final.pdf; Department of Labor, ``Labor Surplus Area 
Classification under Executive Orders 12073 and 10582 2014,'' https://
www.doleta.gov/lsa/Archived/2013-2014/2013-
2014_LSA_Federal_Register_Notice.pdf.
---------------------------------------------------------------------------
    FNS therefore is proposing to pick and choose which features of LSA 
designation to adapt to the proposed regulation without discussion of 
why it made this choice, or even acknowledgement of this choice. The 
Department proposes on one hand to implement an unemployment rate floor 
for an area to qualify under the ``20 percent standard,'' but does not 
propose a similar unemployment rate ceiling. In fact, the Department 
proposes what DOL recognizes as a sufficiently high unemployment rate 
to qualify for a ceiling, ten percent unemployment, as a possible 
unemployment rate floor. The Department therefore proposes to ensure 
that unemployment rates in an area must meet a standard to demonstrate 
they are high but is not proposing a means of limiting this threshold 
during a recession to ensure that the unemployment rate threshold does 
not provide too high a bar that it would substantially bar areas 
suffering from a recession. The Department does not explain why it 
chose to adopt an unemployment rate floor similar to that used in LSA 
criteria but not an unemployment rate ceiling, and it does not mention 
the LSA ceiling. This oversight is particularly perplexing given our 
research review which indicates if anything, the LSA criteria as is are 
very stringent for waiver criteria given the barriers to employment 
childless adult SNAP participants face, which would recommend no floor 
or a very low floor if any at all.
    Without any explanation, it appears that by imposing a floor and 
not a ceiling for the 20 percent standard, the Department considers for 
the purposes of measuring whether an area lacks adequate jobs for 
childless adult SNAP participants that there is a level of unemployment 
that is low enough to ensure that adequate jobs are available, but not 
a level high enough to signify that there likely are not enough jobs. 
Again, without this information, it is difficult to assess the evidence 
that the Department used in proposing an unemployment rate floor, 
especially given that the Department's choice seems to contradict all 
available economic evidence indicating that unemployment rates are a 
poor proxy for jobs for this population.
D. Citation of ``Natural Rate of Unemployment'' Incorrectly Assumes 
        This Is a Fixed and Accurately Measurable Concept
    The Department uses the macroeconomic concept of the ``natural rate 
of unemployment'' to justify its proposed unemployment rate threshold. 
This use of this concept inappropriately applies a macroeconomic 
concept and inaccurately displays economic consensus. The preamble 
states:

          The Department believes a floor should be set for the 20 
        percent standard so that areas do not qualify for waivers when 
        their unemployment rates are generally considered to be normal 
        or low. The ``natural rate of unemployment'' is the rate of 
        unemployment expected given normal churn in the labor market, 
        with unemployment rates lower than the natural rate tending to 
        result in inflationary pressure on prices. Thus, unemployment 
        rates near or below the ``natural rate of unemployment'' are 
        more indicative of the normal delay in unemployed workers 
        filling the best existing job opening for them than a ``lack of 
        sufficient jobs'' in an area. Generally, the ``natural rate of 
        unemployment'' hovers around five percent. The Department 
        believes that only areas with unemployment rates above the 
        ``natural rate of unemployment'' should be considered for 
        waivers.\150\
---------------------------------------------------------------------------
    \150\ NPRM, p. 984.

    The Department appears to be suggesting that the natural rate of 
unemployment is a specific unemployment rate figure that can be used in 
setting a waiver floor to examine job opportunities for childless adult 
SNAP participants. This reasoning is deeply flawed.
    First, the so-called natural rate of unemployment is not a known or 
even an observable jobless rate. It is a concept that derives from the 
theoretical construct that there exists an unemployment rate that is 
consistent with stable inflation. If unemployment falls below this 
``natural rate,'' inflation would rise, and vice versa. More 
colloquially, too low an unemployment rate, where ``too low'' means the 
actual rate is below the natural rate, and the economy will overheat; 
too high a jobless rate relative to the natural rate and inflation will 
fall.
    In theory, an estimate of the natural rate should be derivable from 
observing the (negative) correlation between changes in the rate of 
unemployment and that of inflation. However, because this correlation 
appears to have moved toward zero over time, our ability to reliably 
identify a policy-relevant natural rate, meaning one that could 
fruitfully be referenced as the Department suggests in terms of their 
proposal, is much diminished.
    Note, for example, a recent article about this problem by economics 
journalist Neil Irwin. In the article, former Fed Vice-Chairman Alan 
Blinder notes that the ``confidence interval''--the band of statistical 
uncertainty around the estimate--is such that the concept cannot be 
usefully employed as a policy benchmark: ``If your range is 2.5 to 7, 
that doesn't tell you anything.'' \151\
---------------------------------------------------------------------------
    \151\ Neil Irwin, ``How Low Can Unemployment Really Go? Economists 
Have No Idea,'' New York Times, Feb. 28. 2018, https://www.nytimes.com/
2018/02/28/upshot/how-low-can-unemployment-really-go-economists-have-
no-idea.html.
---------------------------------------------------------------------------
    Figure 3.3 below reveals the problem using a standard statistical 
procedure to measure the inflation/unemployment correlation. The figure 
represents the coefficient from a regression of core inflation on 
lagged inflation and the gap between the unemployment rate and the 
Congressional Budget Office's estimate of the natural rate. The 
estimates are made using ``rolling regressions,'' meaning we estimate 
the model over 20 year periods, beginning with 1959-79, and advance the 
sample 1 year at a time. We then plot the coefficient on the 
unemployment gap variable.
    In this area of economics, the measure is considered to be the 
slope of the Phillips Curve, which is the curve that plots the 
unemployment/inflation tradeoff. The two lines surrounding the estimate 
represent the bounds of a 95 percent confidence interval around the 
estimate. When these lines include zero, as they do for most of the 
figure, the estimate of the slope is insignificant. In other words, in 
these years, the ``natural rate'' cannot be reliably distinguished from 
a range of rates that includes zero.
Figure 3.3
The ``Natural Rate'' of Unemployment Is Not Identifiable

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: CBPP Analysis of Congressional Budget Office, Bureau 
        of Labor Statistics, and Bureau of Economic Analysis Data.

    In other words, at least by this conventional approach, the 
Department cannot reliably use five percent (or any other level) as an 
estimate of the natural rate, because the diminished correlation 
between unemployment and inflation renders such an estimate 
statistically insignificant.
Department Proposes Arbitrary Unemployment Rate Floor
    Not only does the Department improperly support its proposed 
unemployment rate floor with a flawed discussion of the ``natural rate 
of unemployment,'' it proposes a floor that is arbitrarily and 
significantly higher than what it states is the unemployment rate 
consistent with the natural rate of unemployment. The Department 
proposes an unemployment rate floor that is 40 percent above what it 
states is ``generally'' considered the natural rate of unemployment, of 
five percent. This difference is not insignificant. With a labor force 
of about 160 million people, a difference of two percentage points is 
equivalent to more than three million people nationwide, employed or 
not. In a recent 24 month period of January 2017 through December 2018, 
795 counties (or equivalent entities) had unemployment rates above five 
percent, while only 155 counties had 24 month unemployment rates above 
seven percent.
    Considering the difference between those two metrics, it is not 
clear how the Department used the concept of the natural rate in 
developing this floor. These two unemployment rates, five and seven 
percent, are so different that it is difficult to understand how they 
are linked without more information. Furthermore, if the Department 
were actually basing the proposed unemployment rate floor on a concept 
that describes the level of unemployment at which inflation increases, 
it would need to demonstrate how this concept relates to the specific 
population or the purpose of establishing waiver criteria, which is to 
interpret the ``insufficient jobs'' criterion in the law targeted 
towards a disadvantaged group of individuals.
    The Department's proposal for this seven percent unemployment floor 
therefore suggests that it did not in fact use the natural rate of 
unemployment to develop the seven percent unemployment rate floor 
proposal. Either the Department used economic data relating the goals 
of the unemployment rate floor to the natural rate, in which case it 
lacked transparency by not providing this research, or the rate is an 
arbitrary selection unrelated to the statute that the rule is 
interpreting, in which case the discussion of the natural rate is 
irrelevant to the actual proposal. Without an explanation of how and 
why the Department used the natural rate concept to come up with a 
seven percent unemployment rate floor that is related to the purposed 
of the underlying statute, it is impossible to meaningfully comment.
E. Evidence Suggests That a Seven Percent Unemployment Floor Is 
        Inappropriately High for This Population
    Not only does the Department not provide economic evidence to 
support its proposed seven percent floor, evidence shows why this floor 
would be inappropriate for this population. While this comment argues 
that we cannot assume that any particular unemployment floor will 
provide the necessary labor market opportunities to some groups of 
workers, the proposed floor of seven percent is surely too high. As 
Figure 3.4 below shows, using national BLS data, there were 106 months 
since 1972 when the overall unemployment was between 6.5 and 7.5 
percent. The average rate was 7.1 percent, about the level of the 
proposed floor. But unemployment for African American and Latino 
workers was a much higher 13.9 percent and 10.2 percent. White 
unemployment was 6.2 percent.\152\
---------------------------------------------------------------------------
    \152\ The results are very similar: blacks, 13.5 percent; 
Hispanics, ten percent--when looking at minority unemployment rates 
conditional on the 2 year average of overall unemployment centered on 
seven percent to more closely simulate the proposed rule.
---------------------------------------------------------------------------
Figure 3.4
A 7 Percent Unemployment Floor Is Substantially Higher for Black and 
        Latino Workers
Unemployment rates by race/ethnicity when national unemployment is 
        between 6.5 and 7.5 percent
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Bureau of Labor Statistics, 1972-2018.

    Consider that at the depth of the Great Recession (2009-10), 
broadly recognized as the deepest recession since the Depression, the 
overall unemployment rate hit ten percent. This was widely, and 
correctly, seen as evidence of a huge, negative demand shock, one 
requiring an aggressive response from both fiscal and monetary 
authorities. And yet, the proposal suggests an unemployment floor that 
corresponds historically to a black unemployment rate well above the 
overall rate at the worst of the recession.
    Other rates that have been floated suffer from the same problem 
that even in the best of overall labor markets, certain groups face a 
much less welcoming set of job opportunities. Table 3.1 below shows the 
results from a simple regression of total rates on a constant and rates 
for black and Latino workers. Even at five percent unemployment, the 
African American jobless rate is predicted to be almost ten percent, 
and the Latino rate is at seven percent. At rates above seven percent, 
both black and Latino workers have jobless rates close to and in double 
digits.

                                Table 3.1
     Unemployment Rates and Predicted Black and Latino Unemployment
------------------------------------------------------------------------
                                   Predicted Unemployment Rates
      Unemployment       -----------------------------------------------
                                   Black                  Latino
------------------------------------------------------------------------
               5%                      9.6%                    7.0%
               7%                     13.2%                    9.8%
              10%                     18.5%                   13.9%
------------------------------------------------------------------------
Note: Black and Latino unemployment rates are predicted by regressing
  total unemployment rate on race-specific rates.

    Turning from minorities to other less advantaged groups in the 
labor market reveals similarly wide disparities between their 
unemployment rates and the floor in the proposed rule.

   Since 2008, the BLS has tracked unemployment among those who 
        self-report as disabled. In months when the unemployment rate 
        averaged seven percent, the disabled rate was 13 percent.

   Doing the same comparison by education level (for job 
        seekers 25 and older, per BLS published data), yields jobless 
        rates of 11 percent for those with less than high school 
        degrees.
F. Local Areas With High Unemployment Rates for Sub-Groups
    We can also see how in local areas, when the overall unemployment 
rate is five, six, or seven percent, some sub-groups face much higher 
unemployment rates. Of the 239 large labor market areas (metropolitan 
areas) with average overall unemployment rates below seven 
percent,\153\ 85 have unemployment rates that are at least 14 percent 
for particularly vulnerable groups, such as adults ages 25 to 64 with 
very low education, people with self-reported disabilities, and/or 
black and Latino residents. The data this analysis includes are 
published Census Bureau figures from the American Community Survey and 
average together 5 years of data from 2013 through 2017 in order to 
increase reliability.\154\
---------------------------------------------------------------------------
    \153\ U.S. Census Bureau, American Community Survey Table S2201. 
The United States has 389 metropolitan areas overall, without regard to 
unemployment rate.
    \154\ To further increase reliability, we impose two additional 
requirements on the comparisons. First, subgroup unemployment rates in 
a given metropolitan area are only included in our analysis if they are 
statistically significantly higher than the overall unemployment rate 
for the area. Second, we only include estimated unemployment rates that 
are at least twice as large as their margins of error.
---------------------------------------------------------------------------
    15 metropolitan areas in 2013-2017 had unemployment rates of seven 
percent or less overall but at least 14 percent for the least educated 
workers. For example, in the Springfield, IL metro area, for example, 
the overall unemployment rate was 6.7 percent, but was 21.2 percent for 
those with less than a high-school diploma. (Table 3.2.)

                                                    Table 3.2
 Metropolitan Areas With 5 Year Average Unemployment Rates Less Than 7 Percent Overall but Greater Than or Equal
                        to 14 Percent for Workers With Less Than a High School Education
----------------------------------------------------------------------------------------------------------------
                                                                Overall  Unemployment     Unemployment Rate for
     State                    Metropolitan Area                Rate for Population 16    Workers With Less Than
                                                                   Years and Over        High School  Education
----------------------------------------------------------------------------------------------------------------
            FL  Pensacola-Ferry Pass-Brent, FL                                     6.9                      16.3
           ID   Pocatello, ID                                                      6.3                      15.2
            IL  Springfield, IL                                                    6.7                      21.2
            IL  Peoria, IL                                                         6.6                      15.6
           IN   Kokomo, IN                                                         6.5                      14.2
           KY   Elizabethtown-Fort Knox, KY                                        6.7                      16
           MI   Kalamazoo-Portage, MI                                              6.6                      17.2
           MI   Monroe, MI                                                         5.9                      16.7
         MO-IL  St. Louis, MO-IL                                                   6.3                      14.7
           NY   Syracuse, NY                                                       6.6                      14.4
           NY   Elmira, NY                                                         5.3                      14.5
           OH   Canton-Massillon, OH                                               6.6                      15.7
        OH-PA   Youngstown-Warren-Boardman, OH-PA                                  6.9                      14.1
           VA   Blacksburg-Christiansburg-Radford, VA                              5.2                      15.5
           WV   Parkersburg-Vienna, WV                                             6.1                      17.3
----------------------------------------------------------------------------------------------------------------

    59 metro areas in 2013-2017 had unemployment rates of seven percent 
or less overall but at least 14 percent for workers with any 
disability. For example, in the Peoria, IL metro area, the overall 
unemployment rate was 6.6 percent, but was 18.4 percent for those with 
any disability. (Table 3.3.)

                                                    Table 3.3
 Metropolitan Areas With 5 Year Average Unemployment Rates Less Than 7 Percent Overall but Greater Than or Equal
                                  to 14 Percent for Workers With Any Disability
----------------------------------------------------------------------------------------------------------------
                                                                Overall  Unemployment     Unemployment Rate for
     State                    Metropolitan Area                Rate for Population 16        Workers With a
                                                                   Years and Over              Disability
----------------------------------------------------------------------------------------------------------------
            AL  Florence-Muscle Shoals, AL                                         6.7                      16.3
            AL  Decatur, AL                                                        6.6                      17.6
            AL  Birmingham-Hoover, AL                                              6.8                      15.4
            AL  Daphne-Fairhope-Foley, AL                                          5.5                      14.3
           CA   Santa Maria-Santa Barbara, CA                                      6.6                      14.5
           CA   Salinas, CA                                                        6                        14.4
           CA   Napa, CA                                                           5.4                      16
           CT   Hartford-West Hartford-East Hartford, CT                           6.8                      15.6
           DE   Dover, DE                                                          6.7                      14.9
            FL  Deltona-Daytona Beach-Ormond Beach, FL                             6.7                      15.6
            FL  Gainesville, FL                                                    6.7                      14.4
            FL  Pensacola-Ferry Pass-Brent, FL                                     6.9                      15.1
            FL  Tampa-St. Petersburg-Clearwater, FL                                6.8                      16.2
            FL  Orlando-Kissimmee-Sanford, FL                                      6.8                      15.1
            FL  North Port-Sarasota-Bradenton, FL                                  6.1                      14.9
           ID   Pocatello, ID                                                      6.3                      15.3
            IL  Springfield, IL                                                    6.7                      17.4
            IL  Peoria, IL                                                         6.6                      18.4
            IL  Champaign-Urbana, IL                                               5.2                      14.2
           IN   Fort Wayne, IN                                                     5.9                      14
        KY-IN   Louisville/Jefferson County, KY-IN                                 6                        14
             LA Houma-Thibodaux, LA                                                6.5                      16.2
             LA Shreveport-Bossier City, LA                                        6.8                      15.6
             LA Baton Rouge, LA                                                    6.7                      14.9
        MA-CT   Worcester, MA-CT                                                   6.3                      14.5
           MD   Baltimore-Columbia-Towson, MD                                      6                        14.8
           ME   Lewiston-Auburn, ME                                                5.2                      15.2
           ME   Bangor, ME                                                         6.5                      18.1
           MO   Springfield, MO                                                    5.2                      14
         MO-IL  St. Louis, MO-IL                                                   6.3                      14.2
         MO-IL  Cape Girardeau, MO-IL                                              5.4                      14.7
           NC   Burlington, NC                                                     6                        15.2
           NC   Greensboro-High Point, NC                                          6.6                      15.2
           NC   Durham-Chapel Hill, NC                                             6                        14.1
           NY   Utica-Rome, NY                                                     6.5                      17.2
           NY   Syracuse, NY                                                       6.6                      15.4
           NY   Rochester, NY                                                      6.3                      14.8
           NY   Ithaca, NY                                                         4.8                      14
     NY-NJ-PA   New York-Newark-Jersey City, NY-NJ-PA                              6.9                      14.5
           OH   Canton-Massillon, OH                                               6.6                      14.9
     OH-KY-IN   Cincinnati, OH-KY-IN                                               5.8                      14
        OH-PA   Youngstown-Warren-Boardman, OH-PA                                  6.9                      14.5
           OR   Corvallis, OR                                                      6.7                      14.1
        OR-WA   Portland-Vancouver-Hillsboro, OR-WA                                6.2                      14.5
           PA   Pittsburgh, PA                                                     5.7                      14.4
        PA-NJ   Allentown-Bethlehem-Easton, PA-NJ                                  6.7                      14.1
        RI-MA   Providence-Warwick, RI-MA                                          6.9                      15.9
           TX   Sherman-Denison, TX                                                6.5                      14.7
           TX   Tyler, TX                                                          6.5                      15.8
           TX   Waco, TX                                                           5.2                      14.7
           VA   Richmond, VA                                                       6.3                      14.3
           VA   Roanoke, VA                                                        5.4                      14.6
           VA   Blacksburg-Christiansburg-Radford, VA                              5.2                      16.1
           WA   Walla Walla, WA                                                    6.1                      14
           WA   Kennewick-Richland, WA                                             5.9                      15.1
           WI   Sheboygan, WI                                                      4.4                      15.4
           WI   Janesville-Beloit, WI                                              6.4                      14.1
        WI-MN   La Crosse-Onalaska, WI-MN                                          4.4                      14.9
     WV-KY-OH   Huntington-Ashland, WV-KY-OH                                       6.6                      16
----------------------------------------------------------------------------------------------------------------

    29 metro areas in 2013-2017 had unemployment rates of seven percent 
or less overall but at least 14 percent for African American residents. 
For example, in the Canton-Massillon, OH metro area, the overall 
unemployment rate was 6.6 percent, but was 16.9 percent for black 
workers. (Table 3.4.)

                                                    Table 3.4
 Metropolitan Areas with 5 Year Average Unemployment Rates Less Than 7 Percent Overall but Greater Than or Equal
                                to 14 Percent for Black/African American Subgroup
----------------------------------------------------------------------------------------------------------------
                                                                Overall  Unemployment
     State                    Metropolitan Area                 Rate for Population 16   Black/African  American
                                                                    Years and Over          Unemployment Rate
----------------------------------------------------------------------------------------------------------------
           AR   Jonesboro, AR                                                      5.9                      14.3
           IA   Dubuque, IA                                                        3.9                      17
           IA   Waterloo-Cedar Falls, IA                                           4.9                      19.7
            IL  Springfield, IL                                                    6.7                      16.4
            IL  Peoria, IL                                                         6.6                      18.1
           IN   Elkhart-Goshen, IN                                                 5.6                      15
           IN   Terre Haute, IN                                                    6.9                      16.9
           IN   Fort Wayne, IN                                                     5.9                      15.2
        IN-MI   South Bend-Mishawaka, IN-MI                                        6.6                      14.3
           KY   Owensboro, KY                                                      6.1                      15.6
           MA   Pittsfield, MA                                                     6.8                      19.6
           ME   Lewiston-Auburn, ME                                                5.2                      17.7
           ME   Bangor, ME                                                         6.5                      28.9
           MI   Kalamazoo-Portage, MI                                              6.6                      15.3
           MN   Mankato-North Mankato, MN                                          4.2                      23.3
           MN   Rochester, MN                                                      3.9                      20.2
           MN   St. Cloud, MN                                                      4.5                      17.2
           NY   Utica-Rome, NY                                                     6.5                      15.2
           NY   Syracuse, NY                                                       6.6                      15.1
           NY   Rochester, NY                                                      6.3                      14.8
           OH   Canton-Massillon, OH                                               6.6                      16.9
        OH-PA   Youngstown-Warren-Boardman, OH-PA                                  6.9                      17.2
           PA   Altoona, PA                                                        5.2                      18.7
           PA   Scranton-Wilkes-Barre-Hazleton, PA                                 6.4                      19.2
           PA   Pittsburgh, PA                                                     5.7                      14.1
           PA   Williamsport, PA                                                   6.2                      23
           WI   Janesville-Beloit, WI                                              6.4                      17
           WI   Green Bay, WI                                                      4.4                      18.1
        WV-OH   Wheeling, WV-OH                                                    6.1                      15.8
----------------------------------------------------------------------------------------------------------------

    Six metro areas had unemployment rates of seven percent or less 
overall but at least 14 percent for Hispanic or Latino adults. For 
example, in the Bloomsburg-Berwick, PA metro area, the overall 
unemployment rate was 4.9 percent, but it was 26.9 percent for 
Hispanic/Latino workers (who may be of any race). (Table 3.5.)

                                                    Table 3.5
Metropolitan Areas With 5 Year Average Unemployment Rates Less Than 7 Percent Overall but Greater or Equal to 14
                                      Percent for Latino/Hispanic Subgroup
----------------------------------------------------------------------------------------------------------------
                                                                 Overall  Unemployment
     State                     Metropolitan Area                 Rate for Population 16     Hispanic or Latino
                                                                     Years and Over         Unemployment Rate
----------------------------------------------------------------------------------------------------------------
           NY   Utica-Rome, NY                                                      6.5                     15.5
           OH   Canton-Massillon, OH                                                6.6                     20.1
           PA   Bloomsburg-Berwick, PA                                              4.9                     26.9
           PA   Erie, PA                                                            6.5                     15.8
           PA   Chambersburg-Waynesboro, PA                                         5.7                     14.8
           PA   Lebanon, PA                                                         5.8                     14.2
----------------------------------------------------------------------------------------------------------------

    Fourteen percent unemployment, averaged over 5 years, is a 
strikingly high rate that matches the unemployment rate estimated for 
the overall labor force in 1937 during the Great Depression.\155\ Under 
the proposed rule, some communities would be ineligible for a waiver 
where some individuals subject to the time limit are likely to face 
unemployment rates of this level.
---------------------------------------------------------------------------
    \155\ The U.S. Department of Labor estimates 7.7 million persons 
(or about 14.2 percent) were unemployed in 1937 out of a labor force of 
54.0 million. U.S. Department of Labor, ``Labor Force, Employment, and 
Unemployment, 1929-39: Estimating Methods,'' Technical Note, Monthly 
Labor Review, July 1948, https://www.bls.gov/opub/mlr/1948/article/pdf/
labor-force-employment-and-unemployment-1929-39-estimating-methods.pdf.
---------------------------------------------------------------------------
    A recent analysis by the Center on Poverty and Social Policy at 
Columbia University also demonstrates how sub-populations face 
significantly higher unemployment rates than the area average. This 
analysis looked at over 200 metropolitan areas that could potentially 
lose waiver eligibility: areas with unemployment rates 20 percent above 
the national average, but below seven percent unemployment rates. As 
the authors stated:

          The median unemployment rate for non-white individuals is 
        closer to ten percent, and in some metro areas the unemployment 
        rate is greater than 20 percent for non-white individuals. For 
        those with a high school education or less, over \3/4\ of all 
        metropolitan areas have a higher unemployment rate than the 
        seven percent floor, which is a particularly dire statistic for 
        the relevant population of ``lower-skilled'' workers. Given 
        that close to \3/4\ of ``ABAWDs'' have a high school education 
        or less and close to \1/2\ are non-white, this evidence 
        suggests that the proposed rule would disqualify many areas 
        where the individuals subject to the time limit face 
        substantially higher unemployment rates than seven percent.

    This analysis also looked at complementary labor force metrics, 
finding that non-white individuals and workers with a high school 
education or less had significantly lower employment-population ratios, 
with over \1/2\ of individuals with less education living in areas with 
employment-population ratios lower than 50 percent. This research shows 
how the unemployment rate hides the variation for sub-groups.\156\ We 
strongly encourage FNS to review these data.
---------------------------------------------------------------------------
    \156\ Robert Paul Hartley, Christopher Wimer, and Jane Waldfogel, 
``Limiting States' Ability to Waive Federal SNAP Work Requirements: A 
Closer Look at the Potential Implications,'' Columbia University Center 
on Poverty and Social Policy Research Brief, Vol. 3 No. 4, March 25, 
2019. https://static1.squarespace.com/static/5743308460b5e922a25a6dc7/
t/5c9a2d43652dea22023c
9492/1553608003521/
Poverty+%26+Social+Policy+Brief+3_4_+SNAP+Work+Requirements+
and+Unemployment.pdf.
---------------------------------------------------------------------------
    Given this evidence, the proposal to restrict states' ability to 
waive areas except with very high overall unemployment rates will have 
a disproportionate impact on subgroups with rates much higher than 
overall unemployment, including groups belonging to protected classes 
under 7 U.S.C.  2020(c).
Many ``Distressed Communities'' Have Relatively Low Unemployment
    Another way of considering how areas with relatively low 
unemployment may provide insufficient jobs for individuals subject to 
the time limit is to look at other economic indicators, which provide 
other information that could indicate a paucity of jobs. The nonprofit 
organization Economic Innovation Group calculates a measure of 
community well-being called the ``Distressed Community Index'' that 
combines seven metrics for the 2012-2016 period: the share of adults 
ages 25 and up without a high school diploma; the percent of habitable 
housing that is unoccupied; the share of the prime-age (25-64) 
population that is not employed; the poverty rate; median household 
income as a percent of the state's median households income; the change 
in employment; and the change in business establishments.
    While these measures do not strictly measure job availability, they 
do provide a snapshot of economic health, and present a snapshot of how 
divergent the economic conditions are, and recovery from the Great 
Recession has been, at the local level. For example, most of the job 
growth from 2007 to 2016 has occurred in the [ZIP C]odes in the top 
quintile of the index. Over \2/3\ of [ZIP C]odes in that quintile, 
termed ``prosperous,'' added jobs since 2007 (adding an average of 
1,300 jobs); meanwhile, over \2/3\ of [ZIP C]odes in the lowest 
quintile, called ``distressed,'' have fewer jobs since 2007, and those 
that did add jobs only added an average of 400 over the period studied. 
While about \1/5\ of prime-age adults were out of work in 
``prosperous'' [ZIP C]odes, that share was double for adults in 
``distressed'' [ZIP C]odes.\157\
---------------------------------------------------------------------------
    \157\ Economic Innovation Group, ``From Great Recession to Great 
Reshuffling: Charting a Decade of Change Across American Communities,'' 
October 2018, https://eig.org/wp-content/uploads/2018/10/2018-DCI.pdf.
---------------------------------------------------------------------------
    We looked at counties that had a waiver in 2018 but would not have 
qualified if the proposed rule were in place because they did not meet 
the seven percent unemployment rate threshold. Of these over 600 
counties, over 100 were considered ``distressed.'' Todd County, South 
Dakota, for example, had a 6.6 unemployment rate for the January 2016-
December 2017 period. According to this index, nearly \1/2\ of the 
residents in this county lived in poverty in the 2012-2016 period, and 
close to \1/2\ of prime-age adults were not employed. The number of 
jobs in this county declined by over \2/5\ from 2012 to 2016, and the 
number of business establishments also declined by over five percent. 
Another example is Stewart County, Georgia, where over \1/3\ of adults 
have less than a college degree, household median income is only about 
\2/5\ of the state's median income, and over \2/3\ of prime-age adults 
are not employed. Stewart County also lost both jobs and business 
establishments between 2012 and 2016.
    While an area's unemployment rate may mask differences between 
unemployment rates within that area, it also may fail to reflect 
economic conditions more broadly, which may contribute to job 
availability for the individuals potentially subject to the time limit. 
Here again, we are concerned that FNS did not appear to offer any 
evidence to support its contention that unemployment rates are a 
reliable predictor of jobs available for low-income individuals.
Underemployment Rates Also Higher For Sub-Groups
    Along with the impossibility of identifying an unemployment rate 
that reliably implies the absence of available jobs, it is also the 
case that the unemployment rate is an insufficient indicator of labor 
market slack. For one, it leaves out those who have left the labor 
market, in some cases due to slack labor demand or to personal labor 
market barriers, including skill deficits and discrimination. Second, 
the unemployment rate leaves out a significant group of part-time 
workers who would prefer to be full-timers. Such workers are literally 
under-employed, as they want to work more hours than their current job 
offers them. For families with low incomes, working too few hours can 
put pressure on family budgets and lead to nutritional hardship.
    Table 3.6 shows underemployment rates associated with unemployment 
rates of five, seven, and ten percent for all workers and by race/
ethnicity (see note under the table for methodology). At the proposed 
rule's suggested level of seven percent unemployment, overall 
underemployment is predicted to be 12.5 percent, with rates of about 20 
and 18 percent for African American and Latino workers, respectively. 
In other words, the rule suggests that SNAP waivers should be 
disallowed in places where about \1/5\ of black and Latino workers 
could be un- or underemployed.
    Higher unemployment of course corresponds to even higher 
underemployment rates, but even at five percent unemployment, black and 
Latino underemployment is around 15 and 13 percent, respectively.

                                Table 3.6
Predicted Underemployment Rates at Different Unemployment Rates, by Race/
                                Ethnicity
------------------------------------------------------------------------
                                Predicted Underemployment
 Unemployment  ---------------------------------------------------------
                     All           White          Black       Hispanic
------------------------------------------------------------------------
          5%            9.1%           7.4%          14.9%         13.0%
          7%           12.5%          10.2%          20.4%         17.9%
         10%           17.7%          14.4%          28.8%         25.2%
------------------------------------------------------------------------
Note: Rates for ``all'' are derived from regression of U-6
  underemployment rate on the overall unemployment rate. Racial
  underemployment rates are then derived from ratios of the overall
  unemployment to underemployment rates by race using Economic Policy
  Institute data from 1994-2018.

    While the ``20 percent standard'' currently uses unemployment 
rates, which do not capture aspects such as labor force participation 
or part-time work, FNS proposes making these criteria even less 
responsive to economic conditions by requiring a specific unemployment 
rate. Given the severe racial disparities that exist in labor force 
measures, the fact the Department did not address whether it considered 
how an unemployment rate varies in relation to other labor force 
metrics is another reason why we cannot comment on how it supported 
this rule.
G. Unemployment Rate Floor of Seven Percent Fails to Protect Areas 
        During Recessions
    The Department does not discuss how the seven percent unemployment 
rate floor would affect waiver eligibility during an economic downturn, 
or at any other point in the business cycle besides a time of 
relatively low unemployment. An unemployment rate of seven percent is 
relatively high for any area; for example, during the 2001 recession, 
the national unemployment rate never reached seven percent.\158\ An 
area with a 24 month unemployment rate averaging at least seven percent 
signifies that an area has experienced a prolonged depression. The 
Department does not acknowledge the unemployment rate of seven percent 
in relation to other economic indicators, but also does not discuss how 
the length of time its proposal would require such a high unemployment 
rate to qualify for a waiver would affect states entering an economic 
recession.
---------------------------------------------------------------------------
    \158\ ``National Bureau of Labor Statistics, U.S. Business Cycle 
Contractions and Expansions,'' https://www.nber.org/cycles.html. 
Monthly unemployment rates are from BLS.
---------------------------------------------------------------------------
    When unemployment rates rise rapidly when the economy is entering 
into a recession and jobs are quickly declining, individuals likely 
face many challenges finding or keeping work. By requiring a very high 
2 year average unemployment rate, the proposed rule, however, would 
keep many areas from qualifying for a waiver during this time. The 
proposed rule would continue to allow an area to qualify for a waiver 
when it qualifies under any of the criteria (including optional 
criteria) for Extended Benefits (EB) in the Unemployment Insurance 
program, which would often allow states with rapidly rising 
unemployment to qualify for a waiver.\159\ Among other criteria, under 
EB, states can qualify for a waiver if they meet optional indicators 
that include a 3 month unemployment rate of 6.5 percent that is at 
least 110 percent of the same 3 month period in either of the previous 
2 years.\160\ We agree with the Department's proposal to continue to 
allow states to request waivers when they qualify for EB, as these are 
times when unemployment rates are high and rising and individuals 
likely have difficulty finding jobs. Because the seven percent 
unemployment rate floor is so high and because the NPRM would prohibit 
states from requesting statewide waivers based on the 20 percent 
standard, many states would experience a gap between when their 
unemployment rates begin to rise during a recession and when they 
qualify for a waiver based on meeting the EB criteria.
---------------------------------------------------------------------------
    \159\ NPRM, p. 992.
    \160\ While the criteria mentioned are the criteria to establish 
additional weeks of Extended Benefits under an optional trigger (called 
the Total Unemployment Rate trigger), FNS guidance establishes that 
states only have to qualify based on the unemployment measures, even if 
the state chooses not to provide EB benefits under these criteria.
---------------------------------------------------------------------------
    For example, consider the experience of two states, South Carolina 
and Oregon, who would have been left with a gap between their 
unemployment rates rising and their qualification for a waiver based on 
the EB criteria in the beginning of the Great Recession. During the 
beginning of the Great Recession, which officially started in December 
2007, these states both had unemployment rates that were rising and 
would have qualified for 2007 statewide waivers under the existing 20 
percent standard (as in, without the seven percent unemployment rate 
floor).\161\ These states, like others, would have had to wait several 
months before they would have qualified for a waiver under Extended 
Benefits: South Carolina qualified for a waiver based on Extended 
Benefits beginning in August 2008 and Oregon qualified for a waiver 
based on Extended Benefits in November 2008.\162\ Under the proposed 
rule, they could not have requested statewide waivers based on the ``20 
percent standard,'' as we discuss in Chapter 5, as the rule would only 
allow statewide waivers based on EB. Even if they could have requested 
statewide waivers, however, these states would have been well into the 
recession before they qualified for a waiver based on having 24 month 
unemployment rates above seven percent. Therefore, under the proposed 
rule, at least some areas in both states would have been ineligible for 
a waiver at a time when unemployment was high and rising.
---------------------------------------------------------------------------
    \161\ Oregon would not have qualified for a 2008 waiver.
    \162\ In November 2008, Congress passed a temporary expansion of 
Extended Benefits, called the Emergency Unemployment Compensation 
program, that in January 2009 the Bush Administration stated could 
qualify states for a 1 year statewide waiver. These states would have 
qualified for statewide waivers immediately then, as both qualified 
beginning with the first EUC trigger notice November 23, 2008 (https://
oui.doleta.gov/unemploy/euc_trigger/2008/euc_
112308.pdf). In addition, Congress passed the American Recovery and 
Reinvestment Act (ARRA), which waived all states statewide beginning in 
April 2009. Because these are both temporary measures requiring action 
by Congress, there is no guarantee under the proposed rule that they 
would be available in the future, and as we note, they weren't 
available in the beginning of the recession in early 2008.
---------------------------------------------------------------------------
    The Department repeatedly explains how its rulemaking would prevent 
states from requesting waivers when unemployment is low. For example, 
it states:

          Right now, nearly \1/2\ of ABAWDs live in areas that are 
        covered by waivers despite a strong economy. The Department 
        believes waiver criteria need to be strengthened to better 
        align with economic reality. These changes would ensure that 
        such a large percentage of the country can no longer be waived 
        when the economy is booming and unemployment is low.\163\
---------------------------------------------------------------------------
    \163\ NPRM, p. 981.

    The Department has clearly considered the role of waivers at a time 
when national unemployment is low, though this analysis of course does 
not take into account the fact that unemployment rates can vary across 
the country and even with low unemployment rates, individuals subject 
to the time limit may lack available jobs. Even more concerning, 
however, is that the Department did not indicate whether it considered 
the effect of the proposed rule at different parts of the business 
cycle, such as entering into a recession, and how climbing unemployment 
rates affect job availability. (The Regulatory Impact Analysis also 
fails to include analyses of the impact of the provision using 
historical data to assess how it would fare differently in different 
economic times.) Without such a discussion, it is impossible to assess 
the economic considerations it made in proposing a policy that would 
result in many areas remaining ineligible for waivers at a time of 
rising unemployment.
H. Department Does Not Explain Claim That Suggested Floor Is ``Designed 
        Specifically for ABAWDS''
    The Department proposes a seven percent unemployment floor for 
areas to qualify for a waiver under the 20 percent standard, 
significantly higher than the floor used by Labor Surplus Areas. The 
Department suggests this floor would be more ``targeted'' towards the 
specific individuals subject to the time limit, but provides no 
evidence to support this assertion. The preamble states:

          The Department believes that amending the waiver regulations 
        to include an unemployment floor is a critical step in 
        achieving more targeted criteria. While the 20 percent standard 
        is similar to the calculation of an LSA, the Department 
        believes it is appropriate to request public comment to explore 
        a floor that is designed specifically for ABAWD waivers.\164\
---------------------------------------------------------------------------
    \164\ NPRM, p. 984.

    The Department suggests that having a higher unemployment rate 
floor than that used by DOL in its identification of LSAs would be more 
appropriate for this population than the general LSA floor of six 
percent unemployment. Evidence shows that the childless adult SNAP 
participants face labor market disadvantages, and likely experience 
higher unemployment rates than their area. This evidence would 
recommend against a specific unemployment rate floor, given the 
difficulty in assessing a specific rate that would reflect available 
jobs for this population. For example, a city or county may have an 
unemployment rate of seven percent, but the unemployment rate for 
childless adult SNAP participants is 12 or 14 percent. The difficulty 
in establishing available jobs is especially true in local areas where 
local labor market conditions may yield differing opportunities for 
this population for different levels of unemployment. For example, even 
if two areas had the same unemployment rate, an area where individuals 
live close to jobs that match their skills will have more opportunities 
than an area where there is considerable spatial mismatch.
    Given this evidence, if the Department did want a floor ``designed 
specifically for ABAWD waivers,'' it would follow that they would want 
to explore a floor that is considerably lower than that used by the 
Department of Labor in designating Labor Surplus Areas. By suggesting 
that a higher floor would be ``designed specifically for ABAWD 
waivers,'' the Department is suggesting that unless unemployment is at 
a relatively high level, substantially higher than what the Department 
of Labor considers sufficiently high in designating Labor Surplus 
Areas, there are sufficient jobs available for childless adult SNAP 
participants. All available evidence suggests the opposite is true: 
unemployment has to fall to very low levels before more disadvantaged 
workers can find jobs. The Department does not provide any evidence to 
support its conclusion that a seven percent unemployment rate bears any 
relationship to available jobs for this specific population. By 
referencing that such a floor would be ``targeted,'' the Department 
indicates that there are considerations that it took when establishing 
a floor substantially higher than the LSA floor. Without any discussion 
of those considerations, however, it is impossible to follow the 
Department's logic and thus meaningfully comment on it. The robust 
review we did to understand the availability of jobs as related to the 
unemployment rate finds that the unemployment substantially overstates 
job opportunities for the individuals subject to the time limit, which 
would recommend flexibility, not imposing a specific unemployment rate 
floor. The Department is claiming an opposite finding without providing 
any evidence to support its conclusions.
I. Reasoning for Specific Unemployment Rate Floor Not Consistent With 
        Congressional Intent
    The Department's stated rationale for proposing the seven percent 
unemployment rate floor for waivers for areas with unemployment rates 
20 percent above the national average is to ensure that time limit 
waivers cover a reduced population compared to current standards. This 
rationale is inconsistent with the intent of Congress. Congress 
intended for FNS to develop criteria to measure a lack of jobs, and did 
not specify intended limits to the usage of waivers, provided they 
reflect economic conditions.
    In the preamble, the Department establishes that the justification 
for the proposed rule is not based on an analysis of the relationship 
of the unemployment rate floor to job availability for childless adult 
SNAP participants. The Department states it instead weighed the effect 
of the proposal on the breadth of waiver coverage:

          The Department seeks to establish a floor that is in line 
        with the Administration's effort to encourage greater 
        engagement in work and work activities. The Department believes 
        that the seven percent floor for the 20 percent standard would 
        strengthen the standards for waivers so that the ABAWD work 
        requirement would be applied more broadly and fully consider 
        the ``lack of sufficient jobs'' criteria in the statute.\165\
---------------------------------------------------------------------------
    \165\ NPRM, p. 984.

    The Department therefore states that the goal of the NPRM is to 
apply the time limit to more childless adults. The Department states 
that applying the time limit to more unemployed adults would ``fully 
consider the `lack of sufficient jobs' criteria in the statute,'' but 
does not explain how restricting areas and would better reflect 
employment opportunities for this population. Moreover, this reasoning 
is completely contrary to Congressional intent, which was to allow 
states to waive areas with insufficient jobs without imposing limits on 
the share of areas covered by waivers by state or nationally. We are 
confused as to why FNS believes it has the authority to purposefully 
expose more people to the time limit as a rationale.
    Available evidence suggests that restricting waivers to only areas 
with very high unemployment would actually make waivers less likely to 
reflect available jobs for this population, given that it would exclude 
from eligibility many areas where these individuals lack jobs (such as 
an area with an unemployment rate of 6.7 percent, but unemployment 
rates well above ten percent for childless adult SNAP participants).
    Congress intended for the Administration to develop economic 
criteria to measure job opportunities for childless adult SNAP 
participants. Congress did not propose any measure to limit waivers 
based on the number or share of individuals subject to the time limit. 
If Congress had intended for waivers to be limited so that a specific 
share of childless adults live in an area with a waiver, it could have 
written legislation to achieve that goal. For example, in the 2017 Tax 
Cuts and Jobs Act (P.L. 115-97), Congress created Opportunity Zones, 
which are low-income Census tracts designated by the chief executive of 
a state that are eligible for tax incentives for investment. While 
Congress created several criteria to identify areas that are nominated, 
such as the poverty rate or median family income, Congress also limited 
the number of potential eligible Opportunity Zones each state is 
allowed to designate based on the total number of low-income 
communities in the state. For example, in areas with over 100 low-
income census tracts, no more than 25 percent of the number of those 
low-income tracts can be designated as Opportunity Zones.\166\ This law 
serves as an example of one way that Congress can establish criteria to 
limit a particular sub-state designation, if that is indeed its intent. 
In establishing waiver criteria in the welfare reform law, Congress did 
not establish any mechanism to limit the scope of waivers, which it 
could have done by various means, had that been its goal. Instead, the 
law allows for the Department to develop measures to evaluate available 
jobs for childless adult SNAP participants, which are not limited in 
scope. The number of areas lacking jobs can expand or contract with 
economic conditions, and Congress allowed for states to waive areas in 
response to these changing economic conditions. Congress has not 
changed this approach since the original 1996 legislation.
---------------------------------------------------------------------------
    \166\ Congressional Research Service, ``Tax Incentives for 
Opportunity Zones: In Brief,'' R45152, November 20, 2018, https://
fas.org/sgp/crs/misc/R45152.pdf.
---------------------------------------------------------------------------
    Furthermore, the Department uses provisions of the House-passed 
version of H.R. 2 to support its proposed unemployment rate floor, 
ignoring that Congress ultimately rejected such provisions. In 
providing support for the seven percent unemployment rate floor, the 
preamble states, ``Furthermore, this aligns with the proposal in the 
Agriculture and Nutrition Act of 2018, H.R. 2, 115th Cong.  4015 (as 
passed by House, June 21, 2018).'' \167\ While this bill did contain a 
similar provision, the Senate bill did not include this provision, and 
the Conference Committee chose to align with the Senate version, 
passing both chambers without any restrictions on waivers. While the 
Department may consider Congressional bills, offering this as support 
while ignoring that these provisions were ultimately excluded from the 
final bill, the Department is offering an incomplete interpretation of 
Congressional intent.
---------------------------------------------------------------------------
    \167\ NPRM, p. 984.
---------------------------------------------------------------------------
Department Provides Little Explanation to Support Stated Goal to Limit 
        Waiver Coverage
    The Department establishes that its intent is to limit waiver 
coverage and therefore expand the time limit to the extent possible. 
The Department does not explain how this goal is related to the intent 
of the statute to identify areas that lack jobs for childless adults. 
Even if it had clarified how its stated goal related to the underlying 
statute it is interpreting, the Department also does not provide clear 
explanation of the assumptions used in determining the metric used 
repeatedly to support its conclusions, the share of ``ABAWDs'' living 
in a waived area. Without any explanation of the analysis used to 
understand what it believes the relationship between the unemployment 
rate floor and waiver coverage is, and how waiver coverage relates to 
the ``insufficient jobs'' law, the Department has limited our ability 
to comment on these specific assertions.
    The Department explains that the principal criteria it considered 
when proposing the specific seven percent unemployment rate floor was 
not based on an economic argument about the relationship between the 
general unemployment rate and jobs available for disadvantaged 
individuals, but rather a desire to limit waivers of the time limit:

          As stated previously, the Department seeks to make the work 
        requirements the norm rather than the exception to the rule 
        because of excessive use of ABAWD time limit waivers to date. 
        Using the proposed rule's seven percent floor for this 
        criterion and eliminating waiver approvals based on an LSA 
        designation (as well as utilizing the proposed limit on 
        combining areas discussed below), an estimated 11 percent of 
        ABAWDs would live in areas subject to a waiver. Currently, 
        approximately 44 percent of ABAWDs live in a waived area. The 
        Department views the proposal as more suitable for achieving a 
        more comprehensive application of work requirements so that 
        ABAWDs in areas that have sufficient number of jobs have a 
        greater level of engagement in work and work activities, 
        including job training.\168\
---------------------------------------------------------------------------
    \168\ NPRM, p. 984.

    The Department suggests that not only would limiting waivers be 
preferable to keeping the current regulations, but also suggests that 
the more the rules result in limited waivers, the more SNAP 
participants will be led towards self-sufficiency. (The Department also 
makes the claim that current waiver coverage is ``excessive without 
providing explanation of the criteria used to judge appropriate waiver 
coverage.'') It states that a seven percent floor, which it indicates 
would result in a decline from 44 percent of ``ABAWDs'' living in a 
waived area to 11 percent, would be ``more suitable'' than the current 
rules. In the proposed rule, the Department asks for feedback on 
alternative floors to the seven percent unemployment rate floor of six 
percent or ten percent (see Section I, below), explaining how the 
higher the unemployment rate floor, the more preferable according to 
---------------------------------------------------------------------------
their standards:

          Based on the Department's analysis, nearly 90 percent of 
        ABAWDs would live in areas without waivers and would be 
        encouraged to take steps towards self-sufficiency if a floor of 
        seven percent was established. In comparison, a six percent 
        floor would mean that 76 percent of ABAWDs would live in areas 
        without waivers and a ten percent floor would mean that 98 
        percent of ABAWDs would live in areas without waivers. A higher 
        floor allows for the broader application of the time limit to 
        encourage self-sufficiency.\169\
---------------------------------------------------------------------------
    \169\ NPRM, p. 984.

    The Department therefore believes that expanding the time limit to 
more people is a desirable outcome. The Department states that the 
greater the unemployment rate threshold, the fewer childless adults 
will live in waived areas, suggesting that its goal is to minimize 
waiver coverage to the extent possible. Setting aside the issue that 
there is no evidence that applying the time limit more broadly 
encourages self-sufficiency, which we address comprehensively in 
Chapters 6 and 11, the Department leaves several unanswered questions 
with regards to how this rulemaking will further the intent of the law 
---------------------------------------------------------------------------
in defining areas with insufficient jobs:

   The Department does not explain how imposing a higher 
        unemployment rate floor would better approximate a lack of 
        jobs. The purpose of the regulation is to define areas that 
        lack ``a sufficient number of jobs to provide employment'' to 
        childless adult SNAP participants. To interpret this 
        regulation, it would follow that the specific waiver criteria 
        the Department develops would best allow states to identify 
        areas lacking jobs, and best enable the Department to approve 
        those waivers based on consistent criteria. Operating under the 
        framework that these regulations interpret the statute, the 
        appropriate amount of waiver coverage is related to the share 
        of this population facing limited employment opportunities 
        (i.e., a lack of sufficient jobs) in their area. For example, 
        if about \1/3\ of counties did not have sufficient job 
        opportunities for childless adults, then about \1/3\ of 
        counties would be eligible for a waiver, if there were a way to 
        perfectly capture job availability for this population. If this 
        share rises during a recession to 75 percent, then the share of 
        the country eligible for a time limit waiver could also rise 
        accordingly.

      The Department is therefore proposing an alternative 
        interpretation of the statute, though it is not clear what this 
        interpretation is or what the authority it has to drastically 
        change this interpretation. The Department does not explain 
        whether it believes that there is an economic argument 
        supporting limiting waivers, or instead if it believes that the 
        goal of limiting waivers is separate from establishing areas 
        with insufficient jobs, and if so, the authority under which it 
        can establish new criteria for waivers that are not found in 
        the statute. Without more explanation as to why its goal for 
        limiting waivers is relevant to this rulemaking, it is 
        difficult to assess the merits of the underlying arguments.

   The Department does not explain whether there are any 
        parameters to its stated goal of limited waivers, and under 
        what criteria it judges the appropriate level of waiver 
        coverage. In regard to the proposed six percent unemployment 
        rate floor, the preamble states that ``the Department is 
        concerned that too many areas would qualify for a waiver of the 
        ABAWD time limit with a six percent floor and that too few 
        individuals would be subject to the ABAWD work requirements.'' 
        \170\ The language of ``too many'' or ``too few'' implies that 
        there is a desired level of waiver coverage, and that the 
        waiver coverage that they estimate a six percent unemployment 
        rate floor would yield (24 percent of ``ABAWDs'' living in 
        waived areas) is too high. The Department therefore has 
        implicit criteria by which it is judging an appropriate share 
        of individuals living in a waived county that it does not 
        explain. While it is not clear how the share of ``ABAWDs'' 
        living in a waived area is relevant to the rulemaking, even if 
        it were, the Department does not allow commenters the ability 
        to provide input on this metric without establishing the 
        criteria it is using to judge the appropriate level.
---------------------------------------------------------------------------
    \170\ NPRM, p. 984.

   Relatedly, the Department does not explain if it believes 
        that limiting waivers would be an equally important goal during 
        an economic recession, when the share of areas with limited 
        jobs would expand considerably. Again, if the Department states 
        that 24 percent of ``ABAWDs'' in waived areas is ``too high,'' 
        would that also be true during an economic recession, if most 
        areas of the country offered few jobs to those individuals, and 
        a majority of childless adults lived in an area covered by a 
        waiver? Without explaining how its stated goal of limiting 
        waivers is related to assessing the economic conditions in an 
        area, it is impossible to tell if the Department considers this 
        goal to be a relative goal (as in, it believes it is acceptable 
        to expand the time limit in response to higher unemployment), 
        or if it believes there is a desired percentage of ``ABAWDs'' 
---------------------------------------------------------------------------
        living in waived areas regardless of economic conditions.

   Finally, the Department's calculation of the share of 
        ``ABAWDs'' living in a waived area does not take several 
        important factors into account.

     When the Department calculates the share of what it 
            terms ``ABAWDs'' living in waived areas under the different 
            scenarios it lays out, it is not clear if it is considering 
            how this share will change as the overall denominator 
            changes and what other assumptions are embedded in its 
            analysis. In a time when fewer areas are waived, childless 
            adults will be more disproportionately concentrated in 
            waived areas, as they will lose benefits in non-waived 
            areas. At any time, childless adults include a combination 
            of participants who are living in waived areas; exempt from 
            the time limit (but who the data does not allow us to 
            identify as exempt); in their first 3 months of SNAP 
            participation; or working or complying with the 
            requirements through training. Other variables at the local 
            level, such as state or county implementation of the time 
            limit, the composition of childless adults (for example, in 
            some areas, there may be proportionately more disadvantaged 
            individuals), and the amount of job or training 
            opportunities that are suitable for childless adults, will 
            also affect how likely childless adults are to continue 
            participating in SNAP in non-waived areas. Therefore, what 
            share of childless adult SNAP participants living in waived 
            areas is not just a function of the share of counties 
            covered by a waiver, but also how they are distributed 
            among those counties. It is not clear if the Department is 
            using this statistic as a proxy for measuring overall 
            waiver coverage, or if it is meant to convey an analysis 
            modeling these dynamic variables.

        Consider a simplified example. Here, we will look to see how 
            the distribution of childless adults living in certain 
            waived areas changes as overall nationwide waiver coverage 
            changes, using eight states, Guam and Virgin Islands that 
            had statewide waivers in 2017, and 13 states that had 
            statewide waivers at least from 2010 through 2013 but had 
            dropped them by 2017. (This illustrative analysis therefore 
            is not looking at childless adults living in all waived 
            areas, but rather choosing to look at states when they had 
            a statewide waiver or no waiver in 2017 to simplify the 
            analysis.) In 2010, almost all areas of the country were 
            waived in the aftermath of the Great Recession, which 
            prompted Congress to include a provision in the Recovery 
            Act (P.L. 111-5) that waived the time limit nationwide. 
            (About five states continued to implement the time limit in 
            parts of their state, but they offered work opportunities 
            for individuals subject to the time limit.) About 89 
            percent of the general population lived in an area that was 
            waived.\171\
---------------------------------------------------------------------------
    \171\ Center on Budget and Policy Priorities, ``States Have 
Requested Waivers from SNAP's Time Limit in High Unemployment Areas for 
the Past Two Decades,'' https://www.cbpp.org/research/food-assistance/
states-have-requested-waivers-from-snaps-time-limit-in-high-
unemployment.

        Because of this widespread waiver coverage, the share of 
            childless adult SNAP participants living in waived areas 
            would be expected to be very similar to the share of all 
            SNAP participants living in those areas. To the extent that 
            this distribution differed, it would likely be due to 
            compositional differences, such as areas with greater 
            shares of children or elderly individuals. Waiver coverage 
            would not be the main driver of differences, as waiver 
            coverage was similar nationwide. Indeed, in 2010, about 20 
            percent of the total U.S. population lived in states that 
            had statewide waivers continuously through 2017, and a 
            slightly smaller share, 17 percent, of SNAP participants 
            lived in those states. (This may be because these states' 
            populations had slightly smaller shares of individuals with 
            income below SNAP's income eligibility limits, reduced 
            access to SNAP, or increased barriers for eligible people, 
            or other reasons.) The share of all adults ages 18-49 in 
            childless households who lived in those states in 2010 was 
            similar to the distribution of SNAP participants; about 18 
            percent of childless adults lived in those eight states. 
            Similarly, for states that had earlier had statewide 
            waivers, but dropped them by 2017, the share of childless 
            adults living in those states was similar to the share of 
            SNAP participants living in those states in 2010. Those 
            states had about 24 percent of the total U.S. population, 
            but a slightly higher share of SNAP participants (26 
            percent), and a slightly higher share of childless adults 
            (28 percent). (Table 3.7.)

                                                                        Table 3.7
                           Distribution of Childless Adults Living in Waived Areas Changes as Overall Waiver Coverage Changes
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                              SNAP
                                                                                                                          Participants       Share of
                                                                                            SNAP                          Ages 18-49,      Participants
                                                         Total        Share of Total    Participants    Share of SNAP       Without        Ages 18-49,
                                                       Population       Population     (program data,    Participants    Disabilities,       Without
                                                       (millions)                        millions)                        in Childless    Disabilities,
                                                                                                                           Households      in Childless
                                                                                                                           (millions)       Households
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                              Fiscal Year 2010: 89% of U.S. Population Lives in Waived Area
--------------------------------------------------------------------------------------------------------------------------------------------------------
States with statewide waivers in at least 2010-                72.9              24%             10.4              26%              1.1              28%
 2013 and no waivers 2017
States with statewide waivers in at least 2010-                62.1              20%              6.7              17%              0.7              18%
 2013 and 2017
                                                   -----------------------------------------------------------------------------------------------------
  Total all states                                            309.3             100%             40.3             100%              3.9             100%
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                              Fiscal Year 2017: 36% of U.S. Population Lives in Waived Area
--------------------------------------------------------------------------------------------------------------------------------------------------------
States with statewide waivers in at least 2010-                77.0              24%             10.6              25%              0.6              19%
 2013 and no waivers 2017
States with statewide waivers in at least 2010-                64.7              20%              8.3              20%              0.9              29%
 2013 and 2017
                                                   -----------------------------------------------------------------------------------------------------
  Total all states                                            325.1             100%             42.1             100%              3.2             100%
--------------------------------------------------------------------------------------------------------------------------------------------------------
Notes: The states with statewide waivers from the time limit in 2013 but no waivers at all in 2017 (which represented about \1/4\ of SNAP participants
  in 2013) were Alabama, Arkansas, Florida, Indiana, Iowa, Kansas, Maine, Mississippi, Missouri, North Carolina, Oklahoma, South Carolina, and
  Wisconsin. The states with statewide waivers in both 2013 and 2017 (which represented about 20 percent of SNAP participants) included Alaska,
  California, District of Columbia, Illinois, Louisiana, Nevada, New Mexico, Rhode Island, Guam, and Virgin Islands.
Sources: CBPP analysis of FY 2010 and FY 2017 SNAP household characteristics data; USDA program data; Census population estimates as of July 1st 2010
  and 2017.

        By 2017, the share of waived areas had declined dramatically as 
            the economy improved. About 36 percent of the U.S. 
            population lived in an area that was waived.\172\ Given 
            that the eight states who continued to waive the time limit 
            statewide were the only states remaining with statewide 
            waivers and overall waiver coverage was much lower, it 
            would be likely that childless adults would be 
            disproportionately living in states that continued to be 
            waived statewide, as they would be more likely to be 
            subject to the time limit in other states with partial or 
            no waivers. Similarly, we would expect the share of 
            childless adults living in states without waivers to have 
            declined relative to the share of all SNAP participants in 
            those states. As Table 3.7 shows, the share of childless 
            adult SNAP participants who lived in states with statewide 
            waivers (29 percent) was about 50 percent greater than the 
            share of overall SNAP participants who lived in those areas 
            (19 percent). This share also represents a significant 
            increase from the 2010 share of 18 percent. We see the 
            opposite trend for states that had no waiver by 2017: while 
            in 2010, states that had no waivers by 2017 had a slightly 
            greater share of SNAP participants and childless adults 
            than they did of overall U.S. population, by 2017, 
            proportionately fewer childless adults lived in those 
            states. While about 25 percent of SNAP participants lived 
            in those states in 2017, only 19 percent of childless 
            adults lived in states with the time limit statewide in 
            2017.
---------------------------------------------------------------------------
    \172\ Ibid.
---------------------------------------------------------------------------
        The distribution of childless adults essentially flipped 
            between these two groups of states between 2010 and 2017, 
            as the overall number of areas waived declined and 
            childless adults became more concentrated in states with 
            waivers and much less so in areas without waivers.
        It is not clear if the Department incorporated these factors 
            into its analysis, or assumed a more static relationship. 
            The complexity of analyzing childless adults in waived 
            areas raises the question of why the Department chose this 
            particular statistic to measure waiver coverage, 
            particularly given that it provided little explanation of 
            some of these assumptions behind this calculation. Without 
            more information, it is difficult to evaluate how relevant 
            this statistic is to their overall point, which is to 
            measure childless adults in a waived area as a measure of 
            the proposed rule's effect.

     Similarly, as we explain in our comments on the 
            Regulatory Impact Analysis in Chapter 11, the term 
            ``ABAWD'' lacks specificity, particularly when describing 
            changes in waiver coverage. The data do not allow us to 
            tell which of the larger group of adults without dependent 
            children, ages 18 to 49, without disabilities, might be 
            exempt from the time limit, so many of these adults are not 
            subject to the time limit. Others are only subject to the 
            time limit if they live in an area without a waiver. When 
            fewer areas are waived, more of these adults will be 
            subject to the time limit and lose benefits, and the 
            overall number of these adults participating in SNAP will 
            decline. Therefore, when the Department describes how 11 
            percent of ``ABAWDs'' would live in a waived area with a 
            seven percent floor and 24 percent would live in a waived 
            area with a six percent floor, it is unclear if the 
            Department is considering the decline in overall childless 
            adults participating in SNAP that would occur with the 
            reduction in waivers.

     As we also discuss in our comments on the Regulatory 
            Impact Analysis in Chapter 11, it is unclear why the 
            Department used the number of participants in non-public 
            assistance households in the FNS-388 form as a proxy for 
            childless adults in estimating the share of ``ABAWDs'' 
            living in waived areas under different scenarios.

    Instead of providing evidence that the Department's proposal will 
interpret the statute in a more effective manner by improving on the 
measurement of jobs available for low-income childless adult SNAP 
participants, the Department instead uses a confusing and unexplained 
metric to support its proposal, the share of childless adults living in 
an area covered by a waiver. This metric is seemingly unrelated to the 
intent of the statute. Because the Department provided little evidence 
to support the assumptions made in estimating this metric and to 
explain why it is relevant to the underlying law, it is impossible to 
provide more detailed discussion.
J. Proposed Alternative Unemployment Rate Floors Also Problematic
    In addition to the Department's preferred unemployment rate floor 
of seven percent, the Department also sought comment on unemployment 
rate floors of six or ten percent. Both of these proposals are flawed, 
demonstrating why selecting a specific unemployment rate floor to 
reflect available jobs for this population is a misguided approach.
Department's Proposed Six Percent Floor Demonstrates Why No Specific 
        Unemployment Rate Floor Is Appropriate
    The Department explains that a six percent unemployment rate floor 
would both be consistent with Labor Surplus Areas and bears a similar 
relationship to what it erroneously considers to be the natural rate of 
unemployment, stating ``As previously noted, the ``natural rate of 
unemployment'' generally hovers around five percent, meaning that 20 
percent above that rate is 6.0 percent.'' \173\ The Department 
therefore at least provides evidence that is somewhat more consistent 
with current standards such as relying on Department of Labor criteria, 
though undermines its seven percent unemployment rate proposal, for 
which it does not provide any such evidence.
---------------------------------------------------------------------------
    \173\ NPRM, p. 984.
---------------------------------------------------------------------------
    This unemployment rate floor would still exclude many areas where 
childless adult SNAP participants face considerably higher unemployment 
or underemployment rates and where they will not have access to jobs, 
however. As stated above, even at five percent unemployment rates, 
black and Latino workers nationally face unemployment rates of 9.6 
percent and seven percent, respectively, and underemployment rates of 
14.9 percent and 13 percent, respectively. Some local metropolitan 
areas had 2013-2017 average unemployment rates below six percent, but 
unemployment rates for sub-populations well above 14 percent, the 
unemployment rate of the Great Depression. (See Tables in Section E 
above.) For example, Monroe, MI, had an unemployment rate of 5.9 
percent, but workers without a high school degree faced unemployment 
rates of 16.7. In Fort Wayne, Indiana, the area unemployment rate was 
5.9 percent, but workers with a disability had unemployment rates of 14 
percent. In the Pittsburgh, PA metro area, while the unemployment rate 
was 5.7 percent, the unemployment rate among African American workers 
was 14.1 percent. Again, these figures demonstrate why it is impossible 
to set a specific unemployment rate threshold at which it can be 
reasonably assured that childless adults subject to the time limit can 
readily find a job with steady hours. Evidence shows that this group is 
likely to face unemployment rates much higher than their local area, 
and the unemployment rate floor is an inadequate proxy to measure jobs 
available to them.
    The Department's suggestion that a ten percent unemployment rate 
floor is in any way a reasonable proposal highlights many of the 
internal inconsistencies and inadequate explanations in this proposed 
rule.
Ten Percent Floor Inconsistent with Congressional Intent and Based on 
        Obscure and Inconsistent Reasoning
    Congress clearly designated a ten percent unemployment rate as one 
way for a state to qualify for a waiver, and a second, more flexible 
and targeted criterion of ``insufficient jobs'' as an alternative to 
demonstrating a ten percent unemployment rate. Had Congress intended 
for ten percent unemployment to be the only way for a state to qualify 
for a waiver, it would not have included an alternative. This proposal 
therefore runs afoul of Congressional intent.
    The Department also ignores the LSA standard's ten percent ceiling, 
demonstrating how its reasoning is inconsistent. As explained 
elsewhere, the Department picks and chooses when and how it will aim to 
be consistent with the DOL's approach in assessing unemployment. The 
Department of Labor clearly considers ten percent to be a sufficiently 
high level of unemployment that an area with ten percent unemployment 
over 24 months demonstrates a surplus of labor. The Department ignores 
this fact by proposing a ten percent unemployment rate floor for the 20 
percent standard, while also citing the LSA standard as support for the 
unemployment rate floor concept in general.
    As with the proposed seven percent floor, the Department suggests 
that the ten percent standard would achieve its goal of greatly 
curtailing waiver coverage, which as explained above, is a goal not 
aligned with the intent of the underlying statute and for which it does 
not offer a transparent rationale. The Department states, ``the 
Department estimates that a ten percent floor would reduce waivers to 
the extent that approximately two percent of ABAWDs would live in 
waived areas.'' \174\ While the Department may consider reducing the 
population of ``ABAWDs'' living in waived areas a priority, the 
Department provides little explanation of how this priority relates to 
the underlying statute and identifies areas lacking jobs for childless 
adults. It also provides little transparency with regards to the 
assumptions used in estimating the effects of these unemployment rate 
floors on the population living in waived areas.
---------------------------------------------------------------------------
    \174\ NPRM, p. 984.
---------------------------------------------------------------------------
Ten Percent Floor Would be Duplicative of Existing Ten Percent Criteria
    This proposal would also be largely duplicative of existing 
criteria. The Department does discuss how the time frame used would be 
different from the existing regulations regarding waivers based on ten 
percent unemployment rates: ``the ten percent unemployment floor would 
be attached to the 20 percent standard, which would mean an area would 
require an average unemployment rate 20 percent above the national 
average for a recent 24 month period and at least ten percent for the 
same period; the other similar, but separate standard requires an area 
to have an average unemployment rate of over ten percent for a 12 month 
period.'' \175\ Ten percent unemployment is extremely high at the 
national level. Only during a deep recession, such as immediately 
following the Great Recession, would 20 percent above the national 
average be close to or above ten percent, which would require the 
national average to be at least 8.4 percent for a 24 month period. 
Since the BLS began tracking monthly unemployment statistics since 
1948, out of 830 total 24 month periods there have been only 64 24 
month periods when the national average would have met this standard. 
There were 28 periods from November 1980 through January 1985, during 
and following the 1981-1982 recession, and 36 24 month periods around 
the Great Recession of 2007-2009 and its long, slow recovery, from May 
2008 through March 2013.
---------------------------------------------------------------------------
    \175\ NPRM, p. 984.
---------------------------------------------------------------------------
    For most of the time barring these prolonged economic crises, then, 
this regulation would simply extend the time frame for demonstrating 
ten percent unemployment, essentially eliminating the 20 percent 
standard most of the time. There may be some areas that are recovering 
from a deep economic shock that have more recent unemployment rates 
just below ten percent, but unemployment over the past 2 years high 
enough over that rate to nudge the average up above ten percent. 
Because ten percent is such a high level of unemployment, however, it 
is unlikely that many areas would qualify that would not have otherwise 
qualified under the ten percent criterion.
    For example, we analyzed all 12 month time periods that a state 
could use to examine waiver eligibility based on either having a 12 
month unemployment rate over ten percent or a 24 month unemployment 
rate 20 percent above the national average but at least ten percent 
from 2008 to 2019. With the exception of the years capturing peak 
unemployment rates immediately following the Great Recession, from 2014 
to 2016, when capturing a longer time frame allowed for more months 
during peak unemployment, only a handful of counties would qualify 
under the 24 month average but not the 12 month average. (Even during 
those years from 2014-2016, fewer than 150 counties, or less than five 
percent of counties, would have qualified under the 24 month but not 
the 12 month standard.)
    It is not clear if the Department more fully considered the 
practical differences between these measures, such as comparing how 
many areas would have qualified in past years based on having 12 months 
of ten percent unemployment or 24 months of ten percent unemployment. 
Its rationale for essentially replacing the 20 percent standard, which 
measures high unemployment relative to the national average, with 
additional criteria for the ten percent standard, is therefore not 
transparent.
Department's Alternative Floors Highlight Arbitrary Choice of Seven 
        Percent Floor
    The Department's discussion of alternate unemployment rate floors 
also demonstrates how its proposed floor of seven percent is an 
arbitrary figure. When proposing the six percent and ten percent floors 
as alternatives to the seven percent floor, the Department gives little 
discussion of the relevance to these floors to the underlying statute, 
which is to identify areas where individuals subject to the time limit 
do not have access to enough jobs. For example, the Department could 
have provided economic evidence that indicates how these specific 
unemployment rates relate to job availability for childless adult SNAP 
participants. The only such discussion Department includes is when it 
explains a relationship between the natural rate of unemployment and 
the six percent floor by stating that it the six percent floor is 
roughly 20 percent above five percent, which the Department 
inaccurately states is consistent with the ``natural rate of 
unemployment'' concept.\176\ The Department therefore uses no economic 
discussion to support its proposed seven percent floor but gives some 
discussion to explain the six percent floor, which is its alternate 
proposal. The discussion of the relationship of the six percent floor 
to the natural rate of unemployment therefore undermines the 
Department's proposal for the seven percent floor, as it demonstrates 
that the Department either did not consider any economic evidence, or 
did not provide any evidence to allow us to meaningfully comment.
---------------------------------------------------------------------------
    \176\ NPRM, p. 984.
---------------------------------------------------------------------------
    The only discussion the Department gives to justify choosing the 
seven percent floor (or to support the alternative of ten percent) is 
to limit the share of the population covered by a waiver, which as 
discussed above, is not what Congress intended when creating the waiver 
authority. Without any discussion to explain how the seven percent 
unemployment rate is an appropriate measure of available jobs for the 
individuals subject to the time limit, the seven percent floor appears 
to be a completely arbitrary choice.
    The Department proposes alternative unemployment rate floors for 
the ``20 percent standard.'' These proposed floors would also be 
problematic, as would any specific unemployment floor, because it is 
impossible to demonstrate that an area with an unemployment rate below 
a specific threshold lacks jobs for the individuals subject to the time 
limit.
K. Conclusion: Proposal for Unemployment Rate Floor Is Deeply Flawed
    The Department proposes to change one of the most frequently used 
standards for waiver approval, the ``20 percent standard,'' to require 
a minimum unemployment rate. With this proposal, a state could request 
a waiver with an unemployment rate 20 percent above the national 
average for a 24 month period only if it was above this unemployment 
rate floor. The Department proposed an unemployment rate floor of seven 
percent, but also sought input on proposed floors of six or ten 
percent.
    The Department provides little economic evidence to support the 
claim that imposing this floor would better interpret the statute, 
which allows states to request waivers for areas that lack ``a 
sufficient number of jobs to provide employment'' for the individuals 
subject to the time limit. The Department instead appears to work 
backwards from its stated goal of applying the time limit to more SNAP 
participants by limiting waivers, and proposes an unemployment rate 
floor as a means of achieving this goal. The Department does not 
explain how this goal relates to the purpose of the law it is 
interpreting, or how the specific floors it proposes would better 
reflect available jobs for participants.
    Research shows that the childless adults who may be subject to the 
time limit if they are not exempt or living in a waived area tend to 
have many characteristics that are associated with higher unemployment 
rates. The majority have lower levels of educational attainment, they 
are disproportionately people of color, many have health conditions or 
barriers such as unstable housing that limit their ability to work, and 
many likely experience spatial mismatch and lack access to the jobs 
that are available in their communities. Because of these features, it 
is difficult to find a labor force metric that accurately portrays the 
job opportunities available to these individuals. Current regulations 
allow states to show that an area has elevated unemployment compared to 
the national average. The current ``20 percent standard'' therefore 
already disqualifies many areas with unemployment similar to or below 
the national average where there are not enough jobs for these 
individuals to find employment that are not reflected in the 
unemployment rate.
    While current regulations could be improved, this proposal would 
substantially worsen the existing inadequacies. The proposal would 
require an area has an unemployment rate consistent with weak labor 
markets for the overall labor force, seven percent, to qualify for a 
waiver. Given that the unemployment rates for the group of these 
individuals are likely substantially higher than their area, this 
proposal would disqualify many areas where individuals face much higher 
rates than six or seven percent unemployment. The proposal would not 
align with Congressional intent, which purposefully did not specify a 
specific unemployment rate to signify that an area lacks jobs in 
recognition that the unemployment rate cannot capture job availability 
for this specific group.
    States frequently request waivers based on the current ``20 percent 
standard,'' given that data are readily available and consistent across 
states, and FNS standards for approval are transparent and consistently 
applied. While the current standard falls short of accurately 
reflecting jobs available for this population, we believe the proposed 
unemployment rate floor would be inconsistent with the intent of 
Congress and would make the current criteria substantially less 
effective at measuring insufficient jobs. We therefore urge FNS to drop 
the unemployment rate floor proposal, keep the 20 percent standard as 
it is, and explore metrics based on evidence that would more 
effectively reflect jobs available to the population in recognition of 
the likely substantially higher unemployment rates they face.
Chapter 4. Dropping Several Key Criteria From Waiver Criteria Is 
        Inconsistent With the Statute
    The NPRM proposes several significant changes to longstanding SNAP 
policy that would restrict states to one limited measure of labor 
market conditions, the unemployment rate, when providing evidence of 
lack of sufficient jobs. It would eliminate the ability of states to 
use valuable, readily available labor market indicators, such as a low 
and declining employment-to-population ratio, a lack of jobs in a 
declining industry, or an academic study or other publication(s) that 
describes an area's lack of jobs. The NPRM fails to provide reasons for 
limiting states' ability to use widely accepted labor market measures 
to support requests for waivers, to discuss the implications of relying 
on a single measure of labor market conditions, or to acknowledge the 
valuable information provided by other measures. Without knowing what 
evidence justifies this change in longstanding policy and an adequate 
discussion of alternative methods for assessing labor market 
conditions, it is impossible to assess the potential impact of the 
changes on SNAP participants and their ability to achieve self-
sufficiency. The sections below provide an overview of existing 
statutes, regulations, and guidance, address limitations of the general 
unemployment rate, and discuss alternative measures of labor market 
conditions.
A. Current Statute, Regulations, and Guidance Acknowledge That There Is 
        No Perfect Measure of an Area's ``Lack of Sufficient Jobs''
    According to the statute, a state may waive the applicability of 
the work requirement ``to any group of individuals in the state if the 
Secretary makes a determination that the area in which the individuals 
reside has an unemployment rate above 10% or does not have a sufficient 
number of jobs to provide employment for the individuals.'' The statute 
does not limit the type of information that can be used to support a 
claim of lack of sufficient jobs.
    According to the current rule (7 CFR  273.24 (f)(2)(ii)), states 
are not limited to using unemployment rates to support a claim of lack 
of sufficient jobs. States may provide evidence that an area has a low 
and declining employment-to-population ratio, has a lack of jobs in 
declining occupations or industries, or is described in an academic 
study or other publications as an area where there are lack of jobs. In 
the preamble to the current rule, FNS stated that ``State agencies 
could submit requests with no limit on the supporting documentation, 
and every request would be weighed on its own individual merits.'' The 
final rule included a non-exhaustive list of the kinds of information a 
state agency may submit to support a claim of ``lack of sufficient 
jobs.''
    Below are excerpts from FNS guidance and rulemaking that give 
states flexibility in using other types of data to provide evidence of 
a lack of sufficient jobs, acknowledging that unemployment rates may 
not adequately capture the local labor market prospects of individuals 
subject to the time limit.

   December 3, 1996 guidance: According to FNS, the statute 
        ``recognizes that the unemployment rate alone is an imperfect 
        measure of the employment prospects of individuals with little 
        work history and diminished opportunities. It provides states 
        with the option to seek waivers for areas in which there are 
        not enough jobs for groups of individuals who may be affected 
        by the new time limits.'' \177\
---------------------------------------------------------------------------
    \177\ USDA, ``Guidance for states Seeking Waivers for Food Stamp 
Limits,'' December 3, 1996.

      ``Lack of jobs due to lagging job growth. Job seekers may have a 
        harder time finding work in an area where job growth lags 
        behind population growth. A falling ratio of employment-to-
        population may be an indicator of an adverse job growth rate. 
        When the number of jobs in an area grows more slowly than the 
        working age population, the local economy is not generating 
        enough jobs.
      ``The employment-to-population ratio complements measures of 
        unemployment by taking into account working age persons who may 
        have dropped out of the labor force altogether. The ratio can 
        be computed by dividing the number of employed persons in an 
        area by the area's total population. A decline in this ratio 
        over a period of months could indicate an adverse job growth 
        rate for the area . . .
      ``Lack of jobs in declining occupations or industries. Employment 
        markets dominated by declining industries could lead to the 
        presence of large numbers of people whose current job skills 
        are no longer in demand. This can be especially true in 
        smaller, rural areas where the loss of a single employer can 
        immediately have a major effect on local job prospects and 
        unemployment rates.''

   1999 proposed rule: In the preamble, FNS noted that ``the 
        legislative history does not provide guidance on what types of 
        waivers the Department should approve under this standard, and 
        there are no standard data or methods to make the determination 
        of the sufficiency of jobs. States requesting waivers are 
        therefore free to compile evidence and construct arguments to 
        show that in a particular area, there are not enough jobs for 
        individuals who are affected by the time limit.'' \178\ FNS 
        reiterated that one possible indicator that an area has 
        insufficient jobs is a falling ratio of employment-to-
        population, but that ``no particular approach is required.''
---------------------------------------------------------------------------
    \178\ Food Stamp Program: Personal Responsibility Provisions of the 
Personal Responsibility and Work Opportunity Reconciliation Act of 
1996; 64 Federal Register 242, p. 70946 (December 17, 1999) (to be 
codified at 7 CFR pts. 272 and 273).

   August 2006 guidance: ``Waivers may also be submitted based 
        on the following criteria: (1) areas having a low and declining 
        employment-to-population ratio; (2) areas having a lack of jobs 
        in declining occupations or industries; (3) areas described in 
        an academic study or other publications as an area where there 
        is a lack of jobs. The state may submit whatever data it deems 
        appropriate to support requests based on this data. FNS will 
        evaluate the data and determine if it is acceptable to justify 
        a waiver.'' \179\
---------------------------------------------------------------------------
    \179\ USDA, ``Guidance on Requesting ABAWD Waivers,'' August 2006.

   December 2016 guidance: FNS provided additional detail on 
        ``other potential types of waiver requests'' beyond those based 
        on the LSA designation or unemployment rates: \180\
---------------------------------------------------------------------------
    \180\ USDA, ``Guide to Supporting Requests to Waive the Time Limit 
for Able-Bodied Adults without Dependents (ABAWD),'' December 2, 2016.

      A low and declining employment-to-population ratio. Employment-
        to-population (ETP) ratio can be a meaningful economic 
        indicator for an area where the unemployment rate may not 
        provide a complete picture of the labor market due to people 
        leaving the workforce--but demographic changes, such as an 
        aging population, can influence these data. Historically, low 
        and declining ETP data have been used successfully to waive 
        Indian reservations or Tribal lands where unemployment 
        statistics and other economic data are limited or unavailable. 
        ETP data can also be used to request waivers for non-Tribal 
        areas, such as counties, but it is uncommon because BLS 
        unemployment data is readily available for these areas. 
        Therefore, FNS has approved requests based on ETP data for non-
        Tribal areas, such as rural counties, on a limited basis when 
---------------------------------------------------------------------------
        the state has demonstrated that the area's ETP ratio is:

     Low: at least one percentage point below the national 
            average for the most recent year of the reference period;

     Declining: best demonstrated by a decline year after 
            year;

     Covering at least a 4 year reference period, ending no 
            earlier than 2 years prior to the year in which the waiver 
            is effective; and

     Complemented by a recent 24 month unemployment rate at 
            least ten percent above the national average in the 
            requested area.

      A lack of jobs in declining occupations or industries. Employment 
        markets dominated by declining industries could impact large 
        numbers of people whose current job skills are no longer in 
        demand. This can be especially true in smaller, rural areas in 
        which the loss of a single job provider, such a major 
        manufacturing plant or mining industry, can have a major effect 
        on local job availability. The state might consider providing 
        studies, reports, or other analysis from credible sources in 
        demonstrating that an area has a lack of jobs in declining 
        occupations or industries.
      Description in an academic study or other publication as an area 
        where there is a lack of jobs. The state might consider 
        providing an academic study or other credible publication that 
        documents a lack of sufficient jobs in an area.
      The state may submit whatever data or evidence it deems 
        appropriate to support these types of requests. FNS will 
        evaluate such requests on a case-by-case basis and will approve 
        those that provide compelling support of a lack of sufficient 
        jobs in the area. FNS strongly encourages the state to work 
        closely with its regional offices for technical assistance if 
        it is considering requesting a waiver based on the less common 
        support mentioned above.
B. The Proposed Rule Would Restrict the Evidence to Support Lack of 
        Sufficient Jobs to a Single, Imperfect Measure of Labor Market 
        Conditions
    The NPRM says the proposed core standards would not include other 
labor market information, such as a low and declining employment-to-
population ratio, a lack of jobs in a declining industry, or an 
academic study or other publication(s) that describes an area's lack of 
jobs. It would eliminate the ability of states to support a waiver 
request using other available information about the labor market, 
unless BLS unemployment data for the area is limited or unavailable, 
such as a reservation area or U.S. territory. FNS proposes to eliminate 
these other criteria on the grounds that they are ``rarely used, 
sometimes subjective, and not appropriate when other more specific and 
robust data are available,'' but does not provide further 
substantiation of this claim.
    The proposed rule would replace an approach that allows for 
multiple measures to capture labor market conditions experienced by 
individuals subject to the time limit with a single, limited, and 
imperfect measure, the unemployment rate. FNS has stated in its 
guidance that using the unemployment rate is an imperfect measure for 
the job prospects for individuals subject to the time limit. Labor 
market researchers routinely use other labor market measures in 
addition to, or instead of, the unemployment rate, such as the 
employment-to-population ratio.
Other Measures, Including the Employment-To-Population Ratio, Provide 
        Important Information About Labor Market Conditions That the 
        General Unemployment Rate Does Not
    The employment-to-population ratio is a well-defined and widely 
used measure that is far from subjective. The employment-to-population 
ratio is the proportion of the civilian noninstitutional population 
aged 16 and over that is employed. As the 1996 guidance describes, 
employment data for areas is available from BLS. Population estimates 
for areas are available from the Bureau of Census. The calculation of 
the employment-to-population ratio is a standard BLS procedure, which 
is a measure it reports on a regular basis at the regional and state 
level.\181\ In many instances, researchers use employment-to-population 
ratio as a more appropriate measure for labor market conditions for 
low-skill workers who face serious barriers to employment.
---------------------------------------------------------------------------
    \181\ See for instance: https://www.bls.gov/news.release/
srgune.nr0.htm.
---------------------------------------------------------------------------
    Current regulations allow states to demonstrate that an area lacks 
sufficient jobs by showing that it has a low and declining employment-
to-population ratio. The rule proposes eliminating this criterion as a 
means for an area to qualify for a waiver. This would be a mistake, as 
it would throw away valuable information about the state of the labor 
market and the likely availability of jobs that cannot be gleaned from 
the unemployment rate alone. The unemployment rate is the number of 
people actively looking for a job as a percentage of the labor force 
(the number of people who have a job plus the number of people who 
don't have a job but are actively looking for one). In a job market 
with limited job opportunities for any of a number of reasons, such as 
weak demand due to a national economic recession, a local business 
slump, or the closing of a major plant, there could be a number of 
people who would like to work but for reasons such as discouragement 
due to a failed job search, experience with discrimination, or a 
general sense that their job prospects are limited haven't looked 
recently enough to be counted as in the labor force but unemployed.
    These individuals are classified as ``marginally attached to the 
labor force'' and are included in broader measures of labor market 
underutilization, including the U-6 measure, which includes the 
unemployed, the marginally attached to the labor force, and those who 
are working part-time but want to be working more hours.\182\
---------------------------------------------------------------------------
    \182\ See for instance ``How the Government Measures 
Unemployment,'' Bureau of Labor Statistics, online at https://
www.bls.gov/cps/cps_htgm.htm#unemployed.
---------------------------------------------------------------------------
Figure 4.1
Job Market Indicators in the Great Recession

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    In a national recession, a local economic slump, or in localities 
with limited job opportunities, the unemployment rate can paint a very 
incomplete picture of the availability of jobs. This was illustrated 
dramatically at the national level in the Great Recession. Between the 
start of the recession in December 2007 and early 2010, the share of 
the population with a job (the employment-to-population ratio) fell 
sharply. That was mostly due to the sharp rise in the unemployment 
rate, but some of it reflected a drop in labor force participation as 
the number of people marginally attached or otherwise not in the labor 
force rose.
    The unemployment rate then began a long decline but the labor force 
participation rate continued to fall as well. As a result, the share of 
the population with a job remained depressed and did not begin to rise 
again until 2014 (see Figure 4.1, above).
    The U-6 measure of unemployment came down more slowly than the 
official unemployment rate as jobs, especially full-time jobs, remained 
scarce. Even as the unemployment rate dropped below seven percent, the 
employment-to-population ratio remained well below where it was at the 
start of the recession.
Researchers Routinely Use Employment-to-Population Ratio to Measure 
        Local Labor Market Conditions
    Researchers routinely use the employment-to-population ratio in 
addition to, or instead of, the unemployment rate to measure labor 
market conditions. According to Bartik, it is ``unclear whether the 
availability of labor is best measured by employment-to-population 
ratios or employment to labor force ratios.'' \183\ Bartik finds that 
employment-to-population ratios are more strongly related to job growth 
than employment to labor force ratios.\184\
---------------------------------------------------------------------------
    \183\ The unemployment rate can be derived from the employment to 
labor force ratio by subtracting the latter from 1.
    \184\ Timothy J. Bartik, ``How Do the Effects of Local Growth on 
Employment Rates Vary with Initial Labor Market Conditions,'' Upjohn 
Institute Staff Working Paper 09-148 (Nov. 4, 2006), pp. 1-35, https://
www.econstor.eu/bitstream/10419/64401/1/607052678.pdf.
---------------------------------------------------------------------------
    For individuals subject to the time limit, the employment-to-
population ratio may be more appropriate than the unemployment rate. 
According to Western and Pettit, for groups who are weakly attached to 
the labor market and who face significant barriers to labor force 
participation, like young men with little education, economic status is 
often measured by the employment-to-population ratio. This measure 
counts as jobless those who have dropped out of the labor market 
altogether. The unemployment rate is more restrictive and does not 
account for individuals who are not currently in the labor force.\185\ 
A study by Cadena and Kovak illustrates this approach, using 
employment-to-population ratios to estimate the probability of 
employment in the less-skilled labor market.\186\
---------------------------------------------------------------------------
    \185\ Bruce Western and Becky Pettit, ``Incarceration and Social 
Inequality,'' D#dalus Journal of the American Academy of Arts & 
Sciences (Summer 2010), pp. 8-19, https://www.mitpressjournals.org/doi/
pdf/10.1162/DAED_a_00019%20.
    \186\ Brian C. Cadena and Brian K. Kovak, ``Immigrants Equilibrate 
Local Labor Markets: Evidence From the Great Recession,'' National 
Bureau of Economic Research (August 2013), https://www.nber.org/papers/
w19272.pdf.
---------------------------------------------------------------------------
    An improved (or deteriorating) unemployment rate does not directly 
correspond to an improvement (or deterioration) of the employment 
situation, because it does not take into account changes in the labor 
force participation rate due to the movement of discouraged jobseekers 
in and out of the labor market. Only a stable participation rate allows 
for unambiguous conclusions from a rising (or falling) unemployment 
rate. Unemployed people who have been adversely affected by economic 
restructuring may give up hope of working again and withdraw from the 
labor force. Job booms may only be a boom for certain kinds of workers. 
Watson argues that a more useful indication of the quantity of 
employment in the economy is provided by employment-to-population 
ratios, which remove the confounding influence of labor force 
participation and give a more accurate indication of the amount of 
employment available to the population.\187\
---------------------------------------------------------------------------
    \187\ Ian Watson, ``Beyond the Unemployment Rate: Building a Set 
Indices to Measure the Health of the Labour Market,'' Australian 
Bulletin of Labour (September 2000), pp. 175-190, http://
www.ianwatson.com.au/pubs/health%20of%20labour%20market.pdf.
---------------------------------------------------------------------------
    Hoynes estimated the effect of local labor markets on Aid to 
Families with Dependent Children participation in California using 
several measures of labor market conditions, including unemployment 
rates, log of employment, employment-to-population ratios, and 
earnings. Results showed that higher unemployment rates, lower 
employment growth, lower employment-to-population ratios, and lower 
wage growth are associated with longer welfare spells and shorter 
periods off welfare. Models that controlled for labor market conditions 
using employment-based measures, such as employment-to-population 
ratios, performed better than unemployment rates. ``Unemployment rates 
are less desirable measures of labor market opportunities because they 
fluctuate not only with employment but also with changes in labor force 
participation.'' \188\
---------------------------------------------------------------------------
    \188\ Hilary W. Hoynes, ``Local Labor Markets and Welfare Spells: 
Do Demand Conditions Matter?'' The Review of Economics and Statistics 
(August 2000), pp. 351-368, https://gspp.berkeley.edu/assets/uploads/
research/pdf/Hoynes-RESTAT-2000.pdf.
---------------------------------------------------------------------------
    Dranove, Garthwaite, and Ody used employment-to-population ratio to 
examine the impact of the economic slowdown that began in 2007 on the 
rate of growth in health spending. They used the employment-to-
population ratio, rather than unemployment rate, because it is not 
affected by decisions to enter the labor force and instead provides a 
local measure of changes in economic activity resulting from the 
slowdown. Their results were broadly consistent with results using the 
local unemployment rate instead of employment-to-population ratio.\189\
---------------------------------------------------------------------------
    \189\ David Dranove, Craig Garthwaite, and Christopher Ody, 
``Health Spending Slowdown is Mostly Due to Economic Factors, not 
Structural Change in the Health Care Sector,'' Health Affairs (Aug. 
2014), pp. 1399-1406, https://www.healthaffairs.org/doi/pdf/10.1377/
hlthaff.2013.1416.
---------------------------------------------------------------------------
The General Unemployment Rate May Not Adequately Measure Weak Labor 
        Demand at the State and Sub-State Level
    During this period when the national employment-population ratio 
was flat, there were many local and regional labor markets where labor 
market conditions remained weak even as the general unemployment rate 
fell.
    In a 2017 speech that partially focused on the geographical 
variance of labor markets across the country (and on policies to 
ameliorate such differences), then Federal Reserve Chair Janet Yellen, 
pointed out the following: \190\
---------------------------------------------------------------------------
    \190\ Janet Yellen, ``Addressing Workforce Development Challenges 
in Low-Income Communities,'' Federal Reserve Board of Governors, March 
28, 2017, https://www.federalreserve.gov/newsevents/speech/files/
yellen20170328a.pdf.

          While the job market for the United States as a whole has 
        improved markedly since the depths of the financial crisis, the 
        persistently higher unemployment rates in lower-income and 
        minority communities show why workforce development is so 
        essential. For instance, unemployment rates averaged 13 percent 
        in low- and moderate-income communities from 2011 through 2015, 
        compared with 7.3 percent in higher-income communities . . . . 
        The challenges for workers in minority communities are even 
        greater. The average unemployment rate across all census tracts 
        where minorities made up a majority of the population averaged 
---------------------------------------------------------------------------
        14.3 percent from 2011 through 2015.

    Labor economist Danny Yagan added an important insight about the 
geographical dispersion of employment conditions following the 
historically large, negative demand shock from the Great 
Recession.\191\ As Figure 4.2 below shows, states that were harder hit 
by the downturn saw significantly larger losses in employment rates, 
even years after the recession was over. Yagan argues that his findings 
provide evidence of ``hysteresis,'' meaning lasting economic damage to 
persons and communities from periods of economic weakness. As he 
summarizes, ``These findings reveal that the Great Recession imposed 
long-term employment and income losses even after unemployment rates 
signaled recovery.'' \192\
---------------------------------------------------------------------------
    \191\ Danny Yagan, ``Employment Hysteresis from the Great 
Recession,'' NBER Working Paper No. 23844, August 2018, https://
www.nber.org/papers/w23844.
    \192\ Ibid.
---------------------------------------------------------------------------
    As shown in the next section, even at low rates of national and 
regional unemployment (meaning rates well below seven percent), there 
are areas of the country where economic weakness persists. Yagan's 
findings suggest that these areas may suffer from more lasting damage 
to workers' ability to find gainful jobs. In the context of the 
proposed rule, such dynamics speak to the importance of taking a much 
more nuanced approach to the waiver process, examining local labor 
markets from both the demand side (i.e., the extent of job 
availability, both quantity and quality) and the supply side (i.e., the 
skills and abilities of members of the local workforce to respond to 
labor demand).
Figure 4.2
State-Level Great Recession Employment Shocks and 2007-2015 Employment 
        Rate Changes
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Note: Yagan defines employment shocks as the sum of state-
        level employment growth forecast errors for 2008 and 2009. 
        These forecast errors represent the difference between each 
        state's actual employment growth and its predicted employment 
        growth based on pre-recession trends. Values on the x-axis 
        represent the inverse of 2007-2009 employment growth forecast 
        errors.

    Geographical Variation of Weak Labor Demand, Even at Low 
Unemployment

    Echoing the Yagan findings referenced above, a recent paper by 
Austin, et al., illustrates that labor demand, particularly for low-
wage workers, varies significantly from place to place.\193\ In their 
recent analysis of regional disparities, these authors find pockets of 
persistently weak labor markets across America, citing what they label: 
``a hardening of America's geographic divisions.'' Their paper 
identifies three findings particularly germane to the shortcomings of 
the new rule: ``the decline of geographic mobility,'' ``increased 
sorting by skills across space,'' and ``persistent pockets of non-
employment.'' The combination of these three negative developments 
imply a larger share of lower-wage workers stuck in various locations 
without enough work. We find these disparities very much present in the 
labor market over the current expansion, even at historically low rates 
of unemployment.
---------------------------------------------------------------------------
    \193\ Benjamin Austin, Edward Glaeser, and Lawrence Summers, ``Jobs 
for the Heartland: Place-Based Policies in 21st Century America,'' 
March 8, 2018, https://www.brookings.edu/wp-content/uploads/2018/03/
AustinEtAl_Text.pdf.
---------------------------------------------------------------------------
    Many labor economists consider the prime-age employment rate to be 
a proxy for labor demand. As part of their ``Distressed Community 
Index,'' the Economic Innovation Group (EIG) provides county-level data 
on non-employment rates, or 1 ^ the employment rate. Thus, higher non-
employment rates correspond to weaker labor demand.
    Between 2012 and 2016, the average national non-employment rate for 
prime-age workers was 23 percent, meaning 77 percent of such workers 
had jobs. EIG's data, to which we appended county-level unemployment 
data from the BLS, reveal that in counties with unemployment rates 
between 6.5 and 7.5 percent, the average non-employment rate for prime-
age adults was about 34 percent, more than ten percentage points above 
the national average.\194\ Note that even at the worst of the Great 
Recession, the non-employment rate peaked at about 25 percent.\195\
---------------------------------------------------------------------------
    \194\ To be clear, employment (and non-employment) rates are 
mechanically correlated as higher unemployment means lower employment. 
Our focus here, however, is on the levels of these variables and what 
they imply for labor demand.
    \195\ EIG and BLS have slightly different definitions for ``prime-
age''--BLS uses adults 25-54, and EIG uses adults 25-64.
---------------------------------------------------------------------------
    The scatterplot in Figure 4.3 below shows the correlation between 
un- and non-employment at the county level. Note that the scatterplot 
expands at higher unemployment, implying greater dispersion of labor 
demand across counties at higher rates of unemployment. For example, 
the plot shows that at ten percent county unemployment, there are some 
counties with quite low non-employment rates and some with very high 
rates. This dispersion further underscores the need to avoid the single 
number approach proposed in the rule. Second, the scatterplot shows 
that at seven percent unemployment, as noted above, non-employment is 
above 30 percent.
Figure 4.2
County-Level Unemployment and Non-Employment

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Using the same procedure employed in the previous section, a 
regression of county-level non-employment rates on the county's 
unemployment rate, predicts that at five, seven, and ten percent 
unemployment rates, county-level non-employment rates would range from 
27 to 41 percent. (Table 4.1.) In other words, such high levels of non-
employment demonstrate significant labor market slack at the jobless 
rates proposed by the Department.

                                Table 4.1
          Predicted County-Level Prime-Age Non-Employment Rate
------------------------------------------------------------------------
                                      Predicted Prime-Age Non-Employment
          Unemployment Rate                          Rate
------------------------------------------------------------------------
                     5%                                   27%
                     7%                                   33%
                    10%                                   41%
------------------------------------------------------------------------
Note: County-level prime-age non-employment rates are predicted by
  regressing county-level unemployment rates on non-employment rates.

    The Federal Reserve recognized that there was still considerable 
``slack'' in the labor market not captured by the unemployment rate and 
kept short-term interest rates effectively at zero until December 2015 
before it began to raise them cautiously in small increments.
C. Information About Declining Occupations or Industries Can Help 
        Identify Smaller Areas Experiencing a Lack of Sufficient Jobs
    According to current regulations and guidance, states can support a 
claim of lack of sufficient jobs by providing evidence of a lack of 
jobs in declining occupations or industries. This can be especially 
true in smaller, rural areas in which the loss of a single job 
provider, such a major manufacturing plant or mining industry, can have 
a major effect on local job availability. In the December 1996 
guidance, FNS suggested that states could use BLS monthly data 
published in the ``Employment and Earnings'' report on state and sub-
state employment figures by major industry.\196\ A declining trend 
within a particular industry or sector may be taken as evidence of 
declining employment prospects for persons with experience in or skills 
appropriate to that sector.
---------------------------------------------------------------------------
    \196\ https://www.bls.gov/opub/ee/home.htm.
---------------------------------------------------------------------------
    Although states have not frequently used occupation or industry 
employment data to support claims of lack of sufficient jobs, FNS has 
approved them on a limited case-by-case basis. For example, FNS 
approved waivers for a county (Polk) in Arkansas and a county (Coos) in 
New Hampshire that were significantly affected by plant closures during 
the recession that started in 2001. The state agencies provided 
evidence of the adverse labor force impacts due to a major factory or 
plant closing, such as the number of workers affected by layoffs and 
rapidly increasing unemployment rates (ten percent and higher) over a 
short period of time. The impact of a plant closure may not show up in 
24 month unemployment rates until several months, or even a year, have 
passed. Information indicating the decline of particular industries, 
such as significant plant closures, gives states the ability to quickly 
adapt their waiver policy to respond to rapidly deteriorating labor 
market conditions.
D. Eliminating Criteria of Three-Month Average Unemployment Rate Over 
        Ten Percent and Historical Seasonal Unemployment Rate Over Ten 
        Percent Is Inconsistent With the Statute
    The proposed rule would restrict states' ability to use an 
unemployment rate over ten percent as the basis for waiver approval. It 
would limit the use of the criterion of a recent 3 month average 
unemployment rate over ten percent to ``exceptional circumstances'' and 
eliminate the criterion of an historical seasonal unemployment rate 
over ten percent.\197\ This would leave just one criterion--having a 12 
month average unemployment rate over ten percent--as the basis for 
approval using an average unemployment rate over ten percent. These 
changes are inconsistent with the statute and regulations that clearly 
establish that areas with an unemployment rate over ten percent qualify 
for a waiver. If the Department proceeds to publish a final rule it 
must reject these changes to be consistent with the statute.
---------------------------------------------------------------------------
    \197\ NPRM, p. 985, 987.
---------------------------------------------------------------------------
    According to the statute, a state may waive the applicability of 
the work requirement ``to any group of individuals in the state if the 
Secretary makes a determination that the area in which the individuals 
reside has an unemployment rate above 10% or does not have a sufficient 
number of jobs to provide employment for the individuals.'' \198\ The 
statute clearly establishes the ten percent unemployment rate criterion 
as a basis for approval. The statute does not specify requirements 
regarding the duration of time that an area must have an unemployment 
rate above ten percent.
---------------------------------------------------------------------------
    \198\ Food and Nutrition Act, 7 U.S.C.  2015(o)(4). This language 
is identical to the language in P.L. 104-193, PRWORA.
---------------------------------------------------------------------------
Three-Month Average Unemployment
    In guidance issued in December 1996 and then reinforced in the 
preamble of the 1999 proposed rule,199-200 the Department 
stated that it would not require a 12 month average to approve a waiver 
because of two shortcomings. ``A 12 month average will mask portions of 
the year when the unemployment rate rises above or falls below ten 
percent. In addition, requiring a 12 month average before a waiver 
could be approved would necessitate a sustained period of high 
unemployment before an area became eligible for a waiver.'' To address 
these shortcomings and to ensure that waivers are granted as quickly as 
possible where needed, the Department explained that ``states have 
several options. First, a state might opt to use a shorter moving 
average. A moving average of at least 3 months is preferred. In periods 
of rising unemployment, a 3 month average provides a reliable and 
relatively early signal of a labor market with high unemployment. A 
state might also consider using historical unemployment trends to show 
that such an increase is not part of a predictable seasonal pattern to 
support a waiver for an extended period (up to 1 year).'' \201\
---------------------------------------------------------------------------
    \199\ USDA, ``Guidance for states Seeking Waivers for Food Stamp 
Limits,'' December 3, 1996.
    \200\ Food Stamp Program: Personal Responsibility Provisions of the 
Personal Responsibility and Work Opportunity Reconciliation Act of 
1996; 64 Federal Register 242 (December 17, 1999) (to be codified at 7 
CFR pts. 272 and 273).
    \201\ USDA, ``Guidance for states Seeking Waivers for Food Stamp 
Limits,'' December 3, 1996.
---------------------------------------------------------------------------
    In the preamble to the proposed rule, the Department expressed its 
preference that waivers reflect current economic conditions.\202\ Yet 
by eliminating the ability of states to use a recent 3 month average 
unemployment rate over ten percent as the basis for waiver approval, it 
is eliminating one of the criteria that most closely aligns with 
current economic conditions and signals deteriorating labor market 
conditions in an area.
---------------------------------------------------------------------------
    \202\ NPRM, p. 986.
---------------------------------------------------------------------------
Historical Seasonal Unemployment
    In guidance issued in December 1996 and in the preamble of the 1999 
proposed rule,203-204 the Department confirmed the 
applicability of waivers to ``areas with predictable seasonal 
variations in unemployment.'' The Department provided a detailed 
example:
---------------------------------------------------------------------------
    \203\ USDA, ``Guidance for states Seeking Waivers for Food Stamp 
Limits,'' December 3, 1996.
    \204\ Food Stamp Program: Personal Responsibility Provisions of the 
Personal Responsibility and Work Opportunity Reconciliation Act of 
1996; 64 Federal Register 242 (December 17, 1999) (to be codified at 7 
CFR pts. 272 and 273).

          States may use historical trends to anticipate the need for 
        waivers for certain periods. For example, if the pattern of 
        seasonal unemployment is such that an area's unemployment rate 
        typically increases by two percentage points in January, 
        February, and March, and the area's unemployment rate is 
        currently nine percent, a state may request a waiver for this 
        area based on its current rate and historical trends. The 
        period covered by the waiver will then coincide with the period 
        of high unemployment.
Aligning the Period Covered by the Waiver and the Period of Projected 
        High Unemployment Does Not Require Data of a Particular 
        Duration
    The 2001 final rule codified criteria related to unemployment rates 
over ten percent at 7 CFR  273.24(f)(2)(i) and provided flexibility to 
meet these criteria using data of varying duration. ``To support a 
claim of unemployment over ten percent, a state agency may submit 
evidence that an area has a recent 12 month average unemployment rate 
over ten percent; a recent 3 month average unemployment rate over ten 
percent; or an historical seasonal unemployment rate over ten 
percent.''
    The intent of current regulations was to align the period covered 
by the waiver to the period when unemployment is high, rather than 
designate an arbitrary duration requirement:

          Therefore, the Department is proposing that in general, the 
        duration of a waiver should bear some relationship to the 
        documentation provided in support of the waiver request. FNS 
        will consider approving waivers for up to 1 year based on 
        documentation covering a shorter period, but the state must 
        show that the basis for the waiver is not a seasonal or short 
        term aberration.\205\
---------------------------------------------------------------------------
    \205\ Food Stamp Program: Personal Responsibility Provisions of the 
Personal Responsibility and Work Opportunity Reconciliation Act of 
1996; 64 Federal Register 242 (December 17, 1999) (to be codified at 7 
CFR pts. 272 and 273).

    In the preamble of the NPRM, the Department arbitrarily adds a 
duration requirement of 12 months to the ten percent criterion.\206\ 
Only areas with a recent, 12 month average unemployment rate over ten 
percent would be considered for approval. Under the proposed rules, the 
Department may approve a waiver for an area with a recent 3 month 
average unemployment rate over ten percent only if an ``exceptional 
circumstance has caused a lack of sufficient number of jobs.'' \207\
---------------------------------------------------------------------------
    \206\ NPRM, p. 983.
    \207\ NPRM, p. 985, 992.
---------------------------------------------------------------------------
    The Department does not discuss the rationale for restricting the 
ten percent criterion to a 12 month duration. It does not adequately 
explain what represents an exceptional circumstance and what economic 
measures might signal this circumstance. It does not discuss what 
economic measures a state might be able to use as an alternative to the 
3 month average unemployment rate, which it has described as a 
``reliable and relatively early signal of a labor market with high 
unemployment'' in past guidance.
    The Department argues for eliminating the historical unemployment 
rate criterion because it does not demonstrate ``prolonged'' lack of 
sufficient jobs, that it is ``limited to a relatively short period of 
time each year,'' and that it is ``cyclical rather than indicative of 
declining conditions.'' \208\ The Department acknowledges that, by 
definition, historical seasonal unemployment is contradictory with 
prolonged duration. Rather than drop the newly introduced and 
contradictory requirement on duration (which is inconsistent with 
existing statute and regulation), the Department argues for the 
elimination of the historical seasonal unemployment criterion (which is 
upheld in existing statute and regulations).
---------------------------------------------------------------------------
    \208\ NPRM, p. 987.
---------------------------------------------------------------------------
    The Department also proposes to eliminate the historical seasonal 
unemployment criterion because it has not approved a waiver using this 
criterion. This is not sufficient ground for the proposed change, as 
the Department has no way of knowing if states intend to use this 
criterion in the future. To maintain consistency with the statute, we 
urge the Department to leave the regulation as is and retain the 3 
month average unemployment rate over ten percent and historical 
seasonal unemployment rate over ten percent as criteria for waiver 
approvals.
E. We Recommend Rejecting Proposed Changes That Would Ignore Important 
        Information About Labor Market Conditions Not Captured by the 
        General Unemployment Rate
    These studies and analyses illustrate how the unemployment rate 
alone may not tell the full story of how abundant or scarce jobs are in 
the labor market. FNS would be mistaken to rely solely on unemployment 
rates as the basis for demonstrating that an area has a lack of 
sufficient jobs. The unemployment rate does not account working-age 
persons who may have dropped out of the labor force altogether. Other 
labor force measures, such as the employment-to-population ratio or 
industry-specific employment data, complement unemployment rates in 
capturing the labor market conditions faced by individuals subject to 
the time limit, who often face significant barriers to labor force 
participation. In the Great Recession a low or depressed employment-to-
population ratio was often a better measure of labor market slack and 
lack of job opportunities than the unemployment rate. Thus, a low or 
falling employment-to-population ratio is a valuable indicator and data 
are available for local areas.
    The proposed rule is based on insufficient reasons to change 
current regulations by prohibiting states from using average 
unemployment rates over ten percent and other available information 
about labor market conditions, except for areas that have limited or 
unavailable unemployment data from BLS or a BLS-cooperating agency. It 
fails to discuss the reasons why it is restricting the use of average 
unemployment rates over ten percent during periods of acute or seasonal 
high unemployment or the limitations of the general unemployment rate 
in assessing the labor market conditions, particularly those faced by 
individuals subject to the time limit. It does not acknowledge the 
valuable information that will be lost if measures such as the 
employment-to-population ratio are excluded as evidence of lack of 
sufficient jobs. Given the lack of supporting information, the public 
has an insufficient opportunity to comment meaningfully on the proposed 
rule and we recommend rejecting the proposed changes to the rules.
Chapter 5. Restricting State Flexibility on Grouping Areas Is Counter 
        to Evidence
    The NPRM proposes several significant changes to longstanding SNAP 
policy that would significantly restrict state flexibility to develop 
and implement waiver policy that aligns with state operations, 
priorities, and resources. The NPRM fails to provide reasons for 
limiting states' ability to consider relevant factors when grouping 
areas covered by waivers, fails to acknowledge decades of state 
discretion in grouping areas (including statewide areas) for waivers, 
and fails to identify the data and evidence that justify the 
elimination of statewide waivers and the use of one narrow, inflexible, 
federally prescribed method for grouping areas. Without knowing what 
evidence justifies such a drastic change in longstanding policy and an 
adequate discussion of alternative methods for grouping, it is 
impossible to assess the potential impact of the changes on SNAP 
participants and their ability to achieve self-sufficiency. The 
sections below provide an overview of existing statutes, regulations, 
and guidance, and discuss factors that states consider when grouping 
areas, alternative grouping methods used by states to group areas, and 
limitations of the method for grouping proposed by the Department.
A. States Have Had Broad Discretion to Define Areas for More Than Two 
        Decades
    Since the passage of the 1996 welfare law (P.L. 104-193) and the 3 
month time limit, FNS has given states broad discretion to determine 
which geographic areas the state would like to waive from the 3 month 
time limit, including an area spanning the entire state or sub-state 
areas. While every state-defined area as a whole must meet the waiver 
eligibility criteria as set forth in 7 CFR  273.24(f), states may 
define areas that best align with local and regional labor force 
conditions, resources, and administrative needs. The Federal rules do 
not limit waivers to specific sub-state areas, such as cities or 
counties. As states define areas to request waivers for, they often 
consider a range of factors within geographic regions, such as labor 
market characteristics, job opportunities, availability of SNAP 
Employment and Training (E&T) services, housing and transportation, 
workforce and economic development resources and strategies, and SNAP 
agency administrative capacity.
    States have had discretion to define areas in accordance with the 
law, regulation, and guidance over the past 22 years. For nearly as 
long, USDA has approved waivers for entire states or those that group 
sub-state areas.\209\ The proposed rule would take flexibility away 
from states to define what areas they wish to waive and restrict states 
to one narrow, inflexible, federally prescribed criteria. The proposed 
criteria are likely outdated and disconnected from local and regional 
factors that states consider when developing and implementing policies 
to connect low-skilled workers to job and training opportunities. By 
prohibiting states from grouping areas according to their needs, FNS 
would hamper their ability to deliver integrated support to SNAP 
participants in gaining the skills and work experience needed to secure 
jobs leading to self-sufficiency. The proposed rule would severely 
restrict, and potentially eliminate, state flexibility to define areas 
for waivers, without providing evidence that the changes would help 
increase self-sufficiency among SNAP participants.
---------------------------------------------------------------------------
    \209\ In 2000, FNS approved a waiver requested by Florida for the 
combined area of Broward and Dade Counties, which belong to the same 
Metropolitan Statistical Area.
---------------------------------------------------------------------------
    According to the statute, a state may waive the applicability of 
the work requirement ``to any group of individuals in the state if the 
Secretary makes a determination that the area in which the individuals 
reside has an unemployment rate above 10% or does not have a sufficient 
number of jobs to provide employment for the individuals.'' The statute 
does not identify or require a geographic definition of ''area.'' The 
2018 Farm Bill did not change this and the House bill proposal that 
sought to limit states' ability to define areas was rejected.
    According to the current rule (7 CFR  273.24(f)(6)), ``States may 
define areas to be covered by waivers. We encourage state agencies to 
submit data and analyses that correspond to the defined area. If 
corresponding data does not exist, state agencies should submit data 
that corresponds as closely to the area as possible.'' The current rule 
gives states broad discretion in defining regions, requiring only that 
the data and analysis that states submit to support the waiver request 
correspond to the defined area.
    Below are excerpts from USDA guidance and rulemaking that uphold 
state flexibility in defining areas:

   December 3, 1996 guidance: The initial USDA guidance on 
        waivers from the 3 month time limit gives states flexibility to 
        define areas and goes a step further by encouraging states to 
        consider combining sub-state areas. ``USDA will give states 
        broad discretion in defining areas that best reflect the labor 
        market prospects of program participants and state 
        administrative needs.'' \210\ While USDA encouraged states to 
        consider sub-state waivers over statewide, the flexibility was 
        left completely to states.
---------------------------------------------------------------------------
    \210\ USDA, ``Guidance for states Seeking Waivers for Food Stamp 
Limits,'' December 3, 1996.

   1999 proposed rule: In the preamble, USDA noted its intent 
        to ``balance the competing goals of ensuring consistent 
        national application of these requirements, and providing state 
        agencies with appropriate implementation flexibility'' to 
        implement the time limit.\211\ ``The Department is allowing 
        states broad discretion in defining areas that best reflect the 
        labor market prospects of Program participants and state 
        administrative needs. In general, the Department encourages 
        states to consider requesting waivers for areas smaller than 
        the entire state. Statewide averages may mask slack job markets 
        in some counties, cities, or towns. Accordingly, states should 
        consider areas within, or combinations of, counties, cities, 
        and towns. The Department also urges states to consider the 
        particular needs of rural areas and Indian reservations. 
        Although the Department is proposing to allow states 
        flexibility in defining areas to be covered by waivers, the 
        supporting data must correspond to the requested area (e.g., a 
        county-wide waiver must be supported by county-wide data). In 
        other words, states may define areas to be covered by waivers, 
        but the data and analysis used to support the waiver must 
        correspond to the defined area.'' [Emphasis added.]
---------------------------------------------------------------------------
    \211\ Food Stamp Program: Personal Responsibility Provisions of the 
Personal Responsibility and Work Opportunity Reconciliation Act of 
1996; 64 Federal Register 242 (December 17, 1999) (to be codified at 7 
CFR pts. 272 and 273).

   2001 final rule: In the preamble of the final rule, USDA 
        noted that it had ``proposed that state agencies have complete 
        discretion to define the geographic areas covered by waivers so 
        long as they provide data for the corresponding area'' and that 
        most of the comments received supported this proposal. USDA 
        explained that, ``for simplicity sake, we encourage states to 
        define areas for which corresponding data exists. We believe 
        this is very easily done, especially since unemployment data 
        goes down to the census tract level.'' \212\
---------------------------------------------------------------------------
    \212\ Food Stamp Program: Personal Responsibility Provisions of the 
Personal Responsibility and Work Opportunity Reconciliation Act of 
1996; 66 Federal Register 11 (January 17, 2001). (to be codified at 7 
CFR pts. 272 and 273).

   August 2006 guidance: ``Jurisdictions or a cluster of areas 
        or counties may be combined to waive an area larger than one 
        county. States have authority to define the cluster of areas to 
        be combined. If a state defines its own jurisdiction or cluster 
        of areas, the boundaries or clusters must be thoroughly 
        documented to expedite review of the waiver request. The 
        Department of Commerce, Bureau of Economic Analysis (BEA) is 
        one source that can be used to identify economic areas. This 
        data may be found at the website www.bea.gov/bea/regional/docs/
        econlist.cfm. These areas define the relevant regional markets 
        surrounding metropolitan or micropolitan statistical areas. 
        They consist of one or more economic nodes--metropolitan or 
        micropolitan statistical areas that serve as regional centers 
        of economic activity--and the surrounding counties that are 
        economically related to the nodes. Other sources or methods may 
        be used to combine a cluster of areas.'' \213\ The guidance 
        also provided an example illustrating the use of the Department 
        of Commerce economic areas to create groups of counties in 
        Montana. It also explained that ``the state could request a 
        waiver for all counties or a sub-area of the economic areas as 
        long as the data for the combined area meets the waiver 
        criteria.''
---------------------------------------------------------------------------
    \213\ USDA, ``Guidance on Requesting ABAWD Waivers,'' August 2006.

   December 2, 2016 guidance: In its most recent guidance on 
        waivers, USDA repeated that ``the state agency has discretion 
        to define the area(s) in which it requests to waive the time 
        limit.'' According to this latest guidance, ``the state can 
        request that a waiver apply statewide or at the sub-state 
        level, as statewide averages may mask slack job markets in some 
        counties, cities, or towns. However, in order to receive FNS 
        approval to waive the ABAWD time limit the state must support 
        its request with evidence that corresponds to the requested 
        area (e.g., a county-wide waiver must be supported by county-
        wide data). The state must also clearly identify which areas 
        are being requested and under which criteria. Unemployed and 
        labor force data from individual areas can be combined to waive 
        a larger group of areas, whether based upon a recent 
        unemployment rate over ten percent or a 24 month unemployment 
        rate 20 percent above the national average.'' \214\
---------------------------------------------------------------------------
    \214\ USDA, ``Guide to Supporting Requests to Waive the Time Limit 
for Able-Bodied Adults without Dependents (ABAWD),'' December 2, 2016.

      USDA provided guidance on how states could combine areas. The 
        guidance requires that combined areas must be contiguous or 
        must belong to an economic region. The guidance provides 
        flexibility in defining an economic region. ``In order to be 
        combined, the areas must be contiguous or considered parts of 
        the same economic region. For example, two or more contiguous 
        counties could be grouped together in order to consider their 
        aggregate average unemployment rate. If the counties in the 
        sub-area all belong to the same region, they do not need to be 
        contiguous to be defined as an area. The state has discretion 
        to define the group of areas to be combined, provided that the 
        areas are contiguous or can be considered to be part of an 
        economic region. If the state defines its own group, the 
        rationale for the boundaries of the group must be thoroughly 
        documented. For example, state or local labor departments often 
        have defined economic regions based upon shared industries or 
        other factors. Other sources, methods, or rationale to support 
        that areas share an economic region may also be considered.'' 
        The guidance repeated the example from the August 2006 guidance 
---------------------------------------------------------------------------
        of using BEA economic areas as a guide for grouping counties.

    Without providing justification or rationale, the proposed rule 
would end over 2 decades of consistent guidance and support for state 
flexibility to determine the geographic scope of waivers that best 
aligns with state SNAP policies and capacity, training and workforce 
service delivery, funding and resources, and regional planning and 
strategies. The Department did not discuss or reference over 2 decades 
of consistent regulation and guidance it has issued on grouping areas. 
The Department did not go back to review the comments it received on 
the 1999 proposed rule supporting the proposal to give states complete 
discretion to define areas. The Department did not elaborate on any 
shortcomings it believes exist with the current flexibility that states 
have to define geographic areas. This makes it difficult for people who 
wish to comment to critique the Department's proposal to restrict the 
ability of states to define areas they would like to waive from the 3 
month time limit.
    The proposed rule significantly restricts the ability of states to 
waive groups of areas. Without providing a discussion, the Department 
arbitrarily eliminates the ability of states to waive the entire state 
even when statewide unemployment rates have risen significantly during 
an economic downturn, except for situations when the state qualifies 
for extended unemployment benefits.
    The Department introduces a specific definition of labor market 
areas as the only acceptable method for grouping areas and does not 
acknowledge past guidance it has issued that encouraged states to 
explore different sources and methods for grouping areas. It proposes 
these labor market areas to ensure that grouped areas are economically 
tied, yet this approach only captures one way (commuting patterns) that 
areas might be economically tied. This proposal ignores all the other 
ways areas may be economically integrated, such as through workforce 
development initiatives, economic development investments, employer 
recruiting practices, and migration patterns.
    For example, guidance issued in 2006 and 2016 both use the BEA 
economic areas (either entire economic areas or sub-areas) to 
illustrate how a state can combine unemployment data to support a 
waiver for grouped areas. The guidance suggests that states explore 
other sources or methods for combining areas, including economic 
regions defined by state or local labor departments. Even if the 
Department had provided reasons for requiring a very specific method 
for combining areas, it is difficult for the public to understand why 
the Department would disregard or minimize other economic or 
administrative factors, such as SNAP E&T service provision, that it 
currently gives great consideration to in other aspects of program 
operations.
B. States Use the Current Flexibility to Align SNAP Policies With 
        Administrative Needs, Job Opportunities, Training Funding and 
        Resources, and Economic and Workforce Development Strategies
    States consider multiple factors when grouping areas for waivers to 
align resources, administrative policies and capacity, and service 
delivery. A state may consider a range of local, sub-state (regional), 
and statewide factors:

   SNAP E&T service delivery

   TANF work programs

   Workforce Innovation and Opportunity Act (WIOA) regional 
        workforce development funding and strategies

   Office locations (SNAP, workforce development centers)

   Community college locations

   Employer and industry recruiting patterns

   Regional economic development funding and strategies

   Commuting patterns

   Housing and transportation patterns
Aligning With SNAP E&T Coverage
    By limiting state flexibility to define areas, the proposed rule 
would restrict a state's ability to allocate and coordinate SNAP E&T 
resources and service delivery to meet the needs of its SNAP 
participants. States have used their discretion to define areas to help 
align the geographic scope of waivers with areas where they are unable 
to provide sufficient work or training opportunities to work 
registrants, including those subject to the time limit. A state that 
can only provide SNAP E&T slots in certain counties may request waivers 
for (eligible) counties where SNAP E&T slots are not available or 
guaranteed. As Maryland was preparing to lose its statewide waiver in 
January 2016, the state agency requested a waiver for multiple sub-
state areas, including the nine-county Eastern Shore recognized by the 
state as an economic region. In areas not covered by waivers, Maryland 
offered SNAP E&T services to individuals subject to the time 
limit.\215\ Similarly, Colorado operated a mandatory SNAP E&T program 
for all work registrants, including individuals subject to the time 
limit, in 40 of its 64 counties. Individuals in the remaining counties 
were not subject to the time limit because of waivers or the use of 
individual exemptions, but could still participate in the E&T program 
on a voluntary basis.\216\
---------------------------------------------------------------------------
    \215\ Maryland Department of Human Resources, ``Maryland 
Supplemental Nutrition Assistance Program (SNAP) Employment and 
Training (E&T) Program: state Plan of Operations,'' Revised September 
22, 2015, https://dhr.maryland.gov/documents/Data%20and%20Reports/FIA/
YR2016%20SNAP%20E&T%20State%20Plan%20of%20Operations%20(revised).pdf.
    \216\ Colorado Department of Human Services, ``Colorado SNAP E&T 
State Plan: Federal Fiscal Year 2017,'' 2016, http://
coemploymentfirst.org/wp-content/uploads/2017/01/Colorado-SNAP-E_T-
State-Plan-FFY17.pdf.
---------------------------------------------------------------------------
Aligning With Workforce and Economic Development Regions
    Restricting or eliminating waivers for grouped areas would deny 
states the ability to align SNAP and Workforce Innovation and 
Opportunity Act (WIOA) regional service delivery, funding, and planning 
efforts. Coordinating service delivery with WIOA can help SNAP agencies 
make more qualified work activities available to SNAP participants 
because participation in a WIOA program is considered a qualifying 
activity for purposes of meeting work requirements for individuals 
subject to the time limit.
    Both USDA and the Department of Labor (DOL) recognize the 
opportunity to coordinate these two programs to integrate services and 
resources and avoid duplication. In a joint letter issued in March 
2016, USDA and DOL encouraged SNAP and the workforce system to work 
together to develop shared strategies to better connect SNAP 
participants, specifically individuals subject to the time limit, to 
job and training services through WIOA American Job Centers 
(AJCs).\217\ The letter cited the shared goal of helping low-skilled, 
low-income, or low-wage individuals find work through training 
activities and workforce programs.
---------------------------------------------------------------------------
    \217\ U.S. Department of Agriculture and U.S. Department of Labor, 
``Partnering to Help Connect Low-Income Able-bodied Adults to the 
Public Workforce System,'' March 31, 2016, https://fns-
prod.azureedge.net/sites/default/files/snap/USDA-DOL-joint-ABAWD-
letter.pdf.
---------------------------------------------------------------------------
    A state may want to align its waivers and SNAP E&T service delivery 
with WIOA regions, workforce development regions, or economic 
development regions in order to best plan and coordinate service 
delivery related to training and job opportunities for the population 
subject to the time limit. States may be able to maximize 
administrative capacity by aligning service delivery, case management, 
and data tracking by multi-county regions, such as WIOA Local Workforce 
Development Areas.
    For example, Tennessee SNAP E&T services are delivered through the 
workforce system. SNAP E&T participants are referred to the WIOA 
program for training provided through partnerships with technical and 
community colleges. SNAP participants have access to on-the-job 
training (OJT) opportunities not available outside of the WIOA-SNAP E&T 
partnership.\218\ Tennessee organizes its workforce activities into 
three regions: East, Middle, and West. These regions are further broken 
down into Local Workforce Development Areas (LWDAs). In 2007, before 
Tennessee eventually became eligible for a statewide waiver during the 
most recent recession, the state requested waivers for groups of 
counties belonging to the same LWDAs (known as Local Workforce 
Investment Areas before the passage of WIOA in 2014).
---------------------------------------------------------------------------
    \218\ U.S. Department of Agriculture, ``SNAP to Skills: Policy 
Brief 8,'' June 2018, https://snaptoskills.fns.usda.gov/sites/default/
files/2018-06/Brief_June2018_508comp.pdf.
---------------------------------------------------------------------------
    States sometimes adjust the regional alignment of programs to 
reflect changes to the labor force, resources, service delivery, and 
administrative capacity. Federal agencies may not be aware of these 
changing circumstances or be able to make adjustments in a timely 
manner. In 2018, Tennessee realigned its LWDAs by consolidating 13 
areas into nine.\219\ The Tennessee Workforce Development Board 
realigned the LWDAs to match the regional organization of other 
programs, such as the Tennessee Department of Economic and Community 
Development base camps, Tennessee Reconnect Communities, and Tennessee 
Pathways regions.\220\
---------------------------------------------------------------------------
    \219\ Tennessee Department of Labor and Workforce Development, 
``Map of Realignment of Local Workforce Development Areas,'' 2018, 
https://www.tn.gov/content/dam/tn/workforce/documents/
ProgramManagement/RealignmentMaps.pdf.
    \220\ ``TN Realigns Workforce Development Areas,'' The 
Chattanoogan, June 28, 2018, https://www.chattanoogan.com/2018/6/28/
371092/TN-Realigns-Workforce-Development-Areas.aspx.
---------------------------------------------------------------------------
Alignment With Other Regional Factors
    Beyond the alignment opportunities between SNAP E&T and WIOA, there 
are many other reasons why a state might group sub-state areas. States 
may combine different streams of funding to delivery SNAP E&T services 
regionally. Some funding opportunities may be available as regional 
grants, such as some CDBG grants or workforce development grants. In 
Portland, Oregon, the regional Workforce Development Board integrates 
WIOA, SNAP E&T, Community Development Block Grant, and other funding 
streams, to provide workforce development activities serving SNAP 
recipients and others in the Portland region.\221\
---------------------------------------------------------------------------
    \221\ U.S. Department of Agriculture, ``SNAP to Skills: Policy 
Brief 8,'' June 2018, https://snaptoskills.fns.usda.gov/sites/default/
files/2018-06/Brief_June2018_508comp.pdf.
---------------------------------------------------------------------------
    States may have administrative reasons for grouping areas. 
According to a 2016 report by the USDA Office of Inspector General, the 
requirements related to the time limit are difficult to implement.\222\ 
Some state officials said that waivers can help reduce the burden of 
tracking individuals subject to the time limit. A state may request a 
waiver to cover areas that have reduced administrative capacity and 
give areas more time to acquire staff, training, or upgrade case 
management or data systems. For example, San Francisco County in 
California used the time while it was covered by the waiver to upgrade 
its data systems and secure new E&T partnerships.\223\
---------------------------------------------------------------------------
    \222\ U.S. Department of Agriculture, Office of Inspector General. 
``FNS Controls Over SNAP Benefits for Able-Bodied Adults Without 
Dependents,'' Audit Report 27601-0002-31, https://www.usda.gov/oig/
webdocs/27601-0002-31.pdf.
    \223\ U.S. Department of Agriculture, ``State Highlights: 
California,'' Retrieved Feb. 15, 2019, https://
snaptoskills.fns.usda.gov/state-highlights/state-highlights-california.
---------------------------------------------------------------------------
    States may want to align waivers with the geographic scope of other 
resources. A state may align Information Technology (IT) systems, such 
as eligibility, case management, or data tracking systems within 
geographic regions. Counties in California are grouped into eligibility 
system consortia (with 40 counties belong to the CalACES consortium and 
18 counties belonging to the CalWIN consortium). Within each 
consortium, counties are further organized into regions (eight CalACES 
regions and four CalWIN regions). Each of these consortia systems 
support TANF work programs, SNAP E&T activities, and county-specific 
employment programs. Waivers for groups of counties could be organized 
by consortia regions to help align service delivery, case management, 
and data tracking.
    The Department did not explain why it was eliminating states' 
ability to use relevant methods for grouping areas, such as workforce 
development service delivery, to inform how they group areas covered by 
waivers. From 2 decades of experience reviewing state waiver requests, 
the Department is aware of how states use their existing flexibility to 
balance multiple priorities, resources, and policies, such as SNAP E&T 
policies and services, housing and transportation planning, and 
workforce and economic development strategies. The Department did not 
provide reasons for ignoring these other considerations, making it 
difficult for the public to comment on the proposed changes. As a 
Federal agency, USDA may not be aware of all the local and sub-state 
factors that impact the development and delivery of employment and 
training services. The proposed rule makes sweeping and arbitrary 
changes that will hamper states' ability to integrate and coordinate 
resources to provide employment and training to SNAP recipients. By 
prohibiting states from grouping sub-state areas, the agency would 
limit states' ability to coordinate and align SNAP ABAWD policies with 
training opportunities and resources, workforce and economic 
development strategies, and other factors within the state.
C. Eliminating a State's Ability to Adjust to Rising Unemployment 
        Across the State
    The proposed rule would eliminate statewide waivers when sub-state 
unemployment data is available, except for situations when a state 
qualifies for extended unemployment benefits. The Department provides 
no discussion of the rationale for eliminating this flexibility, other 
than asserting that the use of sub-state unemployment data helps target 
particular areas with high unemployment. This ignores the statistical 
principle of weighted averages--in order for an entire state to qualify 
under current rules, unemployment rates throughout the states must have 
risen dramatically, particularly in the most populous areas of the 
state.
    While we commend the Department for retaining the qualification of 
a state for extended unemployment benefits (EB) as a core standard for 
approval, this criterion does not adequately detect states with high 
unemployment rates that are not rising rapidly. States must meet both 
the minimum 3 month unemployment threshold of 6.5 percent and have 
rising unemployment over at least 1 year in order to qualify for 
extended benefits. States that only meet one of these conditions would 
not be able to obtain a waiver. For example, a state with a consistent 
unemployment rate of eight percent over time would not qualify for 
extended benefits because its rate, by definition, is not rising. 
Similarly, a state with rapidly rising unemployment, but whose rate has 
not yet reached 6.5 percent, would also not qualify. In both examples, 
states can have high or worsening unemployment and would not be able to 
obtain a waiver to help SNAP participants working in these economic 
conditions.
    To illustrate this, consider the experience of South Carolina and 
Oregon in 2007, prior to the Great Recession. These states had high 
unemployment in the months preceding the recession, but under the 
proposed rule would have had no options for statewide waivers until 
well into the recession. South Carolina qualified for extended benefits 
in August 2008, 8 months into the recession that started in December 
2007. Had the state sought a waiver for 2007 based on statewide 
unemployment rates,\224\ it would have qualified based on 24 month 
average unemployment rates that ranged between 6.6 and 6.8 percent for 
the various periods relevant for such a waiver. Similarly, the state 
would have qualified for a 2008 statewide waiver under existing rules, 
based on 24 month average unemployment rates that ranged between 6.1 
and 6.6 percent for the various periods relevant for such a waiver.
---------------------------------------------------------------------------
    \224\ South Carolina had a 2 year statewide waiver that expired in 
February 2009.
---------------------------------------------------------------------------
    Oregon qualified for extended benefits in November 2008, 11 months 
after the start of the recession. Had Oregon sought a waiver for 2007 
based on statewide unemployment rates,\225\ it would have qualified 
based on 24 month average unemployment rates that hovered between 5.8 
and 6.7 percent for the relevant period.
---------------------------------------------------------------------------
    \225\ Oregon was waived under a 2 year statewide waiver that ended 
in April 2008.
---------------------------------------------------------------------------
    The proposed rule only allows waivers for sub-state geographies, 
and not the entire state, until statewide labor market conditions 
become so dire that the state qualifies for extended benefits. The 
Department argues that statewide unemployment figures may include areas 
in which unemployment rates are relatively low and that eliminating 
statewide waivers will help target areas in which unemployment rates 
are high. The Department does not discuss the dynamic nature of labor 
market conditions across time and across geographic areas. Unemployment 
rates do not change uniformly within a state. A state may include areas 
with persistently high unemployment, areas with relatively low but 
rapidly rising unemployment rates, areas with high unemployment rates 
that are slowly creeping higher, as well as areas with relatively low 
unemployment rates.
    Current rules allow a state to detect deteriorating economic 
conditions across the state even before it qualifies for extended 
unemployment benefits. The Department makes an arbitrary decision to 
eliminate statewide unemployment analysis because of potential 
variation in unemployment rates within the state. Variation in 
unemployment rates exists at all geographic levels, including at small 
scale Census tract and Census block group levels. The Department's 
failure to provide a robust assessment of the impact of this change on 
the ability of states to cushion the blow of deteriorating economic 
conditions across their borders makes it difficult to comment on the 
proposed rule.
Arbitrary Standard for Grouping Areas
    The proposed rule would limit the ability of states to request 
waivers for groups of geographic areas, such as multi-county areas, 
except for areas that are ``economically tied.'' The Department 
provides a limited definition of an ``economically tied'' area based on 
commuting patterns--an area within which individuals can reside and 
find employment within a reasonable distance or can readily change 
employment without changing their place of residence. The preamble says 
that ``existing general conditions for grouping of areas--that the 
areas must be either contiguous and/or share the same economic region--
were intended to ensure that the areas grouped together are 
economically tied.'' Yet in statutes, regulations, and guidance over 
the past 2 decades, USDA has given states broad discretion to define 
areas and has never expressed the requirement of that grouped areas be 
``economically tied'' based solely on commuting patterns. The proposed 
rule arbitrarily imposes this requirement without providing 
justification or acknowledging the many other ways areas can be 
economically tied apart from commuting patterns, such as employer 
recruiting practices, regional workforce development strategies, and 
regional economic development and investment patterns.
    More specifically, the proposed rule would limit grouping to areas 
that are designated labor market areas (LMAs) based on a narrow 
statistical definition used by the Bureau of Labor Statistics (BLS). 
USDA has requested public comment on whether it should be even more 
restrictive and prohibit grouping entirely. USDA proposes taking away 
state discretion to define areas, arguing that the application of 
waivers on a more limited basis will encourage more individuals subject 
to the time limit to take steps towards self-sufficiency, but does not 
explain how restricting the grouping of areas will help achieve this 
goal. USDA did not offer any other alternative frameworks for grouping 
areas for discussion, even the Bureau of Economic Analysis economic 
areas that it has used as an example for grouping in past guidance.
D. Using a Narrow, Statistical Definition of Labor Market Areas
    The Department uses the BLS definition of a labor market area, 
which is an area within which individuals can reside and find 
employment within a reasonable distance or can readily change 
employment without changing their place of residence.\226\ It defines 
an ``economically tied area'' the same way. By using the same 
definition for ``labor market area'' and ``economically tied area,'' 
the Department is conflating the two concepts and makes it confusing 
and challenging for the public to respond. The Department appears to be 
taking the BLS definition of labor market areas and applying it to the 
more general concept of ``economically tied'' areas. Using the 
relatively narrow definition of labor markets to also define 
economically tied areas ignores the various ways areas can be connected 
economically beyond commuting patterns. For example, areas can share 
economic ties through investment patterns, service delivery models, and 
migration patterns.
---------------------------------------------------------------------------
    \226\ Bureau of Labor Statistics, ``Local Area Unemployment 
Statistics Geographic Concepts: Labor Market Areas,'' 2018, https://
www.bls.gov/lau/laugeo.htm#geolma.
---------------------------------------------------------------------------
    The proposed rule establishes BLS labor market areas as the only 
standard by which sub-state areas may be grouped together and covered 
by a waiver from the 3 month time limit. To delineate the entire United 
States into mutually exhaustive and exclusive labor market areas, BLS 
uses a narrow operational definition that relies solely on measures of 
population size and commuting flows between counties. BLS LMAs include 
both the metropolitan and micropolitan areas defined by the Office of 
Management and Budget (OMB) and the small labor market areas maintained 
by the BLS Local Area Unemployment Statistics (LAUS) program.
    Major labor market areas include Core Based Statistical Areas 
(CBSAs), which can be either Metropolitan Statistical Areas or 
Micropolitan Statistical Areas. Metropolitan Statistical Areas have at 
least one urbanized area with a population of 50,000 or more, along 
with adjacent territory that has a high degree of social and economic 
integration with the core (as measured by commuting patterns). 
Micropolitan Statistical Areas have at least one urban cluster with a 
population of at least 10,000 but less than 50,000, along with adjacent 
territory that has a high degree of social and economic integration 
with the core (as measured by commuting patterns).
    An outlying county is combined with the central county or counties 
of the CBSA if it meets the following commuting requirements:

   At least 25 percent of the workers living in the county work 
        in the central county or counties of the CBSA; or

   At least 25 percent of the employment in the county is 
        accounted for by workers who reside in the central county or 
        counties of the CBSA.

    Metropolitan and Micropolitan Statistical Areas are delineated in 
terms of whole counties (or equivalent entities) and counties can only 
belong to one CBSA.\227\
---------------------------------------------------------------------------
    \227\ Office of Management and Budget, Revised Delineations of 
Metropolitan Statistical Areas, Micropolitan Statistical Areas, and 
Combined Statistical Areas, and Guidance on Uses of the Delineations of 
These Areas, 2015. https://www.whitehouse.gov/sites/whitehouse.gov/
files/omb/bulletins/2015/15-01.pdf.
---------------------------------------------------------------------------
    For counties that do not belong to metropolitan or micropolitan 
areas, counties are combined into a small LMA if either or both of the 
following conditions are met: \228\
---------------------------------------------------------------------------
    \228\ Bureau of Labor Statistics, ``Local Area Unemployment 
Statistics Geographic Concepts: Labor Market Areas,'' 2018, https://
www.bls.gov/lau/laugeo.htm#geolma.

   At least 25 percent of the employed residents of one county 
---------------------------------------------------------------------------
        commute to work in another county; and

   At least 25 percent of the employment (persons working) in 
        one county are accounted for by workers commuting from another 
        county.

    Labor market areas can vary in geographic scope, ranging from a 
single county to multi-county metropolitan areas. LMAs can also span 
multiple states and in New England, they are composed of cities and 
towns. The proposed rule requires states to group all areas within an 
LMA together (leaving no areas out), but multi-state LMAs would require 
states to treat areas within their state borders separately from the 
rest of its LMA.
    Based on the conditions described earlier, the BLS labor market 
area definition only considers aggregated commuting patterns between 
county of residence and county of employment--and does not take into 
account sub-county variations by industry or by an individual's 
socioeconomic and demographic characteristics. It also does not take 
into account the many other dynamics beyond commuting patterns that may 
impact an individual's ability to find and secure a job, such as 
housing and transportation, the location of new or future employment 
opportunities, the location of training providers (such as community 
colleges), industry-specific recruitment practices, or regional 
workforce or economic development strategies. It also does not reflect 
the ability of some individuals to relocate within a state to pursue a 
job opportunity.
    According to OMB guidance, the purpose of the Metropolitan and 
Micropolitan Statistical Area standards is to provide nationally 
consistent delineations for collecting, tabulating, and publishing 
Federal statistics for a set of geographic areas. OMB establishes and 
maintains these areas solely for statistical purposes and does not take 
into account or attempt to anticipate any non-statistical uses that may 
be made of the delineations, nor will OMB modify the delineations to 
meet the requirements of any non-statistical program. OMB cautions that 
Metropolitan Statistical Area and Micropolitan Statistical Area 
delineations should not be used to develop and implement Federal, 
state, and local non-statistical programs and policies without full 
consideration of the effects of using these delineations for such 
purposes.\229\ While this does not preclude the Department or states 
from using LMAs to inform the grouping of areas covered by waivers, the 
Department did not explain the rationale for, and effect of, using LMAs 
as the only framework for grouping areas and its reasons for excluding 
other methods for grouping.
---------------------------------------------------------------------------
    \229\ Office of Management and Budget, ``Revised Delineations of 
Metropolitan Statistical Areas, Micropolitan Statistical Areas, and 
Combined Statistical Areas, and Guidance on Uses of the Delineations of 
These Areas,'' 2015, https://www.whitehouse.gov/sites/whitehouse.gov/
files/omb/bulletins/2015/15-01.pdf.
---------------------------------------------------------------------------
    In the following section, we discuss some of the other approaches 
for grouping areas that capture regional dynamics that BLS LMAs don't 
account for and that FNS did not explain whether it considered. We also 
discuss how FNS failed to address why it believes LMAs are preferable 
to the many possible alternatives.
BLS LMA Definition Based on Outdated Data
    BLS LMAs are limited in their ability to capture current and local 
workforce dynamics. They are relatively static and do not account for 
sub-county variation. The BLS LMAs are revised each decade following 
the census. The current list of BLS LMAs are based on population data 
from the 2010 Census and commuting data from the American Community 
Survey 5 year dataset for 2006-2010 (issued on February 28, 2013, 
through OMB Bulletin No. 13-01 (https://www.whitehouse.gov/sites/
whitehouse.gov/files/omb/bulletins/2013/b13-01.pdf)).\230\ In other 
words, current BLS LMAs reflect population data from 9 years ago and 
commuting data from 9 to 14 years ago. The BLS LMAs are not updated 
frequently enough to capture current or recent labor market trends and 
may not line up with more current labor force patterns.
---------------------------------------------------------------------------
    \230\ Office of Management and Budget, ``Revised Delineations of 
Metropolitan Statistical Areas, Micropolitan Statistical Areas, and 
Combined Statistical Areas, and Guidance on Uses of the Delineations of 
These Areas,'' 2013, https://www.whitehouse.gov/sites/whitehouse.gov/
files/omb/bulletins/2013/b13-01.pdf.
---------------------------------------------------------------------------
    The BLS LMAs are based on population and commuting data aggregated 
at the county level. The commuting patterns between counties may vary 
depending on the industry or type of occupation. For instance, 
commuting flows for workers working at an automotive assembly plant 
(which may be relatively focused around the plant location) will tend 
to vary from commuting flows for workers in food retail (relatively 
dispersed). Goetz and Han note that a given county may belong to 
multiple commuting sheds and give the example of a commuter county on 
the east coast with residents who commute to Washington, D.C., 
Philadelphia, and Baltimore and may be located in the border region 
between the cores of multiple LMAs.\231\ Another example is Mercer 
County, Pennsylvania, which could be considered part of Philadelphia's 
LMA or New York's LMA based on the commuting patterns of residents. In 
these situations, it is not obvious which commuting regions or labor 
market areas the county should be considered a part of and will vary 
depending on the industry or type of worker.
---------------------------------------------------------------------------
    \231\ Stephan J. Goetz and Yicheol Han, ``Identifying Labor Market 
Areas Based on Link Communities,'' paper prepared for presentation at 
the 2015 Agricultural & Applied Economics Association and Western 
Agricultural Economics Association Annual Meeting, San Francisco, CA, 
July 26-28, 2015, https://aese.psu.edu/nercrd/publications/published-
papers/identifying-labor-market-areas-based-on-link-communities.
---------------------------------------------------------------------------
    Some BLS LMAs include areas from more than one state. For example, 
the Philadelphia LMA includes counties belonging to Pennsylvania, New 
Jersey, Delaware, and Maryland. The proposed regulatory language says 
that ``the state agency may only combine data from individual areas 
that are collectively considered to be a Labor Market Area by DOL.'' 
\232\ The Department did not discuss how states should handle multi-
state LMAs, making it difficult for the public to comment on the 
implications of using only BLS LMAs to group areas. Consider an LMA 
that spans two states, but with 90 percent of its population residing 
in one state. Suppose the county with ten percent of the population has 
an unemployment rate that is slightly lower than the threshold needed 
to qualify for a waiver, but the unemployment rate for the entire LMA 
exceeds the threshold needed to qualify. Would that county qualify for 
a waiver, recognizing that the LMA that it belongs to may lack 
sufficient jobs?
---------------------------------------------------------------------------
    \232\ NPRM, p. 992.
---------------------------------------------------------------------------
Alternative Definitions of Labor Market Areas
    Unlike a county or state, which are political and administrative 
units with defined borders, a labor market area is an analytical 
concept and the definition used by BLS is only one of several ways that 
labor economists and other researchers approximate labor market areas. 
``Researchers examining labor markets in the United States often turn 
to one of several standard geographic definitions that are widely known 
and compatible with publicly available economic data, including: 
states, metropolitan areas, and counties.'' \233\ Definitions of labor 
market areas that are based on single or multiple counties, such as the 
one used by BLS, have the advantage of having unemployment data readily 
available. However, the political or administrative boundaries that are 
used to delineate labor market areas may not always align well with the 
notion of a labor market as ``a set of relationships between employers 
and workers that are spatially bounded by places of work and 
residence.''234-235
---------------------------------------------------------------------------
    \233\ Andrew Foote, Mark J. Kutzbach, and Lars Vilhuber, 
``Recalculating . . . How Uncertainty in Local Labor Market Definitions 
Affects Empirical Findings,'' Center for Economic Studies Working Paper 
CES 17-49, August 2017, https://www2.census.gov/ces/wp/2017/CES-WP-17-
49.pdf.
    \234\ Erik Scherpf, et al., ``Participation in USDA's Supplemental 
Nutrition Assistance Program (SNAP): Effect of Local Labor Market 
Conditions in Oregon,'' United [S]tates Department of Agriculture 
(September 2018), pp. 1-50, https://www.ers.usda.gov/webdocs/
publications/90038/err-257.pdf?v=0.
    \235\ Charles M. Tolbert and Molly Sizer, ``U.S. Commuting Zones 
and Labor Market Areas: A 1990 Update, AGES-9614,'' U.S. Department of 
Agriculture, Economic Research Service, 1996, https://usa.ipums.org/
usa/resources/volii/cmz90.pdf.
---------------------------------------------------------------------------
    For example, a definition of a local labor market area that 
attempts to better capture an area in which individuals both live and 
work is the Commuting Zone (CZ). CZs group counties based on commuting 
flow data and hierarchical cluster analysis.\236\ Noting that there is 
no consensus definition of LMAs, economists at the Economic Research 
Service, Scherpf, et al., tested multiple definitions of LMAs (BLS 
LMAs, Commuting Zones, and Workforce Innovation Areas) in their 
examination of the relationship between labor market area conditions 
and length of SNAP participation spell.\237\ Their preferred definition 
of LMAs used the CZ definition and had the largest estimated effects. 
They found that a ten percent increase in county-level employment 
raised the share of recipients who finished their SNAP spell in 12 
months or less by about 5.3 percentage points (or about 8.8 percent). 
Using alternative definitions of labor market areas resulted in 
smaller, but still positive, estimated effects: a ten percent increase 
in county-level employment raised the probability that a SNAP recipient 
would finish a spell in 12 months or less by between 1.5 and 2 
percentage points (or between two and three percent).
---------------------------------------------------------------------------
    \236\ Hierarchical cluster analysis is a method for exploring 
similarities between objects. An algorithm is used to group similar 
objects into a cluster. Each cluster is distinct from other cluster and 
the objects within each cluster share similar features.
    \237\ Erik Scherpf, et al., ``Participation in USDA's Supplemental 
Nutrition Assistance Program (SNAP): Effect of Local Labor Market 
Conditions in Oregon,'' United [S]tates Department of Agriculture 
(September 2018), pp. 1-50, https://www.ers.usda.gov/webdocs/
publications/90038/err-257.pdf?v=0.
---------------------------------------------------------------------------
    Although CZs are delineated using more sophisticated analytical 
methods, they share a similar limitation to BLS LMAs. Like BLS LMAs, 
CZs are based on commuting patterns and do not account for other 
administrative or economic factors a state would like to consider when 
grouping areas covered by waivers. In proposing the BLS LMAs as the 
only acceptable framework for grouping areas, the Department did not 
discuss alternative frameworks for grouping and the implications of 
using a framework based on commuting data to shape SNAP policy.
Commuting Patterns Are Not the Only Factor Connecting Labor Markets
    BLS LMAs only look at commuting patterns and ignore other economic 
factors that may be related to spatial correlation of unemployment. 
Spatial correlation is a measure of the relationship between ``close'' 
spatial units, such as neighboring counties. Using county-level monthly 
price data from the real estate service Zillow, Fogli, Hill, and Perri 
examined trends in housing prices across geographic areas. They 
observed that the housing price decline from early 2007 to early 2009 
appeared to follow the same spatial patterns as unemployment.\238\ 
Looking at the example of Florida, they show that early 2007 prices 
fell in scattered locations around the coasts and over time prices fell 
in nearby locations until they reached a uniformly low level across the 
state. Spatial diffusion of housing prices and unemployment were 
strikingly similar across the whole period of housing boom and bust, 
suggesting that housing prices might be one of the factors that states 
may implicitly or explicitly consider when grouping areas.
---------------------------------------------------------------------------
    \238\ Alessandra Fogli, Enoch Hill, and Fabrizio Perri, ``The 
Geography of the Great Recession,'' National Bureau of Economic 
Research Working Paper 18447, October 2012, https://www.nber.org/
papers/w18447.pdf.
---------------------------------------------------------------------------
    The Department did not provide an explanation why it is restricting 
states from considering other rationale that might be relevant to their 
SNAP population when requesting waivers. We strongly encourage the 
Department to review the research we have outlined in this chapter 
related to commuting patterns and to explain its rationale relative to 
the evidence that demonstrates the flaws of this approach.
WIOA Regions
    As discussed earlier, states may seek to align SNAP waivers with 
workforce development regions. Under WIOA, states are required to 
identify regions for regional workforce planning. States shall identify 
regions after consultation with elected officials and Local Workforce 
Development Boards and take into account the following factors:

  1.  The extent to which regions are consistent with labor market 
            areas in the state;

  2.  The extent to which regions are consistent with regional economic 
            development areas in the state; and

  3.  An assurance that regions have available the Federal and non-
            Federal resources necessary to effectively administer 
            activities under subtitle B and other applicable provisions 
            of the WIOA, including whether the areas have the 
            appropriate education and training providers, such as 
            institutions of higher education and area career and 
            technical education schools.

    Under WIOA, states have discretion to define regions and are 
encouraged to take an integrated approach to account for a range of 
different factors. States are encouraged to use LMAs as the starting 
point for determining workforce development regions, but also need to 
consider workforce and economic development framework, funding streams, 
and service (training) delivery.
E. States Have Used Their Discretion to Create Groupings Informed by 
        Multiple Factors
    States use grouping methods, such as BLS LMAs and WIOA Local 
Workforce Development Areas, as the starting point for developing 
workforce development plans and policy, but modify them based on 
multiple additional factors. Below are some examples of how and why 
states group areas using a variety of factors.

   When designating areas for workforce development planning, 
        the Virginia Board of Workforce Development considers the BLS 
        LMAs, regional economic development areas, funding streams and 
        service providers for training, community college regions, and 
        industry- and sector-specific strategies.\239\
---------------------------------------------------------------------------
    \239\ Virginia Board of Workforce Development, ``Designation of 
Regions and Planning Requirements,'' June 2016, https://
virginiacareerworks.com/wp-content/uploads/Policy-200-06-Designation-
of-Regions-and-Planning-Requirements-FINAL-Signed.pdf.

   Rhode Island has treated the entire state as a region for 
        the purposes of workforce development planning. The state 
        considered geographic boundaries, LMA analysis, and funding and 
        resource realities in determining the geographic scope of its 
        workforce development plan. From its labor market area 
        analysis, it found that of the 39 cities and towns in Rhode 
        Island, 36 fell within the ``Providence-Warwick, RI-MA 
        Metropolitan NECTA'' LMA as determined by the Bureau of Labor 
        Statistics. One additional community was its own LMA due to the 
        fact it is an island, and two additional communities fall 
        outside of the Providence-Warwick, RI-MA Metropolitan NECTA. 
        The governor concluded that the entire state will be a single 
        planning region for workforce development purposes.\240\
---------------------------------------------------------------------------
    \240\ Rhode Island, Governor's Workforce Board, ``Regional Planning 
Policy,'' March 16, 2017, https://gwb.ri.gov/wp-content/uploads/2017/
06/17-01-3-16-2017.pdf?189db0.

   As discussed earlier, Tennessee implements SNAP E&T in 
        coordination with WIOA and Local Workforce Development Areas. 
        The delineation of these areas is different from the BLS LMAs. 
        For instance, Lauderdale County, located on the western border 
        of Tennessee north of Memphis, is its own BLS LMA, but is 
        grouped with Shelby, Fayette, and Tipton Counties into the 
---------------------------------------------------------------------------
        Greater Memphis Local Workforce Development Area.

    The Department did not explain why it was taking away states' 
ability to consider multiple factors when grouping areas covered by 
waivers. It did not discuss how the use of BLS LMAs would improve 
states' ability to meet the needs of individuals subject to the time 
limit and help move them to self-sufficiency.
BEA Economic Areas
    In two separate guidance memoranda (August 2006,\241\ December 2016 
\242\), the Department provided an example of grouping based on BEA 
economic areas, yet the 2019 NPRM preamble did not offer this framework 
for grouping areas for consideration. In the example provided, the 
Department explained that Montana could group the counties of Blaine, 
Cascade, Chouteau, Glacier, Hill, Liberty, Phillips, Pondera, Teton, 
and Toole that comprise the North Central Montana Economic Area (or 
Area 65: Great Falls, MT Economic Area) and analyze the data to see if 
the grouped area would qualify for a waiver. The guidance also 
suggested that the state could consider grouping a sub-area such as 
Glacier, Liberty, and Toole. The Department did not explain why it is 
eliminating this grouping approach, even though this was a grouping 
policy it has suggested that states could use in the past.
---------------------------------------------------------------------------
    \241\ USDA, ``Guidance on Requesting ABAWD Waivers,'' August 2006.
    \242\ USDA, ``Guide to Supporting Requests to Waive the Time Limit 
for Able-Bodied Adults without Dependents (ABAWD),'' December 2, 2016.
---------------------------------------------------------------------------
Requiring Areas to Be Contiguous Ignores the Reality That Proximity to 
        Job Opportunities Is Decreasing
    The proposed rule establishes BLS LMAs as the only scenario where 
states can request waivers for combined geographic areas (counties). In 
doing so, the agency seeks to limit waivers that combine areas that are 
not contiguous. This suggests an assumption that individuals will only 
respond to job opportunities in their county or in counties adjacent to 
their county of residence. From the perspective of workers in search of 
job opportunities, requiring contiguity of geographic areas is an 
assumption that does not hold up under empirical scrutiny. Using 
county-level data for eight states between 1969 to 1994, Khan, Orazem, 
and Otto found that local county population responded to economic 
growth within the county, in adjacent counties, and even two counties 
away.\243\ The effect decreased as the distance from the reference 
county increased. Workers look beyond their county and adjacent 
counties for job opportunities.
---------------------------------------------------------------------------
    \243\ Romana Khan, Peter F. Orazem, and Daniel M. Otto, ``Deriving 
Empirical Definitions of Spatial Labor Markets: The Roles of Competing 
Versus Complementary Growth,'' Journal of Regional Science (2001), pp. 
735-756.
---------------------------------------------------------------------------
    Other research has found that proximity to jobs has decreased in 
metropolitan areas in recent years and that poor, minority residents 
experienced a bigger decline in proximity to jobs compared to non-poor 
white resident. Kneebone and Holmes looked at the number of jobs within 
a typical commuting distance (median within-metro commuting distance) 
for residents of the 96 largest metropolitan areas and found that the 
number of jobs within a typical commuting distance declined by seven 
percent between 2000 and 2012.\244\ Poor residents experienced a 17 
percent decline in nearby jobs compared to six percent for non-poor 
residents. Hispanic residents had a 17 percent decline and black 
residents had a 14 percent decline compared to six percent for white 
residents. Individuals have to look farther from their local 
neighborhoods for job opportunities, requiring longer commutes.
---------------------------------------------------------------------------
    \244\ Elizabeth Kneebone and Natalie Holmes, ``The Growing Distance 
Between People and Jobs in Metropolitan America,'' Metropolitan Policy 
Program at Brookings (March 2015), pp. 1-24, https://www.brookings.edu/
research/the-growing-distance-between-people-and-jobs-in-metropolitan-
america/.
---------------------------------------------------------------------------
Requiring Areas to Be Contiguous or to Comprise Entire LMAs Ignores the 
        Reality That Unemployment Rates Rise and Fall at Different 
        Rates Even in Neighboring Areas
    The general unemployment rate does not account for variations in 
unemployment rates for sub-populations and for variations in the 
increase or decrease in unemployment rates across geographic areas. The 
proposed rule would require that states request waivers for BLS LMAs in 
their entirety, without omitting certain areas. The Department offers 
no rationale for proposing this and no discussion of the implications 
of this arbitrary requirement. This requirement prevents states from 
responding to variations in unemployment patterns within an LMA.
    Fogli, Hill, and Perri examined how the relationship in 
unemployment in neighboring areas changes over time as a recession 
starts and ends.\245\ The relationship in unemployment between 
neighboring areas (spatial correlation) is high overall, falls at the 
start of the recession, increases sharply during the recession, and 
then stabilizes at the end of the recession. They found that 
unemployment does not increase in all counties simultaneously, but 
initially increases in a few specific counties, not necessarily located 
close to each other. As the recession deepens, the geographic 
distribution of unemployment follows an epidemic pattern, with 
unemployment tending to increase in counties that are closer to 
counties initially hit with high unemployment, so that unemployment is 
high in some concentrated areas and relatively low in others, and this 
results in an increase in the degree of spatial correlation and of 
spatial dispersion. As the recession reaches its peak, high 
unemployment is spread all over the country and both the degree of 
spatial correlation and spatial dispersion stabilize (and eventually 
decline).
---------------------------------------------------------------------------
    \245\ Alessandra Fogli, Enoch Hill, and Fabrizio Perri, ``The 
Geography of the Great Recession,'' National Bureau of Economic 
Research Working Paper 18447, October 2012, https://www.nber.org/
papers/w18447.pdf.
---------------------------------------------------------------------------
    The Department does not discuss the implications of requiring that 
states request waivers for BLS LMAs in their entirety, without omitting 
areas. This requirement ignores variations in labor market conditions 
within a labor market area. Consider an LMA that does not qualify for a 
waiver, but has high unemployment everywhere except for one county. The 
proposed rules would prevent a state from requesting a waiver for a 
sub-area of the LMA that lacks sufficient jobs, even if most of the 
residents in that LMA reside in that sub-area and the sub-area had 
unemployment rates that met or exceeded the threshold to qualify for a 
waiver. Even if the LMA qualified for a waiver, the state may have 
reasons why it only wants to waive a sub-area of the LMA. For instance, 
the state may be able to guarantee SNAP E&T slots in most areas of the 
LMA, but wants to request a waiver to cover counties in the LMA where 
it cannot guarantee enough slots. Even if the areas the state wants to 
waive qualified for a waiver based on unemployment rates, because it 
comprises only a part of the LMA, it would not be eligible for a 
waiver.
F. Conclusion: We Recommend Rejecting Proposed Changes That Restrict 
        State Flexibility to Waive Groups of Areas
    The proposed rule is based on insufficient reasons to change 
current regulations by prohibiting states from seeking statewide 
waivers and from grouping areas, except for areas that are designated 
as BLS labor market areas. It fails to discuss the ways that states 
have used existing flexibility to align waiver policy with state 
operations, policy, and resources. It does not discuss the implications 
and limitations of its proposed framework for grouping, nor does it 
address alternative methods for grouping, including those suggested by 
the Department in past guidance. Given the lack of supporting 
information, the public has an insufficient opportunity to comment 
meaningfully on the proposed rule and we recommend rejecting the 
proposed changes to the rules.
Chapter 6. Taking Away Food Benefits from Individuals Who Cannot 
        Document 20 Hours a Week of Work Will Not Increase Labor Force 
        Participation for This Population
    USDA offered little rational for changing the decades-old criteria 
for requesting waivers of the time limit. The primary reasoning it 
provided is that fewer waivers will result in more individuals subject 
to the time limit. The agency believes that if the state threatens to 
withhold food benefits from these individuals, they will work more and 
have higher participation rates in meaningful work activities. The NPRM 
often describes this as a ``belief.'' For example, the NPRM states, 
``the Department believes the local unemployment floor should be set at 
seven percent to best meet its goals of promoting self-sufficiency'' 
\246\ (emphasis added). But the NPRM provides no evidence to support 
the belief that taking away food from unemployed individuals will 
result in higher labor force attachment or greater participation in job 
training. Because the NPRM includes no supporting data or research, 
commentators are left to accept as unequivocally true that the time 
limit has an instrumental role in moving ``ABAWDs'' from non-work to 
work.
---------------------------------------------------------------------------
    \246\ NPRM p. 984.
---------------------------------------------------------------------------
    Yet the claim that subjecting additional individuals to the time 
limit will result in more meaningful work activities is wildly out of 
synch with what we know from the evidence. Research shows that a 
significant share of individuals subject to the time limit work when 
they can find employment (including while on SNAP) and will work after 
leaving SNAP even in the absence of the time limit. The claim also 
ignores research showing that time limits generally fail to encourage 
employment. And, the NPRM does not account for the particular 
challenges facing this population--barriers and challenges to 
employment that differ from those faced by the general public and 
justify the current approach to providing waivers and individual 
exemptions to unemployed childless adults. Because the agency did not 
provide any evidence that would demonstrate the time limit is likely to 
increase employment, earnings, or self-sufficiency, the agency's claim 
that the time limit should be applied to more individuals in order to 
increase labor force attachment is without merit. We offer findings 
from numerous studies to illustrate our points. We strongly recommend 
that the Department review and reflect on each of these studies before 
moving forward with a final rule.
A. Individuals Subject to the Time Limit Already Have Significant Work 
        Effort, Raising Doubt as to Whether the Rate Can Be Increased 
        by Withholding Food
    The NPRM repeatedly claims that a primary goal of the proposed 
changes is to subject more individuals to the time limit. Terminating 
food assistance to more people (some 755,000 more, by the agency's own 
estimates), the NPRM argues, will increase work effort, job placement, 
and earnings among those subject to the rule. But it fails to include 
any information about the employment of individuals subject, or 
potentially subject, to the time limit either while on SNAP or before 
or after participation in the program. The absence of any information 
is striking because research shows that many low-income adults without 
children who receive SNAP in any given month work while on SNAP or soon 
after, regardless of the existence of the time limit. Other factors, 
such as personal circumstances and local economic conditions, play 
important roles in an individual's employment, but are not taken into 
account by FNS in this proposed change.
    In this section, we review the research that describes both the 
significant work effort of childless adults without disabilities, as 
well as the unique challenges they face in the labor market. Because 
workers turn to SNAP during periods of unemployment, employment rates 
among childless adults while receiving SNAP might be expected to be 
relatively low. Still, in a typical month, almost \1/3\ and perhaps as 
many as \1/2\ of all SNAP households with childless adults work. USDA's 
administrative data show that 31 percent of all SNAP households with 
non-disabled childless adults worked in a typical month of 2017.\247\ 
USDA's administrative data may underestimate earnings because some work 
may not be required to be reported for SNAP, either because it is 
irregular or not expected to continue or because, under SNAP's 
``simplified reporting'' rules, changes in circumstances need only be 
reported at 6 month intervals unless they raise household income above 
130 percent of the poverty level. Work that households aren't required 
to report for SNAP purposes may be captured by the other data but not 
the SNAP data.
---------------------------------------------------------------------------
    \247\ Food and Nutrition Service, Characteristics of Supplemental 
Nutrition Assistance Program Households: Fiscal Year 2017, Tables A.14 
and A.16.
---------------------------------------------------------------------------
    In fact, data from other sources suggest the work rate could be 
closer to 50 percent. Tabulations generated by CBPP from the Census 
Bureau's 2008 Panel of the Survey of Income and Program Participation 
(SIPP) show employment rates for SNAP households with non-disabled 
childless adults close to 50 percent in a typical month between 2011 
and 2013.\248\
---------------------------------------------------------------------------
    \248\ Center on Budget and Policy Priorities, ``Unemployed adults 
without children who need help buying food only get SNAP for three 
months,'' https://www.cbpp.org/unemployed-adults-without-children-who-
need-help-buying-food-only-get-snap-for-three-months.
---------------------------------------------------------------------------
    The 2017 data from USDA does not identify whether a childless adult 
is subject to the time limit or not (because he or she is exempt or 
living in a waived area). But work rates among this population have 
been consistent, even as the percentage of the country covered by 
waivers has varied widely. Comparing 2017 data (when 37 percent of the 
population lived in a waived area) to 2014 data (when 75 percent of the 
population did) shows very similar levels of employment. In 2014, when 
the national unemployment rate was over six percent, 35 states had 
statewide waivers and nine had areas waived. That year, 27 percent of 
all SNAP households with non-disabled adults and no children worked in 
a typical month.\249\ That is close to the 31 percent of non-disabled 
adults without children who were working in 2017, when far more of the 
country had the time limit in place. This suggests factors other than 
waivers or the time limit itself play a more important role in the 
ability of adults without children to find work.
---------------------------------------------------------------------------
    \249\ See Food and Nutrition Service, Characteristics of 
Supplemental Nutrition Assistance Program Households: Fiscal Year 2014, 
Table A.16.
---------------------------------------------------------------------------
    Even more important, childless adults have high work rates prior to 
and after a spell on SNAP, but the NPRM fails to account for how 
subjecting more individuals to the time limit would impact this. About 
72 percent of SNAP households with a childless, working-age adult 
worked in the year before or after receiving SNAP.\250\ And many of 
those not working while receiving SNAP were actively looking for work. 
USDA's administrative data suggest that, in a typical month in 2017, 
close to \1/2\ (46 percent) of all childless adults on SNAP who were 
not working were looking for work (and of those who reported not 
participating in the labor force, many likely had health conditions or 
other barriers to employment that prohibited labor force 
participation). Although this statistic should be viewed with caution 
as the data may not be sufficiently reliable to draw firm conclusions, 
it's noteworthy that Urban Institute researchers, using National Survey 
of American Families data, found that \3/4\ of all low-income, able-
bodied adults without dependents (not just those on SNAP) worked in 
1997 and 86 percent were in the labor force (that is, either working or 
actively looking for work).\251\
---------------------------------------------------------------------------
    \250\ Center on Budget and Policy Priorities, ``Unemployed adults 
without children who need help buying food only get SNAP for three 
months,'' https://www.cbpp.org/unemployed-adults-without-children-who-
need-help-buying-food-only-get-snap-for-three-months.
    \251\ See Stephen Bell and Jerome Gallagher, ``Prime-Age Adults 
without Children or Disabilities: The `Least Deserving of the Poor'--or 
Are They? Assessing the New Federalism Policy,'' Urban Institute, 
February 2001, https://www.urban.org/sites/default/files/publication/
61286/310269-Prime-Age-Adults-without-Children-or-Disabilities.PDF.
---------------------------------------------------------------------------
    Given the consistent evidence of work among individuals likely to 
be subject to the time limit regardless of waiver status or general 
unemployment rates, the claim that the time limit itself leads to 
increased employment is not supported. Instead, it suggests that the 
rule's sole purpose is to take away food assistance from struggling 
unemployed or underemployed workers.
    Further undermining the assertion that the time limit is necessary 
to increase work effort among SNAP participants, most childless adults 
on SNAP who work have substantial work. Among SNAP households that 
worked in a typical month while receiving SNAP or worked at some point 
during the following year, nearly \1/2\ (49 percent) worked full time 
(at least 35 hours a week) for 6 months or more of the following year. 
Twelve percent worked at least 20 hours per week for 6 or more months. 
Another 24 percent worked full time in at least 1 month over that 
period. Only about 15 percent worked 20 or more hours per week for less 
than 6 months or worked fewer hours than that.\252\
---------------------------------------------------------------------------
    \252\ Steven Carlson, et al., ``Who Are the Low Income Childless 
Adults Facing the Loss of SNAP in 2016?'' Center on Budget and Policy 
Priorities, February 8, 2016, p. 9, https://www.cbpp.org/sites/default/
files/atoms/files/2-8-16fa.pdf. An updated analysis that looked at all 
SNAP households with an adult without disabilities (with and without 
children), found similar rates of full- and part-time work among the 
broader group that the earlier analysis found. See Brynne Keith-
Jennings and Raheem Chaudhry, ``Most Working-Age SNAP Participants 
Work, But Often in Unstable Jobs,'' March 15, 2018, https://
www.cbpp.org/research/food-assistance/most-working-age-snap-
participants-work-but-often-in-unstable-jobs.
---------------------------------------------------------------------------
    Nonetheless, childless adults participating in SNAP are generally 
low-income, low-skill workers with limited job prospects. More than 80 
percent have no more than a high school education or GED. They are more 
likely than other SNAP participants to lack basic job skills like 
reading, writing, and basic mathematics.\253\ A work experience program 
in Ohio designed to help individuals subject to the time limit find 
work or qualifying work activities found signs of functional illiteracy 
even among those with a high school degree.\254\ As a result, wages of 
childless working-age adults on SNAP are quite low: one study found 
that 90 percent of those aged 25-49 earned less than twice the minimum 
wage, compared to 47 percent of all workers aged 25-49.\255\
---------------------------------------------------------------------------
    \253\ ``Food Stamp Employment and Training Program,'' United States 
General Accounting Office (GAO-3-388), March 2003, p. 17, http://
www.gao.gov/assets/240/237571.pdf.
    \254\ See ``A Comprehensive Assessment of Able-Bodied Adults 
Without Dependents and Their Participation in the Work Experience 
Program in Franklin County, Ohio,'' Ohio Association of Foodbanks, 
2014, http://admin.ohiofoodbanks.org/uploads/news/WEP-2013-2014-
report.pdf.
    \255\ Stephen Bell and Jerome Gallagher, ``Prime-Age Adults without 
Children or Disabilities: The `Least Deserving of the Poor'--or Are 
They? Assessing the New Federalism Policy,'' Urban Institute, February 
2001, https://www.urban.org/sites/default/files/publication/61286/
310269-Prime-Age-Adults-without-Children-or-Disabilities.PDF.
---------------------------------------------------------------------------
    USDA did not provide any information or analysis in the NPRM that 
suggested it had reviewed this evidence. Nor did it offer any research 
to the contrary--and we believe no such research exists. That leaves us 
to wonder whether the Department was aware that the work rates for 
individuals subject to the time limit appear to be similar whether or 
not they live in a waived area.
B. The NPRM Fails to Account for the Distinct Characteristics of 
        Unemployment Childless Adults on SNAP
    Because the group of individuals subject to the rule are, when 
compared to the general public, poorer, less educated, and more likely 
to have medical conditions or other factors affecting their ability to 
find employment, states have long found the 3 month time limit in 
statute harsh and unfair. To mitigate the harm caused by taking food 
assistance away from this group, states have routinely relied on the 
option to request a waiver based on demonstrating a lack of sufficient 
jobs for the individuals affected by the time limit.
    The NPRM seeks to restrict this state option in order to end food 
assistance for unemployed adults without children in the hope that this 
will result in increased employment among this group. The NPRM offers 
no information about the individuals affected by the rule--whether they 
have the skills, training, and support to find and keep employment, and 
whether they disproportionately face barriers to work like undiagnosed 
health conditions, a lack of transportation, or a criminal history. The 
NPRM neglects to address research findings from multiple sources, 
including USDA itself, showing that this population does face a 
different labor market environment than the general public. We are 
surprised that USDA did not draw upon this wide body of research and 
urge the agency to review the studies we cite--all of which are 
included in the appendix.
    We review the research to show that the job prospects for these 
individuals are not accurately captured by the general unemployment 
rate. Even when unemployment is low due to a strong economy, adult SNAP 
participants without children face a very different labor market than 
higher-income adults. This population is also under-served by other 
support programs, often lacks stable housing, and struggles to be hired 
into stable jobs. States wisely use the flexibility provided by law to 
assess the availability of jobs for this population and request waivers 
when there are insufficient jobs.
    Childless adults on SNAP are extremely poor. Like many others, 
childless adults often turn to SNAP for assistance when they are no 
longer able to make ends meet, especially if their jobs are lost, hours 
are cut, or wages hover at the Federal minimum. While participating in 
SNAP, their income averages 33 percent of the poverty line, the 
equivalent of about $4,000 per year for a single person in 2019. 
Average incomes are even lower--just 18 percent of poverty--for those 
not working 20 or more hours a week, who are most likely to be cut off 
due to the 3 month limit.\256\
---------------------------------------------------------------------------
    \256\ CBPP analysis of FY 2017 USDA SNAP Household Characteristics 
data adjusted to FY 2019 dollars.
---------------------------------------------------------------------------
    The struggles of poor adults are vividly portrayed in $2.00 a Day: 
Living on Almost Nothing in America, which draws detailed portraits of 
households with little access to substantial employment and public 
benefits.\257\ The descriptions illustrate the complexities of 
poverty--the psychological and emotional costs, the uncertainty and 
lack of options available, and the work effort of those in poverty. We 
strongly urge the Department to familiarize itself with the real lived 
experiences of those portrayed in the book.
---------------------------------------------------------------------------
    \257\ Kathryn J. Edin and H. Luke Shaefer, $2.00 a Day: Living on 
Almost Nothing in America, 2015, Houghton Mifflin Harcourt. We 
especially highlight descriptions of the struggles of finding and 
keeping work (pp. 42-47 and 112-114), the challenges facing people of 
color in the job market (pp.52-56), and barriers to employment like 
transportation (pp. 51-52, 138-139).
---------------------------------------------------------------------------
    Unemployed childless adults have few resources other than SNAP to 
rely on while looking for work. In general, adults without children are 
not eligible for most government assistance. In the past, state General 
Assistance programs have provided small monthly cash allotments to 
singe adults to meet basic shelter and other needs, but these programs 
have weakened considerably. Few childless adults qualify for 
unemployment insurance, and childless adults are ineligible for 
Medicaid in states that haven't adopted the Affordable Care Act's 
Medicaid expansion. In addition, childless workers are the only 
demographic group that the Federal tax system taxes into, or deeper 
into, poverty, in part because they are eligible only for a tiny Earned 
Income Tax Credit (EITC). Federal income and payroll taxes pushed about 
7.5 million childless workers into or deeper into poverty in 2015.\258\ 
Given how little Federal support is available to the group subject to 
the time limit, it is surprising that USDA believes the group is able 
to survive while avoiding work. Their SNAP benefits are minimal to meet 
basic food needs, let alone housing, health, and other basic expenses.
---------------------------------------------------------------------------
    \258\ Chuck Marr, et al., ``Lone Group Taxed Into Poverty Should 
Receive a Larger EITC,'' Center on Budget and Policy Priorities, April 
19, 2016, https://www.cbpp.org/research/federal-tax/childless-adults-
are-lone-group-taxed-into-poverty.
---------------------------------------------------------------------------
    Many childless adults have disabilities that make working difficult 
or impossible but don't meet the severe disability standard for 
receiving Supplemental Security Income (SSI) or Social Security 
Disability Insurance (SSDI). If not identified as having a physical or 
mental condition that prevents them from working 20 or more hours per 
week, they would be subject to the time limit, yet unable to 
realistically find work in many cases.
    There's more evidence that people subject to the time limit face 
multiple challenges to independence and self-sufficiency, including 
homelessness, physical and mental health limitations, language 
barriers, unstable employment histories, and criminal records. A 
detailed study of individuals identified by the local SNAP agency as 
ABAWDs subject to the time limit who were referred to a work experience 
program in Franklin County (Columbus), Ohio found that: \259\
---------------------------------------------------------------------------
    \259\ See ``Comprehensive Report on Able-Bodied Adults Without 
Dependents, Franklin County Ohio Work Experience Program,'' Ohio 
Association of Foodbanks, 2015, http://admin.ohiofoodbanks.org/uploads/
news/ABAWD_Report_2014-2015-v3.pdf. The Ohio Association of Foodbanks 
gathered the information for the report as a result of a partnership 
with the county SNAP agency to help place individuals identified as 
subject to the time limit in qualifying work activities after screening 
them.

   Many have extremely unstable living situations, evidenced by 
        residence in short-term shelters or with friends and family and 
---------------------------------------------------------------------------
        limited telephone service.

   One-third have a mental or physical limitation, including 
        depression, post-traumatic stress disorders, mental or learning 
        disabilities, or physical injuries. Some of these disabilities, 
        though not severe enough to qualify for Federal disability 
        benefits, may still limit a person's ability to work more than 
        20 hours a week.

   Nearly \1/4\ are non-custodial parents, and 13 percent are 
        caregivers for a parent, relative, or friend.

   More than 40 percent lack access to reliable private or 
        public transportation; 60 percent lack a valid driver's 
        license.

   Fifteen percent need supportive services like language 
        interpretation or help with transportation to obtain 
        employment.

   Nearly \1/4\ have been dismissed from a job in the past and 
        others have gaps in their employment records--both of which can 
        deter potential employers. More than \1/3\ have felony 
        convictions, making it hard to find jobs and pass background 
        checks.

    These individuals face daunting challenges in finding employment 
even when general unemployment rates are low. The Ohio study 
illustrates why Congress gave states the option to waive the time limit 
in areas where there are insufficient jobs for those subject to the 
rule. Without providing any evidence to the contrary, the NPRM proposes 
to limit the ways in which a state can demonstrate a lack of sufficient 
jobs for the individuals subject to the time limit. It does this by 
eliminating Labor Surplus Areas, low and declining employment-to-
population ratios, seasonal unemployment and requiring recent 
unemployment rates to be at least seven percent. But the Department 
fails to explain how it determined that the proposed new standards 
relate to employment opportunities for those subject to the rule, 
particularly given the characteristics outlined in the list above.
    One reason states were given flexibility to define the areas which 
could be waived due to a lack of sufficient jobs for the individuals 
subject to the rule is that even in the late 1990s, a growing body of 
research showed that the labor market situation for low-skilled workers 
had grown worse over time, and that low-skill workers faced limited 
employment options. As summarized by a report commissioned by USDA in 
1998, ``a relatively large body of research indicates that the labor 
market situation of the low-skilled has become considerably worse in 
recent decades and that their current employment prospects are limited. 
This suggests that even if ABAWDs are willing to work, they may be 
unable to do so because there are not enough jobs for low-skilled 
workers.'' \260\ The report, which reviewed studies on the employment 
prospects of low-income adults fitting the ``ABAWD'' description (but 
not necessarily participating in SNAP), also found that:
---------------------------------------------------------------------------
    \260\ Michael Stavrianos and Lucia Nixon, The Effect of Welfare 
Reform on Able-Bodied Food Stamp Recipients, Mathematica Policy 
Research, Inc., July 1998, pp. 56-57.

   Job prospects for ABAWDs do not look promising, due to 
        changes in the U.S. economy that have resulted in the decline 
---------------------------------------------------------------------------
        of industries and skill types in which ABAWDs are concentrated.

   Many ABAWDs face a spatial mismatch between their residence 
        and the location of low-skill jobs, as well as a skills 
        mismatch, especially for urban residents.

   The job prospects of ABAWDs depend significantly on local 
        economic conditions, tied not to county-level unemployment but 
        to the location of employers needing low-skill workers and the 
        quality and availability of local institutions supporting 
        workforce development.\261\
---------------------------------------------------------------------------
    \261\ Michael Stavrianos and Lucia Nixon, The Effect of Welfare 
Reform on Able-Bodied Food Stamp Recipients, Mathematica Policy 
Research, Inc., July 1998, p. 56-57.

    The Department's commissioned reports as well as other research 
paint a clear picture of individuals in this targeted group who have 
common characteristics that distinguish the group from other unemployed 
adults. These characteristics--including high poverty rates, health 
issues, and few supports--make finding and keeping employment a unique 
challenge. The Department simply asserts that the time limit will 
increase employment for this population but does not acknowledge its 
own research showing that this is not the case. While all aspects of 
the rule strike us as arbitrary, this disconnect between the agency's 
basic knowledge of the affected population and the assertions about how 
the proposed policy would increase employment is particularly 
surprising. This is one of numerous reasons why the proposed rule 
should be withdrawn.
C. Reports Commissioned by USDA Show Subjecting More Individuals to the 
        Time Limit Will Not Increase Their Likelihood of Gaining 
        Employment
    In order to understand the impact of the time limit on individuals' 
well-being, USDA, along with the U.S. Department of Health and Human 
Services, funded studies in four states: Arizona, Illinois, Iowa, and 
South Carolina. While the studies varied in scope and focus and were 
not able to identify individuals who left SNAP because of the time 
limit, they do reveal the policy's limited impact on employment 
outcomes, coupled with low earnings and increased hardship.
    A 2001 study of individuals leaving SNAP in Illinois showed that 
far more families than ABAWDs left SNAP due to increased earnings, even 
though the time limit was in effect. The study found that the 
percentage of ABAWDs who left SNAP and remained off the program and 
employed did not vary between counties that had waivers and counties 
that did not.\262\
---------------------------------------------------------------------------
    \262\ See Exhibit ES-1, page ES-4, Philip Richardson, et al., Food 
Stamp Leavers Research Study--Study of ABAWDs Leaving the Food Stamp 
Program in South Carolina, https://naldc.nal.usda.gov/download/45220/
PDF.
---------------------------------------------------------------------------
    The study looked at those who left SNAP in areas where the time 
limit was in effect and areas where it was not (due to a waiver). It 
found that in the sample who left SNAP in 1998-1999, a majority of 
individuals (53 percent) were off SNAP in 2001 and not working or back 
on SNAP. Of those that were working after leaving SNAP, the study did 
not attempt to identify the role of the time limit in employment gains. 
In fact, the study shows that employment rates among ABAWDs leaving 
SNAP were highest in areas exempt from the time limit due to high 
unemployment rates--higher than the employment rates in areas using the 
individual exemptions and areas in which the time limit was in 
effect.\263\ This shows that factors other than the time limit have 
more impact on employment outcomes for ABAWDs. FNS fails to address 
this in the NPRM. Here we again recommend that FNS review and consider 
this research.
---------------------------------------------------------------------------
    \263\ See Exhibit IV-I, Philip Richardson, et al., Food Stamp 
Leavers Research Study--Study of ABAWDs Leaving the Food Stamp Program 
in South Carolina Final Report, page IV-2, https://naldc.nal.usda.gov/
download/45220/PDF.
---------------------------------------------------------------------------
    The Arizona study commissioned by USDA to understand the outcomes 
of people leaving SNAP showed that ABAWDs had worse outcomes on a 
number of employment-based metrics. They were less likely to have 
achieved self-sufficiency, less likely to have improved their 
employment situation, and more likely to be at risk of hardship or 
deprivation. The study grouped SNAP participants in Arizona in 1997 
into three categories: ABAWDs, non-ABAWD individuals on TANF, and non-
ABAWD, non-TANF adults. Upon leaving SNAP, employment rates for each 
subgroup were over 50 percent, but by one year later, ABAWDs who had 
been cut off SNAP had experienced the greatest drop in employment.\264\ 
This suggests that: (1) the 3 month time limit may not be an important 
factor in causing employment, since SNAP participants not subject to 
the time limit had similar levels of employment as those subject to the 
time limit when leaving SNAP; and (2) ABAWDs struggle to maintain 
employment. The report identifies several reasons why the ABAWD group 
might see a sharp decline in employment. In a survey of ABAWDs who were 
no longer on SNAP and not working, more than 60 percent reported being 
ill or having health problems or a disability--circumstances not 
identified by the SNAP agency as qualifying an individual to an 
exemption.\265\
---------------------------------------------------------------------------
    \264\ See Exhibit 2-6, Covered Employment, in Gregory Mills and 
Robert Kornfeld, Study of Arizona Adults Leaving the Food Stamp 
Program, Final Report, Dec. 2000, page 36, https://naldc.nal.usda.gov/
download/45673/PDF.
    \265\ Gregory Mills and Robert Kornfeld, Study of Arizona Adults 
Leaving the Food Stamp Program, Final Report, Dec. 2000, page 51, 
https://naldc.nal.usda.gov/download/45673/PDF. Individuals with a 
medically certified physical or mental condition that prevents them 
from working are exempt from the 3 month time limit, but states have 
struggled with correctly identifying these individuals. The study did 
not look at whether some of the ABAWDs who left SNAP should have been 
exempt from the time limit.
---------------------------------------------------------------------------
    Across all four studies, many who lost SNAP benefits were employed 
but had low earnings. Between 41 and 76 percent of the former 
recipients in the four states were working after leaving SNAP, but 
earnings were low and many remained in poverty.\266\ Most of those who 
were not working in Arizona and Illinois were in poor health or caring 
for a family member in poor health. Given the complexity of SNAP's 
rules governing work effort, disability, and time limits, some may not 
have realized that they may again qualify for SNAP and failed to 
reapply for benefits.\267\
---------------------------------------------------------------------------
    \266\ Elizabeth Dagata, ``Assessing the Self-Sufficiency of Food 
Stamp Leavers,'' Economic Research Service, USDA, September 2002, 
https://www.ers.usda.gov/publications/pub-details/?pubid=46645.
    \267\ Elizabeth Dagata, ``Assessing the Self-Sufficiency of Food 
Stamp Leavers,'' Economic Research Service, USDA, September 2002, 
https://www.ers.usda.gov/publications/pub-details/?pubid=46645.
---------------------------------------------------------------------------
    While the studies of individuals leaving SNAP looked at all types 
of individuals, the studies yielded important results for ABAWDs that 
are relevant to this proposed rule but do not appear in the 
Department's rationale for the NPRM. For example, in Illinois in 1997, 
the single largest category of individuals losing benefits was ABAWDs. 
The impact varied widely among groups. For example, \2/3\ of the ABAWDs 
leaving the program were African American, well above the percentage of 
all leavers (which was 50 percent).\268\ As discussed more fully in 
Chapter 12, the NPRM acknowledges the proposed rule would have a 
disparate impact by noting the ``potential for disparately impacting 
certain protected groups due to factors affecting rates of employment 
of members of these groups.'' \269\ But the Food and Nutrition Act 
makes clear that the regulations implementing Title VI and other civil 
rights statutes are fully applicable to SNAP. These regulations 
prohibit actions in Federal programs that have disparate impacts on 
members of protected groups as well as intentional discriminatory acts. 
Therefore, the proposed rule's disparate impact on these individuals, 
as demonstrated by the research and acknowledged in the NPRM itself, is 
inconsistent with the Act. Given the requirement under 7 U.S.C.  
2020(c)(2)(D) that the Department ensure the protections of Title VI of 
the Civil Rights Act of 1964 apply to all SNAP households, we are 
stunned that the Department did not review its own research results 
that clearly suggest that the proposed policy would have widely 
disparate impacts on African Americans.
---------------------------------------------------------------------------
    \268\ Anu Rangarajan and Philip M. Gleason, Food Stamp Leavers in 
Illinois: How Are They Doing Two Years Later?, Mathematica Policy 
Research, Inc., January 2001, p. 21, https://www.mathematica-mpr.com/
our-publications-and-findings/publications/food-stamp-leavers-in-
illinois-how-are-they-doing-two-years-later.
    \269\ NPRM, p. 990.
---------------------------------------------------------------------------
D. The Loss of SNAP Due to the Time Limit Fails to Raise Income and 
        Increases Hardship
    Childless adults who lose SNAP benefits struggle without food 
assistance benefits. As noted, USDA's most comprehensive assessment of 
former SNAP recipients in four states in the early 2000s suggests that 
their life circumstances are quite difficult.\270\ A significant 
minority don't find work, and among those who are employed after 
leaving SNAP, earnings are low. Most remain poor. Many struggle to 
acquire enough food to meet their needs, lack health insurance, 
experience housing problems, and/or have trouble paying their bills. 
(These studies include people who leave SNAP because of the 3 month 
time limit or for other reasons, for example, because they found a job 
or mistakenly believed they were no longer eligible.)
---------------------------------------------------------------------------
    \270\ Elizabeth M. Dagata, ``Assessing the Self-Sufficiency of Food 
Stamp Leavers,'' Economic Research Service, USDA, September 2002, 
https://ageconsearch.umn.edu/record/262256/files/31106_fanrr26-
8_002.pdf. This is a summary of in-depth studies in Arizona, Illinois, 
Iowa, and South Carolina.
---------------------------------------------------------------------------
    Despite relatively high levels of work effort, between \1/3\ and 
roughly \2/3\ of SNAP leavers in the four states had household incomes 
below the poverty line--well above the overall poverty rate of 13 
percent. Many of these households experienced severe poverty after 
leaving SNAP: about 40 percent of the leavers in two states were below 
half of the poverty line.
    Many struggled to acquire adequate food. Between 17 and 34 percent 
of the SNAP leavers in the four states reported very low food security 
(meaning they had to skip or reduce the size of their meals or 
otherwise disrupt their eating patterns at times during the year 
because they couldn't afford enough food), compared with three percent 
of all households without children.
    USDA's study of adults in Arizona leaving SNAP found that the 
incidence of moderate or severe hunger was greatest among the ABAWD 
subgroup, at 34 percent, compared to 23 percent for the TANF subgroup 
and 18 percent for the non-TANF subgroup.\271\ By comparison, 3.5 
percent of all Arizona households were classified as facing moderate or 
severe hunger. The Arizona study concludes by pointing out that 
individuals who might appear to be self-sufficient or better off after 
leaving SNAP, because they receive fewer public benefits and report 
less private support, might still be facing significant hardship:
---------------------------------------------------------------------------
    \271\ Gregory Mills and Robert Kornfeld, Study of Arizona Adults 
Leaving the Food Stamp Program, Final Report, Dec. 2000, page 85, 
https://naldc.nal.usda.gov/download/45673/PDF.

          The high rate of food insecurity with hunger found among 
        ABAWD exiters--34 percent--is noteworthy. This incidence is 
        more than twice the 1999 national rate of 14 percent estimated 
        by USDA for households at or below 50 percent of the poverty 
        level, even though most ABAWDs have incomes above the poverty 
        level. The ABAWD finding highlights the importance of 
        considering (in this and other exit studies) whether exiters 
        who appear self-sufficient, in terms of their reduced reliance 
        on public and private support, are able to avoid hardship and 
        deprivation.\272\
---------------------------------------------------------------------------
    \272\ Gregory Mills and Robert Kornfeld, Study of Arizona Adults 
Leaving the Food Stamp Program, Final Report, Dec. 2000, page 94, 
https://naldc.nal.usda.gov/download/45673/PDF.

    This cautionary note is not found in the NPRM. The agency simply 
asserts that it believes expanding the number of individuals subject to 
the time limit will improve their self-sufficiency, without 
acknowledging what the agency-funded research revealed--that the 
outcome for ABAWDs leaving because of the time limit may be increased 
hunger and hardship.
    The studies also showed that many lacked health insurance, had 
housing problems, or had trouble paying their utility bills. About 30 
to 40 percent of the SNAP leavers in the four states faced housing 
issues, including falling behind on rent, moving in with relatives, or 
becoming homeless. Between 20 and 65 percent reported problems paying 
for utilities. Just over \1/2\ of the SNAP leavers in two of the states 
were uninsured.\273\ These are all characteristics, as we illustrate 
elsewhere in the comments, associated with much higher rates of 
unemployment. We continue to be baffled about how or whether USDA 
factored its own research into the NPRM. It would appear that the 
Department ignored its own studies.
---------------------------------------------------------------------------
    \273\ Elizabeth M. Dagata, ``Assessing the Self-Sufficiency of Food 
Stamp Leavers,'' Economic Research Service, USDA, September 2002, 
https://www.ers.usda.gov/publications/pub-details/?pubid=46645. This is 
a summary of in-depth studies in Arizona, Illinois, Iowa, and South 
Carolina.
---------------------------------------------------------------------------
E. Evidence From Other Benefit Programs Shows That Time Limits Do Not 
        Increase Employment and Have a Disproportionate Impact on 
        Certain Populations
    The stated rationale for proposing a change to the long-standing 
waiver process is to expose more individuals to the time limit, in the 
belief that this will result in increased labor market attachment. The 
NPRM provides no evidence to support this assertion. But, research on 
the Temporary Assistance for Needy Families (TANF) program, which 
imposes both work requirements and a time limit for benefits, 
undermines the claim that work requirements and time-limited benefits 
increase employment.
    A review of the many studies on families whose TANF monthly direct 
financial support was reduced or taken away due to work requirements 
shows that these policies harm individuals and families, most of whom 
face significant obstacles to employment, while producing few lasting 
gains in employment. While the studies described in this section are 
about families with children--often single-mother-headed households--
the findings are relevant because they review the circumstances of very 
poor households, similar in important respects to households on SNAP. 
In addition, many ABAWDs are parents of non-minor children or have 
children not in the SNAP household. TANF and SNAP households live in 
similar circumstances and face similar challenges finding employment, 
so the outcomes from one group can inform the likely outcomes for the 
other.
    A time limit ignores the fact that public assistance recipients 
often vary in their needs and circumstances; many often live with one 
or multiple significant barriers to employment. Those barriers range 
from low cognitive functioning, to mental and physical health problems, 
to criminal justice issues, to low measures of human capital. These 
barriers make it hard to find or keep a job or fulfill other work 
requirements. Extensive research on the effect of time limits in TANF, 
which provides modest cash assistance and requires a portion of the 
caseload to engage in work activities, shows that the time limit did 
not notably increase employment, but it did result in increased 
hardship. We strongly recommend that USDA read these studies and 
consider the findings when developing final regulations. We are 
confident the results demonstrate the problems with the proposed 
policy.
F. Most Low-Income People Sanctioned From Public Assistance Due to Work 
        Requirements Face Barriers Finding Employment
    Studies show that many parents who lose TANF benefits due to work 
requirements have significant employment barriers. Those losing 
benefits are more likely than other TANF parents to have physical, 
mental health, or substance use issues; to be fleeing domestic 
violence; to have low levels of education and limited work experience; 
or to face significant logistical challenges, such as lack of access to 
or funds to pay for child care and transportation.\274\ Below are 
summaries of several TANF studies finding that many adults losing 
assistance due to sanctions face significant employment barriers:
---------------------------------------------------------------------------
    \274\ LaDonna Pavetti, Michelle K. Derr, and Heather Hesketh, 
``Review of Sanction Policies and Research Studies,'' Mathematica 
Policy Research (March 2003).

   A 2007 study of regional variation of full-family 
        sanctioning practices in Florida's TANF cash assistance program 
        gives evidence that sanctions are significantly related to 
        various barriers to employment. TANF parents with lower income 
        and lower levels of education were more likely to be sanctioned 
        than those with higher income and education levels. For 
        example, recipients with a high school degree were more likely 
        to be sanctioned than those with more education--however, 
        sanctions were the most common among those with less than a 
        high school degree. Community traits can matter too: families 
        were more likely to be sanctioned in counties with higher 
        poverty rates than other counties, after controlling for other 
        characteristics.\275\
---------------------------------------------------------------------------
    \275\ Richard C. Fording, Joe Soss, and Sanford F. Schram, 
``Devolution, Discretion, and the Effect of Local Political Values on 
TANF Sanctioning,'' Social Service Review (June 2007), pp. 285-316.

   A 2006 study of women in Wisconsin receiving TANF found that 
        the state was more likely to sanction mothers with lower levels 
        of education. Specifically, mothers with at least a high school 
        diploma or equivalent were less likely to be sanctioned than 
        mothers with less than a high school diploma, and those with 
        education beyond high school were even less likely to be 
        sanctioned, even when controlling for how long each individual 
        received TANF grants. The study also concluded that those ``who 
        may be least able to succeed in the labor market are most 
        likely to be sanctioned.'' Specifically, the authors examined 
        sanction activity against the mothers' employment status in the 
        2 years preceding entry to the TANF program. The authors' 
        estimates show a monotonic trend with the number of quarters of 
        work: those with no work in the past 2 years were most likely 
        to be sanctioned, those with 1-4 quarters of employment were 
        less likely to be sanctioned, those with 5-7 quarters of 
        employment were even less likely to be sanctioned, and mothers 
        who had been employed for all eight quarters were the least 
        likely to be sanctioned.\276\
---------------------------------------------------------------------------
    \276\ Chi-Fang Wu, et al., ``How Do Welfare Sanctions Work?,'' 
Social Work Research (March 2006), pp. 33-50.

   A 2002 comparison of sanctioned and non-sanctioned TANF 
        recipients in Boston, Chicago, and San Antonio found sanctioned 
        recipients were less likely than other TANF recipients to have 
        a high school degree or its equivalent, a working telephone at 
        home, or a car. They were more likely to report being in fair 
        or poor health, have a substance use issue, or have a partner 
        who interfered with their employment, training, or schooling. 
        They also had less work experience, lived in neighborhoods with 
        undesirable qualities (such as abandoned houses, assaults and 
        muggings, gangs, and open drug dealing), and reported living in 
        housing of poor quality.\277\
---------------------------------------------------------------------------
    \277\ Andrew J. Cherlin, et al., ``Operating within the Rules: 
Welfare Recipients' Experience with Sanctions and Case Closings,'' 
Social Service Review (September 2002), pp. 387-662.

   A study of women in Michigan receiving TANF shows that those 
        with educational barriers to employment were more likely to be 
        sanctioned than those without such barriers. Women with less 
        than a high school education were 2.06 times more likely to be 
        sanctioned than women with higher levels of formal 
        education.\278\
---------------------------------------------------------------------------
    \278\ Ariel Kalil, Kristin S. Seefeldt, and Hui-chen Wang, 
``Sanctions and Material Hardship under TANF,'' Social Service Review 
(December 2002), pp. 642-662.

   The same study of women in Michigan receiving TANF shows 
        those with transportation-related barriers to employment were 
        more likely to be sanctioned than those without such barriers. 
        Specifically, recipients who lacked either a car or a driver's 
        license were disproportionately sanctioned relative to 
        recipients without these transportation barriers. In addition, 
        those with trauma-related barriers to employment were more 
        likely to be sanctioned than those without such barriers. 
        Specifically, women who reported experiencing severe domestic 
        violence (being hit, kicked, shoved, etc.) within the past year 
        were disproportionately sanctioned relative to recipients who 
        did not report experiences of that type.\279\
---------------------------------------------------------------------------
    \279\ Ariel Kalil, Kristin S. Seefeldt, and Hui-chen Wang, 
``Sanctions and Material Hardship under TANF,'' Social Service Review 
(December 2002), pp. 642-662.

   A California study of a random sample of [CalWORKs] (TANF) 
        recipients in four counties (two large urban counties and two 
        large semi-rural counties) found that recipients without a car 
        were roughly 1.5 times more likely to incur sanctions than 
        recipients who owned a car. Recipients who were sanctioned were 
        less likely to own a car (39 percent of respondents) in 
        comparison to non-sanctioned recipients (52 percent of 
        respondents).\280\
---------------------------------------------------------------------------
    \280\ Yeheskel Hasenfeld, Toorjo Ghose, and Kandyce Larson, ``The 
Logic of Sanctioning Welfare Recipients: An Empirical Assessment,'' 
Social Science Review (June 2004), Vol. 78, No. 2, pp. 304-319, http://
repository.upenn.edu/spp_papers/88/.

   In Illinois, parents who had ever been sanctioned were 
        significantly less likely than those never sanctioned to have a 
        high school diploma or its equivalent and more likely to have 
        limited recent work experience. They also were significantly 
        more likely to be dealing with a physical or mental health 
        issue, to have been arrested multiple times, and to have 
        experienced a child care issue. In South Carolina, parents ever 
        sanctioned were significantly more likely to have a physical 
        health problem, show signs of a learning disability, and have a 
        family member or friend with a health care issue or special 
        need. \281\
---------------------------------------------------------------------------
    \281\ LaDonna Pavetti, et al., ``The Use of TANF Work-Oriented 
Sanctions in Illinois, New Jersey, and South Carolina,'' Mathematica 
Policy Research (April 2004).

   A study of 656 TANF leavers from the 1999 and 2002 data of 
        the National Survey of America's Family (NSAF) found that TANF 
        leavers who reached their lifetime limits had a higher chance 
        of having problems with employment due to work barriers or 
        vulnerable characteristics such as old age, physical or mental 
        health problems. They experienced greater hardship because they 
        had less income and less EITC receipt, most had experienced a 
        cutoff of SNAP, and fewer received child care assistance. The 
        author concluded that time limits can lead low-income families 
        to endure greater economic hardships.\282\
---------------------------------------------------------------------------
    \282\ Kyoung Hag Lee, ``Effect on Lifetime Limits and Differences 
between TANF Leavers Who Had Reached Their Lifetime Limits and Those 
Who Had Exited Voluntarily,'' Poverty & Public Policy (2010), pp. 1-22.

   A study that examined time limits on the receipt of welfare 
        in both the United States and British Columbia, Canada found 
        that time limits are an ineffective policy tool as they 
        increase barriers to employment and result in recipients 
        needing more support and access to specific programs. 
        Recipients who exhausted their benefits struggled financially 
        and had a difficult time finding a job. While some recipients 
        had several barriers to work, others who had fewer barriers 
        were still unable to find a job. Low cognitive functioning, 
        limited education, and physical and mental health problems were 
        some of the barriers that recipients faced.\283\
---------------------------------------------------------------------------
    \283\ Dean Herd, Ernie Lightman, and Andrew Mitchell, ``Welfare 
Time Limits: Symbolism and Practice'', U-Toronto: Social Assistance in 
the New Economy (SANE) Project (2008), pp. 1-26.

   A 2006 study of Minnesota's TANF program explored in detail 
        the circumstances of those at or near 60 months, the Federal 
        time limit. It found about 16 percent of cases had a case head 
        with an IQ of less than 80, about 20 percent were caring for 
        ill or incapacitated family members, and about 21 percent were 
        ill or incapacitated themselves for 30 days or more. Long-term 
        TANF recipients also had mental illness (which is often 
        untreated or inadequately treated), were developmentally 
        disabled or learning disabled, were leaving domestic violence 
        situations, or were otherwise ``unemployable.'' Some recipients 
        suffered from chronic and debilitating health problems because 
        they had worked physically demanding jobs.\284\
---------------------------------------------------------------------------
    \284\ LaDonna Pavetti and Jacqueline Kauff, ``When Five Years Is 
Not Enough: Identifying and Addressing the needs of Families Nearing 
the TANF Time Limit in Ramsey County, Minnesota,'' Mathematica Policy 
Research (March 2006), pp. 1-24, https://www.mathematica-mpr.com/our-
publications-and-findings/publications/when-five-years-is-not-enough-
identifying-and-addressing-the-needs-of-families-nearing-the-tanf-time-
limit-in-ramsey-county-minnesota.

   Another report on Minnesota's TANF program shows the percent 
        of persons with a severe mental health diagnosis at 60 months, 
        the Federal time limit, to be about 52 percent and those with a 
        chemical dependency diagnosis to be 27 percent. Among American 
        Indian recipients, about 60 percent of those near the time 
        limit had a mental health diagnosis and/or a chemical 
        dependency diagnosis.\285\
---------------------------------------------------------------------------
    \285\ Dana DeMaster, ``At the Limit: December 2006 Minnesota Family 
Investment Program (MFIP) Cases that Reached the 60 Month Time Limit,'' 
Minnesota Department of Human Services (January 2008), pp. 1-18, 
https://edocs.dhs.state.mn.us/lfserver/Legacy/DHS-5092B-ENG.

   A more recent Washington State study compared families who 
        left TANF due to time limits and those who left for other 
        reasons. The state found that time-limited families were more 
        likely to be unstably housed in the year prior to losing 
        assistance. They were also more likely to have chronic health 
        issues and visit the emergency room. And, they were more likely 
        to have a range of behavioral health needs, from mental health 
        issues to substance abuse disorders.\286\
---------------------------------------------------------------------------
    \286\ Christina McHugh and J. Taylor Danielson, ``TANF Time Limit 
Analysis Comparing Cases Closed Due to Time Limits with Other Case 
Closures,'' Washington State Department of Social and Health Services 
(February 2019).

   In a survey of 276 West Virginia TANF recipients cut off due 
        to time limits, respondents identified several barriers to 
        employment. Most of the respondents were unemployed after 
        leaving TANF. More than \1/2\ (56.2 percent) of respondents 
        were not working because of a physical or mental illness or 
        disability problem; more than \1/3\ (37.1 percent) had no 
        transportation; \1/3\ (32.6 percent) did not have the right 
        education; and a little less than \1/3\ (29.8 percent) simply 
        could not find a job. Most of these respondents had multiple 
        barriers to employment.\287\
---------------------------------------------------------------------------
    \287\ Robert Jay Dilger, et al., ``WV WORKS 2003: Perspectives of 
Former Recipients Who Have Exhausted Their 60 Months of Program 
Eligibility,'' West Virginia University Interdisciplinary Task Force on 
Welfare Reform (Summer 2004), pp. 1-24.

   A California study of the cases that reach the time limit 
        found families often struggle with one major barrier to work 
        and often multiple barriers. One-third of respondents cited 
        major health issues as a big barrier to work. A smaller share 
        said they were caring for a family member with a major health 
        issue. More than \1/5\ of respondents said they suffered from 
        depression or anxiety or had experienced at least one stressful 
        event in the past year. About 11 percent experienced a domestic 
        violence situation. More than \1/2\ of all respondents noted 
        having at least one barrier and 28 percent cited having two or 
        more barriers. Forty-three percent said they had trouble paying 
        rent; 54 percent said they had trouble paying utility bills; 39 
        percent reported having trouble paying for food; and 40 percent 
        noted they had to use a food bank.\288\
---------------------------------------------------------------------------
    \288\ Rebecca A. London and Jane G. Mauldon, ``Time Running Out: A 
Portrait of California Families Reaching the CalWORKs 60-Month Time 
Limit in 2004,'' Welfare Policy Research Project (November 2006), pp. 
1-6. http://escholarship.ucop.edu/uc/item/37g3510z.
---------------------------------------------------------------------------
G. Imposing Time Limits and Sanctions for Failure to Meet Work 
        Requirements Has a Disparate Impact on Communities of Color
    States' application of work requirements in the TANF cash 
assistance program has exacerbated racial inequities, research shows. 
On the whole, research on TANF suggests that policies to take away SNAP 
from individuals who are not working or participating in work 
activities for a specific number of hours each month will hurt, not 
help, the individuals most in need of assistance. Nearly every study 
comparing the race and ethnicity of sanctioned and non-sanctioned TANF 
recipients finds that African Americans are significantly more likely 
to be sanctioned than their white counterparts.\289\ For example:
---------------------------------------------------------------------------
    \289\ LaDonna Pavetti, Michelle K. Derr, and Heather Hesketh, 
``Review of Sanction Policies and Research Studies,'' Mathematica 
Policy Research (March 2003).

   A 2011 study of Minnesota TANF recipients found that the 
        state sanctioned a disproportionate number of American Indian 
        or Alaskan Native recipients and sanctioned them more often 
        than other racial groups during a 24 month observation period. 
        While American Indian or Alaskan Natives only comprised around 
        10.9 percent of the families receiving TANF in Minnesota, they 
        made up 12.2 percent of all families that the state sanctioned. 
        Further, the average number of sanctions was 3.54 for American 
        Indian and Alaskan Native families, while it was only 3.18 for 
        White, non-Hispanic families.\290\
---------------------------------------------------------------------------
    \290\ Anita M. Larson, Shweta Singh, and Crystal Lewis, ``Sanctions 
and Education Outcomes for Children in TANF Families,'' Child & Youth 
Services (September 2011), pp. 180-199.

   A 2007 study of Florida's TANF program showed black families 
        were more likely to be sanctioned than White families after 
        several months of continuous TANF receipt. Specifically, the 
        study estimated that among families who received TANF benefits 
        for at least 9 months continuously, black families were 22 to 
        35 percent more likely to be sanctioned than White 
        families.\291\
---------------------------------------------------------------------------
    \291\ Richard C. Fording, Joe Soss, and Sanford F. Schram, 
``Devolution, Discretion, and the Effect of Local Political Values on 
TANF Sanctioning,'' Social Service Review (June 2007), pp. 285-316.

   A 2006 study of women in Wisconsin receiving TANF found that 
        black women were more likely to be sanctioned than their white 
        counterparts. This result was statistically significant under 
        both a simple analysis and an analysis that took into account 
        the duration of each individual's receipt of TANF grants.\292\
---------------------------------------------------------------------------
    \292\ Chi-Fang Wu, et al., ``How Do Welfare Sanctions Work?,'' 
Social Work Research (March 2006), pp. 33-50.

   A 2002 study of women receiving TANF in Michigan found that 
        black women were disproportionately sanctioned compared to 
        white women. The authors found similar results under two 
        different specifications. One analysis, looking at mean 
        differences, found that black women made up a disproportionate 
        number of the total number of women that the state sanctioned. 
        The other model, a multivariate logistic regression, provided 
        similar evidence: African American women were 1.73 times more 
        likely to be sanctioned than White women.\293\
---------------------------------------------------------------------------
    \293\ Ariel Kalil, Kristin S. Seefeldt, and Hui-chen Wang, 
``Sanctions and Material Hardship under TANF,'' Social Service Review 
(December 2002), pp. 642-662.

   A 2011 study of TANF recipients in Maryland found evidence 
        that African Americans were more likely to lose benefits due to 
        sanctions compared to other recipients. The study found that 
        African Americans were disproportionately represented among 
        families that had been sanctioned as a result of work 
        requirements relative to respondents of other races.\294\
---------------------------------------------------------------------------
    \294\ Sarah Williamson, ``Full-Family Sanctions & Economic 
Recession,'' University of Maryland Family Research and Training Group 
(January 2011). https://familywelfare.umaryland.edu/reports1/
sanctionsbrief.pdf.

   A New Jersey study found that among TANF recipients entering 
        the program between July 2000 and June 2001, 36 percent of 
        African American recipients had their TANF grants reduced and 
        16 percent had their grant eliminated due to a work-related 
        sanction; the comparable figures for white recipients were 27 
        percent and ten percent, respectively.\295\
---------------------------------------------------------------------------
    \295\ LaDonna Pavetti, et al., ``The Use of TANF Work-Oriented 
Sanctions in Illinois, New Jersey, and South Carolina,'' Mathematica 
Policy Research (April 2004).

    Data from the Wisconsin Department of Workforce Development shows a 
consistent pattern of racial and ethnic discrepancies in TANF 
sanctions. Statewide, 42 percent of African American participants and 
45 percent of Hispanic participants were sanctioned, compared to just 
24 percent of white participants.\296\ Researchers in several states 
have looked at the demographics of the share of their caseload 
approaching or at the time limit. In a number of examples, including a 
national survey, families and recipients of color--and particularly 
black recipients--are more likely to lose benefits due to the time 
limit. This evidence suggests that expanding time limits in other 
programs will disproportionally affect black recipients. Other research 
cited below highlight the unique challenges African Americans face in 
the labor market. Loss of assistance, paired with the difficulty of 
securing or maintaining a job, could make the hardship experienced by 
this group even worse.
---------------------------------------------------------------------------
    \296\ Wisconsin Department of Workforce Development, ``Wisconsin 
Works (W-2) Sanctions Study,'' (December 2004).

   The Minnesota TANF agency found that black women, including 
        African Americans, Somali immigrants, and other African 
        immigrants, made up about \1/2\ of the adult recipients who 
        reached 60 months, the Federal time limit. African Americans in 
        particular were the most likely to reach the Federal time 
        limit.\297\
---------------------------------------------------------------------------
    \297\ Dana DeMaster, ``At the Limit: December 2006 Minnesota Family 
Investment Program (MFIP) Cases that Reached the 60 Month Time Limit,'' 
Minnesota Department of Human Services (January 2008), pp. 1-18, 
https://edocs.dhs.state.mn.us/lfserver/Legacy/DHS-5092B-ENG.

   A study in Maryland found most of the caseload consisted of 
        black women and they were the most likely of any racial group 
        to reach the time limit.\298\
---------------------------------------------------------------------------
    \298\ Pamela Ovwigho, Kathryn Patterson, and Catherine E. Born, 
``The TANF Time Limit: Barriers & Outcomes among Families Reaching the 
Limit,'' Family Welfare Research and Training Group (November 2007), 
pp. 1-34, https://familywelfare.umaryland.edu/reports1/tl_barriers.pdf.

   In Virginia, black families were more likely to reach the 
        limit than white families.\299\
---------------------------------------------------------------------------
    \299\ Anne Gordon, et al., ``Experiences of Virginia Time Limit 
Families After Case Closure: An Interim Report,'' Mathematica (2002), 
pp. 1-177, https://www.mathematica-mpr.com/-/media/publications/pdfs/
expvafinal.pdf.

   In Washington State, families cut off by the time limit 
        tended to be black or American Indian.\300\
---------------------------------------------------------------------------
    \300\ Christina McHugh and J. Taylor Danielson, ``TANF Time Limit 
Analysis Comparing Cases Closed Due to Time Limits with Other Case 
Closures,'' Washington State Department of Social and Health Services 
(February 2019).

   Using the Women's Employment Study for one Michigan urban 
        county, researchers analyzed factors associated with increased 
        time on TANF. They found that black women were far more likely 
        to have accumulated more months, and thus be closer to the time 
        limit, than white women.\301\
---------------------------------------------------------------------------
    \301\ Kristin S. Seefeldt and Sean M. Orzol, ``Watching the Clock 
Tick: Factors Associated with TANF Accumulation,'' National Poverty 
Center Working Paper Series (May 2005), http://www.npc.umich.edu/
publications/workingpaper04/paper9/04-09.pdf.

    SNAP participants of color also face discrimination when looking 
for work. Investigations in job discrimination uncovered strong 
employer preferences for white candidates over candidates of color. One 
study found that resumes with white-sounding names are more likely to 
get call-backs than resumes with equal qualifications but with black-
sounding names. Other research shows that generally, those with a 
criminal record are less likely to get call-backs or requests for 
interviews from employers. Furthermore, black applicants without 
criminal records are less likely to receive favorable treatment than 
white applicants without criminal records, but white applicants with a 
criminal record are more likely to receive favorable treatment than 
black applicants with no criminal history.\302\
---------------------------------------------------------------------------
    \302\ Devah Pager, ``The Mark of a Criminal Record,'' American 
Journal of Sociology (2003), pp. 937-975.
---------------------------------------------------------------------------
H. Households That Lose TANF Benefits Because of Sanctions or Time 
        Limits Experience Higher Levels of Material Hardship and 
        Increased Hardship
    People with incomes low enough to qualify for SNAP also often have 
few or no assets to lean on in difficult times and a limited amount of 
cash to meet basic needs like rent and utilities, clothes, personal 
care items, and gas or bus fare, among other things. SNAP helps meet 
food costs, but when that assistance is taken away, individuals 
struggle to make ends meet and some are unable to avoid a downward 
spiral. Studies examining sanctioned TANF families show that many 
people experience increased hardships after facing a sanction: a 2004 
longitudinal study of TANF recipients in Illinois found that sanctioned 
families who faced grant reductions had higher levels of food hardship 
after being sanctioned than those who did not have their grants reduced 
by sanctions. Researchers defined food hardship as sometimes or often 
not having enough to eat. Recipients who had sanctions in the period 
January 1999 to March 2001 reported between February 2002 and September 
2002 a higher incidence of food hardship and perceived overall hardship 
than other TANF recipients who did not experience grant reductions 
resulting from sanctions during the period. A multivariate analysis 
conducted with the same data indicated respondents who had their grants 
reduced due to sanctions were over three times more likely to report 
food hardship in the final period of the study (after the sanction) 
compared to those who were never sanctioned during the study, after 
controlling for demographic and other factors.\303\
---------------------------------------------------------------------------
    \303\ Bong Joo Lee, Kristen S. Slack, and Dan A. Lewis, ``Are 
Welfare Sanctions Working as Intended? Welfare Receipt, Work Activity, 
and Material Hardship among TANF-Recipient Families,'' Social Service 
Review (September 2004), pp. 370-403.
---------------------------------------------------------------------------
    A 2004 longitudinal study of TANF recipients in Illinois found 
those who had their grants reduced due to sanctions reported higher 
levels of perceived overall hardship. Perceived overall hardship was 
determined by the extent to which respondents agreed with statements 
like, ``I worry about having enough money in the future.'' Respondents 
who saw grant reductions as a result of sanctions between January 1999 
and March 2001 showed more perceived overall hardship following the 
sanctions (between February and September 2002) than those who were 
never sanctioned during the former, 51 month period.\304\
---------------------------------------------------------------------------
    \304\ Bong Joo Lee, Kristen S. Slack, and Dan A. Lewis, ``Are 
Welfare Sanctions Working as Intended? Welfare Receipt, Work Activity, 
and Material Hardship among TANF-Recipient Families,'' Social Service 
Review (September 2004), pp. 370-403.
---------------------------------------------------------------------------
    A 2002 study of women in Michigan receiving TANF suggests those who 
were sanctioned were more likely to experience hardship and have to 
prioritize hardship-mediating activities than those who were not 
sanctioned. Specifically, 21 percent of sanctioned families (compared 
to nine percent of non-sanctioned families) reported having their gas 
or electricity turned off in the previous year because they could not 
afford to make their utility payments.\305\ Researchers found that 34 
percent of sanctioned families (compared to 14 percent of non-
sanctioned families) had resorted to hardship-mitigating activities 
such as pawning, stealing food, searching in trash cans, or 
begging.\306\
---------------------------------------------------------------------------
    \305\ Ariel Kalil, Kristin S. Seefeldt, and Hui-chen Wang, 
``Sanctions and Material Hardship under TANF,'' Social Service Review 
(December 2002), pp. 642-662.
    \306\ Ariel Kalil, Kristin S. Seefeldt, and Hui-chen Wang, 
``Sanctions and Material Hardship under TANF,'' Social Service Review 
(December 2002), pp. 642-662.
---------------------------------------------------------------------------
    The 2002 study of Michigan TANF households suggests women who were 
sanctioned are more likely to expect future hardship--such as 
inadequate housing, food, or medical care in the next 2 months--than 
those who were not sanctioned. Women who were sanctioned were 2.41 
times more likely to expect future hardship compared to women who were 
not sanctioned, the researchers found.\307\
---------------------------------------------------------------------------
    \307\ Ariel Kalil, Kristin S. Seefeldt, and Hui-chen Wang, 
``Sanctions and Material Hardship under TANF,'' Social Service Review 
(December 2002), pp. 642-662.
---------------------------------------------------------------------------
    There is evidence from a study of people who frequented a food 
pantry in upstate New York that people who lost TANF benefits due to 
sanctions were more likely to experience hardship than others. 
Specifically, it was more common for people who had been sanctioned--
relative to those who had not been sanctioned--to report having more 
difficulty in the past 6 months paying for food, rent, adult health 
care, and other bills. Similarly, the number of sanctioned respondents 
who indicated they had moved within the past 6 months due to inability 
to pay rent was disproportionately high relative to the responses from 
those who were not sanctioned. Access to a telephone told a similar 
story of increased hardship, with a disproportionate number of 
sanctioned people lacking access to a phone, relative to those who were 
not sanctioned.\308\
---------------------------------------------------------------------------
    \308\ Jean Oggins and Amy Fleming, ``Welfare Reform Sanctions and 
Financial Strain in a Food-Pantry Sample,'' The Journal of Sociology & 
Social Welfare (June 2001), pp. 101-123, https://
scholarworks.wmich.edu/cgi/viewcontent.cgi?article=2725&context=jssw.
---------------------------------------------------------------------------
    A study using data from the Fragile Families and Child Wellbeing 
survey, which surveys mothers from 20 cities in 15 states, evaluated 
the level of hardship among those who had been sanctioned in the prior 
12 months and non-sanctioned mothers who received TANF. Researchers 
found that those who had been sanctioned in the prior 12 months were 85 
percent more likely to report any material hardship compared to non-
sanctioned mothers receiving TANF (42 percent of those sanctioned 
reported one or more material hardships, compared to 27 percent of 
those not sanctioned). Researchers found that those sanctioned were 63 
percent more likely than non-sanctioned mothers to report maternal or 
child hunger and 76 percent more likely to report having their 
utilities shut off in the 12 months prior to the interview. The study 
also found that sanctioned mothers were 79 percent more likely to 
report being unable to receive medical care, for either themselves or a 
child, due to cost. The study controlled for sociodemographic factors 
known to be associated with being sanctioned and controlled for 
hardship levels mothers faced prior to being sanctioned.\309\
---------------------------------------------------------------------------
    \309\ Nancy E. Reichman, et al., ``TANF Sanctioning and Hardship,'' 
Social Service Review (June 2005) Vol. 79 No. 2, pp. 215-236.
---------------------------------------------------------------------------
    A Washington State study using predictive modeling to identify the 
factors likeliest to cause a new spell of homelessness for TANF parents 
found that sanctioned recipients were about 20 percent more likely than 
non-sanctioned parents to begin a new spell of homelessness in the next 
month.\310\
---------------------------------------------------------------------------
    \310\ Melissa Ford Shah, et al., ``Predicting Homelessness among 
Low-Income Parents on TANF,'' Washington State Department of Social & 
Health Services (August 2015).
---------------------------------------------------------------------------
    Underlying the NPRM's proposal to restrict waivers is the claim 
that subjecting more individuals to the 3 month time limit will 
increase employment. But research on public benefit programs that have 
time limits demonstrates that arbitrary time limits do not lead to 
self-sufficiency. Instead, some research finds families cut off TANF 
because of time limits have significant barriers to employment and 
experience hardship. Without cash, the challenge for parents to support 
their children becomes even harder and a downward spiral emerges. 
Finding employment becomes even harder when parents need to scramble to 
make ends meet. The studies below offer examples of time-limited 
families unable to maintain stable housing and/or pay their bills and, 
in some instances, afford enough food.

   A Washington State study comparing families who left TANF 
        due to time limits and those who left for other reasons found 
        that time-limited families were more likely to be unstably 
        housed 1 month after losing assistance.\311\
---------------------------------------------------------------------------
    \311\ Melissa Ford Shah, et al., ``Predicting Homelessness among 
Low-Income Parents on TANF,'' Washington State Department of Social & 
Health Services (August 2015).

   According to a survey of 276 former West Virginia TANF 
        recipients cut off because of time limits, 59 percent reported 
        that they were either worse off or much worse off financially 
        since leaving WV WORKS; 65 percent did not have enough money to 
        pay the electric, gas, or water bill; and 51 percent did not 
        have enough money to pay for heat. These percentages were much 
        lower when the recipients were on WV WORKS. After losing 
        assistance because of the time limit, 61 percent of respondents 
        reported that the amount of stress in their lives was either 
        worse or much worse since being removed from WV WORKS. They 
        were also more pessimistic about their own personal and 
        financial futures. These financial burdens stem from their very 
        low level of employment: only 26.9 percent of recipients were 
        employed and more than \1/2\ of the employed were working part-
        time.\312\
---------------------------------------------------------------------------
    \312\ Robert Jay Dilger, et al., ``WV WORKS 2003: Perspectives of 
Former Recipients Who Have Exhausted Their 60 Months of Program 
Eligibility,'' West Virginia University Interdisciplinary Task Force on 
Welfare Reform (Summer 2004), pp. 1-24.

   A survey of dozens of Maine TANF recipients cut off by the 
        time limit found that families experienced increased reliance 
        on food banks, inability to pay utilities and other bills, and 
        overcrowded housing conditions or reliance on homeless 
        shelters.\313\
---------------------------------------------------------------------------
    \313\ Sandra Butler, ``TANF Time Limits and Maine Families: 
Consequences of Withdrawing the Safety Net,'' Maine Equal Justice 
Partners (April 2013), pp. 1-16, http://www.mejp.org/sites/default/
files/TANF-Study-SButler-Feb2013.pdf.
---------------------------------------------------------------------------
I. Work Requirements in TANF Do Not Work
    The rationale for reducing or eliminating benefits for not meeting 
a work requirement is that this will compel unemployed adults to find 
work. Evidence suggests that work requirements (along with other policy 
changes that accompanied TANF's implementation) contributed to a modest 
increase in employment during the late 1990s, but that work often was 
not steady, a pattern reflected in recent studies as well.
    Research on adults who lost TANF due to sanctions for failure to 
meet a work requirement found that these individuals have trouble 
finding employment after their exit. The personal, family, or community 
barriers that kept them from finding a job while on TANF also prevent 
these parents from finding work after TANF. Findings from TANF suggest 
that even if the NPRM intends to impose work requirements only on 
``work able'' individuals, substantial numbers of SNAP recipients who 
face personal or family challenges would likely fall through the cracks 
and have their benefits reduced or taken away. Evidence from studies 
that show that employment rates tend to be lower for these populations 
include the following:

   A 2004 longitudinal study of TANF recipients in Illinois 
        found that those respondents who were sanctioned during the 
        study period were more likely to be unemployed after the study 
        period than those who were not sanctioned and had, on average, 
        significantly lower levels of earnings post-sanction than those 
        who were not sanctioned. More specifically, respondents who 
        received sanctions that reduced cash grants between January 
        1999 and March 2001 were 44 percent less likely to be engaged 
        in formal work during the period April 2001 through September 
        2001 than the respondents who did not receive sanctions during 
        the preceding period, even after controlling for previous work 
        experience and other characteristics associated with 
        employment. And, TANF families receiving sanctions that were 
        carried out through grant reductions between January 1999 and 
        March 2001 had lower average earnings during the next 6 months 
        than those respondents who were not sanctioned.\314\
---------------------------------------------------------------------------
    \314\ Bong Joo Lee, Kristen S. Slack, and Dan A. Lewis, ``Are 
Welfare Sanctions Working as Intended? Welfare Receipt, Work Activity, 
and Material Hardship among TANF-Recipient Families,'' Social Service 
Review (September 2004), pp. 370-403.

      The study also helps explain why those sanctioned had worse 
        outcomes and were less likely to be working. Working-age adults 
        who have their grants reduced due to sanctions had higher 
        barriers to employment than those who were not sanctioned. The 
        group had higher levels of engagement in job training and other 
        work activities and had a higher incidence of participation in 
        informal work such as babysitting and odd jobs in the final 
        period of the study than those with no grant reductions due to 
        sanctions. This indicates the lower levels of formal employment 
        among sanctioned respondents is not easily attributed to an 
        unwillingness to work, since this population engages more 
        heavily than those who did not experience grant reductions from 
        sanctions in job training and informal work. A stronger 
        explanation is that those who are sanctioned have more 
        significant barriers to formal employment than those who are 
        not.\315\
---------------------------------------------------------------------------
    \315\ Bong Joo Lee, Kristen S. Slack, and Dan A. Lewis, ``Are 
Welfare Sanctions Working as Intended? Welfare Receipt, Work Activity, 
and Material Hardship among TANF-Recipient Families,'' Social Service 
Review (September 2004), pp. 370-403.

   A 2011 study of Maryland TANF recipients who were sanctioned 
        found that these recipients had consistently lower post-exit 
        employment rates relative to those who left TANF for reasons 
        other than work sanctions, throughout the 9 year post-exit 
        period that the study covered. Similarly, the average earnings 
        for the group that left due to work sanctions was lower than 
        the average earnings of those who left for other reasons.\316\
---------------------------------------------------------------------------
    \316\ Sarah Williamson, ``Full-Family Sanctions & Economic 
Recession,'' University of Maryland Family Research and Training Group 
(January 2011), https://familywelfare.umaryland.edu/reports1/
sanctionsbrief.pdf.

   A 2018 study of state-collected data on the employment of 
        Kansas parents leaving TANF cash assistance due to work-related 
        sanctions and time limits between October 2011 and March 2015 
        indicates that a lower share of these parents worked in the 
        year after their exit compared to families exiting TANF for 
        other reasons. They also found it more difficult to find steady 
        work compared to the other families exiting the program. In the 
        average quarter of the year after exiting, only 49 percent and 
        47 percent of the sanctioned and time-limited families, 
        respectively, were working, compared to 72 percent of families 
        exiting due to the income limit and 53 percent for all other 
        reasons. Moreover, only about a quarter of the sanctioned and 
        time-limited families worked between seven and nine quarters in 
        the year before and after their exit, compared to \1/3\ of 
        families overall.\317\
---------------------------------------------------------------------------
    \317\ Tazra Mitchell, LaDonna Pavetti, and Yixuan Huang, ``Life 
After TANF in Kansas: For Most, Unsteady Work and Earnings Below Half 
the Poverty Line,'' Center on Budget and Policy Priorities (February 
2018), https://www.cbpp.org/research/family-income-support/life-after-
tanf-in-kansas-for-most-unsteady-work-and-earnings-below.

   A 2001 study of people who frequented a food pantry in 
        upstate New York found that employment rates were lower for 
        people who had lost TANF benefits due to sanctions. Among those 
        surveyed in 1997, 13 percent of those sanctioned reported 
        having had wage earnings in the previous 6 months, while 22 
        percent of unsanctioned respondents reported earnings over the 
        same period. Moreover, because of the effects of sanctions, 
        sanctioned respondents were far more likely to report being 
        disconnected both from work and from TANF benefits. A full 
        quarter of the sanctioned sample in 1997 reported no work and 
        no TANF benefits, compared with three percent reporting the 
        same in the unsanctioned sample.\318\
---------------------------------------------------------------------------
    \318\ Jean Oggins and Amy Fleming, ``Welfare Reform Sanctions and 
Financial Strain in a Food-Pantry Sample,'' The Journal of Sociology & 
Social Welfare (June 2001), pp. 101-123, https://
scholarworks.wmich.edu/cgi/viewcontent.cgi?article=2725&context=jssw.

   A 3 year study of TANF recipients in two California counties 
        provides evidence that employment rates are lower for those 
        with significant barriers to employment. The study excluded 
        individuals who received disability benefits and focused on 
        mental health issues such as major depression, generalized 
        anxiety disorder, panic attacks, social phobia, or 
        posttraumatic stress disorder. Those who reported functional 
        impairment over the previous month were far more likely not to 
        have worked over the previous year. In the first follow-up 
        year, 54.2 percent of functionally impaired respondents had 
        worked in the previous year, compared to 75.2 percent of those 
        without such difficulties. In the second follow-up year, 58.5 
        percent of those with functional impairments worked in the 
        previous year, compared to 79.2 percent of those without such 
        difficulties. The researchers found a statistically significant 
        association between having a mental health issue and having no 
        earned income over the previous year. The most reasonable 
        interpretation of this result is not that the respondents who 
        were not working had some other source of support, but that 
        they were unable to secure work due to their significant 
        barrier to employment.\319\
---------------------------------------------------------------------------
    \319\ Daniel Chandler, et al., ``Mental Health, Employment, and 
Welfare Tenure,'' Journal of Community Psychology (September 2005), pp. 
578-609.

   A California study of CALWORKS recipients found that 
        recipients who had been sanctioned were much less likely to 
        report obtaining full-time employment over 3 years (38 percent 
        of respondents) compared to non-sanctioned clients (60 percent 
        of respondents). Researchers measured employment history by 
        asking respondents about previous full- or part-time 
        employment, and whether they were without work throughout the 3 
        years prior to the survey.\320\
---------------------------------------------------------------------------
    \320\ Yeheskel Hasenfeld, Toorjo Ghose, and Kandyce Larson, ``The 
Logic of Sanctioning Welfare Recipients: An Empirical Assessment,'' 
Social Science Review (June 2004), Vol. 78, No. 2, pp. 304-319, http://
repository.upenn.edu/spp_papers/88/.

   Studies consistently find lower employment rates among TANF 
        leavers whose cases were closed due to a work-oriented sanction 
        than among families that left TANF for other reasons. For 
        example, in Arizona, 40 percent of sanctioned leavers were 
        working in the first quarter after exit, compared to 55 percent 
        of non-sanctioned leavers.\321\
---------------------------------------------------------------------------
    \321\ Karen L. Westra and John Routley, ``Arizona Cash Assistance 
Exit Study,'' Arizona Department of Economic Security Office of 
Evaluation (January 2000).

   In Maryland, 6 months after exit, 38 percent of sanctioned 
        leavers were employed, compared to 58 percent of non-sanctioned 
        leavers.\322\
---------------------------------------------------------------------------
    \322\ Catherine Born, Pamela Caudill, and Melinda Cordero, ``Life 
After Welfare: A Look at Sanctioned Families,'' University of Maryland 
School or Social Work (November 1999).

   A study of TANF recipients nationwide using the Census 
        Bureau's Survey of Income and Program Participation indicates 
        those who are disconnected from both work and cash assistance 
        are more likely to have a significant barrier to employment 
        than those who either work or receive cash assistance. Those 
        who had both no household earned income or any cash assistance 
        during the survey month were about twice as likely to report 
        having a physical or mental health condition that limits one's 
        ability to work compared to those who either worked or received 
        cash assistance. Results were consistent across geographic 
        regions. In southern states, 15.8 percent of disconnected 
        respondents reported a physical or mental health work-limiting 
        condition, while only 7.4 percent of non-disconnected 
        respondents had such a condition; in non-southern states, 24.3 
        percent of disconnected respondents had a work-limiting 
        condition, while among non-disconnected respondents, 10.4 
        percent reported having such a condition. It should be noted 
        that these respondents with physical and mental health 
        conditions were not receiving support from SSI; SSI recipients 
        (and those who reported school as their major activity) were 
        excluded from the sample.\323\
---------------------------------------------------------------------------
    \323\ Andrea Hetling, ``The Importance of Region and State Welfare 
Rules for Disconnected Single Mothers,'' University of Kentucky Center 
for Poverty Research Discussion Paper Series (September 2011).
---------------------------------------------------------------------------
Most TANF Recipients Who Lose Benefits Due to Time Limits Do Not Find 
        Steady Employment
    Cutting off families because they have reached some arbitrary time 
limit ignores whether they can actually support themselves or if the 
job market is welcoming to them. Several studies have found that 
parents cut off of TANF due to the time limits have trouble finding 
employment. The health, familial, and behavioral circumstances that 
kept them from finding a job while on TANF also prevent these parents 
from finding work after TANF. Black families are not only the most 
likely to be cut off by time limits, but also very likely to be 
discriminated against in the job market, the evidence shows. In some 
instances, parents who can find work may be working inconsistently and 
thus still fall short of a stable income.

   In Washington State, time-limited parents were less likely 
        to be employed in the year before leaving TANF.\324\
---------------------------------------------------------------------------
    \324\ Christina McHugh and J. Taylor Danielson, ``TANF Time Limit 
Analysis Comparing Cases Closed Due to Time Limits with Other Case 
Closures,'' Washington State Department of Social and Health Services 
(February 2019).

   Researchers from the University of Maryland's School of 
        Social Work found that compared to other people leaving TANF, 
        those leaving because they reached the time limit had less 
        employment history while on TANF and worked fewer quarters in 
        the year after leaving assistance.\325\
---------------------------------------------------------------------------
    \325\ Andrea Hetling, Kathryn Patterson, and Catherine Born, ``The 
TANF Time Limit: Comparing Long-Term and Other Welfare Leavers,'' 
Family Welfare Research and Training Group (February 2006) pp. 1-19, 
https://familywelfare.umaryland.edu/reports1/timelimitleavers.pdf.

   Other researchers of Maryland's TANF program found that 
        recipients who reported having a criminal record were more 
        likely to reach the time limit than those who did not report 
        having a criminal background. While recipients with a criminal 
        conviction are as likely to be employed as other recipients, 
        their employment is more unstable. These women are often more 
        likely to have other barriers as well, such as human capital 
        deficits and situational barriers.\326\
---------------------------------------------------------------------------
    \326\ Valerie Head, Catherine Born, and Pamela Ovwigho, ``Criminal 
History as an Employment Barrier for TANF Recipients,'' Family Welfare 
Research Training Group (March 2009), pp. 1-39. https://
pdfs.semanticscholar.org/2c12/
a0663bd8249586c000a82122cc2e6c796e23.pdf?_ga=
2.3188080.1987654092.1552314136-1225174621.1551994502.

   An analysis of the Building Wealth and Health Network pilot 
        program found depression is often a barrier to employment among 
        TANF recipients, and that adverse childhood experiences (ACEs) 
        and exposure to community violence are often associated with 
        depression. The study investigated how resilience affects the 
        relationship between ACEs, community violence, and depression. 
        TANF families have a high prevalence of health impediments and 
        significant barriers to employment, such as domestic violence, 
        food insecurity, utility shut offs, homelessness, child 
        hospitalizations, and child developmental risks.\327\
---------------------------------------------------------------------------
    \327\ Seth L. Welles, Falguni Patel, and Mariana Chilton, ``Does 
Employment-Related Resilience Affect the Relationship between Childhood 
Adversity, Community Violence, and Depression?'' Journal of Urban 
Health (2017), Vol. 94, pg. 233-243, https://www.ncbi.nlm.nih.gov/pmc/
articles/PMC5391326/.
---------------------------------------------------------------------------
Most TANF Recipients Who Lose Benefits Due to Sanctions Do Not Find 
        Steady Employment
    The rationale for reducing or eliminating benefits for not meeting 
a work requirement is that this will compel parents to find work. 
Evidence suggests that work requirements (along with other policy 
changes that accompanied TANF's implementation) contributed to a modest 
increase in employment during the late 1990s, but that work often was 
not steady[,] a pattern reflected in recent studies as well.
    Another consequence of work requirements in TANF raises concerns 
about the NPRM's goal of subjecting more unemployed adults to SNAP's 
time limit. As it has become harder for single mothers to get direct 
financial assistance when they cannot find work, the number with 
neither jobs nor TANF has grown substantially over time. In 1995, the 
number of families receiving cash assistance in an average month 
exceeded the number of jobless single mothers by about a million. By 
2016, the number of families receiving cash assistance in an average 
month was roughly two million below the number of jobless single 
mothers. As a result, work requirements in TANF fueled an increase in 
deep poverty (measured as income at or below \1/2\ of the poverty 
line).
J. Evidence from Medicaid Work Requirements Shows Beneficiaries Lose 
        Benefits But Don't Gain Employment
    Additional evidence that taking benefits away from individuals who 
are unable to meet strict work requirements does not lead to increased 
work rates comes from the recent state waivers to apply such policies 
in Medicaid. In June 2018, Arkansas became the first state to condition 
receipt of Medicaid benefits on meeting a work requirement. Certain 
beneficiaries must participate in and report 80 hours of work or work-
like activity each month.\328\ Those that fail to meet the requirement 
for 3 months in a calendar year lose coverage.\329\
---------------------------------------------------------------------------
    \328\ In 2018, non-disabled adult beneficiaries ages 30 through 49 
who do not have minor children in their homes were subject to the work 
requirement in Arkansas. The state has begun phasing in the requirement 
for beneficiaries 19 through 29 in 2019.
    \329\ Jennifer Wagner, ``Commentary: As Predicted, Eligible 
Arkansas Medicaid Beneficiaries Struggling to Meet Rigid Work 
Requirements,'' Center on Budget and Policy Priorities, July 30, 2018, 
https://www.cbpp.org/health/commentary-as-predicted-eligible-arkansas-
medicaid-beneficiaries-struggling-to-meet-rigid.
---------------------------------------------------------------------------
    In 2018, over 18,000 Arkansas Medicaid beneficiaries--nearly 25 
percent of the total population the state identified as potentially 
subject to the work requirement--lost coverage for failing to meet the 
requirement.\330\ This far exceeds the population that the state's 
articulated policy intended to target with the requirement: 
beneficiaries who were able to work but were not working. Many 
individuals who qualified for an exemption for being unable to work or 
who were working are among those who lost coverage. Many beneficiaries 
were unaware of the requirement, did not understand what they had to do 
to meet the requirement, or were unable to navigate the reporting 
requirement.\331\
---------------------------------------------------------------------------
    \330\ Jennifer Wagner, ``Medicaid Coverage Losses Mounting in 
Arkansas From Work Requirement,'' Center on Budget and Policy 
Priorities, January 17, 2019, https://www.cbpp.org/blog/medicaid-
coverage-losses-mounting-in-arkansas-from-work-requirement.
    \331\ Jennifer Wagner, ``4,109 More Arkansans Lost Medicaid in 
October for not Meeting Rigid Work Requirement,'' Center on Budget and 
Policy Priorities, October 16, 2018, https://www.cbpp.org/blog/4109-
more-arkansans-lost-medicaid-in-october-for-not-meeting-rigid-work-
requirements.
---------------------------------------------------------------------------
    Moreover, there is no evidence that the work requirement has led to 
increased employment. Although the state has cited data from the New 
Hire Database as evidence that beneficiaries are starting new jobs, it 
has not provided any evidence that the work requirement caused these 
new hires; low-income workers frequently begin new jobs or change jobs. 
Further, the data source includes individuals who worked for only a few 
hours or 1 day, doesn't show if the job is temporary (such as seasonal 
work around the holidays), and doesn't indicate if the employee had 
been previously unemployed as opposed to just recently changed 
jobs.\332\
---------------------------------------------------------------------------
    \332\ Jennifer Wagner, ``Fact Checking Arkansas Governor's Claims 
About Jobs and Medicaid Waiver,'' Center on Budget and Policy 
Priorities, January 28, 2019, https://www.cbpp.org/blog/fact-checking-
arkansas-governors-claims-about-jobs-and-medicaid-waiver.
---------------------------------------------------------------------------
    In fact, other evidence from state Medicaid administrative data 
indicates that at most a few hundred people may have found jobs due to 
the Federal waiver. Most Medicaid beneficiaries don't face monthly 
reporting requirements, mainly because they're already working or 
qualified for exemptions. Only the remaining group, which has to report 
hours of work each month, faces any new work incentive due to the new 
policy. And of that group, only a few hundred each month have met the 
requirement by reporting some work hours, the state reports. What's 
more, many of them likely would have found jobs anyway.\333\
---------------------------------------------------------------------------
    \333\ Ibid.
---------------------------------------------------------------------------
    These data are consistent with focus group interviews showing that 
the work requirement isn't changing Medicaid beneficiaries' behavior. 
Beneficiaries already had enough reasons to work: they need to pay 
their bills. But they often struggle with unstable work hours, live in 
rural areas with few jobs, or face other barriers to employment--and 
the state hasn't invested any new money in job training programs, 
services to address barriers, or supports like transportation to help 
beneficiaries connect to jobs.\334\
---------------------------------------------------------------------------
    \334\ Ibid.
---------------------------------------------------------------------------
    Meanwhile, the work requirement has even proved counterproductive 
for some. News reports describe working beneficiaries who struggled 
with the reporting requirement and lost Medicaid coverage. 
Consequently, some have gone without needed medication, worsening their 
health and in some cases costing them their jobs. Moreover, any small 
increase in employment must be viewed in light of the 18,000 
beneficiaries who lost coverage.\335\
---------------------------------------------------------------------------
    \335\ Ibid.
---------------------------------------------------------------------------
    Numerous other states are in the process of implementing Medicaid 
work requirements, and estimates show similar coverage losses are 
likely. For example, the waiver in Michigan may lead up to 27 percent 
of the state's Medicaid expansion population to lose coverage the first 
year.\336\ Kentucky's own projections say its work requirement, which 
has been challenged in court, would cause 95,000 enrollees to lose 
coverage in 5 years.\337\ A group of health care providers and 
advocates filed an amicus brief in the Kentucky litigation pointing out 
that its work requirement will worsen health and won't promote 
work.\338\
---------------------------------------------------------------------------
    \336\ Cindy Mann and April Grady, ``Potential Enrollment Impacts of 
Michigan's Medicaid Work Requirement,'' Manatt Health, February 6, 
2019, https://www.manatt.com/Manatt/media/Media/Images/White%20Papers/
Manatt_MI-Work-Req-Estimates_20190206-Final.pdf.
    \337\ Phil Galewitz, ``Judge Blocks Kentucky Medicaid Work 
Requirement,'' Kaiser Health News, June 29, 2018, https://khn.org/news/
judge-blocks-kentucky-medicaid-work-requirement/.
    \338\ Steward v. Azar amicus brief, January 24, 2019, https://
www.psychiatry.org/File%20Library/Psychiatrists/Directories/Library-
and-Archive/amicus-briefs/amicus-2019-Stewart-v-Azar-DC-No18-152.pdf.
---------------------------------------------------------------------------
    The experience thus far with Medicaid work requirements 
demonstrates that such policies take coverage away from large portions 
of beneficiaries, including those who are working or qualify for an 
exemption but cannot navigate the red tape of the requirement. At the 
same time, they fail to lead to increased work activity or employment.
Chapter 7: Proposed Rule's Requirement That State Waiver Requests Have 
        the Governor's ``Endorsement'' Violates Congressional Intent
    The proposed rule would require that state requests to waive the 
time limit in areas with insufficient jobs ``be endorsed by the State's 
Governor'' \339\ (emphasis added). This change is in direct violation 
of Congressional intent, as clearly expressed less than 2 months before 
publication of the proposed rule. If FNS proceeds to publish a final 
rule it must reject this change and conform to the intent of Congress.
---------------------------------------------------------------------------
    \339\ NPRM, p. 992.
---------------------------------------------------------------------------
    Current regulations regarding state requests to waive the 3 month 
time limit say simply that such requests are made to FNS ``on the 
request of the state agency.'' \340\ The presumption is that state 
agencies will be acting under the direction of their political 
leadership, including the Governor. CBPP has worked with states on 
their waiver requests for more than 20 years. We cannot remember ever 
working with a state agency that knowingly sought a waiver against the 
wishes of the Governor. It is true, however, that Governors are not 
typically aware of every detailed policy option and choice that their 
cabinet Secretaries adopt for SNAP. Similarly, Governors do not 
typically sign waivers, review nutrition education plans, or even 
personally review large scale procurements. Governors serve as chief 
executives rather than detailed policy implementers.
---------------------------------------------------------------------------
    \340\ 7 CFR  273.24(f).
---------------------------------------------------------------------------
    The House-passed 2018 Farm Bill sought to require the ``approval of 
the chief executive officer of the state'' \341\ for waiver requests 
(emphasis added). The final conference agreement on the 2018 Farm Bill 
rejected the House approach, and instead requires ``the support of the 
chief executive officer of the state''\342\ (emphasis added). The 2018 
Farm Bill, The Agriculture Improvement Act of 2018, passed the Congress 
in mid-December and was signed by the President on December 20, 2018.
---------------------------------------------------------------------------
    \341\ Section 4015 of House-passed H.R. 2, https://
www.congress.gov/115/bills/hr2/BILLS-115hr2eh.pdf.
    \342\ Section 4005 of H.R. 2 as enacted, https://www.congress.gov/
115/bills/hr2/BILLS-115hr2enr.pdf.
---------------------------------------------------------------------------
    The conferees in the conference report that accompanied the bill 
were very clear about their intent in making this change:

          The Managers intend to maintain the practice that bestows 
        authority on the state agency responsible for administering 
        SNAP to determine when and how waiver requests for ABAWDs are 
        submitted. In response to concerns that have been raised by 
        some Members that state agencies have not fully communicated to 
        the chief executive their intent to request a waiver under 
        section 6(o), the Managers have included a provision to 
        encourage communication between the state agency and the chief 
        executive officer of the state. The Managers agree that state 
        agencies should have the support of these officials in their 
        application for waiver, ensuring maximum state coordination. It 
        is not the Managers' intent that USDA undertake any new 
        rulemaking in order to facilitate support for requests from 
        state agencies, nor should the language result in any 
        additional paperwork or administrative steps under the waiver 
        process.\343\ (Emphasis added.)
---------------------------------------------------------------------------
    \343\ Conference Report to accompany H.R. 2, pp. 616-617, https://
www.congress.gov/115/crpt/hrpt1072/CRPT-115hrpt1072.pdf.

    Thus, the conferees were clear that they did not intend for FNS to 
engage in new rule-making based on the change and did not want to 
introduce any new ``paperwork or administrative steps.'' State 
Administrators are left on their own to ensure that they have the 
support of their Governor. The change in statute simply clarifies this 
practice for those who were unduly concerned that state agencies were 
acting against the wishes of the Governor.
    By requiring the ``endorsement'' of the state's governor in the 
proposed rule, FNS ignored this expressed intent of Congress and went 
too far. The only explanation FNS gives in the NPRM is a short sentence 
in the preamble:

          The Department proposes clarifying that any state agency's 
        waiver request must have the Governor's endorsement to ensure 
        that such a critical request is supported at the highest levels 
        of state government.\344\
---------------------------------------------------------------------------
    \344\ NPRM, p. 983.

    The Merriam Webster's Collegiate Dictionary definition of 
``endorsement'' suggests that the term implies a signature, which would 
necessarily require additional paperwork. Such a step would directly 
contradict Congressional intent. From other aspects of the NPRM it is 
clear that FNS was aware of the passage of the 2018 Farm Bill.\345\ So 
the only reasonable conclusion is that FNS chose to ignore 
Congressional intent and intends to add paperwork burden and steps to 
the process.
---------------------------------------------------------------------------
    \345\ See, for example, p. 987 of the NPRM: ``The proposed rule 
would end the unlimited carryover and accumulation of ABAWD percentage 
exemptions, previously referred to as 15 percent exemptions before the 
enactment of the Agriculture Improvement Act of 2018. Upon enactment, 
Section 6(o)(6) of the Act provides that each state agency be allotted 
exemptions equal to an estimated 12 percent of ``covered individuals . 
. .''
---------------------------------------------------------------------------
Chapter 8: Proposed Rule Would Make Implementing The Time Limit Harder 
        by Removing Provisions That Give States Certainty Around 
        Approval
    The proposed rule would eliminate the ability of states to 
implement a waiver at the time a request is submitted, requiring FNS 
approval prior to any waiver implementation. The proposed rule would 
also remove language that identifies waivers that meet certain 
standards as ``readily approvable.'' Currently, these two provisions 
give states certainty of approval that enables them to better plan for 
waiver implementation while waiting for approval. Given that FNS can 
substantially delay approval (and recently has done so), this proposal 
would put an undue burden on states preparing for the complex and 
error-prone process of implementing the time limit. FNS also failed to 
articulate a need for these changes, making it difficult for commenters 
to weigh in on any potential benefit. We therefore urge the Department 
to keep current regulations at 7 CFR  273.24(f)(3) and 7 CFR  
273.24(f)(4), which establish the ``readily approvable'' standard and 
allow states to implement waivers upon submission of the waiver request 
in some instances.
    Current regulations have two provisions that give states more 
certainty in the waiver approval process. These provisions allow them 
time to prepare for implementation of the time limit while waiting for 
FNS approval. The first provision, at 7 CFR  273.24(f)(3), establishes 
that waivers that meet certain standards are ``readily approvable.'' A 
readily approvable waiver includes data from the Bureau of Labor 
Statistics showing a 12 month unemployment rate of ten percent, a 24 
month unemployment rate 20 percent above the national average, or 
designation as a Labor Surplus Area (LSA) by the Department of Labor's 
Employment and Training Agency. The final rule, published in 2001, 
stated that the Department decided to designate that it would approve 
those waivers ``to facilitate the waiver process.'' \346\ The second 
provision, at 7 CFR   273.24(f)(4), allows states to implement waivers 
based on having either a 12 month unemployment rate of ten percent or 
LSA designation for the current fiscal year upon waiver submission, 
rather than waiting for FNS approval. With those two provisions, states 
can plan on implementing the waiver submitted under the first provision 
while awaiting FNS approval, and can actually implement prior to 
approval if it is one of the waivers specified in 7 CFR  273.24(f)(4). 
(FNS can contact the state to modify the waiver if needed.)
---------------------------------------------------------------------------
    \346\ 66 Fed. Reg., No. 11, 4438 (January 17, 2001).
---------------------------------------------------------------------------
A. Certainty About Waiver Approval Process Is Crucial Due to Lengthy 
        State Implementation Process
    With a reasonable amount of certainty about FNS waiver approval, 
states can begin to plan earlier than if they had to wait for FNS to 
process waiver approval, which can take months and substantially delay 
planning. Having time to plan for implementation is crucial for states 
because of the demands of thoroughly implementing the time limit. As 
several documents from USDA--including memos, guidance, and a report 
from USDA's Office of the Inspector General (OIG)--make clear, before a 
state can implement the time limit in a new area, states must:

   Identify individuals subject to the time limit: as one FNS 
        memo explains, ``Prior to waiver expiration, states must review 
        case file information to identify individual ABAWDs and 
        determine whether or not the ABAWD is subject to the time 
        limit.'' \347\
---------------------------------------------------------------------------
    \347\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Supplemental Nutrition Assistance Program--Expiration of Statewide 
ABAWD Time Limit Waivers,'' March 4, 2015, https://fns-
prod.azureedge.net/sites/default/files/snap/SNAP-Expiration-of-
Statewide-ABAWD-Time-Limit-Waivers.pdf.

   Inform individuals subject to the time limit: state agencies 
        have minimum requirements for notifying people who are subject 
        or potentially subject to the time limit (such as an individual 
        a state has identified as likely subject to the time limit 
        based on age and other characteristics, but who may be eligible 
        for an exemption). As one memo explains, states must ``inform 
        ABAWD and potential ABAWD households of the time limit, 
        exemption criteria (including exemptions from the general work 
        requirements), and how to fulfill the ABAWD work requirement,'' 
        as well as the requirement to report when work hours fall below 
        20 hours per week.\348\ The law requires caseworkers to explain 
        these rules during the individual's eligibility interview, but 
        given the complexity of the policy, FNS recommends providing 
        written notice to clients at least 30 days before the waiver 
        ends. FNS encourages states to write notices in clear, 
        understandable language, develop public information materials 
        for websites and waiting rooms, and leverage partnerships in 
        the community such as service providers.\349\ To properly 
        implement the time limit, states must therefore train staff to 
        ensure they can effectively explain the requirements to 
        individuals subject to the time limit, develop written 
        notifications, and use other resources such as community 
        partnerships--all well before a waiver ends.
---------------------------------------------------------------------------
    \348\ U.S. Department of Agriculture, Food and Nutrition Service, 
``SNAP--Requirements for Informing Households of ABAWD Rules,'' April 
17, 2017, https://fns-prod.azureedge.net/sites/default/files/snap/
Requirements_for_Informing_Households_of_ABAWD_Rules.pdf.
    \349\ U.S. Department of Agriculture, Food and Nutrition Service, 
``SNAP--Best Practices and Resources for Informing Households of ABAWD 
Rules,'' May 25, 2018, https://fns-prod.azureedge.net/sites/default/
files/snap/BestPracticesforInformingABAWDS.pdf.

   Develop policies: States must develop policies for many 
        aspects of the time limit, such as whether they will use a 
        fixed or rolling clock, what procedures they will use to screen 
        individuals for exemptions and what verifications are required, 
        whether they will count unpaid or volunteer work towards the 
        requirement, and how they will use 15 percent exemptions, among 
        many others.\350\ States must also communicate these policies 
        to caseworkers and other relevant staff, and ensure that 
        computer systems reflect their policy choices. While some of 
        these policy decisions may not change depending on the waiver 
        outcome if the state has developed these policies for areas 
        that already have the time limit, states or counties preparing 
        to implement the time limit for the first time will need time 
        to ensure that policies are ready for implementation prior to 
        the expiration of a waiver.
---------------------------------------------------------------------------
    \350\ USDA Office of Inspector General, FNS Controls Over SNAP 
Benefits For Able-Bodied Adults Without Dependents, September 2016, 
https://www.usda.gov/oig/webdocs/27601-0002-31.pdf; U.S. Department of 
Agriculture, Food and Nutrition Service, ``Supplemental Nutrition 
Assistance Program--Expiration of Statewide ABAWD Time Limit Waivers,'' 
March 4, 2015, https://fns-prod.azureedge.net/sites/default/files/snap/
SNAP-Expiration-of-Statewide-ABAWD-Time-Limit-Waivers.pdf; https://fns-
prod.azureedge.net/sites/default/files/snap/SNAP-Expiration-of-
Statewide-ABAWD-Time-Limit-Waivers.pdf; U.S. Department of Agriculture, 
Food and Nutrition Service, ``Guide to Serving ABAWDs Subject to Time-
limited Participation,'' 2015 https://fns-prod.azureedge.net/sites/
default/files/Guide_to_Serving_ABAWDs_Subject_to_Time_
Limit.pdf; U.S. Department of Agriculture, Food and Nutrition Service, 
``Supplemental Nutrition Assistance Program--ABAWD Time Limit Policy 
and Program Access,'' November 19, 2015, https://fns-
prod.azureedge.net/sites/default/files/snap/ABAWD-Time-Limit-Policy-
and-Program-Access-Memo-Nov2015.pdf.

   Ready computer systems for tracking: as the OIG report 
        explains, ``Each month, the states are responsible for tracking 
        an ABAWD's status; countable months; fulfillment of the work 
        requirement; exemption status with respect to age, pregnancy, 
        and mental or physical capacity to perform work; 15 percent 
        exemption status; and good cause for not meeting the work 
        requirement.'' \351\ Setting up computer systems to accurately 
        perform this complex monthly tracking, which may require states 
        to work with contractors to re-program systems and test for 
        errors, can be a time-consuming process.
---------------------------------------------------------------------------
    \351\ USDA Office of Inspector General, FNS Controls Over SNAP 
Benefits For Able-Bodied Adults Without Dependents, September 2016, 
https://www.usda.gov/oig/webdocs/27601-0002-31.pdf.

   Train caseworkers: states must build in adequate time to 
        ensure that eligibility workers thoroughly understand and can 
        implement related policies, which may take months. For example, 
        workers must be prepared to follow procedures to assess 
        individuals' fitness for work in order to screen for exemptions 
        from the time limit \352\ and must be prepared to explain the 
        requirements to those individuals during the eligibility 
        interview,\353\ among other tasks instrumental to implementing 
        the time limit. The OIG report states, ``FNS national officials 
        informed us that the ABAWD provisions were very complex and 
        that it takes months of extensive training for new staff to 
        fully understand the ABAWD requirements. A state official said 
        the ABAWD laws and regulations are the `most complicated SNAP 
        policy in existence' and are `fraught with the potential for 
        case errors.' '' \354\
---------------------------------------------------------------------------
    \352\ USDA Office of Inspector General, Supplemental Nutrition 
Assistance Program--Able-Bodied Adults without Dependents (ABAWD) 
Questions and Answers--June, 2015, https://fns-prod.azureedge.net/
sites/default/files/snap/ABAWD-Questions-and-Answers-June%202015.pdf.
    \353\ U.S. Department of Agriculture, Food and Nutrition Service, 
``SNAP--Requirements for Informing Households of ABAWD Rules,'' April 
17, 2017, https://fns-prod.azureedge.net/sites/default/files/snap/
Requirements_for_Informing_Households_of_ABAWD_Rules.pdf.
    \354\ USDA Office of Inspector General, FNS Controls Over SNAP 
Benefits For Able-Bodied Adults Without Dependents, September 2016, 
https://www.usda.gov/oig/webdocs/27601-0002-31.pdf.

   Identify providers for qualifying work activities: Most 
        states are not required to provide individuals subject to the 
        time limit with spots in work programs that can fulfill the 20 
        hour a week requirement (called ``qualifying activities''). The 
        exceptions are ``pledge states,'' which receive additional 
        funding for employment and training (E&T) programs if they 
        commit to providing a work training spot to individuals subject 
        to the time limit in their last month of SNAP benefits. FNS has 
        in the past encouraged states to provide qualifying activities 
        to individuals subject to the time limit.\355\ States that do 
        wish to provide these services must identify current E&T 
        providers that can offer work placements for participants, and/
        or develop new relationships with providers to offer 
        placements.\356\
---------------------------------------------------------------------------
    \355\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Supplemental Nutrition Assistance Program--ABAWD Time Limit Policy 
and Program Access,'' November 19, 2015, https://fns-
prod.azureedge.net/sites/default/files/snap/ABAWD-Time-Limit-Policy-
and-Program-Access-Memo-Nov2015.pdf.
    \356\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Guide to Serving ABAWDs Subject to Time-limited Participation,'' 
2015, https://fns-prod.azureedge.net/sites/default/files/
Guide_to_Serving_ABAWDs_Subject_to_Time_Limit.pdf.

    Preparing for implementation is therefore a lengthy and difficulty 
process, given that states must identify and notify individuals subject 
to the time limit, develop policies and guidance to support 
implementation, train workers, ready computer systems, and (if they 
choose) develop slots in work programs. Ensuring that local offices are 
ready to implement when a waiver changes or when a state or county 
implements the time limit for the first time is important not only to 
ensure that needy individuals don't mistakenly lose access to food 
assistance, but also to prevent incorrect implementation of the time 
limit from causing case errors.
Unclear if Proposed Rule's Core Standards Are Different From ``Readily 
        Approvable'' Standard
    The proposed rule would weaken both provisions that currently 
provide states with more certainty of FNS approval. First, the NPRM 
would remove the language establishing that waivers with certain 
criteria are ``readily approvable.'' The preamble to the NPRM explains 
that the waivers requested under the ``core standards'' are likely to 
be approved, stating: ``These revisions would include the establishment 
of core standards that would allow a state to reasonably anticipate 
whether it would receive approval from the Department.'' \357\ While 
the preamble therefore suggests that states may understand that waivers 
requested under the ``core standards'' can be reasonably be expected to 
be approved (provided they include the correct data and are calculated 
accurately), the actual rule lacks the specificity of the ``readily 
approvable'' language in current regulations at 7 CFR  273.24(f)(3). 
The proposed rule states: ``(2) Core standards. FNS will approve waiver 
requests under (1)(i) and (ii) that are supported by any one of the 
following.'' If these core standards are indeed ``readily approvable,'' 
then clarifying that USDA will approve waiver based on those standards 
would enable states to continue to plan for implementation.
---------------------------------------------------------------------------
    \357\ NPRM, p. 983.
---------------------------------------------------------------------------
    The proposed rule would also eliminate the current provision at 7 
CFR  273.24(f)(4) that allows states to implement the waiver upon 
submission. The preamble states:

          The proposed rule would bar states from implementing a waiver 
        prior to its approval. Though rarely used, current regulations 
        allow a state to implement an ABAWD waiver as soon as the state 
        submits the waiver request based on certain criteria. By 
        removing the current pertinent text in 273.24(f)(4), the 
        proposed rule would require states to request and receive 
        approval before implementing a waiver. This would allow the 
        Department to have a more accurate understanding of the status 
        of existing waivers and would provide better oversight in the 
        waiver process. It would also prevent waivers from being 
        implemented until the Department explicitly reviewed and 
        approved the waiver.\358\
---------------------------------------------------------------------------
    \358\ NPRM, p. 987.

    The Department's rationale for eliminating this provision is 
unclear given that the proposed rule also establishes ``core 
standards'' and current regulations require states to submit a detailed 
waiver request before implementing. The Department claims that 
eliminating the provision would allow the Department to ``have a more 
accurate understanding of the status of existing waivers,'' but the 
Department does not explain why it lacks this clarity under current 
rules (given that states must submit waiver requests with the proposed 
waiver date of implementation) and cannot instead clarify requirements 
around informing FNS about implementation, rather than limit states' 
ability to implement a waiver while waiting for FNS approval. The 
Department also states that removing this provision would allow the 
Department to ``provide better oversight in the waiver process,'' but 
again does not explain what current issue this proposal would address. 
If states can only submit waivers based on very clear criteria with 
clear methods, and FNS has the ability to modify the waiver, why does 
the Department suggest it currently lacks oversight in this process? 
The provision does not remove the ability of FNS to review and approve 
waivers, but instead moves up the timeline to give states the ability 
to more effectively implement waivers they know will be approved. FNS 
does not explain what need or deficit this proposal seeks to remedy, 
which makes it very difficult for comments to respond.
Department Does Not Address Impact of Its New, Lengthy Approval Process 
        on State Implementation
    The most problematic aspect of the Department's proposal to remove 
the ability of states to implement waivers prior to approval is that 
the Department does not make any proposal that will ensure that the 
Department approves waivers in a timely enough fashion to give states 
the certainty they need to properly implement the time limit. Current 
regulations and guidelines require waivers to be based on recent 
economic data, which by definition narrows the window of time between a 
state's waiver submission and the implementation date.
    For example, as recent guidance explains, for waivers based on a 12 
month unemployment rate of ten percent, the data must include at least 
1 month in the year prior to implementation; therefore the ``furthest a 
state could look back in requesting a waiver for January 1, 2018, 
implementation would be the 12 month period of February 2016 through 
January 2017.'' \359\ Local unemployment data is generally available 
with a lag of about 2 months, so January 2017 data would be available 
in early March 2017 or so.\360\ In addition, annual revisions from BLS 
that typically occur in April can substantially change recent estimates 
of unemployment and thus substantially alter waiver eligibility, so 
states often have to confirm waiver requests submitted before the April 
revision to ensure that the waiver request reflects the most up-to-date 
data.\361\ The state would therefore have at most 10 months total (or 9 
months if waiting for the BLS update) to: analyze the data, prepare a 
waiver request, receive approval through the state's internal process 
(a process that the deeply flawed proposal to require the Governor's 
endorsement could substantially lengthen), submit the waiver request to 
FNS (including through the regional office, which must review and then 
forward it to the national office), receive approval from FNS after its 
review, and prepare for implementation of the waiver by taken the steps 
outlined above, such as identifying and notifying participants, 
programming computer systems, and training caseworkers.
---------------------------------------------------------------------------
    \359\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Guide to Supporting Requests to Waive the Time Limit for Able-Bodied 
Adults without Dependents (ABAWD),'' December 2, 2016, https://fns-
prod.azureedge.net/sites/default/files/snap/SNAP-Guide-to-Supporting-
Requests-to-Waive-the-Time-Limit-for-ABAWDs.pdf.
    \360\ For reference, on March 15, 2019, the BLS website stated it 
would release February 2019 state data on March 22 and local data (such 
as counties and metropolitan areas) on April 3, 2019, a typical lag of 
around 2 months for local unemployment data. https://www.bls.gov/
lau/, accessed March 15, 2019.
    \361\ BLS updates sub-state data annually, typically in April. 
https://www.bls.gov/lau/launews1.htm.
---------------------------------------------------------------------------
    Given that the waiver preparation and internal review process may 
take at least a month or 2 within a state, if FNS review extends into 
several months, that can leave states with very little time for 
preparation given the complexities of implementation outlined above. 
The certainty that they can implement the requested waiver allows 
states to plan more effectively, while also allowing FNS time to review 
and issue an approval, knowing that its review does not hinder the 
state from preparing for implementation. Given that FNS seeks to 
eliminate the provisions that currently allow states this certainty 
without committing to approval within a certain timeframe, this 
proposed rule will instead make it harder for states to plan 
effectively.
    One recent example of why this ability helps states is California's 
experience with its 2018 waiver, which the state began to implement 
prior to approval when waiting for an extensive and lengthy FNS review 
process. As mentioned in Chapter 2, any uncertainty has arisen due to 
the most recent Administration substantially delaying the waiver review 
process. In September 2017, California submitted a waiver request for 
areas with unemployment rates 20 percent above the national average 
(one of the categories of ``readily approvable'' areas), with an 
implementation date of September 2018. Given that California was 
transitioning off a statewide waiver and implementing the time limit 
for the first time in several years in some counties, and given the 
complexities with a large, county-administered state, the state needed 
at least 6 months to prepare for waiver implementation. By February 
2018, about 5 months after the state had submitted the request, FNS 
still had neither approved nor denied the request. California wrote FNS 
that it would prepare to implement based on its waiver request, given 
that the request was based on data that fit the ``readily approvable'' 
standard, and requested that FNS advise the state by March 2018 if it 
wished to modify the waiver. Though FNS approval took over 5 months in 
this instance, the ``readily approvable'' standard enabled the state to 
properly plan for implementation. Unless FNS plans to impose deadlines 
on its own review and approval process that will ensure a timely 
response to states via the regulation, taking away these provisions 
will result in substantially less certainty for states as they await 
FNS approval.
    The Department claims that the NPRM would improve consistency in 
the waiver approval process, but eliminating these provisions would 
introduce more uncertainty and inconsistency. The Department several 
times makes clear that one of the motivating factors for the NPRM is to 
improve consistency, such as stating, when introducing ``core 
standards,'' that ``The Department proposes updating criteria for ABAWD 
time limit waivers to improve consistency across states.'' \362\ 
Reducing the ability of states to predict approvals and await FNS 
approval, therefore cutting into implementation planning time, would 
result in states' planning becoming more contingent on the length of 
time that various steps in the waiver preparation and approval process 
take. Factors such as the length of the approval process within the 
state and the length of FNS approval would have even more weight on the 
length of time states have to implement. A state waiting 6 months for 
approval would have significantly less effective implementation time 
than a state waiting 3 months. It is not clear if the Department 
considered the effect of the elimination of these provisions on the 
consistency of time limit implementation outcomes. If so, the 
Department did not explain how eliminating these provisions could 
affect implementation and how it weighed those costs against what it 
perceived to be the benefits of improved oversight, for which it did 
not articulate a need.
---------------------------------------------------------------------------
    \362\ NPRM, p. 983.
---------------------------------------------------------------------------
    FNS proposes substantially limiting the ability of states to plan 
for implementation while waiting for waiver review. In proposing this 
change, FNS makes statements about the intended effect of the proposal 
to increase oversight without explaining why this change is necessary 
or acknowledging the substantial burden it could impose on states and 
clients subject to the time limit. We recommend that FNS keep the 
current language in regulation that gives states more certainty around 
approval, which lets states better plan for waivers.
B. Proposed Implementation Date Would Cause Severe Burden for States
    The Department also proposes that the rule take effect in October 
2019, only 6 months after the end of the comment period for the NPRM--
an extremely short period following the final rule's publication. The 
preamble states:

          The Department proposes that the rule, once finalized, would 
        go into effect on October 1, 2019, which is the beginning of 
        Federal Fiscal Year 2020. All waivers in effect on October 1, 
        2019, or thereafter, would need to be approvable according to 
        the new rule at that time. Any approved waiver that does not 
        meet the criteria established in the new rule would be 
        terminated on October 1, 2019. States would be able to request 
        new waivers if the state's waiver is expected to be 
        terminated.\363\
---------------------------------------------------------------------------
    \363\ NPRM, p. 983.

    The Department clearly is not considering the length of time states 
need to prepare a waiver request, the time it takes states and FNS to 
review and approve waivers, or the substantial time it takes states to 
ensure that they can prepare for implementation and properly notify 
individuals subject to the time limit. Given that the comment period 
ends in April 2019, it is not plausible that there would be anywhere 
near enough time for any one step of this process, let alone all of 
them. In the past, when many states have waivers due at the same time, 
this has substantially delayed FNS review; with this rule 
implementation, at least 30 states would likely attempt to submit a 
waiver request at the same time. FNS does not acknowledge the 
additional resources it would need to designate to review these 
requests, which would need to happen within a very short time frame. 
Nor does FNS acknowledge the burden this proposal would place on 
states, which would need to devote resources to quickly analyzing the 
data to put forward new requests and to implementing the time limit in 
new areas, and on the participants who would be harmed by a likely 
chaotic implementation in many states.
    In addition, FNS does not put forward any need that would justify 
this short timeline. As we have explained, the current regulations 
changed little since the 1996 guidelines; in practice, then, states 
have been operating under current waiver criteria for more than 20 
years. FNS proposes to make significant changes to longstanding policy 
without articulating a need for this change, but also proposes an 
extremely short timeline that it does not attempt to justify. The lack 
of explanation for the proposed implementation date suggests that in 
proposing this rule, the Department did not fully grapple with the 
realities of implementing these changes. Without any explanation of why 
such a drastic change would be necessary under such a short timeline, 
and without any consideration of the downside of forcing states to 
implement the time limit in new areas with very little preparation 
time, the Department leaves us with little opportunity to address the 
unstated need motivating this change. Withdrawing the proposed rule 
would be the best solution to avoid a rushed implementation of an ill-
considered and harmful policy.
C. Limiting the Duration of Certain Waivers to the Fiscal Year in Which 
        They Are Implemented is Unnecessarily Restrictive
    The Department proposes that waivers based on the 20 percent 
standard outlined in paragraph (f)(2)(ii) would not be approved beyond 
the fiscal year in which the waiver is implemented. Since most waivers 
are currently and likely would continue to be requested under the 
criteria specified in 7 CFR  273.24(f)(2)(ii), it's likely that this 
shift would mandate that most waivers shift to a fiscal year cycle.
    As of March 2019, 36 states (including Guam and the Virgin Islands) 
have ABAWD time limit waivers. Nine states are on the Federal Fiscal 
Year 2019 cycle, 19 states are on the calendar year 2019 cycle, and 
eight states are covering parts of both Fiscal Year 2018 and 2019. This 
grouping of waivers around calendar year and fiscal year is a 
relatively new phenomenon that is an outgrowth of two pieces of 
legislation that motivated states to pursue waivers along those time 
cycles:

   The American Recovery and Reinvestment Act (ARRA) of 2009. 
        That legislation suspended the time limit, with a state opt 
        out, through FY 2010.

   The Emergency Unemployment Compensation program (EUC) that 
        operated through December 2015. Many states sought statewide 
        waivers through their eligibility for Extended Benefits (EB) 
        under EUC.

    As the statutory suspension of the time limit (set under a fiscal 
year cycle) expired and as state eligibility for waivers under EUC 
phased out in 2016 (set under a calendar year cycle), states sought to 
renew their waivers using alternative criteria but according to the new 
time cycles. Prior to the passage of ARRA and EUC, states waiver cycles 
were spread throughout the year with many running from April to May.
    In the NPRM, FNS claims that the proposed rule would prioritize 
recent data by preventing states from requesting to implement waivers 
late in the Federal fiscal year. This proposal would actually have a 
different outcome because states would have fewer recent periods of 
data available to use under this criterion. Under the proposed rule 
waivers beginning in Fiscal Year 2020 can use unemployment data 
starting no earlier than January 2017, so approximately five 24 month 
time periods would be available to states. In contrast, for a waiver 
starting in January 2020, states would have eight 24 month time periods 
of unemployment data to use (including three more recent than under the 
fiscal year calendar scenario). Shifting states to a fiscal year waiver 
calendar removes the current option that states have to avail 
themselves of the most recent data.
    For example, states typically submit waiver requests 3 to 6 months 
prior to the waiver implementation date to give FNS sufficient time to 
process waiver requests. For a waiver to begin on October 1, states are 
recommended to submit the waiver in June, when approximately five 24 
month time periods would be available. For a waiver to begin on January 
1, states would have about three additional and more recent time 
periods to use.
    The proposed rule would also force states to have short waivers 
under some circumstances. States are permitted to submit a waiver at 
any point during the year. This is an important feature in times of 
rising unemployment when states may wish to submit new waivers for 
newly eligible areas. If a state wants to request a new waiver or 
modify a waiver after October 1 based on more recent unemployment data, 
a waiver would need to be approved for less than a year under the 
proposed rule. This would impose addition paperwork for waiver renewals 
on states during an economic downturn because states that submit new 
waivers during the fiscal year would not get a 12 month approval 
regardless of how distressed their local labor market is.
    Also, this limitation does not give states sufficient time to plan 
and implement waivers. As noted above, states typically submit waiver 
requests well in advance of their start date to allow for needed 
implementation planning as well as FNS' slow processing. States do not 
know their eligible areas until late April when BLS revises historical 
estimates for sub-state areas from the Local Area Unemployment 
Statistics (LAUS) program. These revisions reflect new population 
estimates from the Census Bureau, updated input data, and 
estimation.\364\ If states need to submit a waiver in June (for an 
October 1 start), they would only have 1 to 2 months to plan and 
request the subsequent waivers. This would likely be particularly 
challenging for states pursuing a thoughtful and thorough 
implementation of the new waiver.
---------------------------------------------------------------------------
    \364\ Bureau of Labor Statistics, Annual Revisions, https://
www.bls.gov/lau/launews1.htm.
---------------------------------------------------------------------------
    Consider states that have to shift from a statewide waiver using 
the EB criteria (which would not be limited to a fiscal year cycle 
under the proposed rule) to a 20 percent above the national average 
criterion (which would operate on a fiscal year cycle only under the 
proposed rule). For example, consider a state that has a statewide 
waiver based on the EB criteria running from September to August. For 
the next waiver (which would likely be based on the criteria listed in 
paragraph (f)(2)(ii)), the waiver could only be for 1 month because the 
fiscal year runs through September. If the state wanted continuous 
waiver coverage, it would have to request a one 1 month waiver for the 
month of September, then a 1 year waiver starting in October. These 
multiple requests would create additional administrative work for 
states and FNS.
    Moreover, FNS does not have enough capacity to process waivers if 
they are all on the fiscal year cycle. And, as we comment elsewhere, 
the Department is not imposing a timely review on itself which has 
resulted in delayed approvals. These delays would only grow worse if 
virtually all waivers were on the same cycle for review.
    This proposal is flawed and should not be included in the final 
rule.
D. Limiting Waivers to One Year Would Impose an Unnecessary 
        Administrative Burden on states
    The proposed rule would limit the duration of waiver approvals to 1 
year. We believe this would impose an additional administrative burden 
on states that is unjustified and unnecessary. In the NPRM, the 
Department asserts that limiting waivers to 1 year would ensure that 
the waiver request reflects current economic conditions, but it 
provides no evidence or discussion to support this assertion. This 
makes it difficult to comment on the proposed change and its potential 
impact on both the alignment of waivers with current labor market 
conditions and state agencies. This section provides an overview of 
existing requirements for 2 year waivers and explains why the proposed 
change is unnecessary.
Existing Requirements for Two-Year Waivers Are Already Restrictive
    The Department generally approves waivers for 1 year. Existing 
regulations state that the Department reserves the right to approve a 
waiver for a longer period if the reasons are compelling.\365\ In 
previous guidance, the Department acknowledged the role of 2 year 
waivers in minimizing administrative burdens on states from preparing 
annual requests for waivers covering areas with chronic high 
unemployment.\366\ Areas that qualify for 2 year waivers are those that 
have had chronic high unemployment and are likely to continue to 
experience high unemployment. Two-year waivers have also been used to 
cover states and sub-state areas hit hard by the Great Recession of 
2007 to 2009.
---------------------------------------------------------------------------
    \365\ 7 CFR  273.24(f)(5).
    \366\ USDA, ``2-Year Approvals of Waivers of the Work Requirements 
for ABAWDs under 7 CFR 273.24,'' February 3, 2006. USDA, ``Guidance on 
Requesting ABAWD Waivers,'' August 2006. USDA, ``Guide to Supporting 
Requests to Waive the Time Limit for Able-Bodied Adults without 
Dependents (ABAWD),'' December 2, 2016.
---------------------------------------------------------------------------
    The data requirements to support a request for a 2 year waiver are 
much more restrictive than those required for a 1 year waiver. The area 
must satisfy at least one of the following:

   Have an unemployment rate above ten percent for the 2 year 
        period immediately prior to request;

   Be designated as a Labor Surplus Area for at least 2 
        consecutive years; or

   Have an unemployment rate more than 20 percent above the 
        national average for a 36 month period ending no earlier than 3 
        months prior to the request.

    The data requirements are more restrictive in several ways. First, 
an area eligible for a 2 year waiver must have evidence of high 
unemployment sustained over a significantly longer period of time than 
that required for a 1 year waiver. For instance, under the third 
criterion above, the area must have elevated unemployment over a period 
that is 50 percent longer than that required to support a 1 year waiver 
(36 months compared to 24 months). These are areas with persistent, 
chronic high unemployment and are likely to continue to experience 
adverse labor market conditions. As we saw during the Great Recession, 
areas eligible for 2 year waivers included those that experienced a 
rapid rise in unemployment rates before the rest of the country or 
experienced slower recovery.
    Second, a request for a 2 year waiver must be supported by very 
recent data. To be eligible under the third criterion above, the 36 
month period must end no earlier than 3 months prior to the request. 
Given that there already is a time lag of 1 to 2 months before BLS 
Local Area Unemployment Statistics becomes available at the sub-state 
level, this is a very restrictive requirement.
    To support a 1 year waiver requested on October 2018, for example, 
a state could submit data for the period January 2016 to December 2017. 
This corresponds with the time period used to compile the LSA list for 
FY 2019. To request a 2 year waiver, the state would have to submit 
data no older than June 2018 and the 35 previous months, a period that 
starts July 2015. Data supporting a 2 year waiver incorporates data 
both earlier and later than what is required for a 1 year waiver. To 
qualify, the area would have to have chronic high unemployment and be 
likely to continue having it within the time frame of a 2 year waiver.
Use of Two-Year Waivers Has Been Very Limited
    Under existing rules and guidance, waivers longer than 1 year in 
duration have only been requested and approved under limited 
circumstances, reflecting their more restrictive and extensive data 
requirements. Over the 2 decades of waiver approvals, FNS has approved 
approximately 900 waiver requests.\367\ Of those approved requests, 
only about six percent (50 waiver requests) were based on the 36 month 
unemployment rate criteria.\368\ Nearly \1/2\ of the 50 waiver 
approvals were in effect during Federal Fiscal Years 2007 to 2009, 
helping states that were hit hard by the Great Recession, like Alaska, 
Mississippi, Oregon, and South Carolina, weather the economic 
downturn.\369\
---------------------------------------------------------------------------
    \367\ A waiver request includes one or more jurisdictions in a 
state and may even cover the entire state. Each approved waiver request 
is given a distinct waiver serial number by FNS. States may not 
implement waivers in all areas approved by FNS.
    \368\ A waiver request includes one or more jurisdictions in a 
state and may even cover the entire state. Each approved waiver request 
is given a distinct waiver serial number by FNS. Based on CBPP internal 
records and summary of 2 year waivers from SNAP 3 month time limit. 
Prepared in March 2019.
    \369\ A table with the states that have had 2 year waivers from the 
time limit is included in Appendix B as, Center on Budget and Policy 
Priorities. ``Summary of 2-Year Waivers from SNAP Three-Month Time 
Limit.''
---------------------------------------------------------------------------
    States have also requested 2 year waivers to cover jurisdictions 
and Native American Tribal areas that have had chronic high 
unemployment. For example, Nebraska prepared, applied for, and received 
2 year waivers for Tribal areas in FFY 2002 (waiver in effect May 2002 
to April 2004), FFY 2004 (waiver in effect May 2004 to April 2006), FFY 
2006 (waiver in effect May 2006 to April 2008), and FFY 2008 (waiver in 
effect May 2008 to April 2010). Had the proposed rule been in effect, 
Nebraska would have had to apply for 1 year waivers for these Tribal 
areas every year. The Nebraska state agency only had to apply four 
times instead of eight times to cover these areas with chronic high 
unemployment from May 2002 to April 2010.\370\
---------------------------------------------------------------------------
    \370\ The American Recovery and Reinvestment Act went into effect 
April 1, 2009, suspending the time limit in all states through 
September 30, 2010 unless state agencies chose to impose specific work 
requirements.
---------------------------------------------------------------------------
    The Department has also approved waivers longer than a year on a 
case-by-case basis to accommodate states facing unusual administrative 
constraints. For example, it approved a 17 month waiver (from May 2007 
to September 2008) for Utah to ease administrative burdens while the 
state was transitioning to a new eligibility system.
Limiting Waiver Duration to One Year Is Inefficient
    Given the more restrictive data requirements, areas eligible for a 
2 year waiver are experiencing chronic high unemployment and would 
likely be eligible for 1 year waivers in 2 or more consecutive years. 
By prohibiting waivers longer than a year, the Department would be 
requiring states to prepare and submit waiver requests twice over the 
course of a 2 year period, instead of submitting a request once. Our 
analysis finds that most areas approved for 2 year waivers in FFY16-17 
would have qualified for the second year,\371\ so requiring the state 
to submit--and FNS to review--the information would have been 
inefficient and burdensome.
---------------------------------------------------------------------------
    \371\ The Department approved 2 year waivers covering 19 
jurisdictions (seven states, one island, and eleven Indian 
reservations) in Federal Fiscal Years 2016 and 2017. Of the 19 
jurisdictions, 17 would have been eligible for back-to-back 1 year 
waivers.
---------------------------------------------------------------------------
    The existing data requirements for a 2 year waiver capture high 
unemployment using data that is very current. The Department did not 
substantiate its assertion that a 1 year time frame would ensure that 
waiver requests reflect current economic conditions. Nor did it discuss 
why the proposed change is warranted given that it would add 
administrative burdens both to state agencies preparing waiver requests 
and the Department itself. The option to request a 2 year waiver is 
already very restrictive and limited in use. We therefore recommend 
that the Department abandon its proposal to limit waivers to 1 year and 
keep the existing rules allowing 2 year waivers as they are.
Chapter 9. Eliminating the Carryover of Unused Individual Exemptions 
        Would Cause Hardship and Exceeds Agency Authority
    In addition to significantly restricting the ability of states to 
request waivers of the 3 month time limit, the NPRM proposes to 
eliminate the accrual of unused individual exemptions for more than 1 
fiscal year. As a result, some individuals who might otherwise be 
exempted from the time limit would lose SNAP benefits and the program's 
integrity would be undermined as states would be less able to 
judiciously exempt particularly vulnerable individuals. The NPRM fails 
to define a problem it is addressing with this proposal, incorrectly 
reads the intent of Congress, and proposes a less effective 
alternative.
    Under current law, states can exempt a limited number of 
individuals who are, or would be, subject to the time limit. Each year, 
FNS is required to estimate the number of exemptions available to each 
state, based on a percentage (currently 12 percent as revised from 15 
percent in the 2018 Agricultural Improvement Act) of ``covered 
individuals.'' These ``covered individuals'' are SNAP participants 
subject to the time limit during the fiscal year or individuals denied 
eligibility in SNAP because of the time limit.
    It is disconcerting to note that the NPRM incorrectly describes the 
way in which exemptions are calculated. The preamble describes 
``covered individuals'' as ``the ABAWDs who are subject to the ABAWD 
time limit in the state in Fiscal Year 2020 and each subsequent fiscal 
year.'' But this is not a correct description of ``covered 
individuals.'' Section 6(o)(6)(A)(ii) of the Food and Nutrition Act (7 
U.S.C.  2015(o)(6)(A)(ii)) defines a ``covered individual'' as ``a 
member of a household that receives [SNAP], or an individual denied 
eligibility for [SNAP] benefits solely due to paragraph (2)'' (emphasis 
added), with several additional clarifications. As we discuss in our 
comments on the Regulatory Impact Analysis, the imprecise use of 
``ABAWD'' makes it unclear whether the NPRM is accurately describing 
the group of SNAP participants who form part of the pool that is used 
to determine the number of exemptions, but the NPRM also fails to 
include individuals denied eligibility due to failure to meet the time 
limit requirements. As most ``ABAWDs'' subject to the rule lose 
benefits over time, this can be a significant number of individuals.
A. There Is No Statutory or Legislative Support for the Claim That 
        Unused Exemptions Cannot Be Kept By States
    The NPRM suggests that Congress did not explicitly intend for 
states to maintain and accrue unused exemptions, but this is not 
supported by the record. The NPRM describes the accrual of unused 
exemptions as an ``unintended outcome of the current regulations.'' 
\372\ It further expresses concern that ``such an outcome is 
inconsistent with Congressional intent to limit the number of 
exemptions available to states each year.'' The NPRM does not provide 
any evidence supporting this claim of Congressional intent. We are 
unable to find any record of Congressional intent to limit the 
carryover of unused exemptions. The historical evidence and recent 
actions by Congress show the opposite.
---------------------------------------------------------------------------
    \372\ NPRM, p. 987.
---------------------------------------------------------------------------
    Congressional history shows that exemptions were enacted in 
legislation approximately 1 year after the time limit was enacted 
precisely due to concerns that the policy was too harsh and states did 
not have enough tools to mitigate the impact of the time limit for 
vulnerable individuals living outside of waived areas. Adding this 
resource gave states an additional way to protect vulnerable residents 
not specifically identified in the exemptions from the time limit 
provided under 7 U.S.C.  2015(o)(3), based on the priorities and 
concerns of the state or local agency.
    The Balanced Budget Act of 1997 contained two major changes in SNAP 
to ameliorate the impact of the 3 month time limit. One was an increase 
in funding for ABAWD training slots in the Employment and Training 
(E&T) program. The other was providing states with the authority to 
exempt a limited number of individuals from the time limit.\373\ 
Commonly referred to as hardship exemptions, these gave states the 
ability to continue to provide SNAP to individuals subject to the time 
limit who could not find jobs or training slots after 3 months of 
participation. Just after passage of this change, FNS clarified in an 
October 1997 guidance to states that unused exemptions could be carried 
over or saved for future use.\374\
---------------------------------------------------------------------------
    \373\ Section 1001 of P.L. 105-33.
    \374\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Implementation of the Provisions of the Balanced Budget Act of 1997 
Relating to Exemptions for Able-Bodied Adults without Dependents 
(ABAWDs),'' October 27, 1997.
---------------------------------------------------------------------------
    The current individual exemption policy has been in place for over 
20 years. Congress did recently intend to limit exemptions, but not in 
the way proposed in the NPRM. Instead Congress reduced the percentage 
of exemptions created each year, but explicitly left the longstanding 
accrual policy in place. In the 2018 Farm Bill, Congress reduced the 
annual percentage of exemptions from 15 percent to 12 percent, but 
notably did not propose ending the practice of accruing unused 
exemptions. In fact, the Conference Report to accompany H.R. 2, the 
Agricultural Improvement Act, clarified that ``States will maintain the 
ability to exempt up to 12% of their SNAP population subject to ABAWD 
work requirements, down from 15%, and continue to accrue exemptions and 
retain any carryover exemptions from previous years, consistent with 
current law.'' \375\ (emphasis added). Congressional intent as recently 
as several months ago shows a deliberate expectation that states can 
carryover an unlimited number of unused exemptions.
---------------------------------------------------------------------------
    \375\ House of Representatives, Conference Report to Accompany H.R. 
2, December 10, 2018, p. 616, https://www.agriculture.senate.gov/imo/
media/doc/CRPT-115hrpt1072.pdf.
---------------------------------------------------------------------------
The Statute Clearly Allows States to Accrue Unused Exemptions
    By drastically reducing the way in which states that choose not to 
use exemptions in the year in which they are issued are able to accrue 
these exemptions, the NPRM suggests that the current policy is an 
interpretation of the intent of the underlying statute. However, the 
statute is less confusing than it appears. It authorizes states to 
exempt up to 12 percent of the caseload (formerly 15 percent) but does 
not mandate that states use the exemptions over any particular time 
period. It then, separately, authorizes the Secretary to adjust the 
number of exemptions based on the state's use of exemptions in the 
prior fiscal year. Under the provision, if a state does not use all 
exemptions, the Secretary increases the number of exemptions available 
in the current year. If the state overuses exemptions, then the 
Secretary reduces the number of exemptions available in the current 
year. The statute reads:

          . . . the Secretary shall increase or decrease the number of 
        individuals who may be granted an exemption by a state agency 
        under this paragraph to the extent that the average monthly 
        number of exemptions in effect in the state for the preceding 
        fiscal year under this paragraph is lesser or greater than the 
        average monthly number of exemptions estimated for the state 
        agency for such preceding fiscal year under this paragraph.

    The language sets out that the Secretary adjusts one way for one 
circumstance (too many exemptions used), and in another way for the 
other condition (fewer exemptions used than issued). The Secretary 
shall increase the number of individual exemptions to the extent that 
the average monthly number used in the previous year is less than then 
number estimated for that year. Similarly, the Secretary shall decrease 
the number of individual exemptions to the extent that the average 
monthly number used in the previous year is more than the number 
estimated for that year.
    Note that the Secretary is required to adjust the number of 
exemptions, but that the use of exemptions remains a state option 
(``individuals who may be granted an exemption''). And, if the state 
uses fewer exemptions than allotted in the previous fiscal year, the 
Secretary should increase the number of exemptions in the following 
year. The provision requires the Secretary to ``increase or decrease'' 
exemptions depending on whether the state's use of exemptions is 
``lesser or greater than'' the allotment for the previous year. So, 
it's an increase if the state uses fewer exemptions and a decrease if 
the state uses more exemptions than allotted. This makes sense. By 
decreasing the allotment to a state that overuses the exemptions, the 
statute ensures that states cannot routinely use more than the yearly 
allotted amount. But that means that a state does increase its 
allotment each year that it does not use that year's amount. States 
that repeatedly under-use allotments will accrue a bank of exemptions. 
This approach, codified in the current regulations, is a 
straightforward and fair reading of the statute's directive.
    The proposed rule, in contrast, makes several unsupported 
assertions. First, it claims without support, that the intent was not 
to accrue exemptions for more than 1 year. Second, by eliminating the 
existing supply of unused exemptions, it treats them as having no value 
to the state even though many states have accessed these accrued 
exemptions for a variety of allowable and sensible reasons. Third, it 
fails to explain why the current procedure to adjust exemptions each 
year is a flawed reading of the underlying statute.
Legislative History Demonstrates That Congress Fully Understood and 
        Approved of the Uncapped Accrual of Exemptions
    The guidelines explaining the calculation and use of individual 
exemptions were first promulgated in the September 3, 1999 interim rule 
implementing two SNAP provisions in the Balanced Budget Act of 
1997.\376\ In that interim rule, the Department outlined how it would 
comply with the statutory requirement that the Secretary adjust the 
number of individuals who may be granted an exemption to account for 
any difference between the average of exemption used and the number 
estimated by the Agency for the preceding fiscal year. If a state uses 
more exemptions than estimated, the state's subsequent allocation is 
reduced. Likewise, if a state uses fewer exemptions than estimated in 
the previous year, the state's subsequent allocation is increased by 
the amount not used. As the Department explained ``if this level of 
exemptions is not used by the end of the fiscal year, the state may 
carry over the balance.'' \377\
---------------------------------------------------------------------------
    \376\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Food Stamp Program: Food Stamp Provisions of the Balanced Budget Act 
of 1997,'' Interim rule, Federal Register Vol. 64, No. 171, Sept. 3, 
1999, p. 48246, https://www.govinfo.gov/content/pkg/FR-1999-09-03/pdf/
99-23017.pdf.
    \377\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Food Stamp Program: Food Stamp Provisions of the Balanced Budget Act 
of 1997,'' Interim rule, Federal Register Vol. 64, No. 171, Sept. 3, 
1999, p. 48249, https://www.govinfo.gov/content/pkg/FR-1999-09-03/pdf/
99-23017.pdf.
---------------------------------------------------------------------------
    This longstanding implementation of the statutory directive is 
clear, reasonable, and fair to states. It addresses the reasonable 
concern that an annual allotment of exemptions could be either overused 
or underused. The continual overuse of individual exemptions has an 
impact on overall program integrity because individuals not eligible 
under an exemption are issued benefits, which is an over-issuance and 
error. To address this, the regulation treats this issue in a sensible 
way, by reducing future exemptions. The continual under-use of 
individual exemptions does not create the same problem, and the 
regulation's treatment is similarly reasonable.
    The statute does not direct the Secretary to make adjustments 
beyond a 1 year period. In other words, the statute does not give the 
Secretary the authority to adjust the number of exemptions issued more 
than 1 year prior ago. Combining the requirement that the Secretary 
adjust exemptions from the previous fiscal year with the limitation on 
looking further back to adjust exemptions based on use means that a 
state can accrue unused exemptions in multiple years, and these 
exemptions can accrue over multiple years.
States Have Relied on Current Policy: USDA Has Never Emphasized the 
        Need to Use Exemptions Each Year
    The Department has not, in the past, suggested that unused 
exemptions would not accrue. Developing a reasonable exemption policy 
is difficult--states must identify the circumstances when an individual 
exemption should be used, the procedures for identifying when that 
circumstance has occurred, and a tracking mechanism to ensure that the 
usage does not exceed the allotment. This implementation challenge has 
discouraged states from experimenting with ways of using the 
exemptions. But it does not indicate that states have no need for them. 
Instead of eliminating earned-but-unused exemptions, the Department 
could provide guidance to states on effective ways to use them. The 
Department could take steps to understand states' concerns or problems 
with using exemptions. Such a response would be much more in keeping 
with Congressional intent and the law. Instead, via the NPRM, the 
Department has taken sweeping measures to curtail a state resource 
counter to the law.
States Have Compelling Reasons to Accrue Individual Exemptions
    The recent statutory change from 15 percent to 12 percent makes the 
banked or unused exemptions more important for some states. While not 
every state uses its annual allotment of exemptions, some states do or 
come close to doing so. Many states use the ability to rollover 
exemptions to build a ``bank'' that gives states options that would be 
unavailable if exemptions expired.
    States use individual exemptions for a variety of purposes, as the 
original provision intended. Some identify certain vulnerable 
populations, such as victims of domestic violence, veterans, young 
adults aging out of foster care, or those with acute barriers to 
employment like a lack of education or limited proficiency in English. 
States have also used exemptions to allow individuals in limited areas 
to remain eligible for SNAP, often because of circumstances that are 
not reflected in a way that qualifies the area for a waiver or due to 
administrative demands. In all cases, states must estimate the number 
of individuals who would receive an exemption and for how long in order 
to ensure that the state does not exceed the number of available 
exemptions. Building up some unused exemptions gives states important 
flexibility and confidence to implement these targeted approaches 
without running afoul of over-issuing exemptions. The buffer provided 
by accrued exemptions is critical in that process.
    Because of the recent change in the percentage of exemptions made 
available to states, these states that are using exemptions would be at 
risk of exceeding their allotment and being subject to error 
determinations and overpayments under the NPRM. Between 2014 and 2017, 
28 states used more exemptions than they had been issued for the fiscal 
year, meaning they used at least some of the exemptions they had 
accrued in previous years. During that time period, some states did so 
for more than 1 year. Several of the states used all of their multi-
year exemptions which demonstrates the importance of accruing 
exemptions over several years.
    For example, Washington used 28,886 exemptions in Fiscal Year 
2016.\378\ It had earned no exemptions in the prior year (because it 
had a statewide waiver in 2015). It had accrued 11,530 exemptions in 
previous years. It did not overuse exemptions because it was allocated 
26,784 exemptions that year (meaning that it started 2017 with over 
9,000 unused exemptions). The state relied on exemptions that year 
because it was transitioning off of the statewide waiver and was 
developing training programs and operational procedures for childless 
adults subject to the rule. Other states used banked exemptions in a 
similar way. For example, in 2016, Maryland earned no exemptions for 
the year (based on having a statewide waiver in 2015) but issued 18,871 
exemptions to aid in its transition to the time limit. It could do so 
only because it had a ``bank'' of unused exemptions from prior years of 
18,915. Under the proposed rule, neither state would have had been able 
to take this approach.
---------------------------------------------------------------------------
    \378\ U.S. Department of Agriculture, Food and Nutrition Service, 
``SNAP--FY 2017 Allocations of 15 Percent Exemptions for ABAWDs--Totals 
Adjusted for Carryover,'' March 15, 2017, https://fns-
prod.azureedge.net/sites/default/files/snap/FY2017-ABAWD-15%25-
Exemption-Totals.pdf.
---------------------------------------------------------------------------
    The 3 month time limit is complex and difficult to administer, as 
is documented in the USDA Inspector General's report.\379\ A majority 
of states have used individual exemptions to ensure that particularly 
vulnerable individuals are not inappropriately terminated from the 
program. Allowing states to keep unused exemptions enables states to 
plan in advance and prepare for major events affecting the unemployed 
childless adult population on SNAP (such as an area transitioning from 
waived to unwaived status).
---------------------------------------------------------------------------
    \379\ U.S. Department of Agriculture, Office of Inspector General, 
``FNS Controls Over SNAP Benefits for Able-Bodied Adults Without 
Dependents,'' September 2016.
---------------------------------------------------------------------------
Congress Knows How to Limit Carryover and Has Repeatedly Declined to Do 
        So for Unused Exemptions
    Congress has the authority and ability to limit the carryover of 
allocated resources in the legislation it crafts. This authority is 
exercised frequently, in order to prevent unused funds or resources 
from accruing. In fact, the Food and Nutrition Act demonstrates that 
Congress, when it deems it appropriate, can limit or reallocate 
resources, though it has not done so for individual exemptions. For 
example, in allocating funding for SNAP Employment and Training 
program, Section 16(h) (7 U.S.C.  2025(h)) reads:

          (C) Reallocation.--

                  (i) In General.--If a state agency will not expend 
                all of the funds allocated to the state agency for a 
                fiscal year under subparagraph (B), the Secretary shall 
                reallocate the unexpended funds to other states (during 
                the fiscal year or the subsequent fiscal year) as the 
                Secretary considers appropriate and equitable.

    Here, Congress not only directs the Secretary to reallocate unspent 
funds but indicates when such reallocation occurs. Further subsections 
provide more detail on the mechanics of the reallocation.
    There is no similar provision in the Food and Nutrition Act 
indicating that Congress intended to limit the accrual of unused 
exemptions, or indeed, any directive for the states once exemptions are 
provided.\380\
---------------------------------------------------------------------------
    \380\ Section 6(o) (7 U.S.C.  2015(o)) of the Act does direct the 
Secretary to make limited adjustments to exemptions each year, but 
these are limited to changes in caseload and not based on whether or 
not a state used the exemptions issued. The E&T funding, by contrast, 
is adjusted based on state decisions to spend the allocation.
---------------------------------------------------------------------------
B. The Proposed Rule Change Fails to Provide a Legitimate Reason for 
        the Change
    The NPRM states that the change would result in administering the 
program more efficiently and to further the Department's goal to 
promote self-sufficiency. However, the NPRM provides no explanation or 
information on how the proposed change would achieve either goal. The 
current exemption policy has worked well for 20 years and FNS has never 
identified issues with the efficiency of the policy. Nothing in the 
proposed replacement policy would make it easier for state to 
administer. Indeed, because the safety valve of a bank of exemptions is 
eliminated, states will find it more difficult to fine-tune policies 
that authorize the use of exemptions.
    For example, a state may decide to provide an exemption to any 
individual who is working, but not enough hours to meet the 20 hour per 
week requirement for those subject to the time limit. It can estimate 
the number of individuals, and hence the number of exemptions. But 
inaccuracies in the calculation of the estimate or changes in the 
composition of the group can significantly change the number of 
exemptions needed. The existing policy provides states with a pool of 
unused exemptions a way to adjust; the proposed rule almost completely 
eliminates this ability to adjust. As a result, every state would be 
more at risk of exceeding its annual allocation of individual 
exemptions.
    In the NPRM, the Department references the September 2016 report 
from the Office of Inspector General to support the proposed change in 
the accrual of unused exemptions. While it is true that the report 
notes the large number of accrued exemptions, the report very 
explicitly declines to recommend any change in current policy. The 
report states ``OIG generally agrees that FNS has the discretion to 
interpret and implement the exemption provisions as it has done, so we 
do not have a recommendation for FNS with respect to exemptions.'' 
\381\
---------------------------------------------------------------------------
    \381\ USDA Office of Inspector General, FNS Controls Over SNAP 
Benefits For Able-Bodied Adults Without Dependents, September 2016, p. 
11.
---------------------------------------------------------------------------
The Proposed Method of Calculating Exemptions and Adjusting From Prior 
        Years Will Discourage States From Using Them and Increase the 
        Potential for Errors
    The proposed adjustment procedure in the NPRM is needlessly 
confusing, will discourage the use of exemptions and is likely to 
increase errors. The varied exemption use example provided in the NPRM 
(Example 2 on page 988) shows how this proposed approach would 
discourage the use of allocated exemptions and contradicts the 
statutory requirements. In 2021, the state uses eight exemptions.\382\ 
In 2022, use plummets to two. In 2023, use quadruples to eight, and in 
2024, it drops again to two. While the math works out to meet the 
proposed rule, the implementation of a policy that varies this widely 
in scope is hard to conceive. The state must be able to estimate the 
number of exemptions that would be used each year, design a policy and 
the procedures to implement to meet that target number, correctly train 
staff, and actually implement and track. Then, the following year, the 
state must design a policy that uses four times fewer (or greater) the 
number of exemptions, retrain staff, and properly implement. The NPRM 
offers no assurance that a state could successfully resdesign important 
program elements on a yearly basis. The history of state administration 
of SNAP also offers no assurance.
---------------------------------------------------------------------------
    \382\ NPRM, p. 988.
---------------------------------------------------------------------------
    In 2022, the state does not have its yearly allocation available, 
because in the prior year it tapped into previously earned exemptions. 
The state is paying back exemptions despite not overusing the total 
available exemptions in any year in the example. That conflicts with 
the statutory authority granting states exemptions in each year in 
which individuals are subject to the time limit or ineligible because 
of it. And, averaging over 2 years so that the average is equal to 12 
percent of the ``ABAWDs'' does not fulfill the statutory requirement 
that states can allocate an average of 12 percent per year.
Chapter 10. The Proposed Rule Fails to Provide Sufficient Rationale or 
        Supporting Evidence for the Proposed Policy Change
    The NPRM proposes several significant changes to long-standing SNAP 
policy that would affect an estimated 1.1 million low-income Americans, 
including 755,000 who would lose food assistance and face increased 
financial and food insecurity. Despite the far-reaching impact of the 
proposed rule, and contrary to requirements in the rulemaking process, 
the NPRM fails to provide a meaningful rationale for most of the 
proposed changes and fails to identify or summarize any research and 
data to support the rationale for such sweeping and consequential 
changes. Without knowing what evidence justifies such a drastic change 
in long-standing policy, it is impossible to assess the validity of the 
claim or the soundness of the evidence used to support it.
    The goal of the Department's proposed changes in policy appears to 
be to subject more people to the time limit by shrinking the portion of 
the country that can request waivers from it. To achieve this goal, the 
proposed rule would prohibit waivers of the time limit to those that 
are based on a general unemployment rate of at least seven percent and 
at least 20 percent above the national average and would restrict the 
ways in which a state can define the area it seeks to waive. In 
addition, the NPRM eliminates several ways in which a state can 
demonstrate a locale has an insufficient number of jobs for those 
subject to the rule and ends the ability of states to save unused 
exemptions to the time limit. In each case, the Department fails to 
identify a desired goal or outcome (such as a certain percentage of the 
target group gaining employment) or explain how the proposed rule would 
lead to the desired goal, and consistently fails to provide any 
empirical support for the proposal. These failures prevent the public 
from understanding why the existing rule needs to be modified so 
drastically.
A. The Proposed Rule Fails to Support the Justification for New 
        Rulemaking--That Too Many Unemployed Adults on SNAP Are Not 
        Subject to the Time Limit
    The preamble to the proposed rule states that the Department now 
believes that the time limit for unemployed adults was intended to 
apply to more individuals than it currently does. The Department thus 
believes that too many individuals live in areas that are waived by 
states and not subject to the rule. But the Department fails to make 
several important connections to justify the need for new rulemaking.
    The Department fails to show that the intent of Congress was to 
subject the Department's preferred number of individuals to the time 
limit. In fact, the NPRM fails to establish that Congress had any 
interest in subjecting a target number of individuals to the rule. 
Rather, the goal was to allow states to protect individuals in areas 
without a sufficient number of jobs, regardless of how many individuals 
that would be. We are unable to identify any statutory reference to a 
policy goal of protecting only a certain percentage of individuals 
subject to the rule.
    The legislative record does not reveal Congressional debate over 
the appropriate percentage of individuals subject to the rule. In fact, 
the Members of Congress introducing the proposed time limit emphasized 
that adequate protections were included to ensure that individuals were 
not cut off of SNAP if opportunities for work or workfare were not 
available. Representative Robert Ney, one of the authors, stated on the 
floor of the House of Representatives that his amendment ``provides 
some safety; it provides a course of a safety net [sic], it has the 
ability to have waivers from the state department of human services.'' 
\383\ Representative John Kasich clarified that the key was that the 
time limit applied only in areas where jobs were available to those 
subject to the time limit--otherwise, the time limit would not apply. 
``It is only if you are able-bodied, if you are childless, and you live 
in an area where you are getting food stamps, and there are jobs 
available, then it applies.'' The key issue, of course, is whether jobs 
are available for these individuals, as the statute requires.
---------------------------------------------------------------------------
    \383\ 142 Congressional Record, H7904 (daily ed. July 18, 1996).
---------------------------------------------------------------------------
    We address the serious concerns with the way in which the NPRM 
incorrectly interprets the standard that waivers are available in 
places with insufficient jobs for the individuals subject to the rule 
in Chapter 3. Here, we simply note that the authors of the original 
legislation did not have a targeted number of individuals they thought 
should be subject to the rule; nor did Congress establish, or even 
debate, a targeted percentage of individuals to be subject to the rule. 
Instead, the co-authors of the original legislation were careful to 
point out that there were adequate protections for all individuals if 
jobs were not available. Given this clear history, it is incumbent upon 
the Department to substantiate its claim that the legislative history 
somehow suggests that the current regulations must be changed because 
too many individuals live in waived areas. Without explaining the 
underlying claim, the rule leaves commenters with little ability to 
meaningfully respond.
    We would also note that the temporary nature of setting such a 
coverage goal strongly suggests such a goal is not intended or 
practical. As economic circumstances change, the ability of ABAWDs 
subject to the time limit to find work will change, meaning that at 
different points in the economic cycle, the portion of individuals 
subject to the rule who are able to find 20 hours of work per week will 
change significantly, as will the portion of individuals living in 
areas eligible for waivers under any set of criteria. And other 
factors, besides the existence of a waiver, affect an individual's 
participation in SNAP, such as the accessibility of the application 
process, other eligibility rules and processes, and the availability of 
training opportunities for unemployed adults.
B. Despite Claiming That General Unemployment Rates Are the Best 
        Available Measure of Job Sufficiency for Low-Income Adults on 
        SNAP, the Proposed Rule Fails to Support the Claim with 
        Evidence
    The proposed rule asserts that low general unemployment rates 
indicate sufficient jobs are available for those subject to the time 
limit. But it offers no reason why a general unemployment rate of seven 
percent is a good proxy measure for establishing that there are 
sufficient jobs for the individuals subject to the rule. This lack of 
an explanation makes it difficult for interested parties to critique 
the Department's conclusion that no waivers should be permitted below 
seven percent. In contrast, as discussed in Chapter 3, there is a deep 
body of research that shows unemployment rates are much higher for 
groups that make up SNAP's ABAWD caseload.
    Current regulations allow waiver requests that demonstrate a recent 
local unemployment rate significantly above the national average but do 
not set a minimum unemployment rate. The NPRM fails to explain why the 
reasoning behind the current rule no longer applies, and why it 
believes another rulemaking process would result in a justifiable 
change. Indeed, for the most substantial proposed change--to prohibit 
waivers for areas with unemployment below seven percent--the NPRM seeks 
input for changing the number to six or ten percent but does not 
explain why those thresholds are of particular importance, aside from 
noting that a larger or smaller group of individuals might be protected 
at the different levels. It does not explicitly seek input on other 
levels, such as five or eight percent. The rule offers a very weak 
explanation, unsupported by research, for the seven percent that is 
proposed or the alternatives for which it seeks comment. Since 
interested stakeholders do not have adequate information to determine 
why the unemployment rate floor is set where it is, it is difficult to 
provide useful feedback on the appropriateness of the proposed 
threshold. In the rulemaking process, an agency that promulgates a rule 
change needs to explain why the original rationale is no longer 
sufficient when proposing to change the rule.
The Proposed Rule Makes Arbitrary Changes to Long-Standing Regulations 
        That Were Initially Promulgated Based on Sound Reasons
    Until this proposed rule, FNS has always acknowledged that the 
statute requires several different ways for states to document a lack 
of sufficient jobs for the individuals subject to the time limit. In 
its original guidance, the Department noted, ``[t]he statute recognizes 
that the unemployment rate alone is an imperfect measure of the 
employment prospects of individuals with little work history and 
diminished opportunities.'' \384\ It then proceeds to describe the use 
of Labor Surplus Areas (LSAs) as a reliable waiver criteria. However, 
without providing a reason or evidence that LSAs are not a useful 
measure, the proposed rule eliminates LSAs as a possible way of 
qualifying for a waiver. We are at a loss as to why a measure relied 
upon for so long by so many states is simply eliminated.
---------------------------------------------------------------------------
    \384\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Guidance for States Seeking Waivers for Food Stamp Limits,'' Dec. 3, 
1996.
---------------------------------------------------------------------------
    Areas designated as LSAs by the Department of Labor have been 
eligible for waivers because in order to qualify as an LSA, an area 
must have sufficiently high unemployment (120 percent of the national 
average so long as the area rate is at least six percent). LSAs are 
recognized as weak labor markets. Federal, state, and local government 
use LSAs to target contracts and allocate employment-related assistance 
and training. LSAs provide a reasonable indicator that there is a lack 
of sufficient jobs for unemployed SNAP participants, who 
disproportionately struggle to overcome barriers to employment.
    To support the inclusion of LSAs as a way to demonstrate a lack of 
sufficient jobs for the unemployed adults subject to the rule, the 
original 1999 rulemaking process established that LSAs were a 
reasonable measure of labor market weakness and were based on sound and 
relevant data from a trusted source (the Bureau of Labor Statistics). 
The original 1996 guidance explained one reason why:

          Labor surplus areas are classified on the basis of civil 
        jurisdictions rather than on a metropolitan area or labor 
        market area basis. By classifying labor surplus areas in this 
        way, specific localities with high unemployment rather than all 
        civil jurisdictions within a metropolitan area, (not all of 
        which may suffer from the same degree of unemployment) can be 
        identified. This feature also makes the classification 
        potentially useful to identify areas for which to seek waivers 
        [emphasis added].\385\
---------------------------------------------------------------------------
    \385\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Guidance for States Seeking Waivers for Food Stamp Limits,'' Dec. 3, 
1996.

    The original rulemaking process emphasized the importance of 
relying on BLS data (much as the NPRM does). But the original 
rulemaking identified LSA status as a reliable indicator of 
insufficient jobs based on BLS data and as recent enough to be used to 
meet the waiver criteria. In fact, the preamble to the final rule noted 
that an LSA designation was reliable enough to allow for ``immediate 
implementation of waivers for areas where the Employment and Training 
Administration, U.S. Department of Labor (ETA), has designated such 
areas as LSAs.'' \386\ In other words, the Department made a reasoned 
decision to allow states to immediately implement (before approval by 
FNS) any waiver based on an area's designation as an LSA. The proposed 
rule both eliminates the LSA criteria and the immediate implementation 
of certain waivers without explaining why the current process is flawed 
or could be improved.
---------------------------------------------------------------------------
    \386\ 66 Fed. Reg., No. 11, 4438, ``Food Stamp Program: Personal 
Responsibility Provision of the Personal Responsibility and Work 
Opportunity Reconciliation Act of 1996,'' January 17, 2001, p. 4463.
---------------------------------------------------------------------------
    The proposed rule drops LSAs but provides no explanation for why 
this change is needed; nor does it identify deficiencies in the current 
criteria (aside from determining that there should be a seven percent 
unemployment floor). Because we do not know what faults the Department 
now believes exist with the use of LSAs as credible indicator of a lack 
of sufficient jobs for the individuals subject to the time limit, we 
are unable to assess the validity of the claim. It is unclear what 
information the Department now has that invalidates its decision of 
more than 20 years ago--a decision the Department has followed and 
subscribed to until very recently. This prevents the public from 
providing relevant information that supports or refutes the reasons 
behind the proposed rule.
The Proposed Rule Attempts to Achieve Through Regulation a Policy That 
        Congress Explicitly Rejected
    The Administration's aim with this rule appears to be to do through 
rule-making what Congress rejected through legislation. The Trump 
Administration proposed restricting waivers from the time limit through 
legislation in its Fiscal Year 2018 budget proposal and promoted 
exposing more people to the time limit throughout the 2018 Farm Bill 
process.\387\ In the budget, the Administration proposed restricting 
waivers to just areas with an average unemployment rate of ten percent. 
In that proposal, the Administration described current policy as, 
``States can request waivers from the ABAWD time limit that cover the 
entire state, or only parts of the state where unemployment is 
particularly high. States decide whether or not to request a time limit 
waiver, and generally make this assessment annually.'' The proposed 
policy was described as, ``This proposal limits ABAWD waivers to 
counties with an unemployment rate greater than ten percent averaged 
over 12 months.'' While this proposed legislative policy would be 
stricter than the policy in the proposed rule, the actual near-term 
impact of a seven percent floor would be similar as there are very few 
counties in the country with average unemployment rates between seven 
and ten percent.
---------------------------------------------------------------------------
    \387\ Food and Nutrition Service, 2018 Explanatory Notes for the FY 
2018 President's Budget, pages 32-92 to 32-93 https://
www.obpa.usda.gov/32fnsexnotes2018.pdf.
---------------------------------------------------------------------------
    Throughout the farm bill process, the President and Secretary 
Perdue were quoted in the press as saying that they were frustrated 
that Congress would not expose more individuals to the time limit or 
``work requirement.'' \388\ At the 2018 Farm Bill signing ceremony, the 
President remarked that he wanted to implement policy counter to what 
Congress had decided. He said, ``Therefore, I have directed Secretary 
Perdue to use his authority under the law to close work requirement 
loopholes in the food stamp program. Under this new rule, able-bodied 
adults without dependents will have to work, or look for work, in order 
to receive their food stamps. Today's action will help Americans 
transition from welfare to gainful employment, strengthening families 
and uplifting communities. And that was a difficult thing to get done, 
but the farmers wanted it done; we all wanted it done. And I think, in 
the end, it's going to make a lot of people very happy. It's called 
`work rules.' And Sonny is able, under this bill, to implement them 
through regulation.'' \389\ As noted, the farm bill legislation which 
the President refers to as ``the bill'', did not provide any new 
authority to the Secretary to change waiver policy.
---------------------------------------------------------------------------
    \388\ Emily Birnbaum and Julie Grace Brufke, ``Trump Attacks Dems 
on farm bill'', The Hill, September 13, 2018, https://thehill.com/
homenews/administration/406561-trump-calls-out-dems-for-opposing-farm-
bill-over-work-requirements; Phillip Brasher, ``Farm Bill Delayed, but 
Perdue signals administration support'', Agripulse, December 3, 2019, 
https://www.agri-pulse.com/articles/11703-farm-bill-delayed-but-perdue-
signals-administration-support.
    \389\ Remarks by President Trump at Signing of H.R. 2, the 
Agriculture Improvement A[c]t of 2018. https://www.whitehouse.gov/
briefings-statements/remarks-president-trump-signing-h-r-2-agriculture-
improvement-act-2018/.
---------------------------------------------------------------------------
    Congress expressly rejected the Administration's proposal to 
substantially limit waivers in favor of the Senate approach, 
demonstrating intent to keep the current interpretation of the 
``insufficient jobs'' criterion intact. Given that the agency did not 
put forward a coherent evidenced-based argument, we are left to believe 
that the goal of this rule is to defy Congressional intent and the 
agency's own rulemaking to achieve a failed legislative effort.
C. The Failure to Provide a Relevant Explanation or Supporting Data to 
        Justify a Change in Current Regulations Occurs Repeatedly 
        Throughout the Proposed Rule
    Under the NPRM, USDA would simply eliminate several existing 
criteria for requesting waivers because the Department claims they are 
``rarely used, sometimes subjective and not appropriate when more 
specific and robust data is available.'' Under the proposed rule, 
waivers would not be available for areas with low and declining 
employment-to-population ratios, a lack of jobs in declining 
occupations or industries, or a lack of jobs as demonstrated by an 
academic study or other publication. The claim that the data available 
is not rigorous enough to support a request is not explained, given 
that states can submit a wide range of data to support a request.
    Especially concerning is the elimination of the employment-to-
population (E:P) ratio standard. It is a well-established metric that 
has several features that make it preferable to general unemployment 
rates in assessing the health of the labor market. In some ways and 
under some circumstances, particularly in rural areas, the E:P ratio 
may be a better measure of the availability of sufficient jobs for low-
income adults participating in SNAP. The Department fails to establish 
that the E:P ratio relies on questionable or non-specific data. The 
proposed rule insists that sound data be used in supporting waiver 
requests, emphasizing that data from BLS is the standard to be used in 
requesting waivers.
    The employment-to-population ratio has not been widely used, but 
that, by itself, is not a sufficient reason to eliminate the option for 
states. As BLS itself notes, the ratio is ``especially useful for 
evaluating demographic employment trends.'' \390\ In particular, it is 
important to rural areas, which often have less dynamic job creation 
and fewer resources available for the types of training activities that 
allow ABAWDs subject to the time limit to meet the 20 hour requirement. 
For example, South Dakota has waived both whole counties and 
reservations under the employment-to-population criteria. Even in the 
last few years, other states, like New York and Maryland have waived 
counties. The option may not be frequently used, but it represents an 
important measure of labor market weaknesses in some areas and should 
remain available to states. The NPRM does not explain why frequency of 
request is a meaningful reason to keep or drop criteria.
---------------------------------------------------------------------------
    \390\ Carol Boyd Leon, The employment-population ratio: its value 
in labor force analysis, Bureau of Labor Statistics, Monthly Labor 
Review, February 1981, pp. 36-45, https://www.bls.gov/opub/mlr/1981/02/
art4full.pdf.
---------------------------------------------------------------------------
    Finally, the Department offers no insight into whether it considers 
the employment-to-population ratio to be ``sometimes subjective.'' 
Under each of the listed concerns used to justify dropping the 
employment-to-population criterion (that it is rarely used, sometimes 
subjective, and not appropriate if more specific and robust data is 
available), the NPRM provides no explanation or information that allows 
the general public to respond to the proposal.
D. The Public Input Resulting From Last Year's Advanced Notice of 
        Proposed Rulemaking Does Not Appear to Inform This Proposed 
        Rule
    In March 2018, the Department issued an Advanced Notice of Proposed 
Rule Making (ANPRM), seeking public input on ``potential regulatory 
changes or other changes that might better support states in accurately 
identifying ABAWDs subject to the time limit and providing meaningful 
opportunities for them to move towards self-sufficiency.'' \391\ Tens 
of thousands of comments were submitted, but the agency makes only a 
cursory reference to a subset of the comments and does not adequately 
recognize or summarize the public input. While the preamble of the NPRM 
contains a brief and inconclusive summary of the submitted comments, 
the Department provides no explanation for how the ANPRM informed the 
policy making process or whether the Department chose to ignore input 
provided through the public process. Potential commenters are at a loss 
for how the ANPRM informed the development of the proposed rule, what 
the public response to the ANPRM was, or how to engage without any 
information about the comments.
---------------------------------------------------------------------------
    \391\ Fed. Reg., Vol. 83, No. 37, ``Supplemental Nutrition 
Assistance Program: Requirements and Services for Able-Bodied Adults 
Without Dependents; Advance Notice of Proposed Rulemaking,'' February 
23, 2018, p. 8013.
---------------------------------------------------------------------------
    The failure to respond to ANPRM raises serious concerns about the 
current rulemaking proposal. We are left to wonder whether the bulk of 
the comments sought, or did not seek, a change in policy. There is no 
summary of the reasons for supporting a change. Nor is there a summary 
of the input from commentators who opposed a change in policy, or a 
response from the agency as to why it concluded that these commentators 
were incorrect.
    The ANPRM asked numerous questions about helping ABAWDs gain work. 
But the NPRM only references the questions about waivers. This is 
deeply misleading as it suggests comments were focused only on that 
question.
    The failure to adequately respond to the ANPRM also raises concerns 
that the current rulemaking process will fail to take the comments on 
the NPRM into account as the Department decides whether to proceed with 
the current proposed rule or change or withdraw it. If the public's 
input was ignored or outright dismissed in the previous process, why 
should the public have confidence that the NPRM will not yield the same 
result?
E. Alternatives to the Proposed Rule Are Not Discussed
    Under the rulemaking process, USDA is obligated to explain why the 
particular policy is proposed and why alternative approaches are 
inadequate. Given that the agency estimates 755,000 people will lose 
benefits and provides no estimate for how many will gain employment, 
less harmful alternatives exist and the Department has an obligation to 
consider these alternatives.
    In establishing seven percent unemployment as a floor under which 
no area can qualify for a waiver, the Department claims this is ``more 
suitable for achieving a more comprehensive application of work 
requirements so that ABAWDs in areas that have sufficient number of 
jobs have a greater level of engagement in work and work activities, 
including job training.'' \392\ No information is provided as to why 
current policies are not comprehensive and what the current level of 
engagement in work and work activities is, much less what a ``greater 
level of engagement'' would look like. This makes it difficult for 
commenters to provide input on these unsupported assertions.
---------------------------------------------------------------------------
    \392\ NPRM, p. 984.
---------------------------------------------------------------------------
    The claim that a seven percent floor strengthens the work 
requirement is repeated throughout the preamble. As discussed in detail 
[below], the NPRM fails to adequately support the proposed floor. 
Without knowing what research or data the Department relied upon to 
conclude that seven percent was the appropriate floor, the public is 
unable to directly comment on the validity of the Department's action.
    The Department does seek input on setting the unemployment floor--
at seven, six, or ten percent--but offers no explanation why six or ten 
percent are the two alternatives rather than, say, five or eight 
percent. Aside from a cursory mention of the natural rate of 
unemployment, no discussion or information is provided to inform the 
public's comments.
    Other alternatives to grouping areas together also exist. For 
example, the proposed rule limits waiver requests that group sub-state 
areas together to those based entirely on BLS Labor Market Areas 
(LMAs). There are serious limitations to relying solely on LMAs as the 
basis of such waivers, as discussed in more detail in Chapter 5. LMAs 
rely on older data, use a narrow definition of a labor market area that 
does not reflect the challenges facing low-income SNAP participants, 
and do not account for other factors relevant to ABAWDs subject to a 3 
month time limit--such as the availability of training programs. In 
fact, one of the key components of the current grouping policy--that 
states largely define the area of the waiver request--is largely 
eliminated with no explanation or evidentiary support. Other 
alternatives do exist, but the NPRM fails to provide any reason why 
these alternatives are not appropriate and why the proposed grouping 
change is the best available option.
    The proposed rule is based on insufficient reasons to change 
current regulations, fails to provide evidence supporting the change, 
and lacks any discussions of alternatives considered in developing the 
proposed rule. Given this lack of supporting information, the public 
has an insufficient opportunity to comment meaningfully on the proposed 
rule.
Chapter 11. The Proposed Rule's ``Regulatory Impact Analysis'' 
        Highlights FNS' Faulty Justification and Includes Numerous 
        Unclear or Flawed Assumptions
    The Regulatory Impact Analysis (RIA) \393\ that accompanies the 
proposed rule contradicts the Department's justification for the 
proposed rule. The Department repeatedly asserts in the preamble that 
the proposed rule would ``encourage more ABAWDs to engage in work'' and 
would ``promote self-sufficiency.'' But the RIA finds instead that 
755,000 individuals would be cut from SNAP in 2020 for ``failure to 
engage meaningfully in work or work training,'' \394\ and it provides 
no evidence or estimates that other individuals would be induced to 
work because of the proposed changes or would experience any benefit 
from the changes.
---------------------------------------------------------------------------
    \393\ The Regulatory Impact Analysis (RIA), which includes the 
detailed cost-benefit analysis and information about the methodology, 
is included in a separate online document here: https://
www.regulations.gov/document?D=FNS-2018-0004-6000, p. 4-7 and 18-31. 
The NPRM includes only a short summary of the analysis. Hereafter in 
citations we will refer to the Regulatory Impact Analysis as the 
``RIA.''
    \394\ NPRM, p. 989; RIA, pp. 4, 26, 27.
---------------------------------------------------------------------------
    In addition, the methodology for deriving the impact of the 
proposed rule ignores available research evidence, uses imprecise 
terms, includes numerous unclear or inappropriate assumptions, and 
excludes all together any explanations for several other key 
assumptions. The information that is provided in the RIA is 
fundamentally flawed, imprecise, incomplete, and incoherent.
    The result is that the proposed rule does not provide the 
analytical or conceptual information needed to justify the policy 
change and to evaluate the proposed rule's likely impacts. Because of 
the deficiencies in reasoning and analysis of the RIA, the proposed 
rule fails to answer basic questions related to the impact of the 
change and the people whom the proposed rule would affect, and so does 
not contain the information and data necessary to fully evaluate the 
proposed rule or to comment on key aspects on the Department's 
justification for the rule.
    No agency could explain every nuance and assumption, but the RIA 
that accompanies the proposed changes in this NPRM is so deeply flawed 
that we cannot comprehend the basic reasoning behind it. Because 
individuals who wish to comment on the changes cannot understand or 
follow the agency's justification, this rulemaking and comment process 
is compromised.
A. The RIA Does Not Provide Any Evidence to Support the Proposed Rule's 
        Stated Rationale
    The NPRM argues repeatedly that ``the Department is confident that 
these changes would encourage more ABAWDs to engage in work or work 
activities,'' \395\ implying that, as a result of the changes proposed, 
individuals newly subject to SNAP's 3 month time limit in areas no 
longer qualifying for waivers would be likely to work more, have higher 
earnings, or otherwise be better off. But the NPRM provides no evidence 
to support these assertions, and no estimates of any quantifiable 
benefits for any individuals resulting from the changes. The NPRM's 
failure to justify the stated rationale is a serious deficiency and 
makes it impossible for commenters to assess the impact of the proposed 
rule or to comment on the Department's justification.
---------------------------------------------------------------------------
    \395\ NPRM, pp. 981, 982, 987.
---------------------------------------------------------------------------
    The RIA, which is included as supplementary materials accompanying 
the NPRM, contradicts the Department's stated rationale. The analysis 
in the RIA provides that the only benefit of the rule is budgetary 
savings from lower SNAP benefits resulting from 755,000 individuals 
``not meeting the requirements for failure to engage meaningfully in 
work or work training.'' \396\ The RIA does not claim that any 
individuals would be induced to find work or have increased earnings as 
a result of the proposed rule. The RIA does provide a confusing 
assertion that a higher share of ``ABAWDs'' would be working in 2020 
(34 percent) than in 2016 (26 percent), and estimates the impact under 
a different scenario where that increase does not occur.\397\ According 
to the RIA, however, the assumed increase in employment (from 26 
percent with any earnings to 34 percent working at least 20 hours a 
week) is ``based on the projected decline in the unemployment rate'' in 
the President's 2019 budget forecast, not on more work among low-income 
households because of the regulatory change.
---------------------------------------------------------------------------
    \396\ NPRM, p. 989.
    \397\ RIA, pp. 4, 26.
---------------------------------------------------------------------------
    The loss of SNAP benefits as a result of fewer areas qualifying for 
waivers from the time limit is included in the RIA as a benefit because 
of the reduction in Federal spending, but the RIA does not quantify any 
benefits to individuals from the change. There is only a small mention 
in the RIA of the harm, or cost that might occur for low-income 
individuals who lose SNAP:

          To the extent that ABAWDs newly subject to the time limit are 
        unable to find work or otherwise meet work requirements, and 
        thus lose SNAP there may be increases in poverty and food 
        insecurity for this group. However, those ABAWDs who become 
        employed will likely see increased self-sufficiency and an 
        overall improvement in their economic well-being. The 
        Department believes that a number of those affected by 
        strengthened work requirements are able to secure employment in 
        a wide range of different industries.\398\ [Emphasis added.]
---------------------------------------------------------------------------
    \398\ RIA, p. 28.

    Thus, the analysis included in the RIA asserts that the Department 
believes people are likely to get jobs because of the rule, but 
provides no evidence to support that belief and quantifies only the 
Federal budgetary savings from the estimated reduction in SNAP benefits 
associated with individuals' ``failure to engage meaningfully in work 
or work training.'' \399\ It further mentions, but does not quantify, 
the secondary effects on SNAP retailers from lower SNAP 
redemptions,\400\ and on community-based organizations (i.e., food 
banks and others that provide emergency assistance) from increased 
demand for food and services.\401\
---------------------------------------------------------------------------
    \399\ NPRM, p. 989.
    \400\ NPRM, p. 990.
    \401\ RIA, p. 28.
---------------------------------------------------------------------------
    Because the RIA and its cost-benefit analysis are lacking in 
internal logic or transparency, the public cannot see clearly how the 
Department arrived at its conclusions about the need for the proposed 
regulation or its impact.
B. Available Research Evidence Contradicts the Articulated Aims of the 
        Proposed Rule
    The research evidence that is available on the question of the 
effects of policies that take food assistance or other benefits away 
from individuals who don't meet rigid work requirements is not 
mentioned in the RIA. This is a serious omission and constrains the 
public from being able to adequately assess and comment on the 
potential impacts of the proposed rule.
    The loss in benefits, and the related increase in poverty and 
hardship that results from policies that take away food assistance or 
other benefits from individuals who do not meet rigid work requirements 
is well-documented in the research. The research also finds very little 
gain in longer-term employment as a result of such policies. In other 
words, the research supports the findings of the RIA that many 
individuals would be cut off SNAP as a result of fewer areas qualifying 
for waivers from the time limit under the proposed rule, but does not 
support the overall stated purpose of the regulatory change.
    The proposed rule ignores strong research evidence from independent 
researchers that contradicts the stated justification for the proposed 
change. Below we discuss the available relevant research on the impact 
of taking benefits away from individuals who are unable to meet rigid 
work requirements.
The NPRM Ignores Research That Finds That the Characteristics of the 
        Low-Wage Labor Market Contribute to Periods of Unemployment
    The proposed rule implicitly assumes that taking away food 
assistance will cause people not currently working to get jobs. But 
this assumption ignores research evidence about the realities of the 
low-wage labor market that contribute to periods of unemployment and 
mischaracterizes the work patterns of many people who need and receive 
assistance.

    Features of the Labor Market Contribute to Periods of Unemployment

    The basic characteristics of low-wage jobs are well-documented: 
low-paid jobs often don't last; low-wage industries that employ workers 
with limited education or work experience tend to expand and shrink 
their workforces frequently based on demand, resulting in part-time 
jobs that have unstable hours and high turnover; and low-paid workers 
often lack the health coverage, paid leave, and reliable child care 
that can help a worker keep her job. These realities help explain why 
many workers in low-wage jobs need assistance while they are working 
and when they are in between jobs. The nature of low-wage jobs can make 
it hard for a worker to meet rigid work requirements.
    Recent work by economists Kristen F. Butcher and Diane Whitmore 
Schanzenbach used the Census Bureau's Current Population Survey to show 
that the occupations of SNAP or Medicaid recipients who work at least 
part of the year feature instability and low wages overall (not just 
for SNAP or Medicaid recipients). These occupations include personal 
care and home health aides, maids and housekeepers, dishwashers, food 
preparers, and laundry and dry cleaning workers. Looking at all workers 
in the ten occupations most prevalent among SNAP recipients, the 
researchers found that these workers faced more periods of joblessness 
and were less likely to be stably employed from year to year than 
better-paid workers in other occupations. The researchers conclude, 
``Together, these results suggest that it will be difficult for 
individuals who work and participate in benefit programs to meet 
proposed work requirements in the private sector alone. Although 
employment levels are high among many of these types of workers, 
employment volatility is also quite high. Much of this volatility 
reflects characteristics of these types of occupations and is not 
necessarily due to decisions made by the workers.'' \402\
---------------------------------------------------------------------------
    \402\ Kristen F. Butcher and Diane Whitmore Schanzenbach, ``Most 
Workers in Low-Wage Labor Market Work Substantial Hours, in Volatile 
Jobs,'' Center on Budget and Policy Priorities, July 24, 2018, https://
www.cbpp.org/research/poverty-and-inequality/most-workers-in-low-wage-
labor-market-work-substantial-hours-in.
---------------------------------------------------------------------------
    Another study of the jobs that are common among SNAP participants 
found that, ``because SNAP participants work in many industries (such 
as retail and hospitality) and occupations (such as service and sales) 
where features such as involuntary part-time work and irregular 
scheduling are common, they may participate in SNAP to supplement their 
low incomes due to insufficient or fluctuating hours. Similarly, 
because workers often cycle in and out of these jobs, workers may 
participate in SNAP during periods of unemployment or 
underemployment.'' \403\
---------------------------------------------------------------------------
    \403\ Brynne Keith-Jennings and Vincent Palacios, ``SNAP Helps 
Millions of Low-Wage Workers,'' Center on Budget and Policy Priorities, 
May 10, 2017, https://www.cbpp.org/research/food-assistance/snap-helps-
millions-of-low-wage-workers.
---------------------------------------------------------------------------
    In addition, there is evidence that low-wage jobs have higher 
turnover and are far less likely to have access to paid sick leave or 
paid family leave.

   According to a 2018 study by the Economic Policy Institute, 
        ``the monthly rate of churn into and out of employment for low-
        wage workers is roughly twice as high as it is for the typical 
        worker in the middle of the wage distribution.'' \404\
---------------------------------------------------------------------------
    \404\ David Cooper, Lawrence Mishel, and Ben Zipperer, Economic 
Policy Institute, April 2018, https://www.epi.org/publication/bold-
increases-in-the-minimum-wage-should-be-evaluated-for-the-benefits-of-
raising-low-wage-workers-total-earnings-critics-who-cite-claims-of-job-
loss-are-using-a-distorted-frame/.

   Data from the Bureau of Labor Statistics shows that low-wage 
        workers are far less likely to have access to paid sick leave 
        or paid family leave.\405\
---------------------------------------------------------------------------
    \405\ Bureau of Labor Statistics, ``National Compensation Survey: 
Employee Benefits in the United States, March 2017,'' Bulletin 2787, 
September 2017.

    The Rationale of the NPRM Ignores Research on The Work Patterns of 
---------------------------------------------------------------------------
People Who Are Low-Skilled, Low-Wage Workers Subject to The Time Limit

    The NPRM states that ``The application of waivers on a more limited 
basis would encourage more ABAWDs to take steps towards self-
sufficiency.'' \406\ Secretary of Agriculture Sonny Perdue on February 
28, 2019 in testimony before the Senate Agriculture Committee defended 
the proposed rule, saying that ``We think the purpose is to help people 
move to independency . . . . We should help people when they are down 
but that should not be interminably.'' \407\
---------------------------------------------------------------------------
    \406\ NPRM, p. 981.
    \407\ ``Perdue Reiterates Need to Restore Original Intent of SNAP: 
A Second Chance, Not a Way of Life,'' Food and Nutrition Service, USDA, 
Press Release USDA 0025.19, February 28, 2019, https://
www.fns.usda.gov/pressrelease/2019/usda-002519.
---------------------------------------------------------------------------
    The belief that unemployed adults who participate in SNAP are 
dependent on SNAP for long periods ignores research that finds that 
large numbers of recipients who are not working at a point in time have 
recently worked or will work soon. A CBPP analysis of SNAP recipients 
shows that in a typical month in mid-2012, some 52 percent of adult 
recipients not receiving disability benefits were working, but that 74 
percent worked in the year before or after that month.\408\
---------------------------------------------------------------------------
    \408\ For similar households with just childless adults, 46 percent 
were working in a typical month and 72 percent worked within a year 
before or after that month. https://www.cbpp.org/unemployed-adults-
without-children-who-need-help-buying-food-only-get-snap-for-three-
months.
---------------------------------------------------------------------------
    The analysis examined adults who weren't receiving disability 
benefits and who participated in SNAP for at least a month in a period 
of almost 3.5 years. This allowed us to observe their work both while 
they participated in SNAP and in the months when they did not, and to 
observe employment among SNAP recipients over a longer period.
    The adults in the analysis worked the majority of the months in the 
analysis, but they were more likely to participate in SNAP in the 
months when they were out of work and their income was lowest. They 
participated in SNAP in about 44 percent of the months that they were 
working and in 62 percent of the months in which they were not working. 
This helps explain why an analysis that only looks at work in a single 
point-in-time month while people are receiving SNAP will show them 
working less than they do over time: many of them are workers who 
temporarily receive SNAP when they are between jobs.\409\
---------------------------------------------------------------------------
    \409\ Brynne Keith-Jennings and Raheem Chaudhry, ``Most Working-Age 
SNAP Participants Work, But Often in Unstable Jobs,'' Center on Budget 
and Policy Priorities, March 15, 2018.
---------------------------------------------------------------------------
    While these figures apply to all adults, not just those without 
children in the household, the research finding about the difference 
between point-in-time employment and employment over several years is 
still relevant and the NPRM does not address it. While individuals 
subject to the time limit may, in any given month, not have sufficient 
work hours to pass a rigid work test, many will be working (or working 
more hours) within a short time, with or without a work requirement. 
Moreover, when SNAP participants are working, it's often unstable work 
with low wages that does not lead to self-sufficiency, contrary to the 
framing included in the NPRM.
Research From the TANF Program Found That Employment Impacts Are Modest 
        and Fade Over Time
    The rigorous random-assignment evaluations of programs that imposed 
work requirements on cash assistance (AFDC/TANF) recipients in the late 
1990s contradicts the stated rationale for the proposed rule, but 
supports the finding from the RIA that under the proposed rule one 
would not expect to see increased employment or earnings. While these 
evaluations generally found modest, statistically significant increases 
in employment early on, the effects faded over time, and people with 
significant barriers to employment were not helped. In fact, many were 
hurt.

   In Portland, Oregon, the site of the largest earnings impact 
        among the evaluations, the share of recipients with stable 
        employment (defined as being employed in 75 percent of the 
        calendar quarters in years 3 through 5 after the pilot project 
        began) rose only from 31.2 to 38.6 percent.\410\
---------------------------------------------------------------------------
    \410\ Gayle Hamilton, et al., ``National Evaluation of Welfare-to-
Work Strategies: How Effective Are Different Welfare-to-Work 
Approaches? Five-Year Adult and Child Impacts for Eleven Programs,'' 
Manpower Demonstration Research Corporation, December 2001, Appendix 
Table C-6.

   Within 5 years, employment among people subject to and not 
        subject to work requirements was about the same in nearly all 
        the programs evaluated.\411\
---------------------------------------------------------------------------
    \411\ Ibid., Table C.1.

   Even when the programs provided specially tailored services, 
        the vast majority of participants facing significant employment 
        barriers did not find employment as a result of work 
        requirements.\412\
---------------------------------------------------------------------------
    \412\ Dan Bloom, Cynthia Miller, and Gilda Azurdia, ``Results from 
the Personal Roads to Individual Development and Employment (PRIDE) 
Program in New York City,'' MDRC, July 2007.

   The California GAIN program, the so-called ``Riverside 
        Miracle,'' which focused on getting recipients into any job as 
        quickly as possible, was outperformed in the long run by 
        programs that focused on increasing participants' skills and 
        building their human capital.\413\
---------------------------------------------------------------------------
    \413\ V. Joseph Hotz, Guido Imbens, and Jacob Klerman, ``Evaluating 
the Differential Effects of Alternative Welfare-to-Work Training 
Components: A Reanalysis of the California GAIN Program,'' Journal of 
Labor Economics, Volume 24 Number 3, 2006, pp. 521-566.

    One TANF expert researcher commented on the House Agriculture 
Committee's work requirement proposals from the 2014 Farm Bill, which 
would have expanded the existing approach for SNAP for childless adults 
to adults with children, that, ``[t]here is no credible evidence to 
suggest that the specific work requirements developed by the House 
Agriculture Committee would `work.' In fact, they are not likely to do 
much in the way of promoting employment and could push millions of 
families/individuals deeper into poverty.'' \414\
---------------------------------------------------------------------------
    \414\ Peter Germanis, ``Who Killed Work Requirements for SNAP in 
the Farm Bill? Answer: Conservative Ideologues,'' January 1, 2019, 
https://mlwiseman.com/wp-content/uploads/2019/01/Farmbill.120118.pdf.
---------------------------------------------------------------------------
    On balance, as we discuss in more detail in Chapter 6, this 
rigorous research supports the findings of the RIA that many people 
lose benefits when required to comply with work requirements, but 
contradicts the stated purpose of the proposed rule to increase self-
sufficiency.
Strong Evidence That Many Subject to the Time Limit Face Employment 
        Barriers and Would Lose Needed Help
    Based on the TANF experience from the 1990s, as well as the 
existing experience with the time limit in SNAP and the early 
experience from Arkansas (the only state so far to terminate Medicaid 
for individuals who fail to document that they are meeting Medicaid 
work requirements), many people subject to work requirements would lose 
benefits, and poverty and hardship would increase. This, again, is 
consistent with the analysis in the RIA, but not with the justification 
for the proposed rule.

   Research shows that many of the people who would be newly 
        subject to the time limit have circumstances that may limit the 
        amount or kind of work that they can do. A large share face 
        physical or mental health conditions or a cognitive impairment 
        that would be difficult for state agencies to identify or for 
        individuals to obtain paperwork to prove.\415\
---------------------------------------------------------------------------
    \415\ Rachel Garfield, Robin Rudowitz, and Anthony Damico, 
``Understanding the Intersection of Medicaid and Work,'' Kaiser Family 
Foundation, January 5, 2018; MaryBeth Musumeci, Julia Foutz, and Rachel 
Garfield, ``How Might Medicaid Adults with Disabilities Be Affected By 
Work Requirements in Section 1115 Waiver Programs?'' Kaiser Family 
Foundation, January 26, 2018; Bauer, Schanzenbach, and Shambaugh; 
Keith-Jennings and Chaudhry.

   Research shows that many TANF recipients who lost financial 
        assistance due to work requirements had serious barriers to 
        employment. They were likelier than other recipients to have 
        physical or mental health issues, have substance use disorders, 
        be victims of domestic violence, have low education and skill 
        levels, have prior criminal justice records, or lack affordable 
        child care.\416\
---------------------------------------------------------------------------
    \416\ LaDonna Pavetti, Michelle K. Derr, and Heather Hesketh, 
``Review of Sanction Policies and Research Studies: Final Literature 
Review,'' Mathematica Policy Research, March 10, 2003.

   The rigorous experiments from the 1990s that required cash 
        assistance recipients to participate in work-related activities 
        found that the resulting loss in benefits raised ``deep 
        poverty'' rates (the share of households with income below \1/
        2\ the poverty line).\417\ Similar results were found with 
        careful non-experimental analyses of leaver studies and 
        household survey data. Moreover, studies of TANF recipients 
        whose assistance was taken away found that they were likelier 
        to experience serious hardship, such as seeing their utilities 
        shut off, becoming homeless, or lacking adequate food.\418\ In 
        line with these findings, numerous scholars using a variety of 
        data and methods have concluded that cash assistance has 
        weakened as a guard against deep poverty under TANF,\419\ and 
        that some families are worse off as a result.\420\
---------------------------------------------------------------------------
    \417\ Stephen Freedman, et al., ``National Evaluation of Welfare-
to-Work Strategies--Evaluating Alternative Welfare-to-Work Approaches: 
Two-Year Impacts for Eleven Programs,'' Manpower Demonstration Research 
Corporation, June 2000; Pavetti, Derr, and Hesketh; Marianne Bitler, 
Hilary Hoynes, and Jonah Gelbach, ``What Mean Impacts Miss: 
Distributional Effects of Welfare Reform Experiments,'' American 
Economic Review, Volume 96, Number 4, 2006, pp. 988-1012.
    \418\ Ariel Kalil, Kristin Seefeldt, and Hui-chen Wang, ``Sanctions 
and Material Hardship under TANF,'' Social Service Review, December 
2002, pp. 642-662; Melissa Ford Shah, et al., ``Predicting Homelessness 
among Low-Income Parents on TANF,'' Washington State Department of 
Social and Health Services, August 2015, https://www.dshs.wa.gov/sites/
default/files/SESA/rda/documents/research-11-224.pdf; Andrew Cherlin, 
et al., ``Sanctions and Case Closings for Noncompliance: Who Is 
Affected and Why,'' Johns Hopkins University, Policy Brief 01-1, 2001.
    \419\ Gene Falk, ``Temporary Assistance for Needy Families (TANF): 
Size of the Population Eligible for and Receiving Cash Assistance,'' 
CRS Report No. R44724, Congressional Research Service, Washington, 
D.C., January 3, 2017.
    \420\ Ron Haskins, ``Welfare Reform at 20: Work Still Works,'' 
Journal of Policy and Management 35(1), 2016, 223-224; Robert A. 
Moffitt, ``The Deserving Poor, the Family, and the U.S. Welfare 
System,'' Demography 52(3), 2015, 729-749; Hilary W. Hoynes and Diane 
W. Schanzenbach, ``Safety Net Investments in Children,'' March 8, 2018, 
Brookings BPEA Article, https://www.brookings.edu/bpea-articles/safety-
net-investments-in-children/; James P. Ziliak, ``Temporary Assistance 
for Needy Families,'' 2015 NBER Working Paper 21038; Kristin S. 
Seefeldt and Heather Sandstrom, ``When There Is No Welfare: The Income 
Packaging Strategies of Mothers Without Earnings or Cash Assistance 
Following an Economic Downturn,'' 2015. Russell Sage Foundation Journal 
of the Social Sciences 1(1), 139-158; H. Luke Shaefer, Kathryn Edin, 
and Elizabeth Talbert, ``Understanding the Dynamics of $2-a-Day Poverty 
in the United States.'' (2015) Russell Sage Foundation Journal of the 
Social Sciences 1(1), 120-138; Christina Paxson and Jane Waldfogel, 
``Welfare Reforms, Family Resources, and Child Maltreatment,'' Journal 
of Policy Analysis and Management 22(1), 2003, 85-113.

   Research has shown that African American TANF recipients are 
        far likelier to have their benefits taken away than white 
        recipients.\421\ Caseworkers' decisions about when to impose a 
        sanction involve some discretion; one study, using fictitious 
        case examples, showed that caseworkers were likelier to 
        sanction African American recipients than white recipients. 
        Recipients of color may also be likelier to be sanctioned 
        because they face greater challenges in the labor market, 
        including discrimination.
---------------------------------------------------------------------------
    \421\ Sanford F. Schram, et al., ``Deciding to Discipline: Race, 
Choice, and Punishment on the Frontlines of Welfare Reform,'' American 
Sociological Review, January 2009; Kalil, et al.; Richard C. Fording, 
Joe Soss, and Sanford F. Schram, ``Devolution, Discretion, and the 
Effect of Local Political Values on TANF Sanctioning,'' Social Service 
Review, June 2007, pp. 285-316; Chi-Fang Wu, Maria Cancian, and Daniel 
R. Meyers, ``Sanction Policies and Outcomes in Wisconsin,'' Focus, 
Volume 23, Number 1, Winter 2004, https://www.irp.wisc.edu/
publications/focus/pdfs/foc231f.pdf; Pavetti 2004.

   Evidence from SNAP and Medicaid shows that administrative 
        hurdles can lead people to lose assistance even when they are 
        working or may qualify for an exemption, because they do not 
        understand or cannot comply with the requirement or because the 
        state agency fails to properly process the paperwork.\422\
---------------------------------------------------------------------------
    \422\ Rachel Garfield, Robin Rudowitz, and MaryBeth Musumeci, 
``Implications of a Medicaid Work Requirement: National Estimates of 
Potential Coverage Losses,'' Kaiser Family Foundation, June 27, 2018; 
Nader S. Kabbani and Parke E. Wilde, ``Short Recertification Periods in 
the U.S. Food Stamp Program,'' Journal of Human Resources, Volume 38, 
2003; David Ribar, Marilyn Edelhoch, and Qiduan Liu, ``Watching the 
Clocks: The Role of Food Stamp Recertification and TANF Time Limits in 
Caseload Dynamics,'' Journal of Human Resources, Volume 43, Number 1, 
2008, pp. 208-238; Mark Edwards, et al., ``The Great Recession and SNAP 
Caseloads: A Tale of Two States,'' Journal of Poverty, Volume 20 Issue 
3, December 11, 2015; Colin Gray, ``Why Leave Benefits on the Table? 
Evidence from SNAP,'' Upjohn Institute Working Paper 18-288, May 21, 
2018; Food and Nutrition Service, U.S. Department of Agriculture, 
``Understanding the Rates, Causes, and Costs of Churning in the 
Supplemental Nutrition Assistance Program (SNAP),'' November 2014.

   Recent evidence from Arkansas's implementation of work 
        requirements for Medicaid is sobering. Arkansas is taking 
        Medicaid coverage away from certain adult beneficiaries who 
        fail to report at least 80 hours of work or work-related 
        activities per month for 3 months. More than 18,000 Arkansans 
        have lost coverage after just 7 months of implementation, and 
        thousands more are at risk over the coming months. Data from 
        the state show that a very small share of those required to 
        report hours of participation (many beneficiaries are exempt 
        from the reporting requirement) have reported their hours, with 
        very few successfully navigating the exemption and ``good 
        cause'' processes.\423\
---------------------------------------------------------------------------
    \423\ Jennifer Wagner, ``Commentary: As Predicted, Arkansas' 
Medicaid Waiver Is Taking Coverage Away From Eligible People,'' Center 
on Budget and Policy Priorities, updated March 12, 2019, https://
www.cbpp.org/health/commentary-as-predicted-arkansas-medicaid-waiver-
is-taking-coverage-away-from-eligible-people.
---------------------------------------------------------------------------
The RIA Cites Only One Study, Which Does Not Support the Proposed Rule
    As mentioned, the NPRM cites no research to support that the 
proposed rule would achieve its purported goal: i.e., that taking food 
assistance away from certain low-income childless adults would 
encourage more self-sufficiency and employment. The one study 
referenced in the entire RIA document instead examines the relationship 
between the duration of unemployment and future employment and 
earnings.\424\ The study finds that long-term unemployment has a 
negative effect on the likelihood of future employment and that the 
fact that someone experiences long-term unemployment is the main reason 
(as opposed to inherent characteristics of individuals who experience 
long-term unemployment.) Strangely, the study offers little support for 
the NPRM and raises important cautionary notes suggesting that the 
proposed rule would worsen, not improve, outcomes for the targeted 
population.
---------------------------------------------------------------------------
    \424\ Katharine G. Abraham, et al., ``The Consequences of Long-term 
Unemployment: Evidence from Linked Survey and Administrative Data,'' 
National Bureau of Economic Research, Working Paper 22665, September 
2016, http://www.nber.org/papers/w22665. The citation appears on p. 3 
of the RIA.

   The one study cited in the RIA does not support the proposed 
        rule. The RIA suggests that because longer unemployment spells 
        are associated with a lower likelihood of future employment, 
        the proposed rule is justified. But the study does not mention 
        SNAP and does not address whether taking food assistance away 
        from low-income individuals would either decrease unemployment 
        spells or directly increase the likelihood of future 
        employment. In fact, it is difficult to understand what 
        connection could be made. The RIA estimates that 755,000 
        individuals would lose SNAP under the proposed rule, but 
        provides no estimate for increased employment. Much of the 
        research in this area shows that those individuals will face 
        increased hardship and may have a more difficult time finding 
---------------------------------------------------------------------------
        work.

   The population studied is not the population subject to the 
        SNAP policy. The study cited looked at long-term bouts of 
        unemployment by looking at a sample of all workers in state 
        unemployment insurance systems. But the childless adult 
        population subject to the SNAP time limit is a distinct group 
        that includes many individuals not included in the study group 
        because many adults who participate in SNAP do not receive 
        unemployment compensation, even if they are working or had 
        worked. A study of ABAWDs subject to the time limit in Ohio 
        found that nearly 80 percent had never been eligible for 
        unemployment benefits.\425\ More importantly, as discussed in 
        detail elsewhere in these comments, other research shows that 
        most childless adults who receive SNAP work when they can find 
        employment. Based on Census Bureau SIPP data, about 75 percent 
        of SNAP households with a childless, working-age adults worked 
        in the year before or after receiving SNAP. Many of these 
        individuals would not be in the pool of adults considered long-
        term unemployed in the study, so the conclusions drawn in the 
        cited study do not directly apply to ABAWDs as a group and do 
        not justify a policy change directed at them.
---------------------------------------------------------------------------
    \425\ Ohio Association of Food Banks, ``Franklin County 
Comprehensive Report on Able-Bodied Adults Without Dependents, 2014-
2015,'' October 14, 2015, p. 15, http://admin.ohiofoodbanks.org/
uploads/news/ABAWD_Report_2014-2015-v3.pdf.

   Finally, the study's findings suggest that support for 
        individuals to improve their skills or participate in work 
        programs would be a better approach. The study finds that ``the 
        longer-term unemployed experience substantially worse 
        employment and earnings losses than the short term 
        unemployed.'' The methodology, ``allows us to rule out the `bad 
        apple' explanation for why the long-term unemployed fare worse 
        . . . and [is] consistent with duration dependence as the 
        explanation for their poorer outcomes.'' This means that it is 
        not the characteristics of the individuals that cause them to 
        be long-term unemployed that are behind the results, but rather 
        the fact of their long-term unemployment. So, if FNS were 
        serious about wanting to help improve the longer-term outcomes 
        for individuals who experience long-term unemployment, it would 
        focus on helping to improve their education and skills or 
        providing slots in work experience programs that allow them to 
        demonstrate their desire to work, rather than cut their food 
        assistance.
C. Reports That Purport to Find Positive Effects From the Time Limit 
        Are Deeply Flawed
    The only studies that claim to find substantial positive impacts 
when low-income individuals are faced with losing food assistance or 
other benefits if they do not meet rigid work requirements are deeply 
flawed. FNS does not cite this research either, but we include here 
some discussion of why FNS should not rely on these kinds of assertions 
in any future policy development. The faulty results come from the 
researchers making causal claims without a random-assignment design (or 
other analytically sound comparison-group methods), ignoring program 
participants' work experience prior to receiving assistance, and 
excluding the impact on households of losing benefits.\426\ Below we 
explain how two such reports, citing data from Kansas and Maine, have 
inaccurately touted the alleged success of reimposing a 3 month time 
limit on SNAP participation for childless adults.\427\
---------------------------------------------------------------------------
    \426\ Dottie Rosenbaum and Ed Bolen, ``SNAP Reports Present 
Misleading Findings on Impact of Three-Month Time Limit,'' Center on 
Budget and Policy Priorities, December 14, 2016; Tazra Mitchell, 
LaDonna Pavetti, and Yixuan Huang, ``Study Praising Kansas' Harsh TANF 
Work Penalties Is Fundamentally Flawed,'' Center on Budget and Policy 
Priorities, updated February 20, 2018.
    \427\ See Jonathan Ingram and Nic Horton, ``The Power of Work, How 
Kansas' Welfare Reform is Lifting Americans Out of Poverty,'' The 
Foundation for Government Accountability, February 16, 2016, https://
thefga.org/wp-content/uploads/2016/02/Kansas-study-paper.pdf; Maine 
Office of Policy and Management, ``Preliminary analysis of work 
requirement policy on the wage and employment experiences of ABAWDs in 
Maine,'' April 19, 2016, http://www.maine.gov/economist/opm/pub/
ABAWD_analysis_final.pdf; and accompanying Maine Department of Health 
and Human Services May 11, 2016 press release and other related 
materials.
---------------------------------------------------------------------------
    When the recession decimated the labor market and unemployment 
spiked, most states, including Kansas and Maine, requested the time 
limit be waived statewide. Kansas reimposed the time limit statewide 
beginning in October 2013 and Maine reinstated the time limit statewide 
in October 2014, even though both states qualified for a statewide 
waiver at the time the time limit returned.
    In both states total SNAP caseloads already were declining, but 
they dropped significantly 4 months after the time limit was put in 
place, as Figure 11.1 shows. Data from Kansas and Maine that are 
limited to the childless adults who were potentially subject to the 
time limit show that SNAP participation fell among that group by 70 to 
80 percent after the time limit returned.
Figure 11.1
States Implementing SNAP Time Limit Experienced Sudden Drops in SNAP 
        Participation
Kansas
SNAP Participants (in thousands)

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Maine
SNAP Participants (in thousands)

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Source: Agriculture Department SNAP participation data.

    The reports assert that, as a result of the SNAP time limit, work 
rates and wages have increased dramatically and the individuals subject 
to the time limit are better off. The reports, however, misrepresent or 
omit data and, as a result, make claims about the impact of the time 
limit on work and earnings that the facts do not support.\428\ The 
analyses also rest on faulty assumptions about why some childless 
adults receiving SNAP are not working.
---------------------------------------------------------------------------
    \428\ Peter Germanis has also written extensively on the 
methodological shortcomings of these types of studies. Peter Germanis, 
``How Do the Foundation for Government Accountability's Evaluations of 
Welfare Reform Measure Up? A Report Card (Hint: The FGA Fails),'' June 
24, 2018, https://mlwiseman.com/wp-content/uploads/2016/05/Evaluating-
Welfare-Reform.pdf.
---------------------------------------------------------------------------
    The reports' three largest problems are:

   They do not take into account that many SNAP recipients 
        already work, or would work soon even without the time limit. 
        The studies attribute rising work rates and earnings to the 
        return of the time limit even though most, if not all, of the 
        changes would have happened without it. The authors fail to 
        acknowledge that many SNAP recipients who are subject to the 
        time limit were working already, or would soon be working, and 
        as a result, their work rates and wages would likely have risen 
        without the time limit. They make claims that can only be 
        identified through a rigorous evaluation that isolates the 
        impacts of the time limit from what would have happened without 
        it (see box below).

   They do not consider the potentially severe impact of the 
        time limit on those cut off SNAP. The studies fail to discuss 
        the circumstances of the individuals who are subject to the 
        time limit and the consequences for increased hardship and food 
        insecurity when they lose SNAP benefits. Without addressing 
        this side of the equation, the studies misrepresent the effect 
        of reinstating the time limit on the well-being of those cut 
        off SNAP. Their figures on the average income of those cut off 
        SNAP are highly misleading because they do not include the loss 
        of SNAP benefits. They do not discuss or attempt to assess what 
        happens to individuals who lose their food assistance and are 
        unable to find employment, who are a large share of those cut 
        off.

   They do not adequately consider the likely explanations for 
        why childless SNAP participants may not work. The authors 
        advance the theory that individuals are avoiding work and 
        remaining in poverty in order to qualify for modest SNAP 
        benefits of only about $5 a day. But research and experience in 
        states with the time limit in effect offer evidence of 
        alternative explanations. Many such individuals do work when 
        they can, but they often face significant barriers to work, 
        such as low education and skills or physical or mental health 
        issues.

    A careful look at the data presented in the reports, taking these 
factors into account, strongly suggests that not much changed related 
to work and earnings when the time limit took effect, but the time 
limit did cause thousands of the states' poorest residents to lose 
essential SNAP benefits.

------------------------------------------------------------------------
 
-------------------------------------------------------------------------
    Conventional Evidence-Based Research Uses a ``Comparison Group''
 
    One of the central tenets of sound, evidence-based research is the
 need to have a ``comparison group'' so that the results can properly
 account for what would have happened in the absence of a change.
    For example, consider researchers who are testing the efficacy of a
 new medicine designed to speed recovery from the common cold. The
 researchers would need to know how fast people would have gotten better
 without the medicine. Without a comparison group there would be no way
 to know what to make of results that showed, for example, that 30
 percent were better after 2 days and 85 percent were better after 5
 days. Many, perhaps all, of these people would have gotten better
 without the medicine.
    The gold standard for comparison groups is ``random assignment,'' an
 experimental approach where people who are otherwise the same are
 randomly assigned to different ``treatment'' groups and the effects of
 the change are measured on each group so the study can isolate the
 effect of the ``treatment.'' These types of studies are expensive,
 though some are underway in SNAP, funded by the 2014 Farm Bill.
    In the case of low-income childless adults, two important factors
 are critical for interpreting the information in the Kansas and Maine
 reports. First, many low-skill, low-wage workers do work, but they work
 in high-turnover jobs with low job security and often experience
 sporadic employment. SNAP acts as a safety net, providing assistance
 during periods of unemployment or when work hours are cut. It is common
 for SNAP recipients to have higher employment and wages in the future.
 Second, both Kansas' and Maine's economies were improving between 2013
 and 2015, the period in which the two states implemented the time limit
 and purport to measure the results. Without controlling for these
 factors, it is difficult to isolate the effects of the time limit on
 employment.
    The authors of the reports for Kansas and Maine could have
 established less complicated and less costly alternative ``comparison
 groups'' by conducting the same analysis in the year before the cutoff
 to observe the work rates and earnings for similar SNAP recipients
 during a period when the time limit was not in effect. Such an approach
 would not have been perfect--it would be impossible to take the
 differences in the labor market and all other factors into account--but
 it would have been a more informative comparison than these reports
 provide.
------------------------------------------------------------------------

Reports Don't Acknowledge That Many SNAP Recipients Subject to the Time 
        Limit Already Work
    The studies from Kansas and Maine assert that reimposition of the 
time limit resulted in higher work rates and earnings for individuals 
who lost SNAP benefits after exhausting 3 months of eligibility.

   For Kansas, the authors claim, ``These reforms immediately 
        freed nearly 13,000 Kansans from welfare on December 31, 2013. 
        Nearly 60 percent of those leaving food stamps found employment 
        within 12 months and their incomes rose by an average of 127 
        percent per year.'' \429\
---------------------------------------------------------------------------
    \429\ Ingram and Horton, op. cit., p. 2.

   For Maine, the Department of Health and Human Services' 
        press release reported that among the individuals whose SNAP 
        was cut off, ``Incomes rose 114 percent within a year of 
        leaving the program,'' and ``nearly \1/2\ (48%) worked at least 
---------------------------------------------------------------------------
        one quarter in 2015.''

    These reports, however, dramatically overstate the increase in work 
rates and wages that resulted from the reimposition of the time limit 
because many of the SNAP recipients affected were working, or would 
have started working anyway, albeit mostly in low-wage jobs with high 
turnover. Moreover, both states reimposed the time limit when their 
unemployment rates were dropping and unemployed individuals were 
somewhat more likely to be able to find work or higher wages as a 
result.\430\
---------------------------------------------------------------------------
    \430\ In Kansas, where the time limit went into effect in October 
2013, the overall unemployment rate fell from 5.3 percent in 2013 to 
4.6 percent in 2014 and 4.2 percent in 2015. In Maine, where the time 
limit was reimposed 1 year later, the unemployment rate fell from 5.6 
percent in calendar year 2014 to 4.4 percent in calendar year 2015. For 
SNAP recipients, especially those with the lowest education and skills, 
employment opportunities are highly sensitive to economic conditions 
and the availability of jobs. See Hilary Hoynes, Douglas Miller, and 
Jessamyn Schaller, ``Who Suffers During Recessions?'' NBER Working 
Paper No. 17591, March 2012, http://www.nber.org/papers/w17951.pdf.
---------------------------------------------------------------------------
Kansas Report Misrepresents Several Key Indicators
    The authors of the Kansas report overstate the degree to which work 
rates increased after the time limit and misrepresent the effect of the 
time limit on numerous outcomes for SNAP recipients while they are 
receiving SNAP.

    Work Rates Were Essentially Unchanged Before and After the Time 
Limit Returned

    The Kansas authors claim that ``nearly 60 percent of those leaving 
[SNAP after the 3 month time limit went into effect] found employment 
within 12 months.'' (The authors' estimate of 60 percent is the share 
who had ever worked in any quarter of 2014.) The claim implies that the 
policy change reimposing the time limit was the reason that these 
people found work, but it's misleading, as explained below.
    However, work rates before and after the time limit were very 
similar, as Figure 11.2 shows. Almost 40 percent of those whose SNAP 
was cut off already worked in each of the last two quarters before the 
time limit returned (the third and fourth quarters of 2013).\431\ The 
share working each quarter in the year after the time limit was 
implemented rose slightly, to just over 40 percent. This modest 
increase could be explained by two factors: (1) low-wage workers are 
more likely to apply for and participate in SNAP when they lose a job 
or their incomes drop, so they often experience improvements in the 
future as their employment situation improves; and (2) Kansas' economy 
was improving during 2014, so a slightly larger share of recipients may 
have been able to find jobs or higher pay. The time limit does not 
appear to have dramatically affected work rates for the group subject 
to it.
---------------------------------------------------------------------------
    \431\ CBPP calculates that almost 40 percent of the people who left 
the program on December 31, 2013 worked in the two quarters just before 
the cutoff based on data on average wages per person from table 8 of 
the Kansas report. (The same calculation is used to estimate the shares 
working in the later quarters as well.) Instead of reporting these 
accurate data, the authors misleadingly report lower rates from 
national data sources as though they applied to the Kansas group. See 
p. 6: ``Currently few able-bodied adults receiving food stamps actually 
work. . . . In 2013 just \1/4\ of childless adult households receiving 
food stamps had any earned income. . . . An analysis of food stamp 
recipients conducted when work requirements first went into effect 
found that fewer than five percent of all able-bodied childless adults 
on the program were meeting those requirements.'' This latter five 
percent is extremely misleading because it excludes a large number of 
individuals who were working more than 30 hours a week as ``exempt'' 
from the time limit. If the authors included the share of non-disabled 
childless adults who were working, the figure would be larger.
---------------------------------------------------------------------------
Figure 11.2
Kansas Work Rates Nearly the Same Before and After Time Limit
Share of non-disabled childless adults cut off SNAP who earned wages in 
        each quarter (Q) of a calendar year
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Note: CBPP derived the share with wages from information on 
        average quarterly wages among those cut off SNAP from table 8 
        of Ingram and Horton.
          Source: Jonathan Ingram and Nic Horton, ``The Power of Work, 
        How Kansas' Welfare Reform is Lifting Americans Out of 
        Poverty.'' The Foundation for Government Accountability. 
        February 16, 20106; Table 8.

    The authors, however, reached the opposite conclusion--that work 
rates grew significantly after the time limit returned. Instead of 
comparing the average work rates in each quarter for this population 
before and after the policy change, they report the share of 
individuals whose SNAP was ended after December 2013 who ever worked in 
a quarter over the following year. This captures typical movement in 
and out of the labor force--given that this group tends to work in 
high-turnover jobs, in any quarter some people lose jobs and some get 
new jobs, so the share that ever worked increases--rather than an 
isolated impact of the policy change. The trends in the share who ever 
worked likely followed a very similar pattern in earlier years when the 
time limit was not in effect (though the authors do not present such 
data). As discussed below, other research about labor force 
participation among childless adults who receive SNAP finds work rates 
over time similar to those in the Kansas report.
Improvements for SNAP Recipients Reflect SNAP Caseload Changes, Not 
        Improved Circumstances
    The Kansas report presents highly misleading information about 
other changes among individuals subject to the time limit who receive 
SNAP. For example:

          [S]ince restoring work requirements, the employment rate 
        among able-bodied adults on food stamps has doubled. As a 
        result their incomes have more than doubled on average, they 
        are spending less time on welfare, and the need for assistance 
        has significantly declined.\432\
---------------------------------------------------------------------------
    \432\ Ingram and Horton, op. cit., p. 8.

    These claims, which the report makes across a range of measures, 
are misleading because the childless adults who remained as SNAP 
participants after the time limit went into effect were significantly 
different from those who participated before because of the policy 
change. The state cut off SNAP those participants who were not working 
at least 20 hours a week, so the work rates, average earnings, and 
other characteristics of those who remained SNAP participants after the 
return of the time limit were better, not because those individuals 
became better off, but because they were better off to begin with and 
were the only ones still eligible for and participating in SNAP.
    Those who may still participate in SNAP are more likely to have 
earnings and, as a result, lower SNAP benefits and appear better off on 
a range of other characteristics. In fact, the number of childless 
adult SNAP recipients working at least 20 hours a week, and thus the 
only non-exempt childless adult SNAP recipients eligible for the 
program, dropped modestly in the year after the time limit took 
effect.\433\
---------------------------------------------------------------------------
    \433\ The drop in the number of childless adults who worked at 
least 20 hours a week and received SNAP could have occurred because 
those individuals who qualified for SNAP (because they were working at 
least 20 hours a week) had recently been cut off SNAP (at a time when 
they were not working at least 20 hours a week) and did not know they 
would be eligible if they reapplied.
---------------------------------------------------------------------------
    As an example of this misleading representation, consider the 
authors' assertion that, ``[p]rior to restoring work requirements, just 
21 percent of childless adults on food stamps were working at all. Two-
fifths were working less than 20 hours per week. But since work 
requirements have gone back into effect, that employment rate has risen 
to nearly 43 percent.'' \434\ The change was driven by a drop in the 
number of SNAP recipients who are childless adults subject to the time 
limit, not an increase in the number of recipients who are working. The 
number of such SNAP recipients who were working fell by more than 40 
percent (from 6,300 to 3,600), as those who were working less than 20 
hours a week were cut off, while the total number of non-disabled 
childless adults receiving SNAP dropped by more than 70 percent (from 
almost 30,000 to 8,500). (See Figure 11.3.)
---------------------------------------------------------------------------
    \434\ Ingram and Horton, op. cit., p. 9.
---------------------------------------------------------------------------
Figure 11.3
Kansas SNAP Benefit Cutoff Did Not Boost Work
Non-disabled childless adult SNAP participants before and after January 
        2014 cutoff of those not working 20+ hours per week
Number of such SNAP participants who worked fell . . . 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

. . . but the work rate rose only because there were fewer such SNAP 
        participants overall
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Jonathan Ingram and Nic Horton, ``The Power of Work, 
        How Kansas' Welfare Reform is Lifting Americans Out of 
        Poverty.'' The Foundation for Government Accountability. 
        February 16, 20106; Tables 1-3.
Maine Also Inappropriately Attributes Changes to the Policy Change That 
        Likely Would Have Occurred Anyway
    Maine's data on work rates and wages among individuals who lost 
SNAP are similar in magnitude to Kansas, and, as in the Kansas report, 
the authors of the Maine report and the accompanying materials from the 
state's Department of Health and Human Services overstate the impact of 
the time limit by failing to take into account the fact that changes 
would have occurred even without it.
    As Figure 11.4 shows, before the reimposition of the time limit in 
October 2014, about 30 percent of the childless adults whose SNAP was 
cut off were working. That proportion peaked in the months after the 
time limit went back into effect at 36 percent in the third quarter 
(the summer, a time when employment in Maine tends to be higher). But, 
though the report includes these quarterly rates, like for Kansas, the 
Maine report and accompanying materials emphasize a different figure: 
that 48 percent had wages some time in 2015 and 58 percent had wages at 
some time ever in 2014 or 2015. But again, like for Kansas, the higher 
numbers count any time anyone had worked in any quarter, and thus 
largely reflect employment instability at a time that the state's 
economy was improving, rather than a change that could be attributed to 
the reimposition of the time limit.
Figure 11.4
Maine Work Rates Nearly the Same Before and After Time Limit
Share of non-disabled childless adults cut off SNAP who earned wages in 
        each quarter (Q) of a calendar year
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: Maine Office of Policy Management, ``Preliminary 
        analysis of work requirement policy on the wage and employment 
        experiences of ABAWDs in Maine of Health and Human Services,'' 
        April 19, 2016; figure 1.
No Consideration of the Well-Being of Those Cut Off or the Support SNAP 
        Provides
    The one-sided pictures in these reports fail to discuss the well-
being of the individuals whose SNAP benefits were cut off. But research 
suggests that many childless adults who lose SNAP as a result of the 
time limit continue to struggle after losing SNAP, in contrast to the 
reports' portrayals of circumstances for recipients who lost benefits. 
The most comprehensive assessment of former SNAP recipients in four 
states in the early 2000s suggests that their life circumstances are 
quite difficult. A significant minority don't find work, and among 
those who are employed after leaving SNAP, earnings are low. Most 
remain poor. Many struggle to acquire enough food to meet their needs, 
lack health insurance, experience housing problems, and/or have trouble 
paying their bills.\435\
---------------------------------------------------------------------------
    \435\ Elizabeth M. Dagata, ``Assessing the Self-Sufficiency of Food 
Stamp Leavers,'' Economic Research Service, USDA, September 2002, 
https://www.ers.usda.gov/publications/pub-details/?pubid=46645, a 
summary of in-depth studies in Arizona, Illinois, Iowa, and South 
Carolina. These studies include people who leave SNAP because of the 3 
month time limit or for other reasons, for example, because they found 
a job or mistakenly believe they are no longer eligible.
---------------------------------------------------------------------------
    In a serious omission the Kansas and Maine reports do not consider 
the impact of the time limit on the large number of people who lost 
SNAP and are among the nation's very poorest adults.

   In Kansas the number of non-disabled childless adults 
        receiving SNAP dropped by 75 percent (from about 30,000 in late 
        2013 to about 7,500 in late 2015).

   The Maine report does not present comparable numbers, but an 
        earlier Heritage Foundation report cited Maine Department of 
        Health and Human Services data showing that the number of 
        ``able-bodied adults without dependents on food stamps'' 
        dropped by 80 percent (from about 13,300 in late 2014 to 2,700 
        in March 2015).\436\
---------------------------------------------------------------------------
    \436\ Robert Rector, Rachel Sheffield, and Kevin D. Dayaratna, 
``Maine Food Stamp Work Requirement Cuts Non-Parent Caseload by 80 
percent,'' The Heritage Foundation, Backgrounder No. 3091, February 8, 
2016, http://www.heritage.org/research/reports/2016/02/maine-food-
stamp-work-requirement-cuts-non-parent-caseload-by-80-percent.

    The individuals whose SNAP was cut off lost about $5 a day, or $150 
to $170 per person per month in SNAP benefits for purchasing food. Many 
of them worked in the year after losing benefits, but for some their 
wages were low enough that they could have continued to qualify for 
SNAP benefits, which could have helped them make ends meet. Some others 
with no earnings for some or all of the subsequent year may have had 
virtually no resources available for food after they were cut off SNAP.
    The Kansas and Maine reports cite average income figures for the 
year after recipients lost SNAP, but they fail to account for the lost 
SNAP benefits. To accurately compare income for a household that used 
to be on SNAP to income after losing SNAP, the lost value of the SNAP 
benefits must be included. When they are accounted for, total income 
does not increase substantially (or actually decreases slightly).
    In Kansas, the authors rest their claim that SNAP recipients were 
better off after the cutoff on a point that the group's income rose by 
127 percent between before the time limit took effect and 1 year later. 
There are three problems with this claim:

   First, as discussed above, it implies that the time limit 
        was responsible for the earnings increase, when most, if not 
        all, of it likely would have occurred anyway;

   Second, it excludes the value of SNAP benefits from the 
        calculation. Total resources available to the household were 
        higher before the time limit because the household received 
        SNAP benefits; and,

   Third, the authors picked a low comparison quarter prior to 
        the time limit returning to exaggerate the increase--wages for 
        the group cut off were more than 30 percent lower two quarters 
        before the cutoff (third quarter of 2013, the quarter used in 
        the report) than they were in the quarter immediately preceding 
        the cutoff (the fourth quarter of 2013). They do not explain 
        why they chose this particular quarter as the baseline.

    It is not possible to adjust for the first issue without a rigorous 
evaluation (see above box), but even adjusting only for the other two 
issues makes a large difference. If $178 a month in SNAP benefits (the 
average SNAP benefit among those cut off in December 2013) is included 
in the base period, and if we compare the quarter immediately before 
the cutoff (the fourth quarter of 2013) as the base period instead of 
the quarter earlier, the total resources (including earnings and SNAP 
benefits) available to SNAP participants who were cut off was three 
percent lower a year after the cutoff, rather than 127 percent higher.
    Had the SNAP recipients who remained income-eligible been able to 
keep receiving SNAP (rather than being cut off by the time limit) more 
of them would be better off because they could have received SNAP while 
working (though their SNAP benefits would be lower because income 
counts in determining SNAP benefit levels).
    A similar contrast applies to the Maine report. The press release 
that accompanied the report claims that total income for those cut off 
rose by 114 percent within a year. However, if the state had included 
the value of SNAP benefits in the base, the increase would be much 
smaller--only about ten to 20 percent.\437\
---------------------------------------------------------------------------
    \437\ The Maine report does not include information about the 
average SNAP benefits received by childless adults who were cut off, so 
the ten to 20 percent range reflects a lower ($150 a month) and higher 
($180 a month) assumption. As with Kansas, we cannot account for the 
large portion of the effect that would have happened anyway and we are 
not including any SNAP benefits for the workers who would have income 
low enough to continue participating in SNAP were there no time limit.
---------------------------------------------------------------------------
No Consideration of Factors Affecting SNAP Participants' Ability to 
        Work
    If a better picture of the data shows that the time limit doesn't 
have a strong role getting people into work, what do we know about why 
this group struggles to find employment? The research that exists on 
this population shows that adults who participate in SNAP work when 
they can, but often in jobs with high turnover and low job security, 
and most struggle with multiple barriers to employment, as discussed 
above.
    Adults on SNAP work when they can. However, the work tends to be 
low wage and unstable, with individuals cycling through periods of work 
and unemployment. Nearly \3/4\ of non-disabled adults who participate 
in SNAP in a typical month work either that month or within a year of 
that month. Over \1/2\ of individuals who were participating in SNAP in 
a typical month in mid-2012 were working in that month. Furthermore, 74 
percent worked in the year before or after that month.\438\ (See Figure 
11.5.) Limited education, lack of training, and a sporadic work history 
make it difficult to compete for anything other than low-skill, low-
wage jobs that do not lift them out of poverty.
---------------------------------------------------------------------------
    \438\ Brynne Keith-Jennings and Raheem Chaudhry, ``Most Working-Age 
SNAP Participants Work, But Often in Unstable Jobs,'' Center on Budget 
and Policy Priorities, March 15, 2018. As mentioned above, for similar 
households with just childless adults, 46 percent were working in a 
typical month and 72 percent worked within a year before or after that 
month. https://www.cbpp.org/unemployed-adults-without-children-who-
need-help-buying-food-only-get-snap-for-three-months.
---------------------------------------------------------------------------
Figure 11.5
Most SNAP Participants and Households Work 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Note: Individuals and households include those who were 
        participating in SNAP in a typical month in mid-2012. A working 
        household refers to a household in which either the household 
        head or spouse worked. Individuals include any non-disabled 
        adult who reported participating in SNAP.
          Source: CBPP analysis of SIPP data from 2011-2013.

    Childless adults on SNAP face barriers. Many low-income childless 
adults face multiple challenges to independence and self-sufficiency, 
including homelessness, physical and mental health limitations, 
language barriers, unstable employment histories, and criminal records. 
A detailed study of childless adults who were referred to a work 
experience program in Franklin County (Columbus), Ohio found that: 
\439\
---------------------------------------------------------------------------
    \439\ See ``Comprehensive Report on Able-Bodied Adults Without 
Dependents, Franklin County Ohio Work Experience Program,'' Ohio 
Association of Foodbanks, 2015, http://admin.ohiofoodbanks.org/uploads/
news/ABAWD_Report_2014-2015-v3.pdf. The Ohio Association of Foodbanks 
gathered the information for the report as a result of a partnership 
with the county SNAP agency to help place individuals identified as 
subject to the time limit in qualifying work activities after screening 
them.

   Many have extremely unstable living situations, illustrated 
        by residence in short-term shelters or with friends and family 
---------------------------------------------------------------------------
        and limited telephone service.

   One-third have a mental or physical limitation, including 
        depression, post-traumatic stress disorder, mental or learning 
        disabilities, or physical injuries. Some of these disabilities, 
        though not severe enough to qualify for Federal disability 
        benefits, may still limit a person's ability to work at least 
        20 hours a week.

   About \1/4\ have less than a high school education, and more 
        than \1/2\ have only a high school diploma or GED.

   Nearly \1/4\ are non-custodial parents, and 13 percent are 
        caregivers for a parent, relative, or friend.

   More than 40 percent lack access to reliable private or 
        public transportation; 60 percent lack a valid driver's 
        license.

   Fifteen percent need supportive services like language 
        interpretation or help with transportation to obtain 
        employment.

   More than \1/3\ have felony convictions, making it hard to 
        find jobs and pass background checks.
D. FNS' Research, Policies, and Practices Show FNS Knows but Ignored 
        That Impact of the Proposed Rule Is Out of Line With the Stated 
        Rationale
FNS' Own Research from 1998 on the Employment Prospects of ``ABAWDs'' 
        Contradicts the Stated Justification of the NPRM
    In 1998 FNS published a study called, ``The Effect of Welfare 
Reform on Able-Bodied Food Stamp Recipients'' to provide information 
that ``[I]s critical to informing policy decisions, issuing guidance to 
states, implementing new policies, as well as estimating effects of the 
[new provisions.]'' It concluded in the forward to the study, ``the 
report offers a sound picture of what able-bodied adult recipients 
without children look like and what will happen to them--they are an 
extremely poor population with limited employment prospects and few 
sources of support outside the Food Stamp Program.'' \440\ This 
research directly contradicts the stated justification for the proposed 
rule.
---------------------------------------------------------------------------
    \440\ Michael Stavrianos and Lucia Nixon, ``The Effect of Welfare 
Reform on Able-Bodied Food Stamp Recipients,'' prepared by Mathematica 
Policy Research for the USDA, Food and Nutrition Service, July 23, 
1998, https://fns-prod.azureedge.net/sites/default/files/finalrep.pdf.
---------------------------------------------------------------------------
    The study used SNAP QC household characteristics data and the 
Census Bureau's Survey of Income and Program Participants (SIPP) to 
describe the characteristics of individuals subject to the time limit, 
including their limited educational attainment and workplace skills, 
their high poverty rates, and their patterns of SNAP participation 
prior to the time limit going into effect. It also estimated how many 
at that time had likely hit the time limit and been cut off SNAP.
    The study also included information about the research available at 
that time on the employment prospects of ``ABAWD'' SNAP participants, 
which is summarized as follows:

          Research indicates that the employment prospects of adults 
        who are demographically similar to ABAWDs are not promising, 
        and so we can assume the same to be true for ABAWDs. Largely 
        for two reasons, job opportunities for less-educated job 
        seekers are severely limited, especially for non-whites and in 
        urban areas, where most ABAWDs live. First, recent research 
        suggests that many large employers of low-skill workers have 
        moved out of the cities to the suburbs. Therefore, many ABAWDs 
        will face a ``spatial mismatch'' between the location of their 
        residence and the location of low-skill jobs. Second, since 
        employment in inner cities has become increasingly concentrated 
        in high-skill jobs, ABAWDs will also likely face a ``skills 
        mismatch'' between what employers require and what ABAWDs can 
        offer.\441\
---------------------------------------------------------------------------
    \441\ Ibid, p. xii and pp. 51-69.

    As we review in Chapter 3, while the nature of spatial mismatch has 
changed, more recent literature has found that there still exists 
mismatch between low-wage jobs and where low-wage workers live, 
particularly with regards to transportation access.
    The 1998 FNS study also points out that low-skilled job seekers in 
many places may have difficulty finding employment even when the 
national unemployment rate is low:

          Implicit in PRWORA's work requirement is the assumption that 
        there are enough employment opportunities for ABAWDs--that is, 
        they can find work if they seek it. . . . However, a relatively 
        large body of research indicates that the labor market 
        situation of the low-skilled has become considerably worse in 
        recent decades and that their current employment prospects are 
        limited. This suggests that even if ABAWDs are willing to work, 
        they may be unable to do so because there are not enough jobs 
        for low-skilled workers.\442\
---------------------------------------------------------------------------
    \442\ Ibid, p. 56-57.

    Despite arguing repeatedly in the preamble and RIA that the low 
national unemployment rate justifies the proposed rule,\443\ FNS does 
not refute this earlier research it published in 1998, nor address 
whether the landscape has changed in the intervening 20 years to 
justify the change in waiver policy.
---------------------------------------------------------------------------
    \443\ NPRM, p. 981, 982, 983; RIA, p. 2, 10.
---------------------------------------------------------------------------
USDA Research on Individuals Who May Have Been Cut off SNAP Because of 
        the Time Limit Does Not Support the Assertion That the Time 
        Limit Improves Self-Sufficiency
    After the 3 month time limit was enacted in 1996, USDA's Economic 
Research Service (ERS) joined with the U.S. Department of Health and 
Human Services to fund studies in four states (Arizona, Illinois, Iowa, 
and South Carolina) that examined the well-being of people who exited 
SNAP in the late 1990s after the time limit went into effect.\444\ The 
studies included people who had left SNAP because of the 3 month time 
limit or for other reasons, for example, because the found a job or 
mistakenly believe they no longer are eligible. Even though the studies 
were not able to isolate the individuals who left SNAP because of the 
time limit, the picture they offer of the hardship such individuals 
face suggest that the time limit has not spurred many to self-
sufficiency, or even resulted in their life circumstances improving 
modestly, and contradict the stated rationale behind the proposed rule.
---------------------------------------------------------------------------
    \444\ Elizabeth Dagata, ``Assessing the Self-Sufficiency of Food 
Stamp Leavers,'' Economic Research Service, USDA, September 2002, 
https://www.ers.usda.gov/publications/pub-details/?pubid=46645, a 
summary of in-depth studies in Arizona, Illinois, Iowa, and South 
Carolina. https://lib.dr.iastate.edu/cgi/
viewcontent.cgi?referer=https://www.google.com/&httpsredir=1&
article=1013&context=card_staffreports--Iowa. https://www.mathematica-
mpr.com/our-publications-and-findings/publications/food-stamp-leavers-
in-illinois-how-are-they-doing-two-years-later--Illinois. https://
naldc.nal.usda.gov/download/45220/PDF--South Carolina.

   Many were employed but had very low earnings. In the four 
        states, employment rates among the individuals who were 
        unemployed childless adults potentially subject to the time 
        limit who left SNAP (or ``leavers'') ranged from 41 percent in 
        Illinois to 76 percent in Iowa, ``but earnings and incomes are 
        low and their poverty rates are high.'' \445\
---------------------------------------------------------------------------
    \445\ Dagata, p. 2.

   Most remain poor. Despite relatively high levels of work 
        effort, between \1/3\ and \2/3\ of SNAP leavers in the four 
        states had household incomes below the poverty line--well above 
        the overall poverty rate of 13 percent at the time. About 40 
        percent of the Illinois and Iowa SNAP leavers had income below 
---------------------------------------------------------------------------
        \1/2\ the poverty line.

   Many struggled to afford adequate food. ``Between 17 and 34 
        percent of the [SNAP leavers in the four states] reported food 
        insecurity with hunger, compared with 11 percent of U.S. low-
        income childless households.'' \446\
---------------------------------------------------------------------------
    \446\ Dagata, p. 4. Food insecurity ``with hunger'' was how USDA 
then referred to the most severe form of food insecurity where 
households had to skip or reduce the size or their meals or otherwise 
disrupt their eating patterns at times during the year because they 
couldn't afford sufficient food.

   Many had housing problems, or had trouble paying their 
        utility bills. About 20 to 40 percent of SNAP leavers faced 
        housing issues, including falling behind on the rent, moving in 
        with relatives, or becoming homeless. Between 20 and 65 percent 
---------------------------------------------------------------------------
        reported problems paying for utilities.

    Finally, the studies raise an important question about exemptions 
from the time limit. ``In two of the studies, the majority of 
nonworking [SNAP leavers] cited health problems as the reason they were 
not working . . . it is important to know whether the standards for 
being categorized as `able-bodied' are set appropriately.'' \447\
---------------------------------------------------------------------------
    \447\ Dagata, p. 5.
---------------------------------------------------------------------------
FNS' SNAP Employment and Training Best Practices and the Department's 
        SNAP to Skills Initiative Do Not Promote Cutting People Off 
        SNAP
    FNS' position that taking food assistance away from people who do 
not meet rigid work requirements will lead to stable employment and 
self-sufficiency also conflicts with the agency's approach to 
employment and training (E&T). For more than 30 years, since the mid-
1980s, SNAP has included an employment and training (E&T) component, 
``for the purpose of assisting members of households participating in 
the supplemental nutrition assistance program in gaining skills, 
training, work or experience that will . . . increase the ability of 
the household members to obtain regular employment.'' \448\ States 
operate the program within Federal rules and FNS oversight, but have 
substantial flexibility about which non-exempt SNAP recipients they 
serve in the program, the services they offer participants, and whether 
the program primarily recruits volunteers who are interested in help 
finding a job or improving their education and skills, or whether the 
program is mandatory, meaning individuals will lose SNAP benefits 
(i.e., be subject to sanction) if they do not comply with the state's 
assigned E&T activity.
---------------------------------------------------------------------------
    \448\ The Food and Nutrition Act of 2008, as amended, 7 U.S.C.  
2015(d)(4)(A)(i).

    E&T Best Practices Research Review Does Not Include Sanctions or 
---------------------------------------------------------------------------
Cutting People Off SNAP

    While over the years Congress has amended and modified the focus of 
SNAP E&T somewhat, it has always included a state option for programs 
that are mandatory and include sanctions for non-compliance. According 
to a 2016 FNS-sponsored literature review of SNAP E&T best practices, 
the only time FNS conducted a rigorous study of the effectiveness of 
SNAP E&T, was in the late 1980s and early 1990s. At the time, SNAP E&T 
was primarily ``a high-volume, `light-touch' program to encourage 
mandatory work registrants to find jobs quickly, primarily by requiring 
participation in job search and job-search training components,'' \449\ 
that is, it consisted primarily of components that included taking SNAP 
benefits away from individuals who did not meet the program's 
requirements. The 2016 literature review described the findings from 
the study that was published in 1994 as follows:
---------------------------------------------------------------------------
    \449\ Deborah Kogan, et al., ``Supplemental Nutrition Assistance 
Program (SNAP) Employment and Training (E&T) Best Practices Study: 
Final Report,'' prepared by Social Policy Research Associates for the 
U.S. Department of Agriculture, Food and Nutrition Service, November 
2016, https://www.fns.usda.gov/snap/snap-employment-and-training-et-
best-practices-study-final-report.

          [T]here was no evidence that the SNAP E&T program--in its 
        high participation/low investment per-participant model--
        increased the likelihood of participants finding jobs. It also 
        found the program had no significant effects on the hourly 
        wages, hours worked per week, or length of job retention for 
        those who did find employment . . .
          The observed effect of a statistically significant decrease 
        in the level of food stamp benefits receipt was described as 
        most likely the result of `voluntary withdrawals and 
        administrative sanctions' rather than of any increase in 
        household income or earnings . . .
          [A]lthough some members of the treatment group did find jobs, 
        members of the control group were also able to obtain similar 
        job search assistance and find employment.\450\
---------------------------------------------------------------------------
    \450\ Ibid, p. III-1.

    Thus, the only rigorous evaluation FNS has ever conducted of 
policies that take food assistance away from individuals who are unable 
to meet work requirements found no improvement in participants finding 
jobs or increases in wages or hours worked, but did find a significant 
decrease in food assistance benefits. The literature review quoted here 
was published in 2016, so was certainly available to FNS as it 
formulated the proposed rule, but FNS ignored its findings. FNS 
promulgated the proposed rule that would take food assistance away from 
people who do not comply with rigid work requirements without ever 
again testing or studying the effectiveness of such an approach in a 
rigorous random assignment study. Moreover, the E&T pilots from the 
2014 Farm Bill, which do incorporate a rigorous random assignment 
evaluation, include three states with mandatory SNAP E&T approaches, 
but FNS promulgated this regulation without waiting for the results of 
those pilots.
    Overall, the 2016 SNAP E&T best practices report had as its 
objective providing ``Congress, FNS, and individual states with 
information that can be used to shape the services provided by the SNAP 
E&T program and thereby improve the employability, self-sufficiency, 
and well-being of individuals receiving nutrition support from SNAP.'' 
\451\ These are the very same goals as the stated intention and 
justification of the proposed rule in the NPRM, and yet the SNAP E&T 
best practices report does not mention taking SNAP benefits away from 
individuals who are unable to meet work requirements as an effective 
strategy. The report includes an annotated bibliography of 160 relevant 
studies from the literature review. The best practices report 
recommendations are summarized in the executive summary:
---------------------------------------------------------------------------
    \451\ Ibid, p. ES-1.
---------------------------------------------------------------------------
    The findings from the research synthesized in this report suggest 
that SNAP recipients will benefit most from SNAP E&T-funded services if 
the services offered by state programs

   Are based on an individualized assessment of the workforce-
        related strengths and weaknesses of SNAP clients;

   Comprehensively address an individual's need for skills 
        training, basic skills education, and overcoming barriers to 
        employment;

   Help participant earn credentials valued by employers in 
        their chose industry or sector; and

   Develop skills that are closely linked to labor market 
        demands in the local area.

      In view of these findings . . . States that enroll a relatively 
        large number of mandatory work registrants in SNAP E&T services 
        or that emphasize self-reported job search as the most 
        frequently prescribed program activity are less likely to see 
        an increase in self-sufficiency among SNAP participants.\452\
---------------------------------------------------------------------------
    \452\ Ibid, p. ES-4.

    Thus, the most recent research available to FNS about what works 
for SNAP E&T for meeting the objectives of increasing employment, self-
sufficiency, and well-being does not include sanctions or cutting 
people off SNAP. But FNS ignored this evidence in promulgating the 
---------------------------------------------------------------------------
regulation. We urge FNS to consider its own best practices study.

    FNS' Most Recent E&T Efforts Also Have Not Promoted Sanctions or 
Cutting People Off SNAP

    Moreover, in the past several years FNS has placed a new emphasis 
on SNAP E&T. It has created an Office of Employment and Training with 
several additional staff members, brought on consultants and 
collaborated with other partners, formed a ``learning academy,'' 
produced new training materials, and provided additional technical 
assistance to select states. The cornerstone of the E&T efforts has 
been a ``SNAP to Skills'' initiative that follows closely the 
recommendations from the SNAP E&T best practices report cited above.
    Under a ``Why SNAP to Skills'' section of its website, FNS makes 
the following case for its SNAP to Skills approach: \453\
---------------------------------------------------------------------------
    \453\ U.S. Department of Agriculture, ``Why SNAP to Skills,'' 
https://snaptoskills.fns.usda.gov/why-snap-to-skills.

          The need of Supplemental Nutrition Assistance Program (SNAP) 
        participants to secure the education and training required to 
        transition to economic self-sufficiency is growing increasingly 
        urgent. The vast majority of jobs in the future will require at 
        least some education beyond high school . . . yet many SNAP 
        participants have not reached this level of educational 
        attainment. Without the skills to meet rapidly changing labor 
        market demand, the chances of SNAP participants getting a good 
        job and reducing their need for SNAP are extremely 
        low . . .
          The SNAP Employment & Training (SNAP E&T) program, a skills 
        and job training program for SNAP participants administered by 
        the U.S. Department of Agriculture's Food and Nutrition Service 
        (FNS), is a key resource states and their partners can utilize 
        to help SNAP participants meet this urgent need for skills and 
        better jobs. SNAP E&T has historically been under-utilized, but 
        a renewed focus on the program amid greater urgency for job 
        training for SNAP participants has created new momentum for 
        states seeking to build bigger, better, and stronger E&T 
        programs.
          There is no more opportune, or critical, time for states to 
        build robust, job-driven SNAP E&T programs. ``Job-driven'' 
        means that programs are responsive to employer demand so that 
        they place ready-to-work participants in good, available jobs 
        or provide skills training and credentials participants require 
        to obtain these jobs. SNAP E&T is increasingly recognized as a 
        critical part of each state's skilled workforce strategy. USDA 
        and other policy makers herald SNAP E&T as an important part of 
        the national conversation about the need to invest in building 
        a skilled workforce while addressing the nation's growing 
        economic inequality.\454\
---------------------------------------------------------------------------
    \454\ Ibid.

    Like the best practices report cited earlier, FNS' signature 
Employment and Training initiative does not include sanctions or 
cutting individuals off of SNAP as an effective strategy for increasing 
employment or earnings.
E. FNS Knows That Research and Experience Shows That People Newly 
        Subject to the Time Limit Won't Get a Job or Be Better Off, But 
        It Promulgated the Proposed Rule Nonetheless
    As we have laid out in this section, there is a large body of 
research evidence that finds that policies that take benefits away from 
individuals who do not meet rigid work requirements result in lost 
benefits and increased poverty and hardship, but very little gain in 
longer-term employment. The realities of the low-wage labor market, 
including high turnover and lack of sick time and other benefits 
contribute to individuals' turning to SNAP during temporary periods of 
unemployment. Many other individuals face various employment barriers.
    FNS has for more than 20 years supported the waiver policy 
currently in regulation, and there is no new research that contradicts 
or provides new information. The RIA conspicuously lacks any 
countervailing research evidence to justify that the narrowing of 
waivers will improve individuals' work rates or earnings, and the 
impact analysis included in the RIA assumes that 755,000 individuals 
will lose SNAP benefits under the change, but there will be no 
quantifiable increases in earnings or work.
    We believe that the only conclusion one can draw is that contrary 
to the stated rationale in the NPRM, FNS knows that the primary effect 
of the regulatory change, if it were finalized, would be that a large 
number of individuals would lose SNAP assistance, with no, or very 
little positive impact. We strongly urge FNS to review the research 
summarized here and included in the Appendix B.
F. The Proposed Rule Uses an Imprecise Definition of ``ABAWDs,'' and 
        the RIA Includes Numerous other Unsubstantiated Assumptions
    In addition to the fact that the proposed rule is not supported by 
available research, the analysis that is included in the RIA includes 
numerous imprecise, illogical, and unsubstantiated assumptions, 
starting with the use of the term ``ABAWDs.'' Since shortly after the 
passage of the 1996 welfare law, FNS has described the group of SNAP 
participants whose eligibility is at issue because of the 3 month time 
limit as ``ABAWDs,'' or able-bodied adults without dependents. The 
proposed rule uses this term throughout, but never defines it, and 
seems to include in it many individuals who are exempt from the time 
limit, who live in an area that is under a waiver, or who are 
participating in SNAP during periods when they are eligible (for 
example, in their first 3 months or when they are working or 
participating in a qualifying employment and training program.) The 
methodology for assessing the impact of the proposed rule indefensibly 
treats everyone who is an ``ABAWD'' by this broad definition as though 
they would be subject to the time limit under the proposed rule (i.e., 
that they live in an area that would no longer qualify for a waiver.) 
The rule also excludes from the analysis individuals who are ``ABAWDs'' 
subject to the time limit, but who are no longer SNAP recipients 
because they have been cut off.
    The imprecise use of the undefined term ``ABAWD'' is confusing and 
makes it difficult for readers to understand and comment on the 
described impacts of the proposed rule. It also appears that the 
Department in its estimates of the impact of the regulation has derived 
percentages for this entire population of SNAP recipients potentially 
subject to the time limit and then applied those percentages to 
individuals who would be newly subject to the time limit under the 
proposed rule, resulting in a methodology that cannot be substantiated.
    This section will first explain the analytical problem in the way 
the RIA defines and categorizes ``ABAWDs,'' and then provide additional 
examples of specific assumptions in the methodology that cannot be 
substantiated, often because of the imprecise use of the term 
``ABAWD.''
Use of the Undefined Term ``ABAWD'' Is Confusing and Misleading
    Section 6(o) (7 U.S.C.  2015(o)) of the Food and Nutrition Act 
limits SNAP eligibility for certain non-exempt adults who are aged 18-
49 to 3 months of SNAP benefits in any 36 month period if they are not 
working 20 hours a week or participating in a qualifying employment and 
training program, and if the area they live in is not waived from the 
rule because of insufficient jobs.\455\ Because of the rule's 
complexity, the individuals subject to the time limit cannot be 
identified in SNAP or Census Bureau data because much of the 
information that would need to be known is not available in the data. 
Analysts often have modeled SNAP eligibility for these individuals to 
the best of their ability and then described the individuals as 
``potentially subject to the time limit,'' because only state 
eligibility workers have access to the information that is needed to 
make a full assessment. In a 2016 paper we explained:
---------------------------------------------------------------------------
    \455\ Individuals who have been cut off SNAP can qualify for a 
second 3 months of eligibility if they work at least 80 hours (or 
participate in a qualifying program) for a month and reapply for SNAP.

          About 4.7 million non-elderly, non-disabled adults aged 18-49 
        in childless households participated in SNAP in Fiscal Year 
        2014. All were ``subject'' to the time limit in the sense that 
        all could, in theory, have lost benefits after 3 months of 
        participation. The number affected by the time limit in 
        practice, however, is much smaller.\456\
---------------------------------------------------------------------------
    \456\ In the 2016 paper CBPP used this methodology and included a 
box explaining the difference between the larger number potentially 
subject to the time limit and the smaller number affected by the time 
limit in practice. See, Steven Carlson, Dorothy Rosenbaum, Brynne 
Keith-Jennings, ``Who are the Low-Income Childless Adults Facing the 
Loss of SNAP in 2016,'' February 8, 2016, https://www.cbpp.org/sites/
default/files/atoms/files/2-8-16fa.pdf, p. 4.

    We also explained in the 2016 paper the various reasons why the 
larger figure, which we represent as individuals potentially subject to 
the time limit based on existing data, applies to ``the characteristics 
of the larger group of childless adults, all of whom would face the 
time limit if their circumstances or local labor market conditions 
change.'' \457\
---------------------------------------------------------------------------
    \457\ Ibid.
---------------------------------------------------------------------------
    In estimates that FNS published in the years immediately following 
passage of the 1996 welfare law, it is clear that the agency understood 
both the limitations in the data (``The QC database does not contain 
all the information needed to determine whether an individual loses 
eligibility under the able-bodied provisions of PRWORA'' \458\) and the 
distinction between those potentially subject to the time limit and 
those who would actually be affected (``the QC-based estimates 
presented in this chapter may overstate the number of people subject to 
the 3 month time limit.'' \459\)
---------------------------------------------------------------------------
    \458\ Mike Stavrianos, et al., ``Characteristics of Childless 
Unemployed Adult and Legal Immigrant Food Stamp Participants: Fiscal 
Year 1995,'' prepared by Mathematica Policy Research for the USDA, Food 
and Nutrition Service, February 13, 1997, p. 6.
    \459\ Ibid, Stavrianos, p. 6.
---------------------------------------------------------------------------
    But the NPRM, RIA, and Agriculture Department materials that 
accompanied the December posting of the proposed rule portray this 
larger group that is potentially subject to the time limit as the 
number who actually would be newly subject under the proposed rule. 
This misidentification is confusing and results in unsubstantiated 
assumptions. It identifies as newly subject to the time limit many 
individuals who are exempt, complying with the time limit, or living in 
an area that is under a waiver from the time limit, and it excludes 
individuals who would qualify for SNAP but have been cut off after 3 
months in areas that are not under a waiver from the time limit.
    The Department asserts that ``[i]n 2016 there were 3.8 million 
individual ABAWDs on the SNAP rolls.'' \460\ We are able to reproduce 
this number using the public use 2016 SNAP Household Characteristics 
Quality Control (QC) file,\461\ and below recreate how we believe FNS 
derived the number. (See Table 11.1)
---------------------------------------------------------------------------
    \460\ U.S. Department of Agriculture, ``Regulatory Reform at a 
Glance, Proposed Rule: SNAP Requirements for ABAWDs,'' December 2018, 
https://fns-prod.azureedge.net/sites/default/files/snap/
ABAWDSFactSheet.pdf. We believe this 3.8 million is the starting point 
for all of the estimates in the RIA regarding the number of ``ABAWDs,'' 
but the RIA does not make that clear, as we discuss later in our 
comments.
    \461\ The 2016 public use Quality Control Data are available at 
https://host76.mathematica-mpr.com/fns/. All figures we cite based on 
this data are for an average month in the fiscal year.

                               Table 11.1
  CBPP Understanding of FNS' Estimate That 3.8 million SNAP Recipients
                         Were ``ABAWDs'' in 2016
------------------------------------------------------------------------
                                                        Percent of Total
    FNS' 3.8 million ``ABAWD''            Number              SNAP
             estimate                                     Participants
------------------------------------------------------------------------
Total SNAP Participants                  43.5 million               100%
Age 18-49                                15.0 million              34.4%
Not receiving disability benefits        12.8 million              29.5%
 a
No minor children in household            3.8 million               8.8%
------------------------------------------------------------------------


 
Other factors that need
    to be taken into
     account for an
    individual to be
  subject to the time     Is this information available in the QC data?
  limit that were not
   factored into FNS'
       estimate:
------------------------------------------------------------------------
Is the individual:
 
     in his or   Variable not reliable b
     her first 3 months
     of SNAP
     participation out
     of 36?
     physically  Variable not reliable c
     or mentally unfit
     for employment?
     living in   Variable not reliable b
     a waived area?      But may be knowable if state was under a
                          statewide waiver
     working 20  Can be estimated using earnings or other
     hours a week or      variables
     more?
     participat  Variable not reliable d
     ing in a
     qualifying E&T
     activity?
     pregnant?   Not available
     otherwise   Variable not reliable c
     exempt from
     employment and
     training?
     exempted    Variable not reliable b
     by an individual
     exemption?
     in a        Variable not reliable b
     second 3 month
     period after
     requalifying?
------------------------------------------------------------------------
Sources: CBPP analysis of FY 2016 USDA SNAP Household Characteristics
  data and Mathematica Policy Research, ``Technical Documentation for
  the Fiscal year 2016 [or other year] Supplemental Nutrition Assistance
  Program Quality Control Database and the QC Minimodel,'' October 2017.
a The SNAP QC data set includes a personal-level disability variable
  (DISi). An algorithm is used to identify individuals with disabilities
  based on SSI receipt, medical expenses, age, work registration status,
  and other factors. The technical documentation flags that ``DISi
  likely underestimates the number of non-elderly individuals with
  disabilities'' and therefore, the 3.8 million likely overestimates the
  number of adults without disabilities.
b The SNAP QC data includes an individual-level variable called
  ``ABWDSTi'' that is intended to collect this information, but the
  technical documentation ``[recommends] caution when
using . . . due to inconsistencies.''
c For the variable intended to capture exemptions for disability and
  other factors (``WRKREGi'') the documentation states, ``we found
  continued evidence . . . of likely miscoding of this variable.''
d The variable intended to capture participation in employment and
  training (``EMPRGi'') is also among the variables the documentation
  ``recommend[s] using with caution.''

    As Table 11.1 shows, there are many eligibility factors that the 
Department's analysis did not take into account when estimating the 
number of people who are subject to the time limit. As a result, the 
3.8 million individuals the Department classifies as ``ABAWDs'' 
includes many individuals who, in fact, are not subject to the time 
limit. The Department's analysis also excludes many individuals who 
would be SNAP recipients except they have been cut off because of the 
time limit, so they do not appear in SNAP data because they are not 
SNAP recipients.\462\ These are individuals who do not live in waived 
areas but are subject to the time limit because they did not meet any 
of the other criteria in Table 11.1.
---------------------------------------------------------------------------
    \462\ Stavrianos and Nixon (1998, p. 4) and Czajka, et al., (2001, 
p. 30) both include flow charts that makes clear FNS had information 
available that made clear which ``ABAWDs'' would be subject to the time 
limit and the various reasons individuals who might be identified in 
the data as ``ABAWDs'' might not be affected by the time limit.
---------------------------------------------------------------------------
    The Department's imprecise use of the term ``ABAWD'' results in a 
lack of transparency. It is difficult to determine what point FNS (in 
the NPRM and RIA), and the Administration more broadly, in its 
materials about the proposed rule, are making when they call 
individuals ``ABAWDs'' when they are not, in fact, subject to the time 
limit, or when they are complying with the time limit. They are 
implying that far more SNAP participants are subject to the time limit 
and not in compliance with it than in fact is the case, and they are 
not counting people who have been cut off because of the time limit.
    In addition, several of the major assumptions in the RIA's 
methodology for assessing the impact of the proposed rule rely on 
shares of this larger ``ABAWD'' group, as defined nationwide using the 
2016 data, but apply those shares to individuals who would be newly 
subject to the time limit because of the proposed changes in the rules 
for areas to qualify for waivers. For example, the RIA's assumption 
about the share of ``ABAWDs'' who are working is derived from the SNAP 
data for 2016 including both waived areas and not-waived areas. Using 
shares that are derived from a group that includes many individuals who 
are not subject to the time limit, and that excludes many individuals 
who have been cut off SNAP in areas that were not waived in 2016 is 
extremely misleading and illogical. The denominator for these 
percentages matters for assessing the soundness of using certain 
percentages for deriving or estimating the impact of the proposed 
changes.
    To help elucidate the problem, we conducted an analysis comparing 
the number and share of SNAP participants for two different categories 
of states regarding waivers from the time limit in 2 different fiscal 
years. The two types of states were those with statewide waivers in 
Fiscal Year 2013, but no waivers in Fiscal Year 2017 and those with 
statewide waivers in both years.\463\ The difference between the 2 
years for the two types of states can help explain how the denominator 
changes when many people are cut off as a result of the time limit. We 
will use figures from this analysis to help explain some of the serious 
methodological problems with the RIA's analysis of the impact of the 
proposed rule. It is easier to see the issue when considering these two 
types of states, but it is present in other states that have had 
different patterns and scope of waivers.
---------------------------------------------------------------------------
    \463\ The states with statewide waivers from the time limit in 2013 
but no waivers at all in 2017 (which represented about a quarter of 
SNAP participants in 2013) were Alabama, Arkansas, Florida, Indiana, 
Iowa, Kansas, Maine, Mississippi, Missouri, North Carolina, Oklahoma, 
South Carolina, and Wisconsin. The states with statewide waivers in 
both 2013 and 2017 (which represented about 20 percent of SNAP 
participants) included Alaska, California, District of Columbia, 
Illinois, Louisiana, Nevada, New Mexico, Rhode Island, Guam, and Virgin 
Islands.
---------------------------------------------------------------------------
    We use fiscal 2017 instead of Fiscal Year 2016 for this analysis 
because many states reimplemented the time limit beginning on January 
1, 2016, which means that for Fiscal Year 2016 the time limit was in 
effect for just the latter 6 months of the fiscal year (April through 
September, once the 3 months of eligibility for January to March are 
taken into account.) By Fiscal Year 2017 those states had no waiver the 
entire fiscal year. Under the definition of ``ABAWD'' that FNS appears 
to use, the 3.8 million national figure cited in the FNS materials and 
that we recreate above in Table 11.1 fell to 3.2 million in Fiscal Year 
2017.
    Table 11.2, shows the number of SNAP participants in the two types 
of state in each year and the number and share that are ``ABAWDs'' 
under our understanding of how FNS is defining ABAWDs for purposes of 
the RIA--as SNAP participants aged 18-49, with no disability benefits 
and no minor children in the household.

                               Table 11.2
         Waivers Affect the Number and Share of SNAP ``ABAWDs''
 Fiscal Years 2013 and 2017 by whether the state had a statewide waiver
------------------------------------------------------------------------
                     Total SNAP                        ABAWDs as a Share
                  Participants (in   Total ``ABAWDs''       of Total
                       000s)            (in 000s)         Participants
------------------------------------------------------------------------
                            Fiscal Year 2013
------------------------------------------------------------------------
Participants                12,439              1,496                12%
 residing in
 states with
 statewide
 waivers in
 2013 and no
 waivers in
 2017
Participants                 8,346                953                11%
 residing in
 states with
 statewide
 waivers in
 both 2013 and
 2017
                --------------------------------------------------------
  Total                     47,098              4,943                10%
   participants
   all states a
------------------------------------------------------------------------
                            Fiscal Year 2017
------------------------------------------------------------------------
Participants                10,325                609                 6%
 residing in
 states with
 statewide
 waivers in
 2013 and no
 waivers in
 2017
    % change                  ^17%               ^59%
     2013 to
     2017
  Participants               8,164                927                11%
   residing in
   states with
   statewide
   waivers in
   both 2013
   and 2017
    % change                   ^2%                ^3%
     2013 to
     2017
                --------------------------------------------------------
  Total                     41,491              3,221                 8%
   participants
   all states a
                --------------------------------------------------------
      % change                ^12%               ^35%
       2013 to
       2017
------------------------------------------------------------------------
Source: CBPP analysis of FY 2013 and FY 2017 USDA SNAP Household
  Characteristics data
a Includes participants in the two types of states identified above, as
  well as participants residing in other states.

    Table 11.2 shows:

   The number of both SNAP recipients and ``ABAWDs'' declined 
        in both types of states between 2013 and 2017 but fell 
        substantially more in states that had reimposed the time limit 
        by 2017. The number of ``ABAWDs'' potentially subject to the 
        time limit fell by 59 percent in states that reimposed the time 
        limit, from 1.5 million in 2013 to 600,000 in 2017. The economy 
        may have been stronger in these states, and there may be other 
        reasons for a larger drop, but the fact that eligibility rules 
        changed and many people in this group could not participate for 
        more than 3 months was likely a major factor in the larger 
        drop.

   The share of total SNAP participants who were ``ABAWDs'' in 
        the states that reimposed the time limit by 2017 fell from 12 
        percent to six percent but was flat at 11 percent in the states 
        that had a statewide waiver in both years. So, although the 
        number and share of ``ABAWDs'' fell substantially in states 
        that reimposed the time limit by 2017, six percent of SNAP 
        participants still fit into the ``ABAWD'' category as defined 
        by the RIA. These individuals likely meet one of the 
        eligibility criteria in Table 11.1 and so were not cut off.

    The share of ``ABAWDs'' working also changes as the denominator 
changes when individuals are cut off SNAP. In Table 11.3 we compare the 
work rates among SNAP participants who were ``ABAWDs'' under the FNS' 
definition in the same two types of states as above for the same 2 
years.

                                                   Table 11.3
Waivers Affect the Number and Share of SNAP ``ABAWDs'' Who Are Working Fiscal Years 2013 and 2017 by whether the
                                          state had a statewide waiver
----------------------------------------------------------------------------------------------------------------
                                                                                   ``ABAWDs''        Share of
                                  Total          ``ABAWDs''        Share of        working at       ``ABAWDs''
                              ``ABAWDs'' (in    working (in       ``ABAWDs''    least 20 hrs/wk     working at
                                  000s)            000s)           working         (in 000s)     least 20 hrs/wk
----------------------------------------------------------------------------------------------------------------
                                                Fiscal Year 2013
----------------------------------------------------------------------------------------------------------------
States with statewide                  1,496              331              22%              187              13%
 waivers in 2013 and no
 waivers in 2017
States with statewide                    953              165              17%               63               7%
 waivers in both 2013 and
 2017
                            ------------------------------------------------------------------------------------
  Total all states a                   4,943            1,087              22%              587              12%
----------------------------------------------------------------------------------------------------------------
                                                Fiscal Year 2017
----------------------------------------------------------------------------------------------------------------
States with statewide                    609              207              34%              138              23%
 waivers in 2013 and no
 waivers in 2017
States with statewide                    926              219              24%              111              12%
 waivers in both 2013 and
 2017
                            ------------------------------------------------------------------------------------
  Total all states a                   3,221              897              28%              534              17%
----------------------------------------------------------------------------------------------------------------
Source: CBPP analysis of FY 2013 and FY 2017 USDA SNAP Household Characteristics data.
a Includes participants in the two types of states identified above, as well as participants residing in other
  states.

    In the states that had reimposed the time limit by 2017, although 
the number of ABAWDs dropped between 2013 and 2017, the share of 
``ABAWDs'' working went up substantially, from 22 percent in 2013 to 34 
percent in 2016, and the share estimated to be working at least 20 
hours a week nearly doubled, from 13 percent to 23 percent. In part 
this could be a function of a stronger economy in these states, or 
other factors, but the fact that many people who were not working were 
cut off also contributed significantly to the change in work rate. In 
the states that had statewide waivers both years, the share working 
went up, but by less, from 17 percent to 24 percent, and the share 
estimated to be working at least 20 hours a week went from seven 
percent to 12 percent.
    Between 2013 and 2017 the number of ``ABAWDs'' in areas without 
waivers went down in large part because individuals were cut off of 
SNAP in areas without waivers. And because individuals could continue 
to receive SNAP if they were working more than 20 hours a week, the 
share of ABAWDs working at least 20 hours a week went up in areas 
without waivers, in large part because the denominator used in 
calculating the share (the number of ABAWDs who received SNAP) went 
down. In a 2016 report responding to a misleading report that claimed 
the circumstances of SNAP recipients in Kansas improved after they 
reinstated the time limit we explained this math as follows:

          . . . the childless adults who remained as SNAP participants 
        after the time limit went into effect were significantly 
        different from those who participated before because of the 
        policy change. The state cut off SNAP those participants who 
        were not working at least 20 hours a week, so the work rates, 
        average earnings, and other characteristics of those who 
        remained SNAP participants after the return of the time limit 
        were better, not because those individuals became better off, 
        but because they were better off to begin with and were the 
        only ones still eligible and participating in SNAP.\464\
---------------------------------------------------------------------------
    \464\ Dorothy Rosenbaum and Ed Bolen, ``SNAP Reports Present 
Misleading Findings on Impact of Three-Month Time Limit,'' Center on 
Budget and Policy Priorities, December 14, 2016, https://www.cbpp.org/
sites/default/files/atoms/files/12-14-16fa.pdf.

    The RIA methodology includes numerous imprecise, confusing, 
inaccurate, or misleading assumptions, some that push in opposite 
directions. We cannot tell if FNS has intentionally produced an 
analysis that inflates (or deflates) the results or if the individuals 
charged with producing the RIA do not understand the policy and 
therefore were unable to produce a coherent analysis of the population 
subject to the current policy and what the impact of the proposal would 
be. Either way, the lack of transparency and coherency in the RIA 
raises serious questions about the validity of the NPRM process.
The RIA Does Not Explain the Claim That There Would Be 3.4 Million 
        ``ABAWDs'' in 2020
    The methodology evidently assumes that there would be 3.4 million 
``ABAWDs'' in 2020 under current law, but never explains where that 
figure comes from. This is a serious omission because this is the 
starting point FNS uses for all of the subsequent assumptions about the 
number of individuals who would be affected by the proposed changes in 
waiver rules. Excluding information on this foundational point 
compromises all that follows.
    The only time the 3.4 million figure is mentioned, on page 25 of 
the RIA, the document says, ``As noted previously, the Department 
estimated that approximately 44 percent of the projected 3.4 million 
ABAWDs . . . would live in waived areas in FY 2020 if waiver authority 
were unchanged.'' It is possible that the figure comes from the 3.8 
million from the SNAP QC data for 2016 cited above, adjusted to reflect 
the FNS' baseline number of participants for 2020 compared to the 
number of participants in 2020, but we cannot comment on this figure as 
the RIA provides no justification for it whatsoever.
    If the 3.4 million is the 3.8 million from 2016 adjusted only for 
baseline changes, then the FNS has made no further adjustments to 
account for the fact that states qualified for and applied for waivers 
for fewer areas in 2018 and 2019 than in 2016 and will likely qualify 
for still fewer waivers in 2020 even without any changes to the waiver 
rules.
The RIA Assumption That 44 Percent of ``ABAWDs'' Are in Waived Areas Is 
        Based on a Proxy That Is Indefensible
    As one step in its estimate of the impact of the proposed rule, the 
analysis in the RIA assumes that 44 percent of the 3.4 million 
``ABAWDs,'' or 1.5 million individuals, would live in waived areas in 
FY 2020 if the regulation were unchanged. According to RIA (page 19):

          The Department used state-reported data from form FNS-388A to 
        estimate the number of non-public assistance SNAP participants 
        living in currently waived areas. Since the FNS-388A does not 
        report ABAWD participation separately, non-public assistance 
        SNAP participants are used as a proxy when estimating the 
        proportion of ABAWDs living in waived areas.

    The RIA does not explain what the form FNS-388A is or why it is 
appropriate to use it and what its shortcomings might be. Based on a 
review of the July 2018 data that states submitted, it appears that the 
388A does not include all SNAP participants (Oregon and Vermont are 
missing) and that there are no county-level data for several states, 
including all of New England, Alaska, Idaho, Missouri, Montana, 
Nebraska, New York, Utah, Washington, and Wisconsin.\465\
---------------------------------------------------------------------------
    \465\ U.S. Department of Agriculture, Supplemental Nutrition 
Assistance Program Data, https://www.fns.usda.gov/pd/supplemental-
nutrition-assistance-program-snap.
---------------------------------------------------------------------------
    But even if the 388A included county data for all states, it does 
not make sense to use the number of participants in non-public 
assistance households (those that do not receive TANF cash assistance, 
Supplemental Security Income, or General Assistance) as a proxy for the 
number of ABAWDs. ABAWDs are much more likely to be subject to the time 
limit and cut off SNAP than non-public assistance households overall. 
ABAWDs are a small fraction of non-public assistance households (less 
than 15 percent) and their distribution across counties will depend on 
whether the time limit is in effect.
    To show that the time limit matters for the distribution of ABAWDs 
across counties, we again used the SNAP QC data for 2013 and 2017 and 
again compared the states that had a statewide waiver in 2013 and not 
in 2017 to states that had a statewide waiver in both years. As can be 
seen in Table 11.4, in 2013 when both types of states had statewide 
waivers, the share of non-public assistance participants was not a bad 
proxy for the share of ABAWDs--their shares differed by only two to 
three percentage points. But in 2017 using the share of non-public 
assistance individuals as a proxy for ABAWDs would overstate the share 
in areas that were not waived--because many ABAWDs had been cut off in 
2017--and understate the number in waived areas. The share of ABAWDs in 
states with statewide waivers in 2017 was ten percentage points higher 
in 2017 than the share of non-public assistance SNAP participants (29 
percent vs. 19 percent.) Thus, the methodology in the RIA is likely to 
have substantially underestimated the share of ABAWDs living in waived 
areas in 2016, and projected to live in waived areas in 2020.

                                                   Table 11.4
 Waivers Affect the Number and Share of Non-PA SNAP Participants Fiscal Years 2013 and 2017 by whether the state
                                             had a statewide waiver
----------------------------------------------------------------------------------------------------------------
                                                   Non-PA          Share of
                                                Participants   National Non-PA    ABAWDs  (in    Share of ABAWDs
                                               (in millions)     Participants      millions)
----------------------------------------------------------------------------------------------------------------
                                                Fiscal Year 2013
----------------------------------------------------------------------------------------------------------------
States with statewide waivers in 2013 and no             10.9              28%              1.5              30%
 waivers in 2017
States with statewide waivers in both 2013                6.4              16%              1.0              19%
 and 2017
                                             -------------------------------------------------------------------
  Total all states                                       39.0             100%              4.9             100%
----------------------------------------------------------------------------------------------------------------
                                                Fiscal Year 2017
----------------------------------------------------------------------------------------------------------------
States with statewide waivers in 2013 and no              9.1              26%              0.6              19%
 waivers in 2017
States with statewide waivers in both 2013                6.6              19%              0.9              29%
 and 2017
                                             -------------------------------------------------------------------
  Total all states                                       35.2             100%              3.2             100%
----------------------------------------------------------------------------------------------------------------
Source: CBPP analysis of FY 2013 and FY 2017 USDA SNAP Household Characteristics data.
Note: ``Non-PA'' means not pure public assistance (PA) household. A household is considered to be pure PA if
  each member of the household receives Supplemental Security Income, a cash TANF benefit, or General Assistance
  income.

FNS' Methodology for Determining the Share of Areas That Would Lose 
        Eligibility for Waivers Is Incomplete and Confusing
    The RIA includes FNS' estimate of the number and share of currently 
waived areas that would no longer qualify for a waiver under the 
proposed rule (755 areas, representing 76 percent of areas currently 
waived), and the share of the impact that is attributable to each of 
the major proposed changes to waiver rules. It also provides an 
explanation of its methodology for deriving these estimates. However, 
the explanations are incomplete, confusing, and misleading. FNS omits 
fundamental information needed to assess the integrity of its analysis. 
For example, it bases the analysis on inconsistent periods of time, and 
provides unclear explanations of its methodological assumptions. 
Understanding which areas and how many areas would lose waivers under 
the proposal is central to understanding the impact of the proposed 
changes, but the analysis FNS included in the RIA has significant flaws 
and lacks sufficient explanation to allow commenters to understand the 
analysis.
    The proposed rule makes three major changes to the existing rules 
for determining waiver eligibility:

  1.  It sets a minimum unemployment rate of seven percent as a floor 
            for waiver eligibility. It makes areas with average 
            unemployment rates below this floor ineligible for a 
            waiver, even if their unemployment rates are 20 percent 
            above the national average unemployment rate. In contrast, 
            there is no floor under current Federal regulations. (See 
            Chapter 3 for more.)

  2.  It restricts states' flexibility to define combined areas, making 
            federally designated labor market areas the only 
            geographical groups that can be eligible for a waiver. In 
            contrast, under current regulations states have the 
            discretion to combine areas into larger geographic regions 
            that are eligible for a waiver if the regional unemployment 
            rates still meet the eligibility thresholds.\466\ (See 
            Chapter 5 for more.)
---------------------------------------------------------------------------
    \466\ USDA, ``Guide to Supporting Requests to Waive the Time Limit 
for Able-Bodied Adults without Dependents (ABAWD),'' December 2, 2016, 
page 10.

  3.  It narrows the allowable criteria for states to request statewide 
            waivers. Under current Federal regulations and FNS 
            guidance, states can request statewide waivers based on 
            average state-level unemployment rates that are 20 percent 
            above the national average over a recent 24 month period; 
            average statewide unemployment rates above ten percent for 
            a recent 3 month or 12 month period; or based on qualifying 
            as a state for extended unemployment benefits.\467\ In 
            contrast, the proposed rule permits to states to request 
            statewide waivers only when they qualify for extended 
            unemployment benefits. (See Chapter 4 for more.)
---------------------------------------------------------------------------
    \467\ 7 CFR  273.24(f)(2), and FNS guidance, ``Supporting Requests 
to Waive the Time Limit for Able-Bodied Adults Without Dependents,'' 
December 2016, https://fns-prod.azureedge.net/sites/default/files/snap/
SNAP-Guide-to-Supporting-Requests-to-Waive-the-Time-Limit-for-
ABAWDs.pdf.

    The problems with FNS' estimates fall into two main categories: 
first, the methodology is confusing and incomplete, and second, the 
discussion of the relative impact of the different changes on areas' 
---------------------------------------------------------------------------
eligibility for waivers is misleading.

    The Methodology Is Confusing and Incomplete

    Below are specific problems with the RIA's methodology that call 
into question the reliability of its estimated impact of the rule 
provisions on waived counties.

   FNS' use of the term ``currently'' is inconsistent; as a 
        result it is not clear what year FNS used for the analysis. 
        Under both current law and the proposed rule an area's 
        eligibility for a waiver for a given fiscal year is based on 
        whether the area's unemployment rate for a specific earlier 
        time period exceeds a threshold that applies to that same time 
        period. Throughout the RIA's discussion of the methodology for 
        determining the impact of the proposed changes, FNS uses the 
        term ``currently'' to refer to the year on which it is basing 
        its estimates. For example, on page 20, where FNS discusses its 
        estimates of the number of areas that would lose eligibility 
        for waivers under the proposed rule, FNS asserts that ``975 
        counties and county-equivalents currently have a time limit 
        waiver'' (emphasis added).\468\ Since FNS issued the NPRM on 
        February 1, 2019, the start of the fifth month of Fiscal Year 
        2019, it would be reasonable to assume that the term 
        ``currently'' refers to Fiscal Year 2019. However, nine pages 
        later in a discussion of ``uncertainties'' associated with all 
        the estimates in the RIA, FNS notes that ``these estimates are 
        based on current waiver eligibility as of FY 2018.'' \469\ 
        Moreover, as discussed below, it appears that the time period 
        FNS used for the data on local area unemployment rates is the 
        time period that applies to waiver eligibility for FY 2019.
---------------------------------------------------------------------------
    \468\ RIA, p. 20.
    \469\ RIA, p. 29.

      FNS' lack of clarity about the year for which it estimated the 
        change in waiver eligibility calls into question whether it is 
---------------------------------------------------------------------------
        assessing accurately the impact of the proposed changes.

   There are inconsistencies in FNS' methodology to estimate of 
        the number of counties that would lose waivers under the 
        proposed rule. The analysis reveals inconsistencies in the 
        methodology:

    1.  According to current and proposed waiver rules,\470\ the 
            calculation to deter-
              mine whether an area is eligible for a waiver for a given 
            fiscal year looks 
              at the area's 24 month average unemployment rate over a 
            defined 
              earlier time period and compares it to 20 percent above 
            the national aver-
              age for the same earlier 24 month period.\471\ States 
            have to compare 
              their areas' unemployment rates for a 24 month period to 
            an unemployment 
              threshold calculated over the same 24 month period. In 
            addition, states 
              need to use a 24 month time period that falls within an 
            earlier window that 
              is consistent with the year for which the waiver will be 
            implemented.\472\ 
              Contrary to its own guidance, FNS fails to use consistent 
            periods in its 
              analysis:
---------------------------------------------------------------------------
    \470\ 7 CFR  273.24(f)(2)(ii).
    \471\ The proposed rule would add the additional condition that the 
area's unemployment rate under this calculation be at least seven 
percent.
    \472\ USDA, ``Guide to Supporting Requests to Waive the Time Limit 
for Able-Bodied Adults without Dependents (ABAWD), December 2, 2016,'' 
page 7.

        a.  The 24 month period FNS says it used for the unemployment 
            data 
                  is inconsistent with the year FNS says it calculated 
            the number of 
                  waived areas. On page 21, FNS notes that it obtained 
            data ``for the 24 
                  month period from January 2016 to December 2017 for 
            3,077 counties 
                  and county-equivalents.'' As mentioned, elsewhere the 
            RIA asserts that 
                  it based its estimates on eligibility for waivers in 
            2018, but the 24 
                  month period used for the unemployment data is not 
            the correct period 
                  for a waiver implemented in 2018. The January 2016 
            through Decem-
                  ber 2017 period that the Department used applies to 
            waivers imple-
                  mented in 2019. The correct period of unemployment 
            data that applied 
                  to areas eligible for 2018 waivers is January 2015 
            through December 
                  2016. If FNS is estimating eligibility for waivers in 
            2018, it should have 
                  used unemployment data that is applicable to that 
---------------------------------------------------------------------------
            year.

        b.  The Department notes on page 21 that it used unemployment 
            data ``to 
                  identify currently-waived counties [that] did not 
            have a 24 month [un-
                  employment rate] that exceeds the current waiver 
            threshold.'' This 
                  threshold is calculated as 20 percent above the 
            national average unem-
                  ployment rate for a 24 month period. FNS did not 
            specify which 24 
                  month period it used to calculate this ``current'' 
            threshold. This creates 
                  additional confusion, given that it is already 
            unclear which year the 
                  Department is using for waived counties and the 
            potential inconsist-
                  ency with the period for which the unemployment data 
            were collected. 
                  This also matters because the thresholds are 
            different depending on the 
                  waiver year FNS is analyzing. For the 24 month period 
            of January 
                  2016 through December 2017, which corresponds to a 
            2019 waiver, the 
                  threshold is 5.5 percent. For the 24 month period of 
            January 2015 
                  through December 2016, which corresponds to a 2018 
            waiver, the 
                  threshold is higher at 6.1 percent. If FNS is 
            comparing county unem-
                  ployment data that apply to a 2019 waiver to the 
            threshold for a 2018 
                  waiver, instead of the lower threshold for a 2019 
            waiver, then it is po-
                  tentially underestimating the number of waived 
            counties that would 
                  lose eligibility as a result of restricting area 
            combinations. Although as 
                  mentioned above, it should have been using data for a 
            2018 waiver if 
                  it is evaluating eligibility for 2018.

    2.  FNS only collected unemployment data for a single 24 month 
            period, but 
              states are allowed to use any 24 month period that is 
            later than the 24 
              month period FNS used. FNS' estimate assumes that all 
            states would use 
              the same period of data as the basis of their requests, 
            and that this period 
              is representative of the other periods of data that 
            states could use. This is 
              unlikely to be the case as unemployment trends change 
            over the course of 
              a year, and the first 24 month period is unlikely to 
            accurately represent 
              unemployment conditions in other periods that states 
            could use for a waiver 
              request. In addition, the decline in unemployment rates 
            in recent years 
              means that the threshold for eligibility in subsequent 24 
            month periods 
              generally decreased over the course of 2018. As a result, 
            the restriction 
              of area combinations would result in fewer waived 
            counties losing eligibility 
              than would be estimated under a single 24 month period. 
            The Department's 
              omission of multiple periods means that it is potentially 
            overestimating the 
              number of waived counties that would lose eligibility as 
            a result of restrict-
              ing area combinations.

    3.  Both inconsistencies create opposite biases, the net effect of 
            which FNS 
              could demonstrate if it had taken into them into account. 
            The fact that it 
              ignores these factors and does not provide a rationale 
            for doing so shows 
              the serious lack of rigor in its analysis.

   FNS fails to adequately explain its exclusion of certain 
        areas from its analysis. Footnote 8 on page 21 indicates that 
        FNS excluded five New England states (Connecticut, Maine, 
        Massachusetts, New Hampshire, and Vermont) when it compiled the 
        unemployment data from the Bureau of Labor Statistics (BLS), 
        but FNS does not indicate if it also excluded these states from 
        the list of areas that it is counting as currently waived. FNS 
        also does not mention these states in the rest of its analysis. 
        As a result of these exclusions, its estimate of the number of 
        currently waived areas is too low.

      The same footnote further explains that FNS excluded these five 
        states because New England counties (also known as NECTAs) are 
        conceptually dissimilar to counties in the rest of the United 
        States. However, Rhode Island, which contains the Providence-
        Warwick, RI-MA NECTA, does not appear on the list of excluded 
        states. The Department fails to mention why it included Rhode 
        Island, which shares the same New England dissimilarities with 
        the other states.\473\ It briefly notes that ``some NECTAs are 
        quite small'' and ``BLS data was not consistently available for 
        these areas,'' which appears to be a reference to the BLS' 
        discontinuation in 2018 of unemployment data for all cities and 
        towns with populations below 1,000 for all New England 
        states.\474\ As Rhode Island does not contain towns with 
        populations below 1,000, BLS data would be available for all 
        areas. If this were the reason for its inclusion in the FNS 
        analysis, this would be consistent with its rationale. But FNS 
        provides no information to help understand its rationale.
---------------------------------------------------------------------------
    \473\ U.S. Census Bureau, ``New England City and Town Areas (NECTA) 
Maps,'' https://www.census.gov/geo/maps-data/maps/nectas.html.
    \474\ U.S. Bureau of Labor Statistics, ``Local Area Unemployment 
Statistics,'' https://www.bls.gov/lau/laugeo.htm.
---------------------------------------------------------------------------
      In addition, the Department does not explain how it treats Guam 
        and the Virgin Islands in its analysis. Although the Bureau of 
        Labor Statistics does not produce employment data for these 
        U.S. territories, these areas were also waived in 2018 but it 
        is unclear if they are included in the number of currently 
        waived areas or in the number that would lose waivers under the 
        proposed rule.

   FNS fails to adjust the number of waived areas for 2020, the 
        year in which the proposed rule would be in effect if 
        implemented. FNS unrealistically assumes no changes in waivers 
        in future years. It fails to adjust for the fact that the 
        number of areas is likely to be lower in FY 2020, the year in 
        which the proposed rule would first apply if implemented. As 
        unemployment rates have declined, states have applied and 
        qualified for fewer areas in 2017, 2018, and 2019. It would be 
        realistic to assume a decline in waived areas in 2020 and later 
        years as well.

   FNS only examined the impact of the proposal in a year when 
        unemployment rates declined. FNS only examined the impact of 
        its proposal in 2018, a year in which the unemployment rates 
        declined. It did not analyze the proposal's impact during a 
        time of rising unemployment rates, such as prior to or during a 
        recession. FNS did not offer any rationale for this exclusion. 
        Expanding its analysis to many periods with rising and 
        decreasing unemployment trends would have provided a fuller 
        understanding of the impact of the proposed rule in different 
        economic conditions.

   FNS does not explain its estimation of the number of areas 
        losing eligibility due to the narrowing of statewide waivers. 
        FNS provided no details about its methodology for estimating 
        the effect of narrowing the criteria that can be used for 
        statewide waivers, beyond noting that it ``estimated the number 
        of counties and county-equivalents that would lose waiver 
        eligibility due to the elimination of statewide waivers.'' 
        \475\ It identified an additional 39 counties eligible only 
        because of a statewide waiver and subtracted those from its 
        total of waived areas that would still be eligible under the 
        rule after already eliminating the areas that were eligible 
        only based on states' ability to combine data. Based on the 
        explanations in the analysis, it is unclear why FNS needed to 
        subtract out these 39 counties since all of them would already 
        be ineligible for waivers under the rule because their data 
        could not be combined with the data of other areas in the 
        state. The description of the methodology is confusing. It is 
        possible that FNS was estimating the impact of the narrowing of 
        statewide waivers separately from the impact of combining data 
        or the seven percent threshold. Table 3 on page 22 of the RIA 
        presents the results as if FNS included the narrowing of 
        statewide waivers as one step within the estimation. If that is 
        not the case, then the methodology is poorly explained.
---------------------------------------------------------------------------
    \475\ RIA, p. 21.

      On the other hand, if the narrowing of statewide waivers is a 
        step in its overall estimation, then it appears that FNS 
        subtracted the same counties twice. The elimination of counties 
        that did not meet the waiver threshold (described in the second 
        paragraph on page 21) would already have removed counties that 
        are ineligible based on their own unemployment rates. This 
        would not leave any counties that are eligible based only on 
        being in a state with a statewide waiver. This additional 
        subtraction would inflate FNS' estimate of counties that would 
---------------------------------------------------------------------------
        lose their waivers due to narrowing of statewide waivers.

   FNS ignores extended benefits as a standard for qualifying 
        for statewide waivers. FNS does not mention that some states 
        would remain eligible for statewide waivers under the proposed 
        rule, based on qualifying for extended unemployment benefits 
        (EB). Under current FNS guidance on qualifying for a waiver 
        based on qualifying for EB, which the proposed rule does not 
        change, states can request statewide waivers that start no 
        later than 1 year after the date that the state qualified for 
        EB.\476\ Alaska and the District of Columbia qualified for EB 
        in January 2018, and therefore would have been able to request 
        statewide waivers in 2018 (and 2019). For either year these two 
        areas count as an additional nine county-equivalents. FNS 
        omission of this factor also inflates the estimate of counties 
        that would lose their waiver under the proposed rule because 
        these two areas would not have lost their waivers.
---------------------------------------------------------------------------
    \476\ USDA, ``Guide to Supporting Requests to Waive the Time Limit 
for Able-Bodied Adults without Dependents (ABAWD),'' December 2, 2016, 
page 3.

   FNS does not analyze the impact of its rules on Native 
        American reservations or on New England towns. FNS does not 
        provide any analysis of the impact of the proposed rule on 
        Native American reservations or on New England towns and 
        cities. This is a glaring omission, because Native American 
        reservations tend to have high poverty rates well above the 
        national average, and over 200 reservations were waived or were 
        located inside a waived area in 2018. This is an important 
---------------------------------------------------------------------------
        segment of the population that FNS analysis ignores completely.

      Similarly, 281 towns and cities in the New England states of 
        Connecticut, Massachusetts, New Hampshire, Vermont, and Rhode 
        Island were waived in 2018, constituting nearly a quarter of 
        all towns in these states. Although the Bureau of Labor 
        Statistics has stopped publishing unemployment data for New 
        England towns with populations below 1,000 people, it continues 
        to provide data for towns above 1,000 people. It is unclear why 
        the Department does not examine the rule's impact on those 
        towns with higher populations.

    The methodological problems listed above cast serious doubt on the 
reliability of the FNS overall estimate of the impact of the proposed 
rule on ``currently waived'' areas.

    FNS' Estimates of the Relative Impacts of the Rule Provisions on 
Waived Areas Is Misleading

    In Table 3 on page 22 of the RIA, FNS indicates that the three 
different changes in the proposed rule result in a cumulative 76 
percent reduction in the number of waived areas. FNS then describes the 
relative impact of the three different changes in the proposed rule, 
asserting that the change to restrict area combinations reduces the 
number of areas waived by 36 percentage points, the change to statewide 
waivers reduces the number by four percentage points, and the seven 
percent unemployment rate floor reduces the number by 37 percentage 
points. However, the order FNS uses for these calculations presents 
misleading results.
    FNS' presentation suggests that the proposed change to restrict 
area combinations and the seven percent floor have a roughly equal 
impact. The presentation is misleading, however, because, although 
restricting the ability to combine areas would have a substantial 
impact by itself relative to current rules, the seven percent floor has 
a far larger effect on areas' eligibility for waivers because seven 
percent is substantially higher than the national average unemployment 
threshold. As a result, all of the areas that would lose because they 
cannot be combined with adjacent areas would also lose under a seven 
percent floor. An example using FNS' own numbers (despite their flaws) 
can be helpful to understand why:
    Under FNS' estimates, out of 975 counties currently waived, 220 
have ``a 24 month [unemployment average] of at least seven percent and 
would continue to qualify for a waiver under the proposed waiver 
criteria.'' \477\ That means 755 waived counties, or 76 percent of all 
counties currently waived would lose their waiver due solely to the 
seven percent floor. The additional impact of restricting area 
combinations would be zero at that point.
---------------------------------------------------------------------------
    \477\ RIA, p. 22.
---------------------------------------------------------------------------
    This occurs because seven percent is higher than the 20 percent 
above the national 24 month average threshold (which would be 6.1 
percent for 2018 or 5.5 percent for 2019, though as discussed above, 
it's not clear which year FNS used for the analysis.) Implementing a 
seven percent minimum unemployment rate automatically eliminates any 
waived counties with rates below 5.5 and 6.1 percent, and the waived 
counties with rates above 5.5 and 6.1 percent but below seven percent. 
The impact of the floor is therefore greater, as it eliminates counties 
that are eligible based on their own 24 month average unemployment 
rates but do not meet the floor, in addition to the counties waived 
through area combinations.
The RIA Assumes That All ABAWDs in Areas That Lose Waivers Will Lose 
        SNAP; An Assumption That Ignores That Many Will Be Exempt or 
        Able to Participate for Other Reasons
    As mentioned above, in determining the impact of the change in 
waivers on SNAP participants, the RIA assumes that 1.5 million ABAWDs 
would live in areas that would be waived under current rules. Under the 
proposed changes, the RIA assumes that:

          Because waived areas are estimated to be reduced by 76 
        percent under the revised waiver criteria, the department 
        assumes that 76 percent of currently-waived ABAWDs would be 
        newly subject to the time limit. This equals approximately 1.1 
        million of the estimated 1.5 million currently waived 
        individuals.\478\
---------------------------------------------------------------------------
    \478\ RIA, p. 25.

    Under this assumption, no individuals who are defined as ``ABAWDs'' 
(using the Department's definition) in the areas that lose waivers 
---------------------------------------------------------------------------
would be:

   exempted from the time limit because of being physically or 
        mentally unfit;

   pregnant;

   participating because they are eligible during the first 3 
        months of participation or qualify for a second 3 month period;

   exempt using individual ``percentage'' exemptions;

   participating in a qualifying work program; or

   working (or complying with a qualifying work program), for 
        less than the required 20 hours per week.

    This assumption defies logic and ignores the evidence from other 
states that have a time limit in effect. In every state without waivers 
there are ``ABAWDs'' who are able to continue to participate for these 
reasons. As we showed above, in Tables 11.1 and 11.2, in the states 
that reinstated the time limit by 2017, the number of ``ABAWDs'' 
declined substantially, from 1.5 million in 2013 to 600,000 in 2017, 
but even if we subtract out the number of ``ABAWDs'' who were working 
at least 20 hours a week (138,000), that still leaves almost \1/2\ 
million ``ABAWDs'' participating in SNAP in 2017 after the time limit 
went back into effect.
    The assumption that none would continue to participate for any of 
these reasons is a glaring error in the RIA methodology. As a result, 
the public is left not knowing how many people who are potentially 
subject to the time limit will be able to continue to participate and 
for which reasons, and whether the Department knows or cares.
The RIA's Assumption That One-Third (34 Percent) of ABAWDs Subject to 
        the Time Limit in Areas That Lose Waivers Will Remain Eligible 
        Because They Would Be Working 20 Hours a Week Is Flawed
    Of the 1.1 million ``ABAWDs'' the Department estimates would be 
newly subject to the time limit under the proposed rule, the RIA 
assumes \1/3\ would be working and \2/3\ would ``lose their eligibility 
for SNAP for failure to engage meaningfully in work or work training:''

          Using FY 2016 QC data, approximately 26 percent of ABAWDs 
        were working. The Department assumes that this proportion would 
        increase to about 34 percent in FY 2020 if the UR [unemployment 
        rate] declines as projected in the 2019 President's Budget and 
        that these individuals will work at least 20 hours per week. 
        Under this scenario, the Department estimates that 
        approximately \1/3\ of ABAWDs newly subject to the time limit 
        will work and maintain their SNAP eligibility. The remaining 
        \2/3\ (755,000 individuals) would lose their eligibility for 
        SNAP for failure to engage meaningfully in work or work 
        training.\479\
---------------------------------------------------------------------------
    \479\ RIA, p. 26.

---------------------------------------------------------------------------
    This assumption is flawed for several reasons:

   First, the assumption being used (the 26 percent working in 
        2016, rising to 34 percent in 2020) was derived from the entire 
        SNAP population of ``ABAWDs'' nationally, including both areas 
        that currently are waived and those that are not waived. As we 
        demonstrated above, the work rates of SNAP participants who 
        live in areas without waivers are higher simply because many 
        people who are not able to find jobs or document their work 
        have been cut off. It is confusing to apply a percentage 
        derived from the entire caseload, where many states have a 
        large share of individuals who live in area without waivers, to 
        areas with waivers.

   Second, in 2017 only 12 percent of ``ABAWDs'' in states that 
        had statewide waivers in 2017 were working at least 20 hours a 
        week--the threshold for individuals subject to the time limit 
        to remain eligible for SNAP. The RIA assumes a share almost 
        triple that (34 percent) would be able to meet the 20 hours a 
        week threshold, with no explanation for why or how that would 
        occur.
The RIA's Assumes That All Individuals Who Lose Eligibility Will 
        Reapply Every Three Years
    When estimating the effect of the change on Federal spending the 
RIA assumes the impact will be felt over only 9 months in 2020 and 
2023, presumably based on an assumption that people who have been cut 
off because of the 3 month time limit will reapply immediately when 
they become eligible again after 3 years, but it provides no evidence 
that this occurs. To the contrary, a study FNS published in 2001 of 
state implementation of the time limit found:

          Many ABAWDs who left the program have not returned. ABAWDs 
        who used up their time-limited benefits in 1997 became eligible 
        again in 2000, creating the potential for a sharp upswing in 
        participation, yet the trend in participation shows no such 
        change.\480\ (Bold emphasis in original.)
---------------------------------------------------------------------------
    \480\ John L. Czajka, et al., ``Imposing a Time Limit on Food Stamp 
Receipt: Implementation of the Provisions and Effects on Food Stamp 
Program Participation: Final Report,'' prepared by Mathematica Policy 
Research for the USDA, Food and Nutrition Service, September 4, 2001, 
https://fns-prod.azureedge.net/sites/default/files/abawd.pdf, p. xxv.
---------------------------------------------------------------------------
G. The RIA Lacks Transparency About the Reasons Individuals May Lose 
        SNAP and Other Possible Impacts If Waivers Are Narrowed
    As noted, the proposed rule's fundamental rationale is that taking 
away (or threatening to take away) food assistance will cause people 
not currently working to get jobs. The NPRM asserts that ``these 
changes would encourage more ABAWDs to engage in work or work 
activities if they wish to continue to receive SNAP benefits,'' \481\ 
and, ``[t]he application of waivers on a more limited basis would 
encourage more ABAWDs to take steps towards self-sufficiency.'' \482\
---------------------------------------------------------------------------
    \481\ NPRM, p. 981.
    \482\ NPRM, p. 981.
---------------------------------------------------------------------------
    However, the only impact that FNS quantifies in the RIA comes from 
the estimated, ``755,000 individuals . . . [who] would not meet the 
requirements for failure to engage meaningfully in work or work 
training.'' The rationale and the estimated impact are inconsistent. 
Moreover, the assessment lacks transparency about what the impact would 
be and it over-simplifies the possibilities. Other FNS materials 
indicate that FNS is aware that when the time limit is in effect it 
results in different outcomes for different groups of people. Many 
people participate in SNAP for a period of time and then leave when 
their circumstances change. But, when faced with a time limit there are 
a range of possible outcomes and impacts. For example:

   Some individuals may find work or additional hours and as a 
        result their SNAP benefits may go down as a result of income 
        they would not have otherwise had.

   Others may wish to comply but be unable to find a job. For 
        these individuals there is a question of whether a qualifying 
        slot in an Employment and Training program would be available. 
        If there is no slot then that individual would likely lose 
        SNAP. If there is, FNS and states would potentially incur a 
        cost for the E&T services. FNS does not contemplate any changes 
        in E&T that might be caused by the proposed rule.

   Other individuals may qualify for an exemption for 
        ``unfitness for work,'' or another reason, but may lose SNAP if 
        they don't realize they could qualify for an exemption because 
        it was not properly explained to them, or if they are unable to 
        get documentation of their health issue because they lack 
        medical coverage.

   Others may actually be working, but not comply with 
        paperwork requirements to document their hours of work.

   And, although unlikely, other individuals who are able to 
        work may intentionally choose not to comply with the time limit 
        and lose SNAP benefits.

    This list is not exhaustive. But FNS guidance in recent years makes 
clear that it is aware of both the range of possible outcomes and the 
fact that the distribution of outcomes can be influenced by state 
implementation choices. For example, in November 2015, FNS issued 
guidance that reminded states that in addition to tracking months of 
participation, ``States must also carefully screen for exemption from 
the time limit and connect ABAWDs to the information and resources 
necessary to maintain eligibility consistent with Federal 
requirements.'' \483\ The guidance covered several areas where state 
implementation could affect individuals' eligibility, for example:
---------------------------------------------------------------------------
    \483\ U.S. Department of Agriculture, Food and Nutrition Service, 
``Supplemental Nutrition Assistance Program--ABAWD Time Limit Policy 
and Program Access,'' November 19, 2015, https://fns-
prod.azureedge.net/sites/default/files/snap/ABAWD-Time-Limit-Policy-
and-Program-Access-Memo-Nov2015.pdf.

---------------------------------------------------------------------------
   ``Screening for Exemptions and Fitness for Work'';

   ``Maintaining Eligibility through Work Programs and 
        Workfare'':

   ``Maintaining Eligibility through Unpaid or Volunteer 
        Work''; and

   ``Good Cause for Failure to Meet the ABAWD Work 
        Requirement.''

    Another FNS guidance focused on states' responsibilities to 
adequately notify individuals who are potential ABAWDs on the details 
of the time limit, work requirement, exemptions, and their 
responsibility to report changes in work hours.\484\ FNS followed this 
with another guidance that outlined best practices and provided model 
language to ``help state agencies effectively inform Supplemental 
Nutrition Assistance Program (SNAP) households of the requirements for 
able-bodied adults without dependents (ABAWD) and to enrich training 
for eligibility workers.'' \485\
---------------------------------------------------------------------------
    \484\ U.S. Department of Agriculture, Food and Nutrition Service, 
``SNAP--Requirements for Informing Households of ABAWD Rules,'' April 
17, 2017, https://fns-prod.azureedge.net/sites/default/files/snap/
Requirements_for_Informing_Households_of_ABAWD_Rules.pdf.
    \485\ U.S. Department of Agriculture, Food and Nutrition Service, 
``SNAP--Best Practices and Resources for Informing Households of ABAWD 
Rules,'' May 25, 2018, https://fns-prod.azureedge.net/sites/default/
files/snap/BestPracticesforInformingABAWDS.pdf.
---------------------------------------------------------------------------
    The RIA oversimplifies the various impacts of the rule. Research 
has found, and FNS is aware, that in practice work requirements result 
in individuals experiencing benefit cuts for a variety of reasons, 
including when they cannot find jobs, when they should have been found 
exempt from the requirement, and when they are working but fail to 
comply with paperwork requirements. FNS failed to adequately reflect 
the various possible reasons why individuals would lose SNAP under the 
proposed rule and as a result failed to adequately explain or consider 
its impact.
H. The RIA's Estimate of No Impact From Eliminating the Carryover of 
        Exemptions Is Confusing and Misleading
    The proposed rule would eliminate states' ability to ``carry over'' 
exemptions that go unused in 1 year into future years. The RIA includes 
two confusing and misleading assumptions about state use of 
exemptions--one about current state use of exemptions and the other 
about how states would use exemptions under the proposed rule.
    The RIA dramatically overstates the number of exemptions states 
have used in recent years. The RIA methodology includes an assumption 
that ``states use approximately 65 percent of their earned exemptions 
in an average year.'' This assumption implies that many states use a 
large share of their annual exemptions each year. This is a significant 
misrepresentation of the pattern of state use of exemptions.
    A closer look at the data FNS posts about the pattern of state use 
and accrual of exemptions \486\ shows that the actual pattern is that:
---------------------------------------------------------------------------
    \486\ U.S. Department of Agriculture, ABAWD 15 Percent Exemptions 
Data, https://www.fns.usda.gov/snap/abawd-15-percent-exemptions.

---------------------------------------------------------------------------
   many states have not used any exemptions in most years;

   some states have used a small share of the exemptions they 
        earned for that year;

   a few states have used the majority of the exemptions they 
        earned for that year;

   a few states have not earned exemptions for a year but have 
        dipped into their accrued exemptions; and

   a handful of states have used more exemptions than they 
        earned in a given year.

    The last two categories result in the number of used exemptions as 
a share of earned exemptions for that year exceeding 100 percent in 
that state. Across all states this will raise the total share of 
exemptions used because the denominator for the percentage is the 
number of exemptions earned for that year (as it appears to be for the 
FNS assumption.) See Table 11.5, below, for the distribution of states 
across these categories. You can see that the small number of states in 
the last two categories is playing an outsized role in raising the 
total share across all the states.

                               Table 11.5
                    State use of exemptions 2014-2018
------------------------------------------------------------------------
 Number
   of        2014         2015         2016         2017        2018 a
 States
------------------------------------------------------------------------
Using             43           41           21           19          N/A
 no
 exempt
 ions
Using              4            7           15           26          N/A
 less
 than
 50% of
 earned
 exempt
 ions
Using              1            2            0            5          N/A
 51-100
 % of
 earned
 exempt
 ions
Using              2            0            2            1          N/A
 more
 than
 100%
 of
 earned
 exempt
 ions
Used               3            3           15            2          N/A
 exempt
 ions
 but
 earned
 none
 for
 that
 year
Average          93%          27%         149%          23%          N/A
 exempt
 ions
 used
 as a
 share
 of
 earned
        ----------------------------------------------------------------
  Total      230,000      115,000      730,000      300,000    1,300,000
   exem
   ptio
   ns
   used
------------------------------------------------------------------------
Source: CBPP analysis of FNS-posted data on exemptions https://
  www.fns.usda.gov/snap/abawd-15-percent-exemptions.
a According to the RIA (p. 23), ``In FY 2019 state carried over
  approximately 6.1 million unused exemptions from the prior year.''
  Since, according to FNS data, states carried 7.4 million exemptions
  into FY2018 they must have used about 1.3 million exemptions in FY
  2018 (7.4 million^1.3 million = 6.1 million.)

    Thus, FNS' statement that ``States use approximately 65 percent of 
their earned exemptions in an average year'' is highly misleading. In 
fact, across all states and the 4 years between 2014 and 2017 only 
eight states used between 51 and 100 percent of their exemptions. The 
vast majority used none, some used less than 50 percent, and a handful 
used more than 100 percent.
    FNS' assumption that eliminating the exemption carryover would have 
no impact is indefensible. FNS estimates the number of exemptions that 
would be taken away from states and concludes that the rule would 
eliminate 6.6 million case-months of carryover exemptions the first 
year (FY 2020) and 160,000 to 180,000 a year in later years.
    FNS estimates no impact from the proposed change to eliminate 
states' past and future exemptions from prior years, saying:

          It is difficult to estimate the impact of such a change on 
        transfer spending because there is no historical record to 
        support an estimate of if and when such a ``run on the bank'' 
        may occur. Current practice by the states suggests that 
        elimination of the carryover will have no change on transfers 
        as the exemptions that will expire represent exemptions that 
        were not distributed to covered individuals (i.e., no transfer 
        is occurring, so no transfer can be reduced.) However, 
        elimination of the carryover will give the Federal Government 
        greater predictability over potential spending requirements 
        because the number of exemptions subject to the sole discretion 
        of the states is smaller.\487\
---------------------------------------------------------------------------
    \487\ RIA, p. 24.

    The assumption of no change in Federal spending from eliminating so 
many exemptions is highly misleading and contrary to the experience of 
the last 23 years. Virtually every state has used waivers at some point 
since 1996, and most states have used exemptions in some years, making 
clear that in some economic situations and under some political 
leadership states wish to shield some SNAP recipients who are subject 
to the time limit from losing SNAP.
    Some states have used exemptions to suspend the time limit in areas 
where no E&T services are available or to transition counties from 
waiver status to non-waiver status and give the area time to establish 
or expand employment and training opportunities to meet the needs of 
individuals subject to the time limit. Other states have used 
exemptions to continue to provide SNAP to certain SNAP participants who 
would be cut off because of the time limit but who they determine 
should continue to receive SNAP, such as individuals who are working, 
but less than 20 hours in a particular month, or individuals who 
recently have been released from incarceration.
    The information that FNS makes public about exemption usage in 
recent years shows that as many areas have no longer been waived in 
recent years, states have increased their use of exemptions. It seems 
highly likely that if the rule went into effect and states faced losing 
waivers for a large share of the counties in their state with the 
highest unemployment rates, many would opt to draw down more 
exemptions, and, over time, to draw down the balances of their 
exemptions that they have been allowed to carry over.
    Another example of when states may use exemptions is when the 
political leadership in a state changes to be more sympathetic to the 
harshness of the time limit. In that case states might establish a 
policy that begins to use exemptions and use them at a higher rate than 
the number that are accrued each year. The carryover exemptions would 
allow such a state to sustain a larger exemption policy for several 
years.
    The assumption that no carried over exemptions would ever be used 
is indefensible, especially in combination with the changes the 
proposed rule would make to the share of the United States that could 
qualify for waiver.
I. The RIA Fails to Accurately Reflect the Impact of the Proposed Rule 
        on Medicaid and Health Coverage and Other Secondary Impacts
    The Centers for Medicare and Medicaid Services of the Department of 
Health and Human Services, the agency that administers the Medicaid 
program, has required some alignment between SNAP and Medicaid work 
requirements. Specifically, states must count enrollment in SNAP as an 
automatic exemption from Medicaid work requirements since individuals 
enrolled in SNAP are either exempt from or complying with SNAP work 
requirements.\488\
---------------------------------------------------------------------------
    \488\ State Medicaid Director Letter, January 11, 2018, https://
www.medicaid.gov/federal-policy-guidance/downloads/smd18002.pdf.
---------------------------------------------------------------------------
    As a result, the proposed rule's changes to SNAP waiver and 
exemption policy would have a direct ripple effect on individuals' 
Medicaid eligibility and coverage. More people in states with Medicaid 
work requirements would be subject to those work requirements, and a 
large number would very likely lose Medicaid coverage. The per-person 
cost of health coverage often is higher than the monthly SNAP benefit. 
The Federal budget savings and the impact on individual's health 
coverage from this direct link between SNAP and Medicaid under the 
Administration's policies should have been reflected in the RIA's cost-
benefit analysis. The RIA's failure to mention and quantify these 
effects is a serious oversight that fails to accurately reflect the 
full impact of the proposed rule.
    Moreover, the RIA does not mention nor quantify several secondary 
effects that SNAP benefit cuts could have. For example:

   SNAP benefits are one of the fastest, most effective forms 
        of economic stimulus when the economy is weak. Low-income 
        individuals generally spend all of their income meeting daily 
        needs such as shelter, food, and transportation, so every 
        dollar in SNAP that a low-income family receives enables the 
        family to spend an additional dollar on food or other items. 
        Moody's Analytics estimated that every $1 increase in SNAP 
        benefits during 2009, when the economy was in a recession, 
        generated about $1.70 in economic activity.

   SNAP has been found to improve some recipients' health 
        outcomes. SNAP is associated with better health and lower 
        health care costs, according to a growing body of 
        evidence.\489\
---------------------------------------------------------------------------
    \489\ Steven Carlson and Brynne Keith-Jennings, ``SNAP is Linked 
with Improved Nutritional Outcomes and Loer Health Care Costs,'' Center 
on Budget and Policy Priorities, January 17, 2018, https://
www.cbpp.org/research/food-assistance/snap-is-linked-with-improved-
nutritional-outcomes-and-lower-health-care.
---------------------------------------------------------------------------
Chapter 12. The Proposed Rule Would Disproportionately Impact 
        Individuals Protected by Civil Rights Laws, Violating the Food 
        and Nutrition Act's Civil Rights Protections
    According to FNS estimates, under the proposed rule some 755,000 
individuals would lose eligibility for SNAP because of a ``failure to 
engage meaningfully in work or work training.'' \490\ As described in 
detail in Chapter 3, evidence from the research on the impact of work 
requirements and time limits, as well as the disparities in 
unemployment in the labor market make clear that the cuts to SNAP 
eligibility from the proposed rule would fall disproportionately on 
African Americans, Latinos, and people with disabilities. In addition, 
Native Americans also would experience a disproportionate impact from 
the proposed rule because individuals who are Native American, whether 
or not they reside on Indian reservations, also have poverty and 
unemployment rates well above the national average, and many of the 
over 200 reservations that were waived or were located inside a waived 
area in 2018 would likely lose their waivers under the proposed rule.
---------------------------------------------------------------------------
    \490\ NPRM, p. 989.
---------------------------------------------------------------------------
    In the civil rights impact analysis included in the NPRM, FNS 
recognizes the disproportionate impact, citing the rule's ``potential 
for disparately impacting certain protected groups due to factors 
affecting rates of employment of members of these groups.'' \491\ But 
the analysis finds that ``the implementation of mitigation strategies 
and monitoring by the Civil Rights Division of FNS will lessen these 
impacts,'' without providing any evidence or examples of how that 
mitigation could occur. It is not clear how the Civil Rights Division 
of FNS could mitigate an eligibility policy that inherently results in 
a disproportionate impact on certain groups. We cannot comment on the 
potential effectiveness of such efforts when the NPRM does not provide 
any information about what they might be and no similar interventions 
have occurred in the history of the program. If FNS envisions giving 
the Civil Rights Division a role in determining eligibility for 
waivers--which the Division apparently has not had to date--it says 
nothing about that in the NPRM and we cannot readily imagine how that 
would work. Even if it did, without anything in the rule varying the 
effects of the new standards it imposes, states would be unlikely to 
request the kinds of waivers that might mitigate the rule's disparate 
impact on members of protected groups.
---------------------------------------------------------------------------
    \491\ NPRM, p. 990.
---------------------------------------------------------------------------
    Moreover, even if mitigation of the disparate impact were possible, 
the fact that the proposed rule still would have a disproportionate 
impact on these protected groups directly violates Section 11(c)(2) of 
the Food and Nutrition Act (7 U.S.C.  2020(c)(2)). In the 2008 Farm 
Bill Congress reasserted its commitment to nondiscrimination and made 
clear that certain civil rights laws apply to SNAP: \492\
---------------------------------------------------------------------------
    \492\ The Food, Conservation, and Energy Act of 2008 (P.L. 110-
246), section 4117.

---------------------------------------------------------------------------
          (c) Civil Rights Compliance.--

                  (1) In general.--In the certification of applicant 
                households for the supplemental nutrition assistance 
                program, there shall be no discrimination by reason of 
                race, sex, religious creed, national origin, or 
                political affiliation.
                  (2) Relation to other laws.--The administration of 
                the program by a state agency shall be consistent with 
                the rights of households under the following laws 
                (including implementing regulations):

                          (A) The Age Discrimination Act of 1975 (42 
                        U.S.C. 6101 et seq.).
                          (B) Section 504 of the Rehabilitation Act of 
                        1973 (29 U.S.C. 794).
                          (C) The Americans with Disabilities Act of 
                        1990 (42 U.S.C. 12101 et seq.).
                          (D) Title VI of the Civil Rights Act of 1964 
                        (42 U.S.C. 2000d et seq.).\493\
---------------------------------------------------------------------------
    \493\ Section 11(c) of the Food and Nutrition Act of 2008, 7 U.S.C. 
 2020(c).

    Of particular note is that, under this amended language, the 
regulations implementing Title VI and other civil rights statutes are 
fully applicable to SNAP. These regulations prohibit actions in Federal 
programs that have disparate impacts on members of protected groups as 
well as intentional discriminatory acts. Therefore, the proposed rules' 
disparate impact on these individuals, as demonstrated by the research 
and conceded in the NPRM itself, is inconsistent with the Act. Key 
Members of Congress made unmistakably clear that this is what the 2008 
amendments sought to accomplish.
    In his floor statement on the 2008 Farm Bill, Representative Joe 
Baca, who at the time was the Ranking Member of the House Agriculture 
Subcommittee on Departmental Operations, Oversight, Nutrition and 
Forestry, explained:

          . . . this legislation makes clear that the [Agriculture] 
        Department's civil rights regulations are among those which 
        have the full force of law and which households have the right 
        to enforce. Discrimination is not acceptable in any form or at 
        any point in the food stamp certification process. Households 
        should not be assisted, or not assisted, approved or denied for 
        any reason other than an individual assessment of their need 
        for help or their eligibility by the state.\494\
---------------------------------------------------------------------------
    \494\ Congressional Record, May 22, 2008, p. H3814, https://
www.govinfo.gov/content/pkg/CREC-2008-05-14/pdf/CREC-2008-05-14-pt1-
PgH3801-3.pdf#page=13.

    Senator Dick Durbin, a leading Member of the Senate Judiciary 
Committee, in his floor statement on the 2008 Farm Bill similarly 
stated that ``This legislation also makes explicit that various civil 
rights laws are binding in the Food Stamp Program. This is not a 
change--these laws and their regulations have applied since they were 
written, and both have been intended to be fully enforceable.'' \495\
---------------------------------------------------------------------------
    \495\ Congressional Record, May 22, 2008, p. S4747, https://
www.congress.gov/crec/2008/05/22/CREC-2008-05-22-pt1-PgS4743-3.pdf.
---------------------------------------------------------------------------
    Given this clear expression of Congressional intent, FNS may not by 
regulation exacerbate discrimination within SNAP based on race, 
ethnicity, or disability. Since FNS recognizes that the proposed rule 
would have discriminatory effects, it must withdraw the rule.
Chapter 13. The Proposed Rule Fails to Adequately Estimate the Impact 
        on Small Entities
    The Regulatory Flexibility Act (5 U.S.C.  601-612) requires 
agencies to analyze the impact of a proposed rule specifically on small 
businesses and entities through an initial regulatory flexibility 
analysis. The Regulatory Flexibility Act specifically mandates that the 
analysis must contain a series of arguments including, but not limited 
to: why action by the agency is being considered, what the legal basis 
is for the proposed rule, and an estimate to the number of small 
entities to which the rule would apply.\496\ The FNS failed to 
undertake the necessary research regarding the impact of this rule on 
all small entities, with the proposed rule offering only a brief impact 
report with minimal analysis that fails to accurately or adequately 
assess the impact of the proposed rule.
---------------------------------------------------------------------------
    \496\ Regulatory Flexibility Act, 5 U.S.C.  603, https://
www.sba.gov/advocacy/regulatory-flexibility-act.
---------------------------------------------------------------------------
    The FNS claims that aside from program participants, the proposed 
rule would primarily impact state agencies. This assessment leaves out 
a key group of impacted stakeholders--small SNAP retailers, who rely on 
SNAP purchases for consistent and dependable revenue. The Department 
incorrectly assumes that after losing benefits, people would replace 
their monthly SNAP allotment with cash. The individuals impacted by 
this NPRM are a very low-income group, as approximately 70 percent of 
all ``ABAWD's'' are below \1/2\ of the Federal poverty line.\497\ They 
do not tend to have disposable income, and taking away their SNAP 
benefits would take away their ability to purchase food. Additionally, 
SNAP benefits normally run out for most households before the end of 
the month.\498\ Many households spend their benefits rapidly because 
they are funds designated specifically for food. Cash cannot be used to 
replace SNAP because these dollars are needed to pay other expenses 
such as rent, clothing, gasoline, and many other necessities.\499\ The 
Department's primary and false assumption that SNAP is supplemental 
rather than essential lays an untrue foundation for the argument that 
small retailers would not be disproportionately impacted.
---------------------------------------------------------------------------
    \497\ Center on Budget and Policy Priorities, ``Unemployed adults 
without children who need help buying food only get SNAP for three 
months,'' https://www.cbpp.org/unemployed-adults-without-children-who-
need-help-buying-food-only-get-snap-for-three-months.
    \498\ Dottie Rosenbaum, ``Many SNAP Households Will Experience Long 
Gap Between Monthly Benefits Despite End of Shutdown,'' Center on 
Budget and Policy Priorities, revised February 4, 2019, https://
www.cbpp.org/research/food-assistance/many-snap-households-will-
experience-long-gap-between-monthly-benefits.
    \499\ Ibid.
---------------------------------------------------------------------------
    Additionally, the NPRM includes an inaccurate estimate of the 
number of small retailers that would be impacted. This leaves the 
public and stakeholder groups ill-informed about the potential 
implications of the rule. A small retailer at risk of being 
significantly harmed by the proposed rule would not understand the 
importance of the issue solely by reading the NPRM and Regulatory 
Impact Analysis due to the failure to scale the estimation exclusively 
to the impacted areas. This section will review which pieces of the 
Regulatory Flexibility Act were not adequately covered, the impact and 
magnitude of the inaccurate estimation of impacted small businesses in 
the NPRM, and which areas would be disproportionately or significantly 
impacted across the United States.
A. Inadequately Addressed Sections of the Regulatory Flexibility Act
    The primary area of concern within the Regulatory Flexibility Act 
(R.F.A.) is  603--Initial regulatory flexibility analysis.\500\ 
According to the R.F.A.  603, an agency publishing an NPRM is required 
to do the following: \501\
---------------------------------------------------------------------------
    \500\ Regulatory Flexibility Act, 5 U.S.C.  603, https://
www.sba.gov/advocacy/regulatory-flexibility-act.
    \501\ Ibid.

          (b) Each initial regulatory flexibility analysis required 
---------------------------------------------------------------------------
        under this section shall contain--

                  (1) a description of the reasons why action by the 
                agency is being considered;
                  (2) a succinct statement of the objectives of, and 
                legal basis for, the proposed rule;
                  (3) a description of and, where feasible, an estimate 
                of the number of small entities to which the proposed 
                rule will apply;
                  (4) a description of the projected reporting, 
                recordkeeping and other compliance requirements of the 
                proposed rule, including an estimate of the classes of 
                small entities which will be subject to the requirement 
                and the type of professional skills necessary for 
                preparation of the report or record;
                  (5) an identification, to the extent practicable, of 
                all relevant Federal rules which may duplicate, overlap 
                or conflict with the proposed rule.

          (c) Each initial regulatory flexibility analysis shall also 
        contain a description of any significant alternatives to the 
        proposed rule which accomplish the stated objectives of 
        applicable statutes and which minimize any significant economic 
        impact of the proposed rule on small entities. Consistent with 
        the stated objectives of applicable statutes, the analysis 
        shall discuss significant alternatives such as--

                  (1) the establishment of differing compliance or 
                reporting requirements or timetables that take into 
                account the resources available to small entities;
                  (2) the clarification, consolidation, or 
                simplification of compliance and reporting requirements 
                under the rule for such small entities;
                  (3) the use of performance rather than design 
                standards; and
                  (4) an exemption from coverage of the rule, or any 
                part thereof, for such small entities.

    The provided regulatory flexibility analysis in the NPRM fails to 
include a detailed description required by  603(b)(1) and  603(b)(2), 
as the analysis includes no legal basis for the proposed rule or why 
the agency is considering the action. Additionally,  603(b)(3) 
mandates an estimate of the number of small entities impacted. The 
agency provides an estimate, but that estimate is flawed as we show 
below.\502\
---------------------------------------------------------------------------
    \502\ CBPP Internal Analysis of U.S. Department of Agriculture, 
Food Environment Atlas Data 2018.
---------------------------------------------------------------------------
    By inaccurately estimating the number of small entities that would 
be impacted by the proposed rule, the Department assumed it was not 
required to satisfy other requirements in the Act. For example,  
609(a)(1) states that when a rule is introduced that will have a 
significant economic impact on small entities; the respective agency 
must provide a statement or notice to the effect on small 
entities.\503\ Providing an imprecise estimate, the Agency is able to 
argue that no significant impact will be made, keeping small entities 
uninvolved with the rulemaking process. In addition,  603(c) of the 
RFA requires a description of potential alternatives to the proposed 
rule. The NPRM fails to provide any possible alternatives because the 
proposed rule inaccurately asserts that there is no disproportionate 
impact on small entities, falsely excusing them from additional 
requirements. Incorrectly estimating the number of small businesses not 
only represents a lack of specificity, it more importantly exempts the 
Agency from providing an advanced notice to small entities, allowing 
them to submit comments and address concerns in the NPRM.
---------------------------------------------------------------------------
    \503\ Regulatory Flexibility Act, 5 U.S.C.  609(a)(1) https://
www.sba.gov/advocacy/regulatory-flexibility-act.
---------------------------------------------------------------------------
B. Impact of the NPRM on Small SNAP Retailers
    Perhaps the most troubling of the agency's regulatory flexibility 
analysis is the inadequate estimate of the total number of small SNAP 
retailers. The NPRM does accurately estimate that there are nearly 
200,000 retailers that fall under the Small Business Administration's 
gross sales threshold, but it is imprecise to assume that all of these 
stores would be impacted by the proposed rule.\504\ As a result, the 
lost sales per store are too low, failing to signal to small entities 
the magnitude of their losses from the NPRM. An internal analysis at 
the Center on Budget and Policy Priorities has shown that a total of 
639 counties across 28 states would be impacted by the proposed 
rule.\505\ Within these counties, there are only a total of about 
67,000 SNAP retailers of all sizes.\506\ If the same percentage of 
these businesses were considered to be small entities under the Small 
Business Administration gross sales threshold used in the NPRM (76 
percent), then it can be estimated that a total of nearly 51,000 small 
entities would be impacted, significantly less than the NPRM's estimate 
of 200,000.\507\
---------------------------------------------------------------------------
    \504\ NPRM, Regulatory Flexibility Act https://
www.federalregister.gov/d/2018-28059/p-118.
    \505\ CBPP Analysis of BLS Unemployment data, 2019.
    \506\ U.S. Department of Agriculture, Food Environment Atlas Data 
2018 https://www.ers.usda.gov/data-products/food-environment-atlas/
data-access-and-documentation-downloads/.
    \507\ CBPP Internal Analysis of U.S. Department of Agriculture, 
Food Environment Atlas Data 2018.
---------------------------------------------------------------------------
    By making an estimate based off the total number of small SNAP 
retailers in the United States (200,000) versus the number of small 
retailers impacted by the NPRM (51,000), FNS has artificially lowered 
the average of the sales lost by four times.\508\ FNS has conducted a 
cursory analysis regarding the impact of the proposed rule on small 
businesses. Using an estimate of nearly four times the true number of 
stores potentially impacted minimizes the reality that small SNAP 
retailers would face from the NPRM.
---------------------------------------------------------------------------
    \508\ CBPP Internal Analysis of U.S. Department of Agriculture, 
Food Environment Atlas Data 2018.
---------------------------------------------------------------------------
    According to the NPRM, SNAP benefit payments are expected to be 
reduced by about $1.7 billion per year.\509\ By conducting the same 
calculation as FNS in the NPRM while including a more accurate estimate 
of the amount of impacted small businesses ($1.7 billion  15% redeemed 
at small retailers / 51,000 stores losing waivers / 12 months), we can 
estimate that the loss of revenue per small store on average each month 
would be $417, compared to the NPRM estimate of $106.\510\ The NPRM 
subsequently states that the average small store redeemed $3,800 in 
SNAP each month in 2017, making the NPRM estimate representative of 
three percent of monthly store sales.\511\ In evaluating the impact on 
small stores with the true loss of revenue per store ($417/month), the 
average small store would realistically face an 11 percent reduction in 
the amount of SNAP benefits redeemed at each store. Not only is 11 
percent a significant portion of a store's SNAP revenue, more 
importantly it is nearly four times greater than the estimated impact 
from the initial regulatory flexibility analysis of three percent in 
the NPRM.\512\ Using the more accurate estimate would have properly 
signaled the implications of the rule to small entities, allowing them 
the opportunity to comment on the proposed rule.
---------------------------------------------------------------------------
    \509\ NPRM, Regulatory Flexibility Act https://
www.federalregister.gov/d/2018-28059/p-118.
    \510\ Ibid.
    \511\ Ibid.
    \512\ Ibid.
---------------------------------------------------------------------------
Small Businesses Located in Rural Areas Will Be Disproportionately 
        Impacted
    The small business impact assessment in the NPRM claims that small 
retailers are not expected to be disproportionately impacted by the 
proposed rule. This is not correct. Using 2018 as an illustrative year, 
a total of 639 counties or county-equivalents would have lost access to 
the waiver across 28 states.\513\ Of those impacted areas (which 
include counties, reservations, and small cities), 405 areas have a 
population of fewer than 50,000 people, where the Census Bureau defines 
the cutoff for urbanized areas.514-515 Research has shown 
that rural areas with lower population levels often rely on corner and 
convenience stores for food, many of which are individually owned small 
businesses.\516\ When nearly \2/3\ of the counties impacted by an 
arbitrary rule depend on small businesses for SNAP purchases, the rule 
is certain to have a substantial impact on small business in the 
impacted areas. As a result, the claim that small entities will not be 
substantially impacted is incorrect.
---------------------------------------------------------------------------
    \513\ CBPP Analysis of BLS Unemployment data, 2019.
    \514\ U.S. Department of Agriculture, Food Environment Atlas Data 
2018, https://www.ers.usda.gov/data-products/food-environment-atlas/
data-access-and-documentation-downloads/.
    \515\ Michael Ratcliffe, et al., ``Defining Rural at the U.S. 
Census Bureau,'' United States Census Bureau, U.S. Department of 
Commerce, Dec 2016, pp. 1-8, https://www2.census.gov/geo/pdfs/
reference/ua/Defining_Rural.pdf.
    \516\ Joseph Sharkey, et al., ``Association between neighborhood 
need and spatial access to food stores and fast food restaurants in 
neighborhoods of Colonias,'' International Journal of Health 
Geographics, 2009, pp. 1-17.
---------------------------------------------------------------------------
    Individuals living in rural areas must travel longer distances to 
supermarkets and grocery stores than their urban or suburban 
counterparts.\517\ As a result, they visit corner and convenience 
stores frequently for their daily food and nutrition needs.\518\ A 
study conducted by USDA found that new-entrant SNAP households along 
with SNAP households that had participated in the program for less than 
6 months lived, on average, 4 miles from a grocery store and 1.6-1.8 
miles from a convenience store.\519\ While this may not appear to be a 
substantial difference in distance, individuals living in rural areas 
are less likely to have access to public transportation and a private 
vehicle.\520\ Because lack of transportation is a common issue across 
low-income rural areas, residents rely on the closest store to spend 
their monthly SNAP benefits.\521\ As mentioned, these closest stores 
are typically convenience stores or small businesses, offering a 
limited selection of foods.
---------------------------------------------------------------------------
    \517\ Joseph Sharkey, ``Measuring Potential Access to Food Stores 
and Food-Service Places in Rural Areas in the U.S.,'' American Journal 
of Preventive Medicine, April 2009, pp. S151-S155.
    \518\ Renee Walker, Christopher Keane, and Jessica Burke, 
``Disparities and access to healthy food in the United States: A review 
of food deserts literature,'' Health & Place, 2010, pp. 876-884.
    \519\ James Mabli, ``SNAP Participation, Food Security, and 
Geographic Access to Food,'' Food and Nutrition Service, Office of 
Policy Support--U.S. Department of Agriculture, March 2014, pp. 1-50.
    \520\ Kevin Matthews, et al., ``Health-Related Behaviors by Urban-
Rural county Classification--United States, 2013,'' Morbidity and 
Mortality Weekly Report, Centers for Disease Control and Prevention, 
February 2017, pp. 1-12.
    \521\ C. Pinard, et al., ``An integrative literature review of 
small food store research across urban and rural communities in the 
U.S.,'' Preventive Medicine Reports, April 2016, pp. 324-332.
---------------------------------------------------------------------------
    In addition to many of the impacted counties being rural and facing 
issues around food access, a significant portion of the impacted 
counties suffer from extremely low access to food.\522\ Sixty-three 
counties (ten percent of those impacted by the proposed rule) have five 
or less SNAP retailers across the county.\523\ While the impact of the 
proposed rule on retailers across all rural counties would be 
considerable, these 63 counties with such few SNAP vendors would be 
disproportionately impacted. In these cases particularly, rural 
residents are often required to travel longer than the previously 
mentioned average distances to access a supermarket or superstore.\524\ 
Table 13.1 shows some of the impacted rural counties with few or lone 
SNAP retailers:
---------------------------------------------------------------------------
    \522\ Lisa Powell, et al., ``Food store availability and 
neighborhood characteristics in the United States,'' Preventive 
Medicine (2007), pp. 189-195.
    \523\ CBPP Internal Analysis of U.S. Department of Agriculture, 
Food Environment Atlas Data 2018.
    \524\ C. Pinard, et al., ``An integrative literature review of 
small food store research across urban and rural communities in the 
U.S.,'' Preventive Medicine Reports, April 2016, pp. 324-332.

                               Table 13.1
   Examples of Rural Counties Disproportionately Impacted by the NPRM
------------------------------------------------------------------------
                                           SNAP
                                     Participants as    Average Monthly
      State            County          of July 2018    SNAP Retailers in
                                          \525\            2016 \526\
------------------------------------------------------------------------
North Dakota      Eddy County                     167                  1
\525\ U.S.
 Department of
 Agriculture,
 SNAP State
 Issuance and
 Participation
 Estimates (FNS-
 388 & FNS-
 388A), data as
 of July 2018,
 https://
 www.fns.usda.go
 v/sites/default/
 files/pd/SNAP-
 FNS388A.zip.
Kentucky          Robertson County                290                  2
\526\ USDA, Food
 Environment
 Atlas, last
 updated on
 March 27, 2018,
 https://
 www.ers.usda.go
 v/data-products/
 food-
 environment-
 atlas/data-
 access-and-
 documentation-
 downloads/.
Virginia          Charles City                    827                  2
                   County
South Dakota      Mellette County                 655                  3
Nevada            White Pine                    1,131                  5
                   County
West Virginia     Doddridge County              1,187                  5
------------------------------------------------------------------------
Source: CBPP Internal Analysis of U.S. Department of Agriculture, Food
  Environment Atlas Data, 2018.

    Table 13.1 shows that rural counties impacted by the NPRM with few 
SNAP retailers, or a single SNAP retailer, would face significant harm 
if sales were lost. These are solely a few illustrative examples of the 
counties impacted. These examples validate the potential of the NPRM on 
rural counties with minimal SNAP retailers. A loss of sales for any of 
these isolated SNAP retailers would inhibit their ability to provide 
for the surrounding community. A much more robust nationwide assessment 
ought to have been included in the NPRM in order to meet the 
requirements of the law and to allow small entities the opportunity to 
meaningfully engage on what the proposed rule might mean to their sales 
and business. It is not representative of the Department to ignore its 
own research and knowledge of the location of small businesses impacted 
by the rule. This failure and lack of discussion on the impact of small 
businesses leaves that constituency and others unable to comment 
effectively on the proposed rule.
Small Retailers in Urban Areas Will Be Significantly Impacted
    While many of the counties that will lose waivers under the 
proposed rule are rural, a noteworthy portion of the counties that 
would no longer be eligible to be waived is considered urban. Some of 
the counties with the highest population in the country would be 
impacted, subjecting a great number of recipients in a condensed area 
to the proposed rule. The following are a few of the impacted urban 
counties and cities:

                                                   Table 13.2
                          Examples of Urban Locales Significantly Impacted by the NPRM
----------------------------------------------------------------------------------------------------------------
                                                    SNAP                                            Estimated
                                              Participants as  Statewide Share     Estimated         Monthly
    State         County       Primary City     of July 2018     of ``ABAWD''      ``ABAWD''        ``ABAWD''
                                                   \527\          Population      Population *     Benefits **
----------------------------------------------------------------------------------------------------------------
         CA   Los Angeles     Los Angeles           1,055,314            11.3%          120,000      $19,444,000
               County
\527\ U.S.
 Department
 of
 Agriculture
 , FNS 2018
 County
 level SNAP
 data.
          IL  Cook County     Chicago                 813,465            12.3%          100,000      $16,295,000
         MI   Wayne County    Detroit                 416,321            11.4%           47,000       $7,685,000
         NV   Clark County    Las Vegas               352,675            13.9%           49,000       $7,984,000
         PA   Philadelphia    Philadelphia            473,269             6.6%           31,000       $5,040,590
               County
----------------------------------------------------------------------------------------------------------------
Source: CBPP analysis of USDA SNAP Household Characteristics data, FY 2017, where the average monthly ``ABAWD''
  benefit is $162.4.
* Estimated ``ABAWD'' Population is derived from share of ``ABAWDs'' in each state applied to county caseload.
** Estimated Monthly ``ABAWD'' Benefit is calculated from average ``ABAWD'' benefit and ABAWD share of
  population.

    The cities shown in Table 13.2 represent some of the largest cities 
impacted by the proposed rule. Across the five cities listed, an 
internal analysis estimates a total of 347,000 ``ABAWDs'' residing 
within these cities. While not all of these individuals are necessarily 
subject to the proposed rule due to exemptions and other factors, it is 
critical to recognize that the average ``ABAWD'' SNAP monthly benefit 
in FY2017 was $162.4, or an estimated annual contribution of 
$677,383,000 in SNAP benefits from ``ABAWDs'' to the combined economies 
of these cities.\528\
---------------------------------------------------------------------------
    \528\ ``Characteristics of Able-bodied Adults without Dependents,'' 
Food and Nutrition Services, United States Department of Agriculture, 
2016, https://fns-prod.azureedge.net/sites/default/files/snap/
nondisabled-adults.pdf.
---------------------------------------------------------------------------
    Similar to rural residents, individuals living in low-income urban 
areas often depend on convenience stores for groceries because of the 
lack of accessibility to full-sized supermarkets and grocery 
stores.\529\ In 2009, a study found that ``ZIP [C]odes representing 
low-income areas had only 75% as many chain supermarkets available as 
ZIP [C]odes representing middle-income areas.'' \530\ With some urban-
dwelling SNAP recipients essentially being forced to redeem benefits at 
small retailers because of the lack of access, the argument that urban 
small retailers would not be impacted by the proposed rule is 
representative of the lack of consideration that FNS has presented.
---------------------------------------------------------------------------
    \529\ Melissa Nelson Laska, et al., ``Healthy food availability in 
small urban food stores: a comparison of four U.S. cities,'' Public 
Health Nutrition, December 2009, pp. 1031-1035.
    \530\ Nicole Larson, Mary Story, and Melissa Nelson, ``Neighborhood 
Environments: Disparities in Access to healthy Foods in the U.S.,'' 
American Journal of Preventive Medicine, 2009, pp. 74-81.
---------------------------------------------------------------------------
    There have been additional studies conducted concerning food access 
in individual cities that would be impacted by the proposed rule. For 
example, the food environment of Philadelphia has a wealth of research 
demonstrating food access issues across the city. A study from 2014 
interviewed hundreds of low-income individuals in Philadelphia and 
found that residents often shopped at convenience stores because of 
``easy parking, accommodation of physical disabilities or special 
needs, and the integration of food shopping into other daily 
activities.'' \531\ With potentially impacted cities having published 
information about how their residents rely on small food retailers, it 
is remiss of the literature that exists in the field to argue that 
small retailers would not be unduly impacted by the proposed rule.
---------------------------------------------------------------------------
    \531\ Carolyn Cannuscio, et al., ``The social dynamics of healthy 
food shopping and store choice in an urban environment,'' Social 
Science & Medicine, October 2014, pp. 13-20.
---------------------------------------------------------------------------
Conclusion
    As demonstrated, the NPRM fails to adequately address the 
disproportionate and significant impact on small entities across the 
country. First, significant portions of the Regulatory Flexibility 
Analysis were either not completed or completed incorrectly, failing to 
signal to small entities the importance of the NPRM. Second, the 
analysis in the proposed rule incorrectly calculates the number of 
impacted small entities, suggesting that small entities would face an 
impact four times smaller than the reality of what the NPRM would 
prescribe. Lastly, both the rural and urban areas impacted by the rule 
would see significant losses. Rural small retailers would see 
disproportionate losses in sales, while urban small retailers would 
experience substantial losses in sales. Until FNS conducts further 
research regarding the impact of the proposed rule on small entities 
and the communities that house these businesses, implementation of this 
rule would be unwarrantable and detrimental to those who reside within 
the impacted areas.
Appendix A: Center on Budget and Policy Priorities' Contributors to Our 
        Public Comments
    For more than 3 decades, the Center on Budget and Policy Priorities 
has been at the forefront of national and state debates to protect and 
strengthen programs that reduce poverty and inequality and increase 
opportunity for people trying to gain a foothold on the economic 
ladder. The Center is a high-caliber strategic policy organization that 
shapes critical policies for low-income families and individuals. To 
these ends, we conduct highly skilled strategic and analytic work to 
develop and advance specific, actionable proposals, to achieve the 
maximum possible policy gains, and to ensure their effective 
implementation on the ground. We also build effective partnerships and 
help a diverse array of organizations and constituencies to engage more 
effectively in these debates.
    As part of this overall mission, we work to strengthen policies and 
programs that reduce hunger and poverty and improve the lives of the 
nation's poorest families and individuals. Since our founding in 1981, 
we have worked on Federal nutrition programs, most notably the 
Supplemental Nutrition Assistance Program (SNAP), formerly known as 
food stamps. We have extensive experience in and knowledge of SNAP's 3 
month time limit and state waiver authority within that rule. Examples 
of our efforts on this front include:

   Issuing reports and analyses on the time limit and the 
        population impacted by the rule, as well as policy options 
        available to states via waiver and individual exemption 
        policies;

   Assisting states in assessing which areas of their states 
        are eligible for waivers and developing waiver requests that 
        meet the Federal criteria. The Center has supported an average 
        of 30-40 states each year, completing or assisting with a total 
        of over 600 waiver applications since the late 1990s;

   Educating anti-hunger nonprofits and other community 
        organizations about the time limit rule, including about the 
        availability of waivers based on the underlying unemployment 
        trends in the state and individual exemptions; and

   Training state agency officials as well as local anti-hunger 
        and poverty advocates about judicious practices in implementing 
        the time limit policy. For years we have worked to help state 
        officials implement the time limit in a way that conforms with 
        Federal law and protects as many individuals with low incomes 
        as possible and avoids (to the degree possible) cutting off 
        individuals who are not actually subject to the limit.

    Over time, the Center's work on waivers and time limit policy has 
evolved into a multi-faceted and complex approach. The Center has 
provided in-person trainings and direct supervision to state officials 
in nearly ten states, while providing support to advocates and working 
on issues of implementation for many more. The Center has also given 
multiple presentations to FNS staff at several regional meetings across 
the United States regarding waivers, along with a series of annual 
presentations to the American Association of SNAP Directors. Additional 
support from the Center has included assistance regarding program 
integrity and payment accuracy implications, writing of manuals for 
dozens of states, and constantly monitoring all 50 states' statuses and 
ability to apply for ``ABAWD'' waivers. The food assistance team's 
experience in combination with the economic expertise of other members 
of the Center provides CBPP with the greatest amount of experience and 
knowledge concerning ``ABAWD'' waivers and the implications of the 
proposed rule.
    Below is a listing of the individuals at CBPP who contributed to 
these comments:
    Jared Bernstein joined the Center on Budget and Policy Priorities 
in 2011 as a Senior Fellow. From 2009 to 2011, Bernstein was the Chief 
Economist and Economic Adviser to Vice President Joe Biden, Executive 
Director of the White House Task Force on the Middle Class, and a 
member of President Obama's economic team. Bernstein's areas of 
expertise include Federal and state economic and fiscal policies, 
income inequality and mobility, trends in employment and earnings, 
international comparisons, and the analysis of financial and housing 
markets.
    Prior to joining the Obama Administration, Bernstein was a senior 
economist and the director of the Living Standards Program at the 
Economic Policy Institute in Washington, D.C. Between 1995 and 1996, he 
held the post of Deputy Chief Economist at the U.S. Department of 
Labor. Bernstein holds a Ph.D. in Social Welfare from Columbia 
University.
    He is the author and coauthor of numerous books for both popular 
and academic audiences, including Getting Back to Full Employment: A 
Better Bargain for Working People, Crunch: Why Do I Feel So Squeezed?, 
nine editions of The State of Working America, and his latest book, The 
Reconnection Agenda: Reuniting Growth and Prosperity. Bernstein has 
published extensively in various venues, including the New York Times 
and Washington Post. He is an on-air commentator for the cable stations 
CNBC and MSNBC, contributes to the Washington Post's PostEverything 
blog, and hosts On The Economy (jaredbernsteinblog.com).
    Thompson Bertschy received his Master's in Social Work from the 
University of North Carolina and began working with the Center as an 
intern in 2019. In North Carolina, Thompson worked with the Carolina 
Farm Stewardship Association to advocate for sustainable and equitable 
food policy while organizing food policy councils across the state. At 
CBPP, Thompson focuses on research and analysis regarding ``ABAWD'' 
exemptions and state legislative proposals around SNAP, TANF, and 
Medicaid.
    Ed Bolen joined the Center in 2010 as a Senior Policy Analyst. At 
the Center, Ed focuses on SNAP Employment & Training and ``ABAWD'' 
waivers. He has provided trainings to multiple states regarding their 
loss of waivers, including Alabama, California, Michigan, Illinois, 
Alaska, and Mississippi. Additionally, Ed has given presentations to 
Food and Nutrition Service staff at the Northeast and Mid-Atlantic 
regional meetings. Since 2011, Ed has consistently worked with 
advocates and state officials concerning implementation issues in 
Georgia, Illinois, Michigan, California, and other states. He has also 
presented to the American Association of SNAP Directors for 4 
consecutive years and created a toolkit on implementing the ``ABAWD'' 
time limit including technical writing on implementation issues. He has 
had many interactions with media and has written multiple blog posts 
and papers concerning ``ABAWDs.'' Throughout his time at the Center, Ed 
has written multiple pieces concerning Employment and Training for SNAP 
recipients.
    Prior to joining the Center, Bolen was a Senior Policy Analyst at 
California Food Policy Advocates. While there, he advocated for 
administrative and legislative improvements to food assistance programs 
and provided training and technical assistance to community-based 
organizations. He also has worked in public health law, most recently 
consulting on legal strategies to combat childhood obesity with the 
National Policy and Legal Analysis Network. Prior to that, Bolen was 
senior staff attorney at the Child Care Law Center, specializing on 
licensing, subsidy and legislative issues affecting low-income families 
in child care and early education settings.
    Kathleen Bryant is a Research Assistant in the Federal Fiscal team 
at the Center on Budget and Policy Priorities. Previously, she interned 
for EMILY's List, the Economic Policy Institute, and the Center's 
Legislative Affairs team. She also conducted independent research on 
school segregation for her honors thesis and was selected as a fellow 
in the Advanced Empirical Research on Politics for Undergraduates 
Program for this research by the Society for Political Methodology. 
Kathleen has a B.A. in Public Policy from The College of William & 
Mary.
    Ashley Burnside is a Research Assistant with the Center's Family 
Income Support division. Before joining the Center, she was an Emerson 
National Hunger Fellow with the Congressional Hunger Center, where she 
advocated for anti-hunger and anti-poverty solutions, led voter 
registration and community engagement efforts for food pantry clients, 
and conducted research on tax credit programs. Burnside has also served 
as a public policy fellow at AIDS United, a national HIV/AIDS advocacy 
organization. She holds a B.A. in Social Theory and Practice from the 
University of Michigan.
    Lexin Cai is a Research Analyst with the Center's Food Assistance 
Division, where she focuses on data analysis for nutrition assistance 
programs. Lexin first joined the Center in May 2015 as an intern. Prior 
to joining the Center, she interned with the World Wildlife Fund, 
focusing on international agricultural trade. At the Center, Lexin 
conducts analyses, compiles historical data and records, and uses 
mapping and modeling to support states with the waiver application 
process.
    Over the last several years, Lexin has worked with Catlin Nchako in 
completing waiver applications for multiple states and providing 
assistance with the process to others, helping an average of 32 states 
per year. She holds a Master of Science in Social Policy from the 
University of Pennsylvania and a Bachelor of Management from Renmin 
University of China.
    Steven Carlson provided background research for these comments as a 
consultant to CBPP. He was a Federal employee at the Food and Nutrition 
Service for 37 years. During his time there, he led their Office of 
Analysis and Evaluation and oversaw research studies and analysis on 
SNAP.
    Maritzelena Chirinos is a recent graduate of Meredith College and 
began working with The Center as an intern in 2019 on The State Fiscal 
Project Team. She holds a B.A. in Criminology and Sociology. She 
previously worked at The Indivisible Project and Democrats For 
Education Reform.
    Stacy Dean is the Vice President for Food Assistance Policy at the 
Center on Budget and Policy Priorities. She directs CBPP's food 
assistance team, which publishes frequent reports on how Federal 
nutrition programs affect families and communities and develops 
policies to improve them. Dean's team also works closely with program 
administrators, policymakers, and nonprofit organizations to improve 
Federal nutrition programs and provide eligible low-income families 
with easier access to benefits. She brings her deep programmatic and 
operational knowledge along with a strong strategic sense to help 
advance CBPP's priorities.
    Dean has over 20 years of experience in working in great detail 
with the USDA and dozens of states concerning ``ABAWD'' waivers. In 
1997, she began writing and distributing information to a majority 
states regarding specific Labor Surplus Areas in the state and their 
ability to apply for a waiver. She has written multiple papers on 
``ABAWDs'' and waiver requests dating back to the early 2000s. In 
addition to writing the papers, Dean has worked closely with many 
states providing supervision, training, and general policy support for 
both application and implementation.
    In addition to her work on Federal nutrition programs, Dean directs 
CBPP efforts to integrate the delivery of health and human services 
programs at the state and local levels. Dean has testified before 
Congress and spoken extensively to national and state nonprofit groups. 
She has been quoted in such publications as the New York Times, 
Washington Post, Wall Street Journal, and Politico, as well as the 
Associated Press. Dean joined CBPP in 1997 as a Senior Policy Analyst 
working on national policy issues such as the Federal budget, SNAP, and 
benefits for immigrants.
    Previously, at the Office of Management and Budget, Dean served as 
a budget analyst for food stamps. In this role, she staffed the White 
House work on the 1996 food stamp program changes, reviewed and cleared 
the ABAWD waiver guidance, and worked on policy development, regulatory 
and legislative review, and budgetary process and execution for a 
variety of income support programs.
    Ife Floyd joined the Center in June 2011 as a Research Associate 
with the Family Income Support Division, and is now Senior Policy 
Analyst.
    Prior to joining the Center, Floyd served as an AmeriCorp VISTA 
with Culture Connect, Inc. and developed a cultural competency workshop 
series for professional workplaces in the Atlanta area. In addition, 
she also worked with the Atlanta Community Food Bank's Prosperity 
Campaign as a benefits screener to increase access to programs like 
SNAP, TANF, and Medicaid for low-income families. Floyd holds a B.A. in 
Sociology from Northwestern University and a M.P.P. degree from Georgia 
State University.
    Brynne Keith-Jennings obtained her Master's in Public Policy from 
the University of Southern California and joined the Center in June 
2011. Her work focuses on Federal and state SNAP policies and research. 
As a Senior Policy Analyst, Brynne works closely with the USDA and 
regularly offers guidance with ABAWD waiver policy. She provides states 
with technical assistance regarding waiver requests and has developed a 
suite of products around the farm bill, SNAP, and employment. Since 
2013, Keith-Jennings has been one of the primary food assistance team 
members concerned with ``ABAWD'' waivers, assisting with the process of 
completing or assisting with 30-40 waivers each year.
    Prior to joining the Center, she worked as an educator and policy 
analyst for Witness for Peace and as a consultant for other NGOs in 
Nicaragua. She has also worked at the Public Welfare Foundation 
supporting human rights and criminal justice reform organizations, at 
the Government Accountability Office, and at the Tomas Rivera Policy 
Institute, where she analyzed language access policies and other issues 
affecting Latino communities in Southern California.
    Joseph Llobrera is a Senior Policy Analyst on the Food Assistance 
team. As a research associate at the Center between 2002 and 2007, 
Llobrera supported the Food Assistance, State Fiscal, and Housing 
Policy teams. During these years, Joseph managed the Center's ``ABAWD'' 
waiver support process through data analysis and extraction, assisting 
nearly 150 states with the waiver process. He also interacted with and 
provided operational support to state advocates and agencies. Before 
returning to the Center in 2019, he served as an Associate Director of 
Learning and Improvement at Insight Policy Research, providing 
technical assistance and training to Federal, state, and local human 
service agencies that administer SNAP and the Temporary Assistance for 
Needy Families program. During this time, Llobrera reviewed notices 
that states would send to SNAP recipients who were subject to the time 
limit. He also worked as a researcher at IMPAQ International and the 
Urban Institute, focusing on food assistance policy, workforce 
development, and health policy.
    Currently at the Center, Llobrera provides operational support and 
technical assistance to advocates and state agencies in an effort to 
streamline the SNAP and Medicaid enrollment process. Llobrera holds a 
Ph.D. in Nutrition from the Friedman School of Nutrition Science and 
Policy at Tufts University, a master's degree in Geography from the 
University of Washington (Seattle), and a bachelor's degree in 
Mathematics and Urban Studies from Brown University.
    Rachel Mbassa is an intern with the Data team. She graduated with a 
Master of Public Health in Biostatistics from Loma Linda University. 
She then went on to work at the University of California San Francisco 
Medical Center, conducting epidemiological/clinical research centered 
on the effects of cumulative psychosocial stressors on cardiovascular 
disease outcomes. She recently moved to Washington, D.C. to transition 
into a career in policy data analysis. She has a strong interest in 
health economics and policy and hopes to pursue a terminal degree in 
this field.
    Tazra Mitchell joined the Center in 2016 as a Policy Analyst in the 
Family Income Support Division. Previously she worked as a State Policy 
Fellow and Policy Analyst with the North Carolina Budget & Tax Center, 
where she conducted analysis of work and income supports as well as 
fiscal and economic policies that enable low-income people and 
disadvantaged communities to thrive. In addition, she worked as a 
Research Assistant in the nonpartisan Fiscal Research Division of the 
North Carolina General Assembly where she analyzed legislative 
proposals to determine the fiscal impact on state government resources 
and worked directly with legislators to develop the state budget.
    Mitchell holds a B.A. in Political Science from North Carolina 
State University and an M.P.P. from the Sanford School of Public Policy 
at Duke University.
    Catlin Nchako joined the Food Assistance Division as a Research 
Associate in November 2013. His work focuses on data analysis and 
research for the food stamp, school meals, and WIC programs. Over the 
last several years, Nchako has worked with Lexin Cai in completing 
waiver applications for multiple states and aiding with the process to 
others, helping an average of 32 states per year. He has also tracked 
employment benefit trigger notices for all states and territories, 
notifying areas when they qualify for an exemption. He has completed 
multiple analyses and contributed to many of the Center's papers 
concerning ``ABAWDs.''
    Nchako worked for the Center as a Food Assistance intern prior to 
joining the organization on a full-time basis. He previously interned 
for the Center for Law and Social Policy. He also worked for several 
years as a labor researcher for the United Food and Commercial Workers 
Union, where he evaluated wage proposals during labor contract 
negotiations, analyzed companies' financial performance, and provided 
campaign research support. He holds a Master's in Public Policy from 
Georgetown University and a B.A. in Africana Studies from Cornell 
University.
    LaDonna Pavetti is the Vice President for Family Income Support 
Policy at the Center on Budget and Policy Priorities. In this capacity, 
she oversees the Center's work analyzing poverty trends and assessing 
the nation's income support programs, including the Temporary 
Assistance for Needy Families (TANF) program. For the last several 
years, she has been working on a special initiative to identify 
opportunities to build executive function and self-regulation skills in 
TANF and other work programs.
    Before joining the Center in 2009, Dr. Pavetti spent 12 years as a 
researcher at Mathematica Policy Research, Inc., where she directed 
numerous research projects examining various aspects of TANF 
implementation and strategies to address the needs of the hard-to-
employ. She has also served as a researcher at the Urban Institute, a 
consultant to the U.S. Department of Health and Human Services on 
welfare reform issues, and a policy analyst for the District of 
Columbia's Commission on Social Services. Across these positions, Dr. 
Pavetti conducted research on how states were implementing TANF, how 
sanctions were implemented, and conducted a study in St. Paul, 
Minnesota analyzing which populations were primarily impacted by TANF 
work requirements. In recent years, she has worked with Vermont, New 
York, Minnesota, Colorado, California, Connecticut, New Jersey, 
Pennsylvania, and Oregon.
    In addition, for several years she was a social worker in Chicago 
and Washington, D.C. Dr. Pavetti has an A.M. in social work from the 
University of Chicago and a Ph.D. in public policy from Harvard 
University's Kennedy School of Government.
    Dottie Rosenbaum is a Senior Fellow who joined the Center in 2000. 
She has worked extensively on Federal and state issues in SNAP as well 
as issues that involve the coordination of SNAP and other state-
administered health and income security programs, such as Medicaid, 
TANF, and child care. In addition, Rosenbaum has expertise on the 
Federal budget and budget process. With over more than 20 years of 
experience with SNAP, she written multiple papers regarding SNAP and 
SNAP Employment and Training.
    Before joining the Center, Rosenbaum was a budget analyst at the 
Congressional Budget Office. In this role, she conducted the cost 
estimate of the 1996 Personal Responsibility and Work Opportunity 
Reconciliation Act which included the time limit. She projected Federal 
spending and provided Congress with cost estimates for a variety of 
programs including: SNAP, Medicaid, the State Children's Health 
Insurance Program, Child Nutrition, and Elementary and Secondary 
Education. Rosenbaum holds a Master's in Public Policy from Harvard 
University's Kennedy School of Government.
    Andrew Scanlan is a recent graduate of Brown University and began 
working with the Center as an intern in 2019 on the Family Income 
Security team. At Brown, he worked as a research assistant in the 
political science department and received a combined degree in 
philosophy and mechanical engineering. He previously worked in the 
Rhode Island state government and the New York City government.
    Arloc Sherman's work focuses on family income trends, income 
support policies, and the causes and consequences of poverty. He has 
written extensively about the effectiveness of government poverty-
reduction policies, the influence of economic security programs on 
children's healthy development, the depth of poverty, tax policy for 
low-income families, welfare reform, economic inequality, material 
hardship, parental employment, and the special challenges affecting 
rural areas.
    He is a member of the National Academy of Sciences Committee on 
National Statistics Panel to Review and Evaluate the 2014 Survey of 
Income and Program Participation's Content and Design. Prior to joining 
the Center in 2004, Sherman worked for 14 years at the Children's 
Defense Fund and was previously at the Center for Law and Social 
Policy. His book Wasting America's Future was nominated for the 1994 
Robert F. Kennedy Book Award.
    Chad Stone is Chief Economist at the Center on Budget and Policy 
Priorities, where he specializes in the economic analysis of budget and 
policy issues. Stone has conducted notable research around measures of 
economic slack that are indicative of a weaker labor market, primarily 
concerning labor force participation and the employment population 
ratio. He was the acting executive director of the Joint Economic 
Committee of the Congress in 2007 and before that staff director and 
chief economist for the Democratic staff of the Committee from 2002 to 
2006. He was chief economist for the Senate Budget Committee in 2001-02 
and a senior economist and then chief economist at the President's 
Council of Economic Advisers from 1996 to 2001.
    Stone has been a senior researcher at the Urban Institute and 
taught for several years at Swarthmore College. His other Congressional 
experience includes two previous stints with the Joint Economic 
Committee and a year as chief economist at the House Science Committee. 
He has also worked at the Federal Trade Commission, the Federal 
Communications Commission, and the Office of Management and Budget. 
Stone is co-author, with Isabel Sawhill, of Economic Policy in the 
Reagan Years. He holds a B.A. from Swarthmore College and a Ph.D. in 
economics from Yale University.
    Jennifer Wagner joined the Center in 2015 as a Senior Policy 
Analyst with the health team. She focuses primarily on Medicaid 
eligibility and implementation of the Affordable Care Act (ACA), 
including analyzing opportunities to align Medicaid with other low-
income support programs for greater access and efficiency.
    Before joining the Center, Wagner served for 5 years as an 
Associate Director with the Illinois Department of Human Services. In 
that capacity, she oversaw SNAP and cash assistance policy as well as 
the local offices throughout the state that determined eligibility for 
cash, SNAP, and medical assistance. She assisted Illinois with Medicaid 
expansion under the ACA and improved customer service through business 
process re-engineering in the local offices. Prior to that, she was a 
staff attorney at the Sargent Shriver National Center on Poverty Law, 
where she focused on public benefits.
    Wagner received her B.S. from the University of Wisconsin--Madison 
and her J.D. from the Northwestern University School of Law.
Appendix B: Materials Cited in Comments
    Appendix B includes all of the resources and materials cited in 
these comments to help ensure that the Department will have complete 
and simple access to the relevant research. The documents in Appendix B 
are listed in alphabetical order by first author's last name or entity 
where appropriate.

    [The documents are retained in Committee file.]

  1.  5 U.S.C., Chapter 6, The Analysis of Regulatory Functions 
            (Regulatory Flexibility Act) (https://www.govinfo.gov/
            content/pkg/USCODE-2017-title5/pdf/USCODE-2017-title5-
            partI-chap6.pdf)

  2.  7 CFR, Part 273, Certification of Eligible Households (https://
            www.govinfo.gov/content/pkg/CFR-2018-title7-vol4/pdf/CFR-
            2018-title7-vol4-part273.pdf) \1\
---------------------------------------------------------------------------
    \1\ This includes 7 CFR  273.24.

  3.  Regulatory Impact Analysis, 7 CFR Part 273 for Supplemental 
            Nutrition Assistance Program: Requirements for Able-Bodied 
            Adults Without Dependents, FNS-2018-0004 (https://
---------------------------------------------------------------------------
            www.regulations.gov/document?D=FNS-2018-0004-6000)

  4.  Abraham, Katharine G., John C. Haltiwanger, Kristin Sandusky, 
            James Spletzer, The Consequences of Long-Term Unemployment: 
            Evidence from Linked Survey and Administrative Data, 
            National Bureau of Economic Research, NBER Working Paper 
            22665, September 2016 (https://www.nber.org/papers/w22665)

  5.  Andersson, Fredrik, John C. Haltiwanger, Mark J. Kutzbach, Henry 
            O. Pollakowski, Daniel H. Weinberg, Job Displacement and 
            the Duration of Joblessness: The Role of Spatial Mismatch, 
            National Bureau of Economic Research, NBER Working Paper 
            20066, April 2014 (https://www.nber.org/papers/w20066)

  6.  Austin, Benjamin, Edward Glaeser, Lawrence Summers, Jobs for the 
            Heartland: Place-Based Policies in 21st-Century America 
            Brookings Papers on Economic Activity, Spring 2018, p. 151-
            255 (https://www.brookings.edu/wp-content/uploads/2018/03/
            AustinEtAl_Text.pdf)

  7.  Bartik, Timothy J., How Do the Effects of Local Growth on 
            Employment Rates Vary With Initial Labor Market Conditions? 
            Upjohn Institute Staff Working Paper 09-148, November 4, 
            2006 (https://research.upjohn.org/up_workingpapers/148/)

  8.  Bauer, Lauren, The Hamilton Project, Workers Could Lose SNAP 
            Benefits Under Trump's Proposed Rule, Brookings Up Front, 
            December 12, 2018 (https://www.brookings.edu/blog/up-front/
            2018/12/20/workers-could-lose-snap-benefits-under-trumps-
            proposed-rule/)

  9.  Bauer, Lauren, Jay Shambaugh, The Hamilton Project, Workers with 
            Low Levels of Education Still Haven't Recovered From the 
            Recession, September 6, 2018 (http://
            www.hamiltonproject.org/blog/employment_rate_gap_workers_
            with_low_levels_of_education_still_havent_recov)

  10.  Bell, Stephen H., and L. Jerome Gallagher, Prime-Age Adults 
            without Children or Disabilities: The ``Least Deserving of 
            the Poor''--or Are They? The Urban Institute, Series B, No. 
            B-26, February 2001 (https://www.urban.org/research/
            publication/prime-age-adults-without-children-or-
            disabilities)

  11.  Bertrand, Marianne, Sendhil Mullainathan, Are Emily and Greg 
            More Employable than Lakisha and Jamal? A Field Experiment 
            on Labor Market Discrimination, National Bureau of Economic 
            Research, NBER Working Paper 9873, July 2003 (https://
            www.nber.org/papers/w9873)

  12.  Bitler, Marianne P., Jonah B. Gelbach, and Hilary W. Hoynes, 
            What Mean Impacts Miss: Distributional Effects of Welfare 
            Reform Experiments, American Economic Review Vol. 96 No. 4, 
            2006, p. 988-1012 (https://www.ssc.wisc.edu/scholz/
            Teaching_742/Bitler_Gelbach_Hoynes.pdf)

  13.  Bloom, Dan, Cynthia Miller, Gilda Azurdia, Results from the 
            Personal Roads to Individual Development and Employment 
            (PRIDE) Program in New York City, MDRC, (https://
            www.mdrc.org/sites/default/files/full_547.pdf)

  14.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Americans at Age 31: Labor Market Activity, Education and 
            Partner Status Summary, Economic News Release, April 17, 
            2018 (https://www.bls.gov/news.release/nlsyth.nr0.htm)

  15.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Regional and State Unemployment, 2018 Annual Average 
            Summary, Economic News Release, February 28, 2019 (https://
            www.bls.gov/news.release/srgune.nr0.htm)

  16.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Persons with a Disability: Labor Force Characteristics, 
            Economic News Release, February 26, 2019 (https://
            www.bls.gov/news.release/disabl.htm) \2\
---------------------------------------------------------------------------
    \2\ This includes Table 1. Employment status of the civilian 
noninstitutional population by disability status and selected 
characteristics, 2018 annual averages.

  17.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, Leave 
            benefits: AccessTable 32. Leave benefits: Access, private 
            industry workers, Employee Benefits Survey, March 2017 
            (https://www.bls.gov/ncs/ebs/benefits/2017/ownership/
---------------------------------------------------------------------------
            private/table32a.htm)

  18.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Employee Tenure in 2018, Economic News Release, September 
            20, 2018 (https://www.bls.gov/news.release/tenure.nr0.htm)

  19.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, The 
            Employment Situation--March 2019, Economic News Release, 
            April 5, 2019 (https://www.bls.gov/news.release/empsit.htm) 
            \3\
---------------------------------------------------------------------------
    \3\ This includes Household Data Table A-4. Employment status of 
the civilian population 25 years and over by educational attainment.

  20.  Birnbaum Emily, and Juliegrace Brufke, Trump attacks Dems on 
            farm bill, The Hill, September 13, 2018 (https://
            thehill.com/homenews/administration/406561-trump-calls-out-
---------------------------------------------------------------------------
            dems-for-opposing-farm-bill-over-work-requirements)

  21.  Brasher, Philip, Farm bill delayed, but Perdue signals 
            administration support, AgriPulse, December 3, 2018 
            (https://www.agri-pulse.com/articles/11703-farm-bill-
            delayed-but-perdue-signals-administration-support)

  22.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Household Data, Annual Averages, 7. Employment status of 
            the civilian noninstitutional population 25 years and over 
            by educational attainment, sex, race, and Hispanic or 
            Latino ethnicity, Labor Force Statistics from the Current 
            Population Survey, January 18, 2019 (https://www.bls.gov/
            cps/cpsaat07.htm)

  23.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Household Data, Not Seasonally Adjusted, Quarterly 
            Averages, E-16. Unemployment rates by age, sex, race, and 
            Hispanic or Latino ethnicity, Labor Force Statistics from 
            the Current Population Survey, January 4, 2019 (https://
            www.bls.gov/web/empsit/cpsee_e16.htm) \4\
---------------------------------------------------------------------------
    \4\ Note: the table on the webpage has been subsequently updated as 
of April 5, 2019.

  24.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, Labor 
            Market Activity, Education, and Partner Status Among 
            America's Young Adults at 29: Results From a Longitudinal 
            Survey, Economic News Release, April 8, 2016 (https://
---------------------------------------------------------------------------
            www.bls.gov/news.release/archives/nlsyth_04082016.pdf)

  25.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, Local 
            Area Unemployment Statistics, (https://www.bls.gov/lau/)

  26.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Alternative Measures of Labor Underutilization for States, 
            2018 Annual Averages, Local Area Unemployment Statistics, 
            (https://www.bls.gov/lau/stalt.htm) \5\
---------------------------------------------------------------------------
    \5\ Note: the table on the webpage has been subsequently updated as 
of April 26, 2019.

  27.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Geographic Concepts, Local Area Unemployment Statistics, 
---------------------------------------------------------------------------
            March 15, 2019 (https://www.bls.gov/lau/laugeo.htm)

  28.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Important Information, Local Area Unemployment Statistics, 
            March 15, 2019 (https://www.bls.gov/lau/launews1.htm) \6\
---------------------------------------------------------------------------
    \6\ Note: the webpage has been subsequently updated as of April 19, 
2019.

  29.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            National Compensation Survey: Employee Benefits in the 
            United States, March 2017, September 2017, Bulletin 2787 
---------------------------------------------------------------------------
            (https://www.bls.gov/ncs/ebs/benefits/2017/ebbl0061.pdf)

  30.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Number of Jobs Held, Labor Market Activity, and Earnings 
            Growth Among the Youngest Baby Boomers: Results from a 
            Longitudinal Survey Summary, Economic News Release, August 
            24, 2017 (https://www.bls.gov/news.release/nlsoy.nr0.htm)

  31.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Employment & Earnings Online, Office of Publications & 
            Special Studies, February 28, 2019 (https://www.bls.gov/
            opub/ee/home.htm)

  32.  U.S. Bureau of Labor Statistics, U.S. Department of Labor, 
            Persons With a Disability: Barriers to Employment, Types of 
            Assistance, and Other Labor-Related Issues--May 2012, 
            Economic News Release, April 24, 2013 (https://www.bls.gov/
            news.release/archives/dissup_04242013.pdf)

  33.  Bolen, Ed, Dorothy Rosenbaum, Stacy Dean, and Brynne-Keith 
            Jennings, More Than 500,000 Adults Will Lose SNAP Benefits 
            in 2016 as Waivers Expire Affected Unemployed Childless 
            Individuals Are Very Poor; Few Qualify for Other Help, 
            Center on Budget and Policy Priorities, March 18, 2016 
            (https://www.cbpp.org/research/food-assistance/more-than-
            500000-adults-will-lose-snap-benefits-in-2016-as-waivers-
            expire)

  34.  Bolen and Stacy Dean, Ed, Waivers Add Key State Flexibility to 
            SNAP's Three-Month Time Limit, Center on Budget and Policy 
            Priorities, February 6, 2018 (https://www.cbpp.org/
            research/food-assistance/waivers-add-key-state-flexibility-
            to-snaps-three-month-time-limit)

  35.  Born, Catherine E., Pamela J. Caudill, Melinda L. Cordero, Life 
            After Welfare: A Look At Sanctioned Families, University of 
            Maryland School of Social Work, November 1999

  36.  Breitwieser, Audrey, Ryan Nunn, Jay Shambaugh, The Hamilton 
            Project, The recent rebound in prime-age labor force 
            participation, Brookings Up Front, August 2, 2018 (https://
            www.brookings.edu/blog/up-front/2018/08/02/the-recent-
            rebound-in-prime-age-labor-force-participation/)

  37.  Butcher, Kristin F. and Diane Whitmore Schanzenbach, Most 
            Workers in Low-Wage Labor Market Work Substantial Hours, in 
            Volatile Jobs SNAP or Medicaid Work Requirements Would Be 
            Difficult for Many Low-Wage Workers to Meet, Center on 
            Budget and Policy Priorities, July 24, 2018 (https://
            www.cbpp.org/research/poverty-and-inequality/most-workers-
            in-low-wage-labor-market-work-substantial-hours-in)

  38.  Butler, Sandra, TANF Time Limits and Maine Families: 
            Consequences of Withdrawing the Safety Net, Maine Equal 
            Justice Partners, March 2013

  39.  Cadena, Brian C., Brian K. Kovak, Immigrants Equilibrate Local 
            Labor Markets: Evidence From the Great Recession, National 
            Bureau of Economic Research, NBER Working Paper 19272, 
            August 2013 (https://www.nber.org/papers/w19272)

  40.  Cajner, Tomaz, Tyler Radler, David Ratner, and Ivan Vidangos, 
            Racial Gaps in Labor Market Outcomes in the Last Four 
            Decades and over the Business Cycle, Finance and Economics 
            Discussion Series, 2017-071, Divisions of Research & 
            Statistics and Monetary Affairs, Federal Reserve Board 2017 
            (https://www.federalreserve.gov/econres/feds/files/
            2017071pap.pdf)

  41.  Cannuscio, Carolyn C., Amy Hillier, Allison Karpyn, Karen Glanz, 
            The social dynamics of healthy food shopping and store 
            choice in an urban environment, Social Science & Medicine, 
            Volume 122, December 2014, p. 13-20 (https://
            www.sciencedirect.com/science/article/abs/pii/
            S0277953614006510)

  42.  Canon, Maria E., Marianna Kudlyak, Guannan Luo, and Marisa Reed, 
            Flows To and From Working Part Time for Economic Reasons 
            and the Labor Market Aggregates During and After the 2007-
            09 Recession, Economic Quarterly, Volume 100, Number 2, 
            Second Quarter 2014, p. 87-111 (https://
            www.richmondfed.org/-/media/richmondfedorg/publications/
            research/economic_quarterly/2014/q2/kudlyak.pdf)

  43.  Carlson, Steven, and Brynne Keith-Jennings, SNAP Is Linked with 
            Improved Nutritional Outcomes and Lower Health Care Costs, 
            Center on Budget and Policy Priorities, January 17, 2018 
            (https://www.cbpp.org/research/food-assistance/snap-is-
            linked-with-improved-nutritional-outcomes-and-lower-health-
            care)

  44.  Carlson, Steven, Dorothy Rosenbaum, and Brynne Keith-Jennings, 
            Who Are the Low-Income Childless Adults Facing the Loss of 
            SNAP in 2016?, Center on Budget and Policy Priorities, 
            February 8, 2016 (https://www.cbpp.org/research/food-
            assistance/who-are-the-low-income-childless-adults-facing-
            the-loss-of-snap-in-2016)

  45.  Letter from Robert Greenstein, Executive Director, CBPP and 
            attached comments, Comments on Rules Proposed to Implement 
            the Food Stamp Penalty Provisions of the 1996 Welfare Law, 
            February 17, 2000, in response to 64 Fed. Reg. No. 242, 
            page 70920

  46.  Center on Budget and Policy Priorities, SNAP Helps 1 in 10 
            Workers in the United States Put Food on the Table, 
            November 2018, Fact Sheet (https://www.cbpp.org/snap-helps-
            low-wage-workers-in-every-state#National)

  47.  Center on Budget and Policy Priorities, SNAP Time Limits: 
            Waivers from the Time Limit Are Back to Historic Norms, 
            March 24, 2017, Fact Sheet (http://www.cbpp.org/sites/
            default/files/atoms/files/3-24-17fa1.pdf)

  48.  Center on Budget and Policy Priorities, States Have Requested 
            Waivers from SNAP's Time Limit in High Unemployment Areas 
            for the Past Two Decades, May 12, 2017, Interactive Map 
            (https://www.cbpp.org/research/food-assistance/states-have-
            requested-waivers-from-snaps-time-limit-in-high-
            unemployment)

  49.  Center on Budget and Policy Priorities, CBPP Summary of Two-Year 
            Waivers from SNAP Three-Month Time Limit, March 2019

  50.  Center on Budget and Policy Priorities, CBPP Summary of Areas 
            That Would Have Lost Their Waivers from the SNAP Three-
            Month Time Limit in 2018 if the Proposed Rule Were 
            Implemented in 2018, March 2019

  51.  Center on Budget and Policy Priorities, Taking Away Medicaid for 
            Not Meeting Work Requirements Harms People Experiencing 
            Homelessness, March 14, 2019, Fact Sheet (https://
            www.cbpp.org/research/health/taking-away-medicaid-for-not-
            meeting-work-requirements-harms-people-experiencing)

  52.  Center on Budget and Policy Priorities, Unemployed adults 
            without children who need help buying food only get SNAP 
            for three months, April 6, 2018 Infographic (https://
            www.cbpp.org/unemployed-adults-without-children-who-need-
            help-buying-food-only-get-snap-for-three-months)

  53.  Chandler, Daniel, Joan Meisel, Pat Jordan, Beth Menees Rienzi, 
            Sandra Naylor Goodwin, Mental Health, Employment, and 
            Welfare Tenure, Journal of Community Psychology, Vol. 33, 
            No. 5, p. 587-609 (https://onlinelibrary.wiley.com/doi/abs/
            10.1002/jcop.20070)

  54.  Cherlin, Andrew J., Karen Bogen, James M. Quane, Linda Burton, 
            Operating within the Rules: Welfare Recipients' Experiences 
            with Sanctions and Case Closings, Social Service Review, 
            Vol. 76, No. 3 (September 2002), p. 387-405 (https://
            www.jstor.org/stable/10.1086/341181)

  55.  Cherlin, Andrew, Linda Burton, Judith Francis, Jane Henrici, 
            Laura Lein, James Quane, and Karen Bogen, Sanctions and 
            Case Closings for Noncompliance: Who Is Affected and Why, 
            Policy Brief 01-1, Johns Hopkins University (http://
            web.jhu.edu/threecitystudy/images/publications/
            09_18058_Welfare_
            Policy_Brief.pdf)

  56.  Ronnie Maurice Stewart, et al., v. Alex M. Azar II, et al., 
            Civil Action No. 1:18-cv-152-JEB, January 24, 2019, Brief 
            for the American Academy of Pediatrics, the American 
            College of Physicians, the American Medical Association, 
            the American Psychiatric Association, the Catholic Health 
            Association of the United States, March of Dimes, and the 
            National Alliance on Mental Illness as Amici Curiae in 
            Support of Plaintiffs (https://www.psychiatry.org/
            File%20Library/Psychiatrists/Directories/Library-and-
            Archive/amicus-briefs/amicus-2019-Stewart-v-Azar-DC-No18-
            152.pdf)

  57.  Coglianese, John, The Rise of In-and-Outs: Declining Labor Force 
            Participation of Prime Age Men, Harvard University (https:/
            /scholar.harvard.edu/files/coglianese/files/
            coglianese_2017_in-and-outs.pdf)

  58.  Colorado Department of Human Services, Colorado SNAP E&T State 
            Plan Federal Fiscal Year 2017, January 10, 2017 (http://
            coemploymentfirst.org/wp-content/uploads/2017/01/Colorado-
            SNAP-E_T-State-Plan-FFY17.pdf)

  59.  Committee of Conference, H.R. 2, Agriculture Improvement Act of 
            2018, 115th Congress, H. Rept. 115-1072, Agriculture 
            Improvement Act of 2018, December 10, 2018 (https://
            www.govinfo.gov/content/pkg/CRPT-115hrpt1072/pdf/CRPT-
            115hrpt1072.pdf) \7\
---------------------------------------------------------------------------
    \7\ Note: the submission is an excerpt from the conference report, 
p. 616-617.

  60.  Congressional Record, Vol. 142, No. 106, Thursday, July 18, 1996 
            (https://www.govinfo.gov/content/pkg/CREC-1996-07-18/pdf/
            CREC-1996-07-18.pdf) \8\
---------------------------------------------------------------------------
    \8\ Note: the submission is an excerpt from the Congressional 
Record, p. H7904 and H7905.

  61.  Congressional Record, Vol. 164, No. 196, Wednesday, December 12, 
            2018 (https://www.govinfo.gov/content/pkg/CREC-2018-12-12/
            pdf/CREC-2018-12-12.pdf) \9\
---------------------------------------------------------------------------
    \9\ Note: the submission is an excerpt from the Congressional 
Record, p. H10149.

  62.  Congressional Record, Vol. 148, No. 67, Wednesday, May 22, 2002 
            (https://www.govinfo.gov/content/pkg/CREC-2002-05-22/pdf/
            CREC-2002-05-22.pdf) \10\
---------------------------------------------------------------------------
    \10\ Note: the submission is an excerpt from the Congressional 
Record, p. H3814 and S4747.

  63.  Lowry, Sean, Donald J. Marples, Tax Incentives for Opportunity 
            Zones: In Brief, Congressional Research Service, R45152, 
---------------------------------------------------------------------------
            November 20, 2018 (https://www.crs.gov/reports/pdf/R45152)

  64.  Cooper, David, Lawrence Mishel, and Ben Zipperer, Bold increases 
            in the minimum wage should be evaluated for the benefits of 
            raising low-wage workers' total earnings--Critics who cite 
            claims of job loss are using a distorted frame, Economic 
            Policy Institute, April 18, 2018 (https://www.epi.org/
            files/pdf/143838.pdf)

  65.  Couch, Kenneth A., Robert Fairlie, Last Hired, First Fired? 
            Black-White Unemployment and the Business Cycle, 
            Demography, Volume 47, Number 1, February 2010, p. 227-247 
            (https://muse.jhu.edu/article/371702)

  66.  Czajka, John L., Sheena McConnell, Scott Cody, Nuria Rodriguez, 
            Imposing a Time Limit on Food Stamp Receipt: Implementation 
            of the Provisions and Effects on Food Stamp Program 
            Participation, Volume I, Final Report, Mathematica Policy 
            Research, September 4, 2001 (https://fns-
            prod.azureedge.net/sites/default/files/abawd.pdf)

  67.  Dagata, Elizabeth, Assessing the Self-Sufficiency of Food Stamp 
            Leavers, Food Assistance & Nutrition Research Program No. 
            26-8, Economic Research Service, U.S. Department of 
            Agriculture, October 2002 (https://www.ers.usda.gov/
            webdocs/publications/46644/31106_fanrr26-8_002.pdf
            ?v=0)

  68.  DeMaster, Dana, At the Limit: December 2006 Minnesota Family 
            Investment Program (MFIP) Cases that Reached the 60 Month 
            Time Limit, Minnesota Department of Human Services, Program 
            Assessment and Integrity Division, January 2008 (https://
            www.leg.state.mn.us/docs/2008/other/080287.pdf)

  69.  Desmond, Matthew, Carl Gershenson, Housing and Employment 
            Insecurity among the Working Poor, Social Problems, 2016, 
            63, p. 46-67 (https://scholar.harvard.edu/files/mdesmond/
            files/desmondgershenson.socprob.2016.pdf)

  70.  Dilger, Robert Jay, Eleanor Blakely, Melissa Latimer, Barry 
            Locke, L. Christopher Plein, Lucinda A. Potter, and David 
            Williams, WV Works 2003: Perspectives of Former Recipients 
            Who Have Exhausted Their 60 Months of Program Eligibility, 
            The West Virginia Public Affairs Reporter, Volume 21, No. 
            2, Summer 2004

  71.  Donovan, Sarah A., An Overview of the Employment-Population 
            Ratio, Congressional Research Service, R44055, May 27, 2015 
            (https://www.crs.gov/reports/pdf/R44055)

  72.  Dranove, David, Craig Garthwaite, and Christopher Ody, Health 
            Spending Slowdown Is Mostly Due To Economic Factors, Not 
            Structural Change In The Health Care Sector, Health 
            Affairs, August 2014, p. 1399-1406 (https://
            www.healthaffairs.org/doi/full/10.1377/hlthaff.2013.1416)

  73.  Economic Innovation Group, From Great Recession To Great 
            Reshuffling: Charting a Decade of Change Across American 
            Communities: Findings from the 2018 Distressed Communities 
            Index, October 2018 (https://eig.org/wp-content/uploads/
            2018/10/2018-DCI.pdf)

  74.  Edin, Kathryn J., H. Luke Shaefer, $2.00 a Day: Living on Almost 
            Nothing in America, 2015

  75.  Edwards, Mark, Colleen Heflin, Peter Mueser, Suzanne Porter, 
            Bruce Weber, The Great Recession and SNAP Caseloads: A Tale 
            of Two States, Rural Studies Program Working Paper Series 
            No. RSP 1401, 2014 (https://ruralstudies.oregonstate.edu/
            sites/default/files/snaptaleof2states_rsp
            workingpaper1401.pdf)

  76.  Falk, Gene, Temporary Assistance for Needy Families (TANF): Size 
            of the Population Eligible for and Receiving Cash 
            Assistance, Congressional Research Service, R44724, January 
            3, 2017 (https://www.crs.gov/reports/pdf/R44724)

  77.  U.S. Department of Labor, Labor Surplus Area Classification 
            under Executive Orders 12073 and 10582, Federal Register, 
            Vol. 78, No. 205, October 23, 2013 p. 63248-63249 (https://
            www.govinfo.gov/content/pkg/FR-2013-10-23/pdf/2013-
            24829.pdf)

  78.  Food and Nutrition Service, U.S. Department of Agriculture, Food 
            Stamp Program; Personal Responsibility Provisions of the 
            Personal Responsibility and Work Opportunity Reconciliation 
            Act of 1996; Final Rule, Federal Register, Vol. 66, No. 11, 
            January 17, 2001, p. 4437-4471 (https://www.govinfo.gov/
            content/pkg/FR-2001-01-17/pdf/01-1025.pdf)

  79.  Food and Nutrition Service, U.S. Department of Agriculture, Food 
            Stamp Program: Food Stamp Provisions of the Balanced Budget 
            Act of 1997, Federal Register, Vol. 64, No. 171, September 
            3, 1999 p. 48246-48258 (https://www.govinfo.gov/content/
            pkg/FR-1999-09-03/pdf/99-23017.pdf)

  80.  Food and Nutrition Service, U.S. Department of Agriculture, Food 
            Stamp Program: Personal Responsibility Provisions of the 
            Personal Responsibility and Work Opportunity Reconciliation 
            Act of 1996, Federal Register, Vol. Vol. 64, No. 242, 
            December 17, 1999, p. 70920-70954 (https://www.govinfo.gov/
            content/pkg/FR-1999-12-17/pdf/99-32527.pdf) \11\
---------------------------------------------------------------------------
    \11\ Note: the submission is a scanned version of the Federal 
Register including marginalia presumably by submitting organization.

  81.  Food and Nutrition Service, U.S. Department of Agriculture, 
            Supplemental Nutrition Assistance Program: Requirements and 
            Services for Able-Bodied Adults Without Dependents; Advance 
            Notice of Proposed Rulemaking, Federal Register, Vol. 83, 
            No. 37, February 23, 2018, p. 8013-8016 (https://
            www.govinfo.gov/content/pkg/FR-2018-02-23/pdf/2018-
---------------------------------------------------------------------------
            03752.pdf)

  82.  Fogli, Alessandra, Enoch Hill, Fabrizio Perri, The Geography of 
            the Great Recession, National Bureau of Economic Research, 
            NBER Working Paper No. 18447, October 2012 (https://
            www.nber.org/papers/w18447.pdf)

  83.  U.S. Department of Agriculture, 2018 USDA Budget Explanatory 
            Notes for Committee on Appropriations, May 25, 2017 (http:/
            /www.obpa.usda.gov/32fnsexnotes2018.pdf) \12\
---------------------------------------------------------------------------
    \12\ Note: the submission is an excerpt from the Explanatory Notes, 
p. 32-92 to 32-93.

  84.  Foote, Andrew, Mark J. Kutzbach, Lars Vilhuber, Recalculating . 
            . . : How Uncertainty in Local Labor Market Definitions 
            Affects Empirical Findings, CES 17-49, Center for Economic 
            Studies, U.S. Census Bureau, U.S. Department of Commerce, 
            August 2017 (https://www2.census.gov/ces/wp/2017/CES-WP-17-
---------------------------------------------------------------------------
            49.pdf)

  85.  Fording, Richard C., Joe Soss, Sanford F. Schram, Devolution, 
            Discretion, and the Effect of Local Political Values on 
            TANF Sanctioning, Social Service Review, June 2007, p. 285-
            316 (http://urban.hunter.cuny.edu/schram/FSSSSR.pdf)

  86.  Freedman, Stephen, Daniel Friedlander, Gayle Hamilton, JoAnn 
            Rock, Marisa Mitchell, Jodi Nudelman, Amanda Schweder, 
            Laura Storto, Evaluating Alternative Welfare-to-Work 
            Approaches: Two-Year Impacts for Eleven Programs, MDRC, 
            June 2000 (https://www.mdrc.org/sites/default/files/
            full_93.pdf)

  87.  Galewitz, Phil, Judge Blocks Kentucky Medicaid Work Requirement, 
            Kaiser Health News, June 29, 2018 (https://khn.org/news/
            judge-blocks-kentucky-medicaid-work-requirement/)

  88.  Garfield, Rachel, Robin Rudowitz, MaryBeth Musumeci, 
            Implications of a Medicaid Work Requirement: National 
            Estimates of Potential Coverage Losses, Issue Brief, Kaiser 
            Family Foundation, June 27, 2018 (http://files.kff.org/
            attachment/Issue-Brief-Implications-of-a-Medicaid-Work-
            Requirement-National-Estimates-of-Potential-Coverage-
            Losses)

  89.  Garfield, Rachel, Robin Rudowitz and Anthony Damico, 
            Understanding the Intersection of Medicaid and Work, Issue 
            Brief, Kaiser Family Foundation, January 5, 2018 (https://
            www.kff.org/medicaid/issue-brief/understanding-the-
            intersection-of-medicaid-and-work/)

  90.  Germanis, Peter, Who Killed Work Requirements for SNAP in the 
            Farm Bill? Answer: Conservative Ideologues, https://
            mlwiseman.com/, January 1, 2019 (https://mlwiseman.com/wp-
            content/uploads/2019/01/Farmbill.120118.pdf)

  91.  Germanis, Peter, How Do the Foundation for Government 
            Accountability's Evaluations of Welfare Reform Measure Up? 
            A Report Card (Hint: The FGA Fails), https://mlwiseman.com/
            , June 24, 2018 (https://mlwiseman.com/wp-content/uploads/
            2016/05/Evaluating-Welfare-Reform.pdf)

  92.  Goetz, Stephan J., Yicheol Han, Identifying Labor Market Areas 
            Based on Link Communities, Agricultural and Applied 
            Economics Association (AAEA), 2015 AAEA & WAEA Joint Annual 
            Meeting, July 26-28, San Francisco, California, (https://
            ageconsearch.umn.edu/record/204870/files/AAEA%20draft_
            Goetz%20and%20Han.pdf)

  93.  Golden, Lonnie, Still falling short on hours and pay: Part-time 
            work becoming new normal, Economic Policy Institute, 
            December 5, 2016 (https://www.epi.org/files/pdf/114028.pdf)

  94.  Gordon, Anne, Susanne James-Burdumy, Renee Loeffler, Barbara 
            Guglielmo, Carole Kuhns, Experiences of Virginia Time Limit 
            Families After Case Closure: An Interim Report: Final 
            Report, Mathematica Policy Research, January 2002 (https://
            www.mathematica-mpr.com/our-publications-and-findings/
            publications/experiences-of-virginia-time-limit-families-
            after-case-closure-interim-report)

  95.  Gray, Colin, Why Leave Benefits on the Table? Evidence from 
            SNAP, Upjohn Institute Working Paper No. 18-288, May 21, 
            2018 (https://research.upjohn.org/up_workingpapers/288/)

  96.  Weiner, Barbara, Summary of the Proposed Rule Implementing 
            Provisions of PRWORA and Published by USDA on December 17, 
            1999 at 64 Fed. Reg. 70920, et seq., Greater Upstate Law 
            Project, January 26, 2000

  97.  Letter from Joanne Werdel, Policy Analyst/Communications 
            Specialist, Center for Civil Justice, February 14, 2000, 
            Re: 7 CFR Part 273; Proposed Rules Implementing Food Stamp 
            Program Personal Responsibility Provisions of the 1996 
            Personal Responsibility and Work Opportunity Reconciliation 
            Act

  98.  Letter from William Waldman, Executive Director, American Public 
            Human Services Association, and attached comments, Food 
            Stamp Program--Personal Responsibility Provisions of the 
            Personal Responsibility and Work Opportunity Reconciliation 
            Act [PRWORA] of 1996, February 15, 2000, in response to 64 
            Fed. Reg. No. 242, page 70920

  99.  Letter from William Waldman, Executive Director, American Public 
            Human Services Association, and attached comments, Food 
            Stamp Program--Work Provisions of the Personal 
            Responsibility and Work Opportunity Reconciliation Act 
            [PRWORA] of 1996, February 22, 2000, in response to 64 Fed. 
            Reg. No. 246, page 72196

  100.  H.R. 2, Agriculture Improvement Act of 2018, Enrolled, 115th 
            Congress, December 12, 2018 (https://www.congress.gov/115/
            bills/hr2/BILLS-115hr2
            enr.pdf) \13\
---------------------------------------------------------------------------
    \13\ Note: the submission is an excerpt from the enrolled bill, p. 
138-145.

  101.  H.R. 2, Agriculture Improvement Act of 2018, Engrossed in 
            House, 115th Congress, June 21, 2018 (https://
            www.congress.gov/115/bills/hr2/BILLS-115hr2enr.pdf) \14\
---------------------------------------------------------------------------
    \14\ Note: the submission is an excerpt from the engrossed bill, p. 
266-289.

  102.  House Committee on the Budget, 104th Congress, H. Rept. 104-
            651, Welfare and Medicaid Reform Act of 1996, June 27, 1996 
            (https://www.govinfo.gov/content/pkg/CRPT-104hrpt651/pdf/
            CRPT-104hrpt651.
            pdf) \15\
---------------------------------------------------------------------------
    \15\ Note: the submission is an excerpt from the report, p. 67.

  103.  Hamilton, Gayle, Stephen Freedman, Lisa Gennetian, Charles 
            Michalopoulos, Johanna Walter, Diana Adams-Ciardullo, Anna 
            Gassman-Pines, Sharon McGroder, Martha Zaslow, Surjeet 
            Ahluwalia, Jennifer Brooks, How Effective Are Different 
            Welfare-to-Work Approaches? Five-Year Adult and Child 
            Impacts for Eleven Programs, MDRC, December 2001 (https://
            www.mdrc.org/publication/how-effective-are-different-
            welfare-work-approaches/file-full) \16\
---------------------------------------------------------------------------
    \16\ Note: the submission is an excerpt from the report Appendix 
Table C.1, Impacts on Employment in Years 1 to 5, p. 352-354; and 
Appendix Table C.6, Impacts on Longer-Term Employment Stability and 
Earnings Growth, p. 363.

  104.  Hartley, Robert Paul, Christopher Wimer, and Jane Waldfogel, 
            Limiting States' Ability to Waive Federal SNAP Work 
            Requirements: A Closer Look at the Potential Implications, 
            Poverty & Social Policy Brief, Vol. 3 No. 4, March 25, 2019 
            (https://www.povertycenter.columbia.edu/s/Poverty-Social-
            Policy-Brief-3_4_-SNAP-Work-Requirements-and-Unemployment-
---------------------------------------------------------------------------
            zj4a.pdf)

  105.  Hasenfeld, Yeheskel, Toorjo Ghose, Kandyce Larson, The Logic of 
            Sanctioning Welfare Recipients: An Empirical Assessment, 
            Social Service Review, Vol. 78, No. 2, June 2004 p. 304-319 
            (https://www.jstor.org/stable/10.1086/382771)

  106.  Haskins, Ron, TANF at Age 20: Work Still Works, Journal of 
            Policy Analysis and Management, Vol. 35, Issue 1, November 
            2015, p. 224-231 (https://onlinelibrary.wiley.com/doi/full/
            10.1002/pam.21878)

  107.  Head, MPP, Valerie, Catherine E. Born, Ph.D., Pamela C. 
            Ovwigho, Ph.D., Criminal History as an Employment Barrier 
            for TANF Recipients, Family Welfare Research and Training 
            Group, School of Social Work, University of Maryland, March 
            2009 (https://pdfs.semanticscholar.org/2c12/a0663
            bd8249586c000a82122cc2e6c796e23.pdf)

  108.  Herd, Dean, Ernie Lightman, Andrew Mitchell, Welfare Time 
            Limits: Symbolism and Practice, Canadian Review of Social 
            Policy, 2007 No. 59, p. 3-19 (https://
            crsp.journals.yorku.ca/index.php/crsp/article/view/34141)

  109.  Hertz, Tom, Lorin Kusmin, Alex Marre, Tim Parker, Rural 
            Employment Trends in Recession and Recovery, Economic 
            Research Service, U.S. Department of Agriculture, August 
            2014 (https://www.ers.usda.gov/webdocs/publications/45258/
            48731_err172.pdf?v=0)

  110.  Hetling, Andrea, The Importance of Region and State Welfare 
            Rules for Disconnected Single Mothers, University of 
            Kentucky, Center for Poverty Research, Discussion Paper 
            Series, DP2011-04, September 2011 (https://
            uknowledge.uky.edu/cgi/
            viewcontent.cgi?article=1040&context=ukcpr_papers)

  111.  Hetling, Ph.D., Andrea, Kathryn W. Patterson, Catherine E. 
            Born, Ph.D., The TANF Time Limit: Comparing Long-term and 
            Other Welfare Leavers, Family Welfare Research and Training 
            Group, School of Social Work, University of Maryland, 
            February 2006 (https://familywelfare.umaryland.edu/
            reports1/timelimitleavers.pdf)

  112.  Hill, Heather D., Paid Sick Leave and Job Stability, Work and 
            Occupations, May 1, 2013 40(2), 143-173 (https://doi.org/
            10.1177/073088
            8413480893)

  113.  Hotz, V. Joseph, Guido W. Imbens, Jacob A. Klerman, Evaluating 
            the Differential Effects of Alternative Welfare-to-Work 
            Training Components: A Reanalysis of the California GAIN 
            Program, Journal of Labor Economics, 2006, vol. 24, no. 3, 
            521-566 (https://www.journals.uchicago.edu/doi/pdfplus/
            10.1086/505050)

  114.  Hoynes, Hilary Williamson, Local Labor Markets and Welfare 
            Spells: Do Demand Conditions Matter?, Review of Economics 
            and Statistics, August 2000, 82(3): 351-368 (https://
            www.jstor.org/stable/2646797)

  115.  Hoynes, Hilary W., Diane Whitmore Schanzenbach, Safety Net 
            Investments in Children, Brookings Papers on Economic 
            Activity, Spring 2018, p. 89-150 (https://
            www.brookings.edu/wp-content/uploads/2018/03/Hoynes
            Schanzenbach_Text.pdf)

  116.  Hoynes, Hilary, Douglas L. Miller, and Jessamyn Schaller, Who 
            Suffers During Recessions?, Journal of Economic 
            Perspectives, Volume 26, Number 3, Summer 2012, p. 27-48 
            (https://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.26.3.27)

  117.  Ingram, Jonathan, Nic Horton, The Power of Work: How Kansas' 
            Welfare Reform Is Lifting Americans Out of Poverty, 
            Foundation for Government Accountability, February 16, 2016 
            (https://thefga.org/wp-content/uploads/2016/02/Kansas-
            study-paper.pdf)

  118.  Irwin, Neil, How Low Can Unemployment Really Go? Economists 
            Have No Idea, The New York Times, February 28, 2018 
            (https://www.nytimes.com/2018/02/28/upshot/how-low-can-
            unemployment-really-go-economists-have-no-idea.html)

  119.  Jensen, Helen H., Steven B. Garasky, Cory Robert Wessman, Sarah 
            M. Nusser, A Study of Households in Iowa that Left the Food 
            Stamp Program, CARD Reports and Working Papers, 02-SR 97, 
            March 2002 (https://lib.dr.iastate.edu/card_staffreports/
            21/)

  120.  Jin, Jangik, Kurt Paulsen, Does accessibility matter? 
            Understanding the effect of job accessibility on labour 
            market outcomes, Urban Studies, 2018, Vol. 55(1) p. 91-115 
            (https://journals.sagepub.com/doi/abs/10.1177/004209
            8016684099)

  121.  Johnson, Rucker C., Landing a job in urban space: The extent 
            and effects of spatial mismatch, Regional Science and Urban 
            Economics, 36 (2006) 331-372 (https://
            www.sciencedirect.com/science/article/pii/S016604620
            5000888)

  122.  Kabbani, Nader S., Parke E. Wilde, Short Recertification 
            Periods in the U.S. Food Stamp Program, Journal of Human 
            Resources, Vol. 38, Special Issue on Income Volatility and 
            Implications for Food Assistance Programs, 2003, p. 1112-
            1138 (https://www.jstor.org/stable/3558983)

  123.  Kalil, Ariel, Kristin S. Seefeldt, Hui-chen Wang, Sanctions and 
            Material Hardship under TANF, Social Service Review, 
            December 2002, p. 642-662 (https://www.jstor.org/stable/
            10.1086/342998)

  124.  Keith-Jennings, Brynne, Raheem Chaudhry, Most Working-Age SNAP 
            Participants Work, But Often in Unstable Jobs, Center on 
            Budget and Policy Priorities, March 15, 2018 (https://
            www.cbpp.org/sites/default/files/atoms/files/3-15-18fa.pdf)

  125.  Keith-Jennings, Brynne, Vincent Palacios, SNAP Helps Millions 
            of Low-Wage Workers: Crucial Financial Support Assists 
            Workers in Jobs with Low Wages, Volatile Income, and Few 
            Benefits, Center on Budget and Policy Priorities, May 10, 
            2017 (https://www.cbpp.org/sites/default/files/atoms/files/
            5-10-17fa.pdf)

  126.  Kesavan, Saravanan, Camelia M. Kuhnen, Demand fluctuations, 
            precarious incomes, and employee turnover, University of 
            North Carolina at Chapel Hill, Kenan-Flagler Business 
            School, Kenan Institute of Private Enterprise, May 2017 
            (http://public.kenan-flagler.unc.edu/faculty/kuhnenc/
            research/ke
            savan_kuhnen.pdf)

  127.  Khan, Romana, Peter F. Orazem, Daniel M. Otto, Deriving 
            Empirical Definitions of Spatial Labor Markets: The Roles 
            of Competing Versus Complementary Growth, Journal of 
            Regional Science, Vol. 41, No. 4, 2001, p. 735-756 (https:/
            /onlinelibrary.wiley.com/doi/abs/10.1111/0022-4146.00241)

  128.  Kneebone, Elizabeth, Natalie Holmes, The growing distance 
            between people and jobs in metropolitan America, Brookings 
            Institution, March 24, 2015 (https://www.brookings.edu/wp-
            content/uploads/2016/07/Srvy_JobsProx
            imity.pdf)

  129.  Kogan, Deborah, Anne Paprocki, Hannah Diaz, Supplemental 
            Nutrition Assistance Program (SNAP) Employment and Training 
            (E&T) Best Practices Study: Final Report, Social Policy 
            Research Associates, November 22, 2016 (https://
            www.fns.usda.gov/sites/default/files/ops/SNAPEandTBestPrac
            tices.pdf)

  130.  Larson, Anita M., Shweta Singh, Crystal Lewis, Sanctions and 
            Education Outcomes for Children in TANF Families, Child & 
            Youth Services, 32: 2011, 180-199 (https://
            www.tandfonline.com/doi/abs/10.1080/0145935X.2011.
            605305#.Uu6IlPldV8E)

  131.  Larson, Ph.D., M.P.H., RD, Nicole I., Mary T. Story, Ph.D., RD, 
            Melissa C. Nelson, Ph.D., RD, Neighborhood Environments: 
            Disparities in Access to Healthy Foods in the U.S., 
            American Journal of Preventive Medicine, Volume 36, Issue 
            1, January 2009, Pages 74-81 (https://www.ajpmonline.org/
            article/S0749-3797(08)00838-6/pdf)

  132.  Laska, Melissa Nelson, Kelley E. Borradaile, June Tester, Gary 
            D. Foster, Joel Gittelsohn, Healthy food availability in 
            small urban food stores: a comparison of four U.S. cities, 
            Public Health Nutrition, 13(7), July 2010, 1031-1035 
            (https://doi.org/10.1017/S1368980009992771)

  133.  Lee, Bong Joo, Kristen S. Slack, Dan A. Lewis, Are Welfare 
            Sanctions Working As Intended? Welfare Receipt, Work 
            Activity, and Material Hardship among TANF-Recipient 
            Families, Social Service Review, Vol. 78, No. 3, September 
            2004, 370-403 (https://www.journals.uchicago.edu/doi/full/
            10.1086/421918)

  134.  Lee, Kyoung Hag, Effect of Lifetime Limits and Differences 
            between TANF Leavers Who Had Reached Their Lifetime Limits 
            and Those Who Had Exited Voluntarily, Poverty & Public 
            Policy, Vol. 2, Iss. 4, Article 3, August 10, 2012, p. 27-
            48 (https://doi.org/10.2202/1944-2858.1072)

  135.  Leon, Carol Boyd, The employment-population ratio: its value in 
            labor force analysis, Monthly Labor Review, Bureau of Labor 
            Statistics, Office of Current Employment Analysis, February 
            1981, p. 36-45 (https://www.bls.gov/opub/mlr/1981/02/
            art4full.pdf)

  136.  Leparulo, Paul A., Amanda K. Rector, Preliminary Analysis of 
            Work Requirement Policy on the Wage and Employment 
            Experiences of ABAWDs in Maine, Maine Governor's Office of 
            Policy and Management, Paper 1, April 19, 2016 (https://
            digitalmaine.com/ogvn_policy/1/)

  137.  London, Rebecca A., Jane G. Mauldon, Time Running Out: A 
            Portrait of California Families Reaching the CalWORKs 60-
            Month Time Limit in 2004, Welfare Policy Research Project, 
            Policy Brief, November 2006 (https://escholarship.org/uc/
            item/37g3510z)

  138.  Long, MA, M.P.P., David, John Rio, MA, CRC, Jeremy Rosen, 
            Employment and Income Supports for Homeless People, 2007 
            National Symposium on Homelessness Research, September 2007 
            (https://www.huduser.gov/portal//publications/pdf/p11.pdf)

  139.  Loprest, Pamela, Elaine Maag, Barriers to and Supports for Work 
            Among Adults With Disabilities: Results from the NHIS-D, 
            The Urban Institute, October 2001 (https://aspe.hhs.gov/
            basic-report/barriers-and-supports-work-among-adults-
            disabilities-results-nhis-d)

  140.  Mabli, James, SNAP Participation, Food Security, and Geographic 
            Access to Food, Mathematica Policy Research, March 2014 
            (https://fns-prod.azureedge.net/sites/default/files/
            SNAPFS_FoodAccess.pdf)

  141.  Maine Department of Health and Human Services. Media Advisory: 
            LePage Administration Welfare Reform Leads to Increased 
            Wages, Press Release, May 11, 2016 and attached study, 
            Preliminary analysis of work requirement policy on the wage 
            and employment experiences of ABAWDs in Maine (https://
            www.maine.gov/dhhs/press-release/DHHS-OPM-Analyses.pdf) 
            \17\
---------------------------------------------------------------------------
    \17\ Note: the Media Advisory part of the submission is a scanned 
version of the Media Advisory including marginalia presumably by the 
submitting organization.

  142.  Mallik-Kane, Kamala, Christy A. Visher, Health and Prisoner 
            Reentry: How Physical, Mental, and Substance Abuse 
            Conditions Shape the Process of Reintegration, Urban 
            Institute, February 2008 (https://www.urban.org/sites/
            default/files/publication/31491/411617-Health-and-Prisoner-
---------------------------------------------------------------------------
            Reentry.PDF)

  143.  Mann, Cindy, April Grady, Potential Enrollment Impacts of 
            Michigan's Medicaid Work Requirement, Manatt, Phelps & 
            Phillips, LLP, February 6, 2019 (https://www.manatt.com/
            Manatt/media/Media/Images/White%20
            Papers/Manatt_MI-Work-Req-Estimates_20190206-Final.pdf)

  144.  Marr, Chuck, Bryann DaSilva, Childless Adults Are Lone Group 
            Taxed Into Poverty: Expanding Earned Income Tax Credit 
            Would Address Problem, Center on Budget and Policy 
            Priorities, April 19, 2016 (https://www.cbpp.org/sites/
            default/files/atoms/files/3-2-16tax.pdf)

  145.  Maryland Department of Human Resources, Maryland Supplemental 
            Nutrition Assistance Program (SNAP) Employment and Training 
            (E&T) Program: State Plan of Operations, October 1, 2015 
            through September 30, 2016, September 22, 2015 (https://
            dhr.maryland.gov/documents/Data%20and%20
            Reports/FIA/YR2016%20SNAP%20E&T%20State%20Plan%20of%20Opera
            tions%20(revised).pdf)

  146.  Matthews, MS, Kevin A., Janet B. Croft, Ph.D., Yong Liu, MD, 
            MS, Hua Lu, MS, Dafna Kanny, Ph.D., Anne G. Wheaton, Ph.D., 
            Timothy J. Cunningham, Sc.D., Laura Kettel Khan, Ph.D., 
            Ralph S. Caraballo, Ph.D., James B. Holt, Ph.D., Paul I. 
            Eke, Ph.D., M.P.H., Wayne H. Giles, MD, MS, Health-Related 
            Behaviors by Urban-Rural County Classification--United 
            States, 2013, Morbidity and Mortality Weekly Report, 
            Surveillance Summaries, Vol. 66, No. 5, February 3, 2017 
            (https://www.cdc.gov/mmwr/volumes/66/ss/pdfs/
            ss6605.pdf)\18\
---------------------------------------------------------------------------
    \18\ Note: the submission does not include the erratum (https://
www.cdc.gov/mmwr/volumes/66/wr/mm6607a5.htm?s_cid=mm6607a5_w) published 
for this report. However, the report in Committee file does contain the 
erratum.

  147.  McHugh, M.A., M.P.P., Christina, J. Taylor Danielson, Ph.D., 
            TANF Time Limit Analysis: Comparing Cases Closed Due to 
            Time Limits with Other Case Closures, Washington State 
---------------------------------------------------------------------------
            Department of Social and Health Services, February 2019

  148.  Mills, Gregory, Robert Kornfeld, Study of Arizona Adults 
            Leaving the Food Stamp Program, Final Report, Abt 
            Associates Inc., E-FAN-01-001, December 2000 (http://
            azmemory.azlibrary.gov/digital/collection/feddocs/id/1050/
            rec/efan-01-001.pdf)

  149.  Mitchell, Tazra, LaDonna Pavetti, and Yixuan Huang, Study 
            Praising Kansas' Harsh TANF Work Penalties Is Fundamentally 
            Flawed, Center on Budget and Policy Priorities, February 
            16, 2018 (https://www.cbpp.org/sites/default/files/atoms/
            files/1-23-18tanf_rebuttal_0.pdf) \19\
---------------------------------------------------------------------------
    \19\ Note: there is a discrepancy with the date of the report as 
posted on the CBPP website (https://www.cbpp.org/research/family-
income-support/study-praising-kansas-harsh-tanf-work-penalties-is-
fundamentally) and the pdf of the report. The website lists the report 
as being updated on February 20, 2018, while the pdf of the report is 
dated as February 16, 2018.

  150.  Mitchell, Tazra, LaDonna Pavetti, and Yixuan Huang, Life After 
            TANF in Kansas: For Most, Unsteady Work and Earnings Below 
            Half the Poverty Line, Center on Budget and Policy 
            Priorities, February 20, 2018 (https://www.cbpp.org/sites/
---------------------------------------------------------------------------
            default/files/atoms/files/1-23-18kstanf.pdf)

  151.  Modestino, Alicia Sasser, Daniel Shoag, Joshua Ballance, 
            Upskilling: Do Employers Demand Greater Skill When Skilled 
            Workers are Plentiful?, Harvard Kennedy School, Taubman 
            Center for State and Local Government, Policy Brief, May 
            2015 (https://scholar.harvard.edu/files/shoag/files/
            modestino_shoag_and_ballance_022316_final.pdf)

  152.  Moffitt, Robert A., The Deserving Poor, the Family, and the 
            U.S. Welfare System, Demography (2015) 52: p. 729-749 
            (https://www.jstor.org/stable/43699160)

  153.  Musumeci, MaryBeth, Julia Foutz, Rachel Garfield, How Might 
            Medicaid Adults with Disabilities Be Affected By Work 
            Requirements in Section 1115 Waiver Programs?, Kaiser 
            Family Foundation, Issue Brief, January 26, 2018 (http://
            files.kff.org/attachment/Issue-Brief-How-Might-Medicaid-
            Adults-with-Disabilities-Be-Affected-By-Work-Requirements)

  154.  Executive Office of the President, Executive Order 12073--
            Federal procurement in labor surplus areas, August 16, 1978 
            (https://www.govinfo.gov/content/pkg/FR-1978-08-18/pdf/FR-
            1978-08-18.pdf) \20\
---------------------------------------------------------------------------
    \20\ Note: the submission is a pdf of the National Archives webpage 
(https://www.archives.gov/federal-register/codification/executive-
order/12073.html). In addition to the pdf of the webpage a copy of the 
actual Federal Register publishing of the EO is also included in 
Committee file.

  155.  National Bureau of Economic Research, U.S. Business Cycle 
            Expansions and Contractions (https://www.nber.org/
---------------------------------------------------------------------------
            cycles.html)

  156.  National Research Council, Chapter 8, Consequences for 
            Employment and Earnings, in The Growth of Incarceration in 
            the United States: Exploring Causes and Consequences 
            (2014), The National Academies Press, 2014 (https://
            www.nap.edu/catalog/18613/the-growth-of-incarceration-in-
            the-united-states-exploring-causes) \21\
---------------------------------------------------------------------------
    \21\ Note: the submission is an excerpt from the book, p. 233-259.

  157.  Donovan, Shaun, Director, Office of Management and Budget, 
            Revised Delineations of Metropolitan Statistical Areas, 
            Micropolitan Statistical Areas, and Combined Statistical 
            Areas, and Guidance on Uses of the Delineations of These 
            Areas, OMB Bulletin No. 15-01, July 15, 2015 (https://
            obamawhitehouse.archives.gov/sites/default/files/omb/
---------------------------------------------------------------------------
            bulletins/2015/15-01.pdf)

  158.  Oggins, Jean, Amy Fleming, Welfare Reform Sanctions and 
            Financial Strain in a Food-Pantry Sample, Journal of 
            Sociology & Social Welfare, Volume 28, Issue 2, June 2001, 
            p. 101-123 (https://scholarworks.wmich.edu/jssw/vol28/iss2/
            7/)

  159.  Ohio Association of Foodbanks, A Comprehensive Assessment of 
            Able-Bodied Adults Without Dependents and Their 
            Participation in the Work Experience Program in Franklin 
            County, Ohio, 2014, December 8, 2014 (http://
            admin.ohiofoodbanks.org/uploads/news/WEP-2013-2014-
            report.pdf)

  160.  Ohio Association of Foodbanks, A Comprehensive Assessment of 
            Able-Bodied Adults Without Dependents and Their 
            Participation in the Work Experience Program in Franklin 
            County, Ohio, 2015

  161.  Ohio Association of Foodbanks, Franklin County Work Experience 
            Program: Comprehensive Report Able-Bodied Adults Without 
            Dependents 2014-2015, October 14, 2015 (http://
            ohiofoodbanks.org/wep/WEP-2013-2015-report.pdf)

  162.  Ovwigho, Ph.D., Pamela C., Kathryn Patterson, BFA, Catherine E. 
            Born, Ph.D., The TANF Time Limit: Barriers & Outcomes Among 
            Families Reaching the Limit, Family Welfare Research and 
            Training Group, School of Social Work, University of 
            Maryland, November 2007 (https://
            familywelfare.umaryland.edu/reports1/tl_barriers.pdf)

  163.  104th Congress of the United States, Public Law 104-193, 
            Personal Responsibility and Work Opportunity Reconciliation 
            Act of 1996 (https://www.govinfo.gov/content/pkg/PLAW-
            104publ193/pdf/PLAW-104publ193.
            pdf)

  164.  105th Congress of the United States, Public Law 105-33, 
            Balanced Budget Act of 1997 (https://www.govinfo.gov/
            content/pkg/PLAW-105publ33/pdf/PLAW-105publ33.pdf) \22\
---------------------------------------------------------------------------
    \22\ Note: the submission is an excerpt from the Public Law, p. 111 
Stat. 251-111 Stat. 252.

  165.  110th Congress of the United States, Public Law 110-246, Food, 
            Conservation, and Energy Act of 2008 (https://
            www.govinfo.gov/content/pkg/PLAW-110publ246/pdf/PLAW-
            110publ246.pdf) \23\
---------------------------------------------------------------------------
    \23\ Note: the submission is an excerpt from the Public Law, p. 122 
Stat. 1872-122 Stat. 1873.

  166.  Office of the Legislative Counsel, U.S. House of 
            Representatives, Statute Compilation: Food and Nutrition 
            Act of 2008, January 15, 2019 (https://www.govinfo.gov/
---------------------------------------------------------------------------
            content/pkg/COMPS-10331/pdf/COMPS-10331.pdf)

  167.  Pager, Devah, The Mark of a Criminal Record, American Journal 
            of Sociology, Volume 108, Number 5, March 2003, p. 937-975 
            (https://scholar.harvard.edu/files/pager/files/
            pager_ajs.pdf)

  168.  Pager, Devah, Bruce Western, Identifying Discrimination at 
            Work: The Use of Field Experiments, Journal of Social 
            Issues, Vol. 68, No. 2, 2012, p. 221-237 (https://
            scholar.harvard.edu/files/pager/files/pdf_5.pdf)

  169.  Pager, Devah, Bruce Western, Bart Bonikowski, Discrimination in 
            a Low-Wage Labor Market: A Field Experiment, American 
            Sociological Review, 2009, Vol. 74 October 2009, p. 777-799

  170.  Pavetti, LaDonna A., Jacqueline Kauff, When Five Years Is Not 
            Enough: Identifying and Addressing the Needs of Families 
            Nearing the TANF Time Limit in Ramsey County, Minnesota, 
            Mathematica Policy Research, March 30, 2006 (https://
            www.mathematica-mpr.com/our-publications-and-findings/
            publications/when-five-years-is-not-enough-identifying-and-
            addressing-the-needs-of-families-nearing-the-tanf-time-
            limit-in-ramsey-county-minnesota)

  171.  Pavetti, LaDonna, Michelle K. Derr, Heather Hesketh Zaveri, 
            Review of Sanction Policies and Research Studies: Final 
            Literature Review, Mathematica Policy Research, March 10, 
            2003 (https://www.mathematica-mpr.com/our-publications-and-
            findings/publications/review-of-sanction-policies-and-
            research-studies-final-literature-review)

  172.  Pavetti, LaDonna, Michelle K. Derr, Gretchen Kirby, Robert G. 
            Wood, Melissa A. Clark, The Use of TANF Work-Oriented 
            Sanctions in Illinois, New Jersey, and South Carolina: 
            Final Report, Mathematica Policy Research, April 30, 2004 
            (https://www.mathematica-mpr.com/our-publications-and-
            findings/publications/the-use-of-tanf-workoriented-
            sanctions-in-illinois-new-jersey-and-south-carolina)

  173.  Paxson, Christina, Jane Waldfogel, Welfare Reforms, Family 
            Resources, and Child Maltreatment, Journal of Policy 
            Analysis and Management, Vol. 22, No. 1, Winter, 2003, p. 
            85-113 (https://www.jstor.org/stable/3325847)

  174.  Pew Research Center, The State of American Jobs: How the 
            shifting economic landscape is reshaping work and society 
            and affecting the way people think about the skills and 
            training they need to get ahead., October 2016 (https://
            www.pewsocialtrends.org/wp-content/uploads/sites/3/2016/10/
            ST_2016.10.06_Future-of-Work_FINAL4.pdf)

  175.  Pinard, Ph.D., Courtney A., Carmen Byker Shanks, Amy Yaroch, An 
            integrative literature review of small food store research 
            across urban and rural communities in the U.S., Preventive 
            Medicine Reports, 3, April 2, 2016, p. 324-332 (https://
            www.ncbi.nlm.nih.gov/pmc/articles/PMC4929238/pdf/main.pdf)

  176.  Powell, Lisa M., Sandy Slater, Donka Mirtcheva, Yanjun Bao, 
            Frank J. Chaloupka, Food store availability and 
            neighborhood characteristics in the United States, 
            Preventive Medicine, 44, March 2007, p. 189-195 (https://
            doi.org/10.1016/j.ypmed.2006.08.008)

  177.  Quilliana, Lincoln, Devah Pager, Ole Hexela, Arnfinn H. 
            Midtboenf, Meta-analysis of field experiments shows no 
            change in racial discrimination in hiring over time, 
            Proceedings of the National Academy of Sciences, October 
            10, 2017 (https://www.pnas.org/content/114/41/10870)

  178.  Rangarajan, Anu, Philip M. Gleason, Food Stamp Leavers in 
            Illinois: How Are They Doing Two Years Later? Final Report, 
            Mathematica Policy Research, January 30, 2001 (https://
            www.mathematica-mpr.com/our-publications-and-findings/
            publications/food-stamp-leavers-in-illinois-how-are-they-
            doing-two-years-later)

  179.  Ratcliffe, Michael, Charlynn Burd, Kelly Holder, Alison Fields, 
            Defining Rural at the U.S. Census Bureau, ACSGEO-1, U.S. 
            Census Bureau, U.S. Department of Commerce, December 8, 
            2016 (https://www.census.gov/content/dam/Census/library/
            publications/2016/acs/acsgeo-1.pdf)

  180.  Rector, Robert, Rachel Sheffield, Kevin D. Dayaratna, Jamie 
            Bryan Hall, Maine Food Stamp Work Requirement Cuts Non-
            Parent Caseload by 80 Percent, Backgrounder, No. 3091, 
            Heritage Foundation, February 8, 2016 (https://
            www.heritage.org/sites/default/files/2018-04/BG3091.pdf)

  181.  Reichman, Nancy E., Julien O. Teitler, Marah A. Curtis, TANF 
            Sanctioning and Hardship, Social Service Review, 79, no. 2, 
            June 2005, p. 215-236 (https://www.journals.uchicago.edu/
            doi/abs/10.1086/428918)

  182.  Governor's Workforce Board, Regional Planning Policy, State of 
            Rhode Island, June 2017 (https://gwb.ri.gov/wp-content/
            uploads/2017/06/17-01-3-16-2017.pdf?189db0)

  183.  White House, Remarks by President Trump at Signing of H.R. 2, 
            the Agriculture Improvement Act of 2018, press release, 
            December 20, 2018 (https://www.whitehouse.gov/briefings-
            statements/remarks-president-trump-signing-h-r-2-
            agriculture-improvement-act-2018/)

  184.  Ribar, David C., Marilyn Edelhoch, Qiduan Liu, Watching the 
            Clocks: The Role of Food Stamp Recertification and TANF 
            Time Limits in Caseload Dynamics, Journal of Human 
            Resources, Vol. 43, No. 1, Winter 2008, p. 208-238 (http://
            jhr.uwpress.org/content/43/1/208.full.pdf+html)

  185.  Richardson, Philip, Gregg Schoenfeld, Susan LaFever, Frances 
            Jackson, Mark Tecco, Food Stamp Leavers Research Study--
            Study of ABAWDs Leaving the Food Stamp Program in South 
            Carolina: Final Report, E-FAN-03-002, MAXIMUS, Inc., March 
            2003 (https://naldc.nal.usda.gov/download/45220/PDF)

  186.  Rosenbaum, Dorothy, Many SNAP Households Will Experience Long 
            Gap Between Monthly Benefits Despite End of Shutdown, 
            Center on Budget and Policy Priorities, February 4, 2019 
            (https://www.cbpp.org/research/food-assistance/many-snap-
            households-will-experience-long-gap-between-monthly-
            benefits)

  187.  Rosenbaum, Dorothy, The Relationship Between SNAP and Work 
            Among Low-Income Households, Center on Budget and Policy 
            Priorities, January 30, 2013 (https://www.cbpp.org/
            research/the-relationship-between-snap-and-work-among-low-
            income-households)

  188.  Rosenbaum, Dorothy, Ed Bolen, SNAP Reports Present Misleading 
            Findings on Impact of Three-Month Time Limit, Center on 
            Budget and Policy Priorities, December 14, 2016 (https://
            www.cbpp.org/research/food-assistance/snap-reports-present-
            misleading-findings-on-impact-of-three-month-time)

  189.  Ross, Martha, Natalie Holmes, Employment by race and place: 
            snapshots of America, Brookings The Avenue, February 27, 
            2017 (https://www.brookings.edu/blog/the-avenue/2017/02/27/
            employment-by-race-and-place-snapshots-of-america/)

  190.  Rothstein, Richard, The Racial Achievement Gap, Segregated 
            Schools, and Segregated Neighborhoods--A Constitutional 
            Insult, Race and Social Problems, November 12, 2014 
            (https://www.epi.org/publication/the-racial-achievement-
            gap-segregated-schools-and-segregated-neighborhoods-a-
            constitutional-insult/)

  191.  Rowe, Gretchen, Elizabeth Brown, Brian Estes, SNAP Employment 
            and Training (E&T) Characteristics Study: Final Report, 
            Mathematica Policy Research, October 31, 2017 (https://
            www.mathematica-mpr.com/our-publications-and-findings/
            publications/snap-employment-and-training-e-t-
            characteristics-study-final-report)

  192.  Scherpf, Erik, Bruce Weber, Deana Grobe, Mark Edwards, 
            Participation in USDA's Supplemental Nutrition Assistance 
            Program (SNAP): Effect of Local Labor Market Conditions in 
            Oregon, ERR-257, Economic Research Service, U.S. Department 
            of Agriculture, September 2018 (https://www.ers.usda.gov/
            publications/pub-details/?pubid=90037)

  193.  Schram, Sanford F., Joe Soss, Richard C. Fording, Linda Houser, 
            Deciding to Discipline: Race, Choice, and Punishment at the 
            Frontlines of Welfare Reform, American Sociological Review, 
            Vol. 74, No. 3, Jun. 2009, p. 398-422 (https://
            www.jstor.org/stable/27736070)

  194.  Schwandt, Hannes, Till von Wachter, Unlucky Cohorts: Estimating 
            the Long-term Effects of Entering the Labor Market in a 
            Recession in Large Cross-sectional Data Sets, Discussion 
            Paper Series, IZA DP No. 11926, IZA--Institute of Labor 
            Economics, October 2018 (http://ftp.iza.org/dp11926.pdf)

  195.  Seefeldt, Kristin S., Sean M. Orzol, Watching the Clock Tick: 
            Factors Associated with TANF Accumulation, National Poverty 
            Center Working Paper Series, No. 04-9, NPC, University of 
            Michigan, May 2004 (Revised May 2005) (http://
            www.npc.umich.edu/publications/workingpaper04/paper9/04-
            09.pdf)

  196.  Seefeldt, Kristin S., Heather Sandstrom, When There Is No 
            Welfare: The Income Packaging Strategies of Mothers Without 
            Earnings or Cash Assistance Following an Economic Downturn, 
            Russell Sage Foundation Journal of the Social Sciences, 
            Vol. 1, No. 1, Severe Deprivation in America, November 
            2015, p. 139-158 (https://www.jstor.org/stable/10.7758/
            rsf.2015.1.1.08)

  197.  Shaefer, H. Luke, Kathryn Edin, Elizabeth Talbert, 
            Understanding the Dynamics of $2-a-Day Poverty in the 
            United States, Russell Sage Foundation Journal of the 
            Social Sciences, Vol. 1, No. 1, Severe Deprivation in 
            America, November 2015, p. 120-138 (https://www.jstor.org/
            stable/10.7758/rsf.2015.1.1.07)

  198.  Shah, MPP, Melissa Ford, Qinghua Liu, Ph.D., David Mancuso, 
            Ph.D., Barbara Felver, MES, MPA, Predicting Homelessness 
            among Low-Income Parents on TANF, Washington State 
            Department of Social and Health Services, August 2015 
            (https://www.dshs.wa.gov/sites/default/files/SESA/rda/
            documents/research-11-224.pdf)

  199.  Sharkey, Ph.D., MPH, RD, Joseph R., Measuring Potential Access 
            to Food Stores and Food-Service Places in Rural Areas in 
            the U.S., American Journal of Preventive Medicine, Volume 
            36, Issue 4, Supplement: Measurement of the Food and 
            Physical Activity Environments: Enhancing Research Relevant 
            to Policy on Diet, Physical Activity, and Weight, April 
            2009, p. S151-S155 (https://www.ajpmonline.org/article/
            S0749-3797(09)00012-9/pdf)

  200.  Sharkey, Joseph R. Scott Horel, Daikwon Han, John C. Huber, 
            Jr., Association between neighborhood need and spatial 
            access to food stores and fast food restaurants in 
            neighborhoods of Colonias, International Journal of Health 
            Geographics, 8:9, February 16, 2009 (https://ij-
            healthgeographics.biomedcentral.com/articles/10.1186/1476-
            072X-8-9)

  201.  Stacy, Christina, Brady Meixell, Serena Lei, Too Far from Jobs: 
            Spatial Mismatch and Hourly Workers, Urban Institute, 
            February 21, 2019 (https://www.urban.org/features/too-far-
            jobs-spatial-mismatch-and-hourly-workers)

  202.  Stavrianos, Michael, Lucia Nixon, The Effect of Welfare Reform 
            on Able-Bodied Food Stamp Recipients, Mathematica Policy 
            Research, July 23, 1998 (https://fns-prod.azureedge.net/
            sites/default/files/finalrep.pdf)

  203.  Stavrianos, Mike, Scott Cody, Kimball Lewis, Characteristics of 
            Childless Unemployed Adult and Legal Immigrant Food Stamp 
            Participants: Fiscal Year 1995, Mathematica Policy 
            Research, February 13, 1997 (https://fns-
            prod.azureedge.net/sites/default/files/ABAWDOAE.PDF)

  204.  Steadman, Ph.D., Henry J., Fred C. Osher, M.D., Pamela Clark 
            Robbins, B.A., Brian Case, B.A., Steven Samuels, Ph.D., 
            Prevalence of Serious Mental Illness Among Jail Inmates, 
            Psychiatric Services, June 2009 Vol. 60 No. 6, p. 761-765 
            (https://ps.psychiatryonline.org/doi/full/10.1176/
            ps.2009.60.
            6.761)

  205.  Stone, Chad, Congress Should Renew Emergency Unemployment 
            Compensation Before the End of the Year, Center on Budget 
            and Policy Priorities, November 20, 2013 (https://
            www.cbpp.org/research/congress-should-renew-emergency-
            unemployment-compensation-before-the-end-of-the-year)

  206.  Letter from Hon. Bill Haslam, Governor, Tennessee and attached 
            maps, May 24, 2018, Re: Realignment of Local Workforce 
            Development Board Areas (https://www.tn.gov/content/dam/tn/
            workforce/documents/ProgramManagement/RealignmentMaps.pdf) 
            \24\
---------------------------------------------------------------------------
    \24\ Note: the submission is an excerpt from the webpage posting 
the Governor's letter (https://www.tn.gov/content/dam/tn/workforce/
documents/ProgramManagement/SignedLetterRe
alignment_May24.pdf).

  207.  ____, TN Realigns Workforce Development Areas, The 
            Chatanoogan.com, June 28, 2018 (https://
            www.chattanoogan.com/2018/6/28/371092/TN-Realigns-
---------------------------------------------------------------------------
            Workforce-Development-Areas.aspx)

  208.  Thiede, Brian C., Shannon M. Monnat, The Great Recession and 
            America's Geography of Unemployment, Demographic Research, 
            Volume 35, Article 30, September 27, 2016, 891-928 (https:/
            /www.ncbi.nlm.nih.gov/pmc/articles/PMC5486972/pdf/
            nihms867734.pdf)

  209.  Tolbert, Charles M., Molly Sizer, U.S. Commuting Zones and 
            Labor Market Areas: A 1990 Update, Staff Reports No. 9614, 
            Economic Research Service, U.S. Department of Agriculture, 
            September 1996 \25\
---------------------------------------------------------------------------
    \25\ Note: the submission does not include the attached 
appendicies, the copy retained in Committee file does.

  210.  Letter from Brian Neale, Director, Centers for Medicare & 
            Medicaid Services, U.S. Department of Health and Human 
            Services, January 11, 2018, Re: Opportunities to Promote 
            Work and Community Engagement Among Medicaid Beneficiaries, 
            SMD: 18-002 (https://www.medicaid.gov/federal-policy-
---------------------------------------------------------------------------
            guidance/downloads/smd18002.pdf)

  211.  U.S. Census Bureau, U.S. Department of Commerce, Food Stamps/
            Supplemental Nutrition Assistance Program (SNAP): 2012-2016 
            American Community Survey 5-Year Estimates (https://
            factfinder.census.gov/faces/tableservices/jsf/pages/
            productview.xhtml?src=bkmk)

  212.  Letter from Arthur T. Foley, Food and Nutrition Service, USDA, 
            January 8, 2009, Re: SNAP--ABAWD Statewide Waivers--New 
            Criteria for Unemployment Insurance Extended Benefits 
            Trigger

  213.  Letter from Mary Ann Ferris, Chief, State Program Improvement, 
            Food Stamp Program, Northeast Region, Food and Nutrition 
            Service, USDA, April 28, 2004, Re: ABAWD Waivers--New 
            Method for Calculating Average Unemployment Rates

  214.  Letter from Lizbeth Silbermann, Director, Program Development 
            Division, Food and Nutrition Service, USDA, May 25, 2018, 
            Re: SNAP--Best Practices and Resources for Informing 
            Households of ABAWD Rules

  215.  Gray, Kelsey Farson, Shivani Kochhar, Characteristics of 
            Supplemental Nutrition Assistance Program Households: 
            Fiscal Year 2014, Report No. SNAP-15-CHAR, Mathematica 
            Policy Research, December 21, 2015 (https://fns-
            prod.azureedge.net/sites/default/files/ops/
            Characteristics2014.pdf)

  216.  Office of the Inspector General, USDA, FNS Controls Over SNAP 
            Benefits For Able-Bodied Adults Without Dependents, Audit 
            Report 27601-0002-31, September 2016 (Released October 24, 
            2016) (https://www.usda.gov/oig/webdocs/27601-0002-31.pdf) 
            \26\
---------------------------------------------------------------------------
    \26\ Note: this document was included in the 27th attachment for 
Appendix B, as well as in the 28th attachment. However, it was not 
listed in the table of contents for attachment 27.

  217.  Facsimile from Office of Advocacy and Enterprise, Food and 
            Nutrition Service, USDA, April 23, 1997, Re: Time Limit 
---------------------------------------------------------------------------
            Waivers for Able-Bodied Food Stamp Recipients

  218.  Cronquist, Kathryn, Sarah Lauffer, Characteristics of 
            Supplemental Nutrition Assistance Program Households: 
            Fiscal Year 2017, Report No. SNAP-18-CHAR, Mathematica 
            Policy Research, February 3, 2019 (https://fns-
            prod.azureedge.net/sites/default/files/ops/
            Characteristics2017.pdf) \27\
---------------------------------------------------------------------------
    \27\ Note: the submission is an excerpt from the report Table A.14. 
Distribution of participating households, individuals, and benefits by 
household composition, p. 52; and Table A.16. Distribution of 
participating households by countable income type and household 
composition, p. 54.

  219.  Letter from Arthur T. Foley, Director, Program Development 
            Division, Food and Nutrition Service, USDA, February 3, 
            2006, Re: FSP--2-Year Approval of Waivers of the Work 
            Requirements for ABAWDs under 7 CFR 273.24 (https://
            www.fns.usda.gov/sites/default/files/snap/2-
            Year%20Approval%20
            of%20Waivers%20of%20the%20Work%20Requirements%20for%20
---------------------------------------------------------------------------
            ABAWDs.pdf)

  220.  Letter from Sasha Gersten-Paal, Chief, Certification Policy 
            Branch, Program Development Division, Food and Nutrition 
            Service, USDA, March 15, 2017, Re: SNAP--FY 2017 
            Allocations of 15 Percent Exemptions for ABAWDs--Totals 
            Adjusted for Carryover (https://fns-prod.azureedge.net/
            sites/default/files/snap/FY2017-ABAWD-15%25-Exemption-
            Totals.pdf)

  221.  Food and Nutrition Service, USDA, Guidance for States Seeking 
            Waivers for Food Stamp Limits, December 3, 1996 \28\
---------------------------------------------------------------------------
    \28\ Note: the submission is a scanned version of the Guidance 
including marginalia presumably by the submitting organization.

  222.  Food and Nutrition Service, USDA, Guidance on Requesting ABAWD 
---------------------------------------------------------------------------
            Waivers, August 2006

  223.  Food and Nutrition Service, USDA, Guide to Serving ABAWDs 
            Subject to Time-limited Participation: A Guide on Serving 
            Able-Bodied Adults without Dependents (ABAWDs), 2015

  224.  Letter from Arthur T. Foley, Director, Program Development 
            Division, Food and Nutrition Service, USDA, October 27, 
            1997, Re: Implementation of the Provisions of the Balanced 
            Budget Act of 1997 Relating to Exemptions for Able-Bodied 
            Adults Without Dependents (ABAWDs)

  225.  Food and Nutrition Service, USDA, Perdue Reiterates Need to 
            Restore Original Intent of SNAP: A Second Chance, Not A Way 
            of Life, Press Release, February 28, 2019 (https://
            www.fns.usda.gov/pressrelease/2019/usda-002519)

  226.  Food and Nutrition Service, USDA, Regulatory Reform at a Glance 
            Proposed Rule: SNAP Requirements for ABAWDs, December 20, 
            2018, Fact Sheet (https://fns-prod.azureedge.net/sites/
            default/files/snap/ABAWDSFact
            Sheet.pdf)

  227.  Letter from Lizbeth Silbermann, Director, Program Development 
            Division, Food and Nutrition Service, USDA, April 17, 2017, 
            Re: SNAP--Requirements for Informing Households of ABAWD 
            Rules (https://fns-prod.azureedge.net/sites/default/files/
            snap/Requirements_for_Informing_Households_of_
            ABAWD_Rules.pdf)

  228.  Cromartie, John Rural America At A Glance, 2017 Edition, 
            Economic Information Bulletin No. EIB-182, Economic 
            Research Service, USDA, November 16, 2017 (https://
            www.ers.usda.gov/webdocs/publications/85740/eib-
            182.pdf?v=0)

  229.  Letter from Lizbeth Silbermann, Director, Program Development 
            Division, Food and Nutrition Service, USDA, June 26, 2015, 
            Re: Supplemental Nutrition Assistance Program--Able-Bodied 
            Adults without Dependents (ABAWD) Questions and Answers--
            June 2015 (https://fns-prod.azureedge.net/sites/default/
            files/snap/ABAWD-Questions-and-Answers-June%202015.pdf)

  230.  Letter from Lizbeth Silbermann, Director, Program Development 
            Division, Food and Nutrition Service, USDA, March 4, 2015, 
            Re: Supplemental Nutrition Assistance Program--Expiration 
            of Statewide ABAWD Time Limit Waivers (https://fns-
            prod.azureedge.net/sites/default/files/snap/SNAP-
            Expiration-of-Statewide-ABAWD-Time-Limit-Waivers.pdf)

  231.  Letter from Lizbeth Silbermann, Director, Program Development 
            Division, Food and Nutrition Service, USDA, and attached 
            Guidance, December 2, 2016, Re: Supplemental Nutrition 
            Assistance Program--Guide to Supporting Requests to Waive 
            the Time Limit for Able-Bodied Adults without Dependents 
            (ABAWD) (https://fns-prod.azureedge.net/sites/default/
            files/snap/SNAP-Guide-to-Supporting-Requests-to-Waive-the-
            Time-Limit-for-ABAWDs.pdf) \29\
---------------------------------------------------------------------------
    \29\ Note: this submission was duplicated in attachment 29.

  232.  Food and Nutrition Service, USDA, SNAP E&T and WIOA: Partnering 
            to Raise Skills and Employment, Policy Brief 8, June 8, 
            2018 (https://snaptoskills.fns.usda.gov/sites/default/
            files/2018-06/Brief_June2018_508
---------------------------------------------------------------------------
            comp.pdf)

  233.  Food and Nutrition Service, USDA, Why SNAP To Skills? (https://
            snaptoskills.fns.usda.gov/why-snap-to-skills) \30\
---------------------------------------------------------------------------
    \30\ Note: the submission is a ``snapshot'' of the webpage.

  234.  Letter from Lizbeth Silbermann, Director, Program Development 
            Division, Food and Nutrition Service, USDA, November 19, 
            2015, Re: Supplemental Nutrition Assistance Program--ABAWD 
            Time Limit Policy and Program Access (https://fns-
            prod.azureedge.net/sites/default/files/snap/ABAWD-Time-
---------------------------------------------------------------------------
            Limit-Policy-and-Program-Access-Memo-Nov2015.pdf)

  235.  Food and Nutrition Service, USDA, State Highlights: California 
            (https://snaptoskills.fns.usda.gov/state-highlights/state-
            highlights-california) \31\
---------------------------------------------------------------------------
    \31\ Note: the submission is a ``snapshot'' of the webpage.

  236.  Letter from Hon. Kevin W. Concannon, Under Secretary, Food, 
            Nutrition and Consumer Service, USDA; and Portia Wu, 
            Assistant Secretary, Employment & Training Administration, 
            U.S. DOL, March 31, 2016, Re: Partnering to Help Connect 
            Low-Income Able-bodied Adults to the Public Workforce 
            System (https://fns-prod.azureedge.net/sites/default/files/
---------------------------------------------------------------------------
            snap/USDA-DOL-joint-ABAWD-letter.pdf)

  237.  Food and Nutrition Service, USDA, February 26, 2018, Fact 
            Sheet, Characteristics of Able-bodied Adults without 
            Dependents (https://fns-prod.azureedge.net/sites/default/
            files/snap/nondisabled-adults.pdf)

  238.  Mills, Gregory, Tracy Vericker, Heather Koball, Kye Lippold, 
            Laura Wheaton; Sam Elkin,* Understanding the Rates, Causes, 
            and Costs of Churning in the Supplemental Nutrition 
            Assistance Program (SNAP) Final Report, Urban Institute (* 
            MEF Associates), November 4, 2014 (https://fns-
            prod.azureedge.net/sites/default/files/ops/
            SNAPChurning.pdf)

  239.  Employment and Training Administration, U.S. DOL, Conformity 
            Requirements for State UI Laws, Fact Sheet, January 6, 2016 
            (https://oui.doleta.gov/unemploy/pdf/uilaws_extended.pdf)

  240.  Employment and Training Administration, U.S. DOL, Emergency 
            Unemployment Compensation (EUC) Expired on January 1, 2014 
            (https://oui.doleta.gov/unemploy/supp_act.asp) \32\
---------------------------------------------------------------------------
    \32\ Note: the submission is a ``snapshot'' of the webpage.

  241.  Employment and Training Administration, U.S. DOL, EUC 2008 
            Trigger Notice No. 2008-45, Second Tier EUC 2008 Triggers 
            Under P.L. 110-449, November 23, 2008 (https://
---------------------------------------------------------------------------
            oui.doleta.gov/unemploy/euc_trigger/2008/euc_112308.pdf)

  242.  Leberg, Stanley, Labor Force, Employment, and Unemployment, 
            1929-39: Estimating Methods, Monthy Labor Review, July 1948 
            (https://www.bls.gov/opub/mlr/1948/article/pdf/labor-force-
            employment-and-unemployment-1929-39-estimating-methods.pdf)

  243.  Employment and Training Administration, U.S. DOL, Labor Surplus 
            Area Classification under Executive Orders 12073 and 10582, 
            Federal Register, Vol. 76, Iss. 167, August 29, 2011, p. 
            53699-53700 (https://www.govinfo.gov/content/pkg/FR-2011-
            08-29/pdf/2011-22003.pdf)

  244.  Employment and Training Administration, U.S. DOL, Labor Surplus 
            Area: Frequently Asked Questions, April 5, 2018 (https://
            www.doleta.gov/lsa/lsa_faq.cfm) \33\
---------------------------------------------------------------------------
    \33\ Note: the webpage has been subsequently updated as of October 
5, 2018. Note: the submission is a ``snapshot'' of the webpage.

  245.  Nilsen, Sigurd R., Director, Education, Workforce, Income 
            Security Issues, U.S. General Accounting Office, Food Stamp 
            Employment and Training Program: Better Data Needed to 
            Understand Who Is Served and What the Program Achieves, No. 
            03-388, March 12, 2003 (https://www.gao.gov/assets/240/
---------------------------------------------------------------------------
            237571.pdf)

  246.  Virginia Board of Workforce Development, Designation of Regions 
            and Planning Requirements, Policy Memorandum No. 200-06 
            (2016), July 1, 2016 (https://virginiacareerworks.com/wp-
            content/uploads/Policy-200-06-Designation-of-Regions-and-
            Planning-Requirements-FINAL-Signed.pdf)

  247.  Vornholt, Katharina, Patrizia Villotti, Beate Muschalla, Jana 
            Bauer, Adrienne Colella, Fred Zijlstra, Gemma Van 
            Ruitenbeek, Sjir Uitdewilligen & Marc Corbiere, Disability 
            and employment--overview and highlights, European Journal 
            of Work and Organizational Psychology, Vol. 27 No. 1, 
            October 10, 2017, p. 40-55 (https://www.tandfonline.com/
            doi/full/10.1080/1359432X.2017.1387536)

  248.  Wagner, Jennifer, 4,109 More Arkansans Lost Medicaid in October 
            for Not Meeting Rigid Work Requirements, Off the Charts 
            Blog, Center on Budget and Policy Priorities, October 16, 
            2018 (https://www.cbpp.org/blog/4109-more-arkansans-lost-
            medicaid-in-october-for-not-meeting-rigid-work-
            requirements)

  249.  Wagner, Jennifer, Center on Budget and Policy Priorities, 
            Commentary: As Predicted, Eligible Arkansas Medicaid 
            Beneficiaries Struggling to Meet Rigid Work Requirements, 
            Center on Budget and Policy Priorities, July 30, 2018 
            (https://www.cbpp.org/health/commentary-as-predicted-
            eligible-arkansas-medicaid-beneficiaries-struggling-to-
            meet-rigid)

  250.  Wagner, Jennifer, Commentary: As Predicted, Arkansas' Medicaid 
            Waiver Is Taking Coverage Away From Eligible People, Center 
            on Budget and Policy Priorities, March 12, 2019 (https://
            www.cbpp.org/health/commentary-as-predicted-arkansas-
            medicaid-waiver-is-taking-coverage-away-from-eligible-
            people)

  251.  Wagner, Jennifer, Fact Checking Arkansas Governor's Claims 
            About Jobs and Medicaid Waiver, Off the Charts Blog, Center 
            on Budget and Policy Priorities, January 28, 2019 (https://
            www.cbpp.org/blog/fact-checking-arkansas-governors-claims-
            about-jobs-and-medicaid-waiver)

  252.  Wagner, Jennifer,Medicaid Coverage Losses Mounting in Arkansas 
            From Work Requirement, Off the Charts Blog, Center on 
            Budget and Policy Priorities, January 17, 2019 (https://
            www.cbpp.org/blog/medicaid-coverage-losses-mounting-in-
            arkansas-from-work-requirement)

  253.  Walker, Renee E., Christopher R. Keane, Jessica G. Burke, 
            Disparities and access to healthy food in the United 
            States: A review of food deserts literature, Health & Place 
            Volume 16, Issue 5, September 2010, p. 876-884 (https://
            doi.org/10.1016/j.healthplace.2010.04.013)

  254.  Watson, Ian, Beyond the Unemployment Rate: Building a Set 
            Indicies to Measure the Health of the Labour Market, 
            Australian Bulletin of Labour, Vol. 26, No. 3, September 
            2000, p. 176-190 (http://www.ianwatson.com.au/pubs/
            beyond%20the%20unemployment%20rate.pdf)

  255.  Welles, Seth L., Falguni Patel, Mariana Chilton, Does 
            Employment-Related Resilience Affect the Relationship 
            between Childhood Adversity, Community Violence, and 
            Depression?, Journal of Urban Health, April 2017, 94(2): p. 
            233-243 (https://www.ncbi.nlm.nih.gov/pmc/articles/
            PMC5391326/)

  256.  The Pew Charitable Trusts, Collateral Costs: Incarceration's 
            Effect on Economic Mobility, September 28, 2010 (https://
            www.pewtrusts.org/en/research-and-analysis/reports/0001/01/
            01/collateral-costs)

  257.  Western, Bruce, Becky Pettit, Incarceration & social 
            inequality, D#dalus Summer 2010, p. 8-19

  258.  Westra, Ph.D., Karen L., John Routley, Arizona Cash Assistance 
            Exit Study, First Quarter 1998 Cohort, Final Report, 
            Administrative and Survey Data Results, Arizona Department 
            of Economic Security, Office of Evaluation, January 2000 
            (https://aspe.hhs.gov/system/files/pdf/177096/AZ-CAExit
            Study00.pdf)

  259.  Williamson, MPP, Sarah, Full-Family Sanctions & Economic 
            Recession, Family Welfare Research and Training Group, 
            University of Maryland School of Social Work, January 2011 
            (https://familywelfare.umaryland.edu/reports1/
            sanctionsbrief.pdf)

  260.  Wilson, Claire, Brian Estes, Examining the Growth of the Zero-
            Income SNAP Caseload: Characteristics, Circumstances, and 
            Dynamics of Zero-Income SNAP Participants, Volume II: In-
            Depth Interview Findings, Insight Policy Research, Inc., 
            October 2014 (https://fns-prod.azureedge.net/sites/default/
            files/ops/ZeroIncome-Vol2.pdf)

  261.  Wilson, Valerie, Before the State of the Union, a fact check on 
            black unemployment, Working Economics Blog, February 1, 
            2019 (https://www.epi.org/blog/before-the-state-of-the-
            union-a-fact-check-on-black-unemployment/)

  262.  Department of Workforce Development, State of Wisconsin, 
            Wisconsin Works (W-2) Sanctions Study, December 2004 
            (https://www.hhs.gov/sites/default/files/ocr/civilrights/
            activities/examples/TANF/wi_tanf_w2study.pdf)

  263.  Wolkomir, Elizabeth, How SNAP Can Better Serve the Formerly 
            Incarcerated, Center on Budget and Policy Priorities, March 
            16, 2018 (https://www.cbpp.org/research/food-assistance/
            how-snap-can-better-serve-the-formerly-incarcerated)

  264.  Wu, Chi-Fang, Maria Cancian, Daniel R. Meyer, Sanction policies 
            and outcomes in Wisconsin, p. 38-40; and Lee, Bong Joo, 
            Kristen Shook Slack, Dan A. Lewis, Sanctions policies and 
            outcomes in Illinois, p. 41-43: Focus Vol. 23, No. 1, 
            Winter 2004 (https://www.irp.wisc.edu/publications/focus/
            pdfs/foc231f.pdf) \34\
---------------------------------------------------------------------------
    \34\ Note: this submission consists of two sub-articles. The full 
article entitled, How do welfare sanctions work? New findings from 
Wisconsin and Illinois, has a one-page introduction p. 37.

  265.  Wu, Chi-Fang, Maria Cancian, Daniel R. Meyer, Geoffrey L. 
            Wallace, How Do Welfare Sanctions Work?, Social Work 
            Research, Volume 30, Number 1, March 2006, p. 33-50 
---------------------------------------------------------------------------
            (https://www.ssc.wisc.edu/gwallace/Papers/(5).pdf)

  266.  Yagan, Danny, Employment Hysteresis from the Great Recession, 
            National Bureau of Economic Research, NBER Working Paper 
            No. 23844 (https://www.nber.org/papers/w23844.pdf)

  267.  Yellen, Janet L., Chair, Board of Governors of the Federal 
            Reserve System, Addressing Workforce Development Challenges 
            in Low-Income Communities, ``Creating a Just Economy,'' the 
            2017 annual conference of the National Community 
            Reinvestment Coalition, March 28, 2017 (https://
            www.federalreserve.gov/newsevents/speech/files/
            yellen20170328a.pdf)

  268.  Ziliak, James P., Restoring Economic Opportunity for ``The 
            People Left Behind'': Employment Strategies for Rural 
            America, in Part II: Increasing Prime-Age Labor Force 
            Participation, of Expanding Economic Opportunity for More 
            Americans: Bipartisan Policies to Increase Work, Wages, and 
            Skills, Aspen Institute Economic Strategy Group, February 
            2019, p. 100-126 (https://assets.aspeninstitute.org/
            content/uploads/2019/01/2.2-Pgs.-100-126-Restoring-
            Economic-Opportunity....pdf) \35\
---------------------------------------------------------------------------
    \35\ Note: this submission consists of one policy memo in the 
report Expanding Economic Opportunity for More Americans: Bipartisan 
Policies to Increase Work, Wages, and Skills (https://
assets.aspeninstitute.org/content/uploads/2019/01/ESG_Report_Expanding-
Economic-Opportunity-for-More-Americans.pdf). The full report is 
retained in the Committee file.

  269.  Ziliak, James P., Temporary Assistance For Needy Families, 
            National Bureau of Economic Research, NBER Working Paper 
            No. 21038 (https://www.nber.org/papers/w21038)
                                 ______
                                 
 Submitted Comment Letter by Hon. Jefferson Van Drew, a Representative 
 in Congress from New Jersey; Authored by Kate Leone, Chief Government 
                   Relations Officer, Feeding America
April 2, 2019

  Ms. Sasha Gerstan-Paal,
  Chief,
  Certification Policy Branch,
  SNAP Program Development Division,
  Food and Nutrition Service, USDA,
  Alexandria, Virginia

  RE: Proposed Rule: Supplemental Nutrition Assistance Program (SNAP): 
            Requirements for Able-Bodied Adults without Dependents RIN 
            0584-AE57

    Dear Ms. Gerstan-Paal:

    We appreciate the opportunity to comment on USDA's Proposed 
Rulemaking on requirements and services for Able-Bodied Adults Without 
Dependents (ABAWDs). The proposed changes would make it harder for 
individuals facing food insecurity to have the resources they need to 
purchase healthy food. The changes would harm those individuals and 
their community, putting an unnecessary burden on food banks and other 
service providers. We strongly encourage USDA to rescind this rule.
    Feeding America is the nationwide network of 200 food banks that 
leads the fight against hunger in the United States. Together, we 
provide food to more than 46 million people through 60,000 food 
pantries and meal programs in communities across America. Feeding 
America also supports programs that improve food security among the 
people we serve; educates the public about the problem of hunger; and 
advocates for legislation that protects people from going hungry. This 
includes working to ensure people understand and are able to access 
Federal nutrition assistance programs like SNAP, through education and 
outreach, and related support services such as nutrition education and 
work support services, including SNAP Employment and Training.
    Feeding America cares about the 3 month time limit for ABAWDs 
because this policy has cut off food assistance to so many in 
communities across the country who are in need just because they are 
unable to find a reliable 20 hour a week job or otherwise document 
their 20 hours of qualifying activities. Feeding America released a 
statement in April 2016 (online here (http://www.feedingamerica.org/
about-us/press-room/feeding-america-food-banks.html) and below) 
expressing our concern as the state waivers were going away, either 
naturally or due to state decisions, that policies like these that take 
important food resources away from vulnerable individuals will only 
make it harder to ensure community members are nourished and ready for 
the workforce. We subsequently released statements in December 2018 
(online here (https://www.feedingamerica.org/about-us/press-room/
feeding-america-statement-able-bodied-adults-without-dependents-
proposed-rule) and below) and February 2019 (online here (https://
www.feedingamerica.org/about-us/press-room/feeding-america-opposes-
harmful-snap-proposed-rule-released-usda) and below) expressing 
concerns about this notice of proposed rulemaking which we are 
commenting on today.
    We strongly encourage USDA to rescind this rule.
SNAP Matters
    SNAP plays a critical role in addressing hunger and food insecurity 
in our community. It is the first line of defense against hunger.
    Based on USDA Economic Research Service analysis, it is estimated 
that each $1 in Federal SNAP benefits generates $1.79 in economic 
activity. Those dollars help many food retailers operating on thin 
margins to remain in business; something that improves food access for 
all residents.
    Access to healthy food is a critical social determinant of health 
and food insecurity is associated with poorer health outcomes.\1\ Food 
insecurity is associated with higher rates of some of the most serious 
and costly chronic conditions, including hypertension, coronary heart 
disease, cancer, asthma, diabetes, and many other serious health 
conditions. Adults who experience food insecurity are also more likely 
to report lower health status overall.\2\
---------------------------------------------------------------------------
    \1\ World Health Organization, https://www.who.int/hia/evidence/
doh/en/index3.html, see also Craig Gundersen and James P. Ziliak, 
``Food Insecurity and Health Outcomes,'' Health Affairs, November 2015, 
https://www.healthaffairs.org/doi/10.1377/hlthaff.2015.0645.
    \2\ Christian A. Gregory and Alisha Coleman-Jenson, ``Food 
Insecurity, Chronic Disease, and Health Among Working-Age Adults,'' 
United States Department of Agriculture, July 2017, https://nopren.org/
wp-content/uploads/2017/08/ERS-Report-Food-Insecurity-Chronic-Disease-
and-Health-Among-Working-Age-Adults.pdf.
---------------------------------------------------------------------------
    SNAP plays a critical role in addressing hunger and food insecurity 
in our communities: it is the first line of defense against hunger for 
low-income residents. Research shows that SNAP reduces poverty and food 
insecurity, and that over the long-term, these impacts lead to improved 
health and economic outcomes, especially for those who receive SNAP as 
children.
    Limiting access to nutrition assistance could be particularly 
harmful for people with significant health care needs, such as diabetes 
or hypertension, who may also have trouble maintaining their health 
while keeping a job. Many people turn to public assistance programs 
because they face significant health or family challenges that limit 
their ability to work or reduce their ability to compete for a limited 
supply of jobs. Physical and mental health conditions that limit an 
individual's ability to work or limit the amount or kind of work the 
individual can do are much more common among public benefit recipients 
than among the general population, research shows.\3\ Taking access to 
nutrition assistance away from people with serious health conditions 
could negatively affect their health, which could make it even more 
difficult for them to maintain employment.
---------------------------------------------------------------------------
    \3\ Pamela L. Loprest and Elaine Maag, ``Disabilities among TANF 
Recipients: Evidence from the NHIS,'' Urban Institute, May 2009, http:/
/www.urban.org/research/publication/disabilities-among-tanf-recipients-
evidence-nhis.
---------------------------------------------------------------------------
Food insecurity increases the risk of negative physical and mental 
        health outcomes
    Food insecurity is a risk factor for negative psychological and 
health outcomes.\4\ (The U.S. Department of Agriculture defines food 
insecurity as a ``lack of consistent access to enough food for an 
active, healthy life.'' \5\) Food insecurity has deleterious impacts on 
health through increases in the prevalence and severity of diet-related 
disease, such as obesity, type 2 diabetes, heart disease, stroke, and 
some cancers.6-8
---------------------------------------------------------------------------
    \4\ Hartline-Grafton, H. (2017). The Impact of Poverty, Food 
Insecurity, & Poor Nutrition on Health and Well-Being. Washington, 
D.C.: Food Research & Action Center.
    \5\ Economic Research Service, U.S. Department of Agriculture. 
(2018). Definitions of Food Security. Available at https://
www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-
us/definitions-of-food-security.aspx Accessed October 3, 2018.
    \6\ Franklin B. Jones, A., Love, D., Puckett, S., Macklin, J., & 
White-Means, S. (2012). Exploring mediators of food insecurity and 
obesity: a review of recent literature. Journal of Community Health. 
37(1), 253-264.
    \7\ Berkowitz, S., A., Karter, A., J., Corbie-Smith, G., Seligman, 
H. K., Ackroyd, S.A., Barnard, L.S., Atlas, S.J., & Wexler, D.J. 
(2018). Food insecurity, food ``deserts,'' and glycemic control in 
patients with diabetes: a longitudinal analysis. Diabetes Care, 19, 
171981.
    \8\ Gregory, C., A., & Coleman-Jensen, A. (2017). Food insecurity, 
chronic disease, and health among working-age adults. Economic Research 
Report, 235. Washington, D.C.: U.S. Department of Agriculture, Economic 
Research Service.
---------------------------------------------------------------------------
    In addition, because of limited financial resources, those who are 
food-insecure--with our without existing disease--may use coping 
strategies to stretch budgets that are harmful for health, such as 
engaging in cost-related medication underuse or non-adherence; 
9-11 postponing or forgoing preventives or needed medical 
care; 12-13 and forgoing the foods needed for special 
medical diets (e.g., diabetic diets).\14\ Not surprisingly, research 
shows that household food insecurity is a strong predictor of higher 
health care utilization and increased health care 
costs.15-16
---------------------------------------------------------------------------
    \9\ Herman, D., Afulani, P., Coleman-Jensen, A., & Harrison, G.G. 
(2015). Food insecurity and cost-related medication underuse among 
nonelderly adults in a nationally representative sample: American 
Journal of Public Health, 105(10), 48-59.
    \10\ Afulani, P., Herman, D., Coleman-Jensen, A., & Harrison G.G. 
(2015). Food insecurity and health outcomes among older adults: The 
role of cost-related medication underuse. Journal of Nutrition in 
Gerontology and Geriatrics, 34(3), 319-343.
    \11\ Knight, C.K., Probst, J.C., Liese, A., D., Sercy, E., & Jones, 
S.J. (2016). Household food insecurity and medication ``scrimping'' 
among U.S. adults with diabetes. Public Health [Nutrition], 19(6), 
1103-1111.
    \12\ Mayer, V.L., McDonough, K., Seligman, H., Mitra, N., & Long, 
J.A. (2016). Food insecurity, coping strategies and glucose control in 
low-income patients with diabetes. Public Health Nutrition, 19(6), 
1103-1111.
    \13\ Kushel, M.B., Gupta, R., Gee, L., & Haas, J.S. (2006). Housing 
instability and food insecurity as barriers to health care among low-
income Americans. Journal of General Internal Medicine, 21, 71-77.
    \14\ Seligman, H.K., Jacobs, E.A., Lopez, A., Tschann, J., & 
Fernandez, A. (2012). Food insecurity and glycemic control among low-
income patients with type 2 diabetes. Diabetes Care, 35(2), 233-238.
    \15\ Tarasuk, V., Cheng, J., de Oliveira, D., Dachner, N., 
Gundersen, C., & Kurdyak, P. (2015). Association between household food 
insecurity and annual health care costs. Canadian Medical Association 
Journal, 187 (14), E429-436.
    \16\ Berkowitz, S.A., Basu, S., Meigs, J.B., & Seligman, H. (2017). 
Food insecurity and health expenditures in the United States, 2011-
2013. Health Services Research, 53(3), 1600-1620.
---------------------------------------------------------------------------
SNAP decreases food insecurity
    Overall, research shows that SNAP is effective at reducing food 
insecurity.17-19 According to one estimate, SNAP reduces 
food insecurity by approximately 30 percent.\20\ SNAP, therefore, is an 
effective anti-hunger program, and more eligible people need to be 
connected to the program given the current high rates of food 
insecurity in the nation. Nearly one in eight American households 
experience food insecurity during the year.\21\
---------------------------------------------------------------------------
    \17\ Mabli, J., & Worthington, J. (2014). Supplemental Nutrition 
Assistance Program participation and child food security. Pediatrics, 
133(4), 1-10.
    \18\ Ratcliffe, C., McKernan, S.M., & Zhang, S. (2011). How much 
does the Supplemental Nutrition Assistance Program reduce food 
insecurity? American Journal of Agricultural Economics, 93(4), 1082-
1098.
    \19\ Nord, M. (2012). How much does the Supplemental Nutrition 
Assistance Program alleviate food insecurity? Evidence from recent 
programme leavers. Public Health Nutrition, 15(5), 811-817.
    \20\ Ratcliffe, C., McKernan, S.M., & Zhang, S. (2011). How much 
does the Supplemental Nutrition Assistance Program reduce food 
insecurity? American Journal of Agricultural Economics, 93(4), 1082-
1098.
    \21\ Coleman-Jensen, A., Rabbit, M.P., Gregory, C.A. & Singh, A. 
(2018). Household food insecurity in the United States in 2017. 
Economic Research Service Report, 256, Washington, D.C.: U.S. 
Department of Agriculture, Economic Research Service.
---------------------------------------------------------------------------
SNAP is associated with decreased health care costs
    Research demonstrates that SNAP reduces health care utilization and 
costs.22-24 For example, a national study revealed that SNAP 
participation was associated with lower health care costs.\25\ On 
average, low-income adults participating in SNAP incurred nearly 25 
percent less in health care costs in 12 month, including those paid by 
private or public insurance, than low-income adults not participating 
in SNAP.
---------------------------------------------------------------------------
    \22\ Gregory, C.A., & Deb, P. (2015). Does SNAP improve your 
health? Food Policy, 50, 11-19.
    \23\ Berkowitz, S.A., Seligman, H.K., Rigdon, J., Meigs, J.B., & 
Basu, S. (2017). Supplemental Nutrition Assistance Program (SNAP) 
participation and health care expenditures among low-income adults. 
JAMA Internal Medicine, 177(11), 1642-1649.
    \24\ Seligman, H.K., Bolger, A.F., Guzman, D., Lopez, A., & 
Bibbins-Domingo, K. (2014). Exhaustion of food budgets at month's end 
and hospital admissions for hyperglycemia. Health Affairs, 33(1), 116-
123.
    \25\ Berkowitz, S.A., Seligman, H.K., Rigdon, J., Meigs, J.B., & 
Basu, S. (2017). Supplemental Nutrition Assistance Program (SNAP) 
participation and health care expenditures among low-income adults. 
JAMA Internal Medicine, 177(11), 1642-1649.
---------------------------------------------------------------------------
SNAP is associated with improved physical and mental health
    SNAP improves children, adult, and senior health outcomes, 
including physical and mental health.\26\ For instance, SNAP increases 
the probability of self-reporting ``excellent'' or ``good health,'' 
\27\ lowers the risk of poor glucose control (for those with 
diabetes),\28\ and has a protective effect on mental health.\29\ 
Journal SNAP also helps reduce stress for struggling individuals and 
families worried about finances, and stress is highly correlated with 
poor health outcomes.\30\
---------------------------------------------------------------------------
    \26\ Hartline-Grafton, H. (2017). SNAP and Public Health: The Role 
of the Supplemental Nutrition Assistance Program in Improving the 
Health and Well-Being of Americans. Washington, D.C.: Food Research & 
Action Center.
    \27\ Gregory, C.A., & Deb, P. (2015). Does SNAP improve your 
health? Food Policy, 50, 11-19.
    \28\ Mayer, V.L., McDonough, K., Seligman, H., Mitra, N., & Long, 
J.A. (2016). Food insecurity, coping strategies and glucose control in 
low-income patients with diabetes. Public Health Nutrition, 19(6), 
1103-1111.
    \29\ Leung, C.W., Epel, E.S., Willett, W.C., Rimm, E.B., & Laraia, 
B.A. (2015). Household food insecurity is positively associated with 
depression among low-income Supplemental Nutrition Assistance Program 
participants and income-eligible nonparticipants. Journal of Nutrition, 
145(3), 622-627.
    \30\ Juster, R.-P., McEwen, B.S., & Lupien, S.J. (2010). Allostatic 
load biomarkers of chronic stress and impact on health and cognition. 
Neuroscience and Biobehavioral Reviews, 35(1), 2-16.
---------------------------------------------------------------------------
Area Waivers and Individual Exemptions Provide Ways to Modestly 
        Ameliorate the Harsh Impact of Arbitrary Time Limits
    Federal law limits SNAP eligibility for childless unemployed and 
underemployed adults age 18-50 (except for those who are exempt) to 
just 3 months out of every 3 years unless they are able to obtain and 
maintain an average of 20 hours a week of employment. This rule in its 
current form is harsh and unfair. It harms vulnerable people by denying 
them food benefits at a time when they most need it and it does not 
result in increased employment and earnings.\31\ At least 500,000 low-
income individuals nationwide lost SNAP in 2016 due to the time limit. 
This put their food security at risk. And, by time-limiting food 
assistance to this group, Federal law has shifted the burden of 
providing food to these unemployed individuals from SNAP to local 
charities, states, and cities.
---------------------------------------------------------------------------
    \31\ A 2002 study that looked at recipients after leaving SNAP 
found that while many were employed, they had low earnings, and between 
\1/3\ and roughly \2/3\ of SNAP leavers had household incomes below the 
poverty line. (This study did not examine the effects of the time limit 
on employment.) See Elizabeth M. Dagata, ``Assessing the Self-
Sufficiency of Food Stamp Leavers,'' Economic Research Service, USDA, 
September 2002 , https://www.ers.usda.gov/publications/pub-details/
?pubid=46645. More recent research finds small increases in employment, 
but much larger decreases in SNAP participation. For example, one 
recent working paper found that the time limit increased work by two 
percentage points, but decreased participation by ten percentage 
points. (Timothy Harris, ``Do SNAP Work Requirements Work?'' Upjohn 
Institute Working Paper, 19-297, https://research.upjohn.org/cgi/
viewcontent.cgi?article=1315&context=
up_workingpapers.)
---------------------------------------------------------------------------
    Under the law, states have some flexibility to ameliorate the 
impact of the cutoff. They can request a waiver of the time limit for 
areas within the state that have ten percent or higher unemployment 
rates or, based on other economic indicators, have ``insufficient 
jobs.'' Moreover, states have discretion to exempt individuals from the 
time limit by utilizing a pool of exemptions (referred to as ``15 
percent'' exemptions). While the 2018 Farm Bill modified the number of 
exemptions that states can receive each year from 15 percent to 12 
percent, it did not change their ability to carry over unused 
exemptions forward.
Proposed Rule Undermining Law's Safety Valves Should Be Rejected
    Feeding America strongly opposes the proposed rule changes that 
would expose even more people to the arbitrary food cutoff policy by 
limiting state flexibility regarding area waivers and individual 
exemptions. By the Administration's own calculations, the proposed rule 
would take food away from 755,000 low-income Americans, cutting food 
benefits by $15 billion over 10 years. This is the equivalent of around 
8.5 billion meals lost from the tables of individuals. The 
Administration does not estimate any improvements in health or 
employment among the affected population.
    The proposed rule would make it harder for areas with elevated 
unemployment rates to qualify for waivers of the time limit by adding a 
seven percent unemployment rate floor as a condition.
    The proposed rule would make it harder for states to obtain and 
implement area waivers by dropping statewide waivers except when a 
state triggers extended benefits under Unemployment Insurance. It would 
unduly limit the economic factors considered in assessing an area's 
eligibility for a waiver (e.g., by no longer allowing employment to 
population ratios that demonstrate economic weakness to qualify areas 
for waivers). It would undermine efficient state implementation of area 
waivers by limiting their duration to 12 months and delaying their 
start dates until after USDA processes the request. In addition, the 
proposed rule would remove states' ability to use exemptions 
accumulated prior to the rule's implementation as well limit the time 
states' have to use exemptions they receive in the future. [Add any 
examples or details about how these proposals would affect your state, 
or a region in your state.]
    The Department provides little analysis to explain its conclusions 
about the impacts the changes would have on individuals and population 
groups nor of realistic plans to avert harm from those changes. USDA 
merely asserts its expectation that \2/3\ of those individuals made 
newly subject to the time limit ``would not meet the requirements for 
failure to engage meaningfully in work or work training.'' Moreover, 
while the Department concedes that the proposed changes ``have the 
potential for disparately impacting certain protected groups due to 
factors affecting rates of employment of these groups, [it] find[s] 
that implementation of mitigation strategies and monitoring by the 
Civil Rights Division of FNS will lessen these impacts.'' But no 
explanation of the mitigation strategies and monitoring is provided, so 
there is no opportunity for us to comment on whether the acknowledged 
disparate impact will in fact be mitigated.
Proposal Does Not Improve Employment Outcomes--and Would Undermine 
        Investments in Programs that Do
Time Limits and Work Reporting Requirements Do Not Encourage Employment
    The proposed rule does not require states to offer any work 
opportunities or employment and training activities to individuals 
subject to the time limit. Historically, many states have chosen not to 
help people subject to the time limit find qualifying work or training 
activities.\32\ Many people who are willing to participate in such 
activities will lose SNAP if they cannot find a countable activity--
which does not include job search--on their own.
---------------------------------------------------------------------------
    \32\ Nune Phillips, ``SNAP and Work,'' Center for Law and Social 
Policy, January 2018, https://www.clasp.org/sites/default/files/
publications/2018/01/2018_snapandwork.pdf.
---------------------------------------------------------------------------
    Some state and local leaders have worked hard over the past decade 
to intentionally engage SNAP participants in high-quality programs and 
develop partnerships for SNAP Employment & Training (E&T). However, 
these efforts, still in early stages, require substantial resources and 
capacity to deliver outcomes. This investment in quality, high-
intensity programs will likely shift as some states seek to spread 
limited SNAP E&T resources thinly to help more people meet SNAP time 
limit rules. The resulting low-intensity SNAP E&T programs have proven 
to be ineffective in moving SNAP recipients into jobs that will allow 
them to achieve economic security.
    If SNAP recipients do manage to find low-wage jobs to meet work 
reporting requirements, they do not fare any better in the long run 
than those in low-intensity SNAP E&T programs. Lessons learned from 
TANF, SNAP, and other programs demonstrate that work reporting 
requirements are not effective in connecting people to living-wage 
jobs.\33\ As laid out by the Center on Budget and Policy Priorities in 
a review of rigorous evaluations, research shows that employment 
increases among individuals subject to work reporting requirements were 
modest and faded over time. In nearly all of the approximately dozen 
programs evaluated, employment among recipients not subject to work 
reporting requirements was the same as or higher than employment among 
individuals subject to work reporting requirements within 5 years.\34\
---------------------------------------------------------------------------
    \33\ Ladonna Pavetti, Work Requirements Don't Cut Poverty, Evidence 
Shows, Center on Budget and Policy Priorities, June 2016, https://
www.cbpp.org/research/poverty-and-inequality/work-requirements-dont-
cut-poverty-evidence-shows.
    \34\ Ibid.
---------------------------------------------------------------------------
    Work reporting requirements are not only ineffectual but have 
opportunity costs: the time that a SNAP recipient loses in low-
intensity programs or low-wage jobs simply to meet requirements could 
have been spent obtaining skills and credentials, finding a quality 
job, and increasing their earnings. A much better focus for public 
policy is to invest in strategies that support people to develop skills 
and access training that prepares them for jobs that pay living wages 
and foster an economy that creates more quality jobs with fair wages.
Workforce Systems to Serve those Subject to the Time Limit are 
        Underfunded
    Even if states offer services to individuals newly subject to the 
time limits, many will offer low-intensity services, aimed primarily at 
providing recipients with enough hours of participation to meet the 
requirements, rather than high quality services.
    Existing workforce systems, which are chronically under-funded, are 
not often designed to serve the range of needs of all those subject to 
the time limit. Some struggling workers will have substantial 
stabilization needs, e.g., emergency housing, transportation, and 
dependent care. As people surmount those barriers, meeting a need as 
basic as food is paramount. The Government Accountability Office (GAO) 
found that SNAP participants subject to the time limit are more likely 
than other SNAP participants to lack basic job skills like reading, 
writing, and basic math.\35\ People should not be punished for 
grappling with hardship.
---------------------------------------------------------------------------
    \35\ Food Stamp Employment and Training Program: Better Data Needed 
to Understand Who Is Served and What The Program Achieves, U.S. 
Government Accountability Office, March 2003, https://www.gao.gov/
assets/240/237571.pdf.
---------------------------------------------------------------------------
    Instead of penalizing people for needing assistance to put food on 
the table, USDA should consider ways to create a foundation for long-
term economic success. Voluntary SNAP E&T programs, for instance, do 
not subject individuals to sanctions that increase food insecurity. In 
fact, research shows that voluntary programs can significantly increase 
employment, while mandatory SNAP E&T programs withhold basic assistance 
if individuals cannot meet participation requirements in a given 
month.\36\ To attract SNAP recipients to voluntary SNAP E&T programs, 
states can partner with trusted service providers that operate programs 
with a successful track record. Furthermore, mandating participation in 
employment or training programs requires participating organizations to 
spend time tracking attendance and not serving clients with the 
programs they need to succeed.
---------------------------------------------------------------------------
    \36\ Ibid., 9.
---------------------------------------------------------------------------
The Proposed Rule Would Harm People Who Already Struggle to Afford 
        Housing
    The lack of affordable housing is a nationwide crisis, with more 
than eight million households paying more than \1/2\ of their income on 
rent.\37\ People whose rent is unaffordable have less money to spend on 
other necessities, such as food, healthcare, and transportation. For 
example, families that spend more than \1/2\ of their income on rent in 
order to avoid eviction and homelessness spend on average $190 less per 
month on food compared to families that spend less than 30 percent of 
their income on rent.\38\ In fact, from 2001 to 2016, many low-income 
households across the country saw their rents increase as their incomes 
stagnated or decreased, meaning they have less money left over each 
month and are likely at higher risk of food insecurity. People 
experiencing homelessness also face difficulty affording food, which is 
often compounded by not having places to safely store and prepare food 
items. For people struggling to afford a place to live, SNAP is a vital 
lifeline that helps prevent food insecurity.
---------------------------------------------------------------------------
    \37\ ``Worst Case Housing Needs 2017 Report to Congress,'' U.S. 
Department of Housing and Urban Development, August 9, 2017, https://
www.huduser.gov/portal/sites/default/files/pdf/Worst-Case-Housing-
Needs.pdf.
    \38\ ``The State of the Nation's Housing 2018,'' Joint Center for 
Housing Studies of Harvard University, 2018, 31, https://
www.jchs.harvard.edu/sites/default/files/Harvard_JCHS_
State_of_the_Nations_Housing_2018.pdf.
---------------------------------------------------------------------------
    The proposed rule would limit states' flexibility to mitigate the 
harmful effects of the time limit. Without this flexibility, it's 
likely that many people could lose access to nutrition assistance. For 
people with severe housing cost burden--or those who spend 50 percent 
or more of their income on rent--limiting access to nutrition 
assistance could have serious repercussions for their housing 
stability, food security, and health.
    Limiting access to nutrition assistance could be particularly 
harmful for people who face housing instability or homelessness. Many 
people turn to public assistance programs such as SNAP because they 
face significant challenges with affording everyday necessities, 
including a safe place to live. Three in four people who qualify for 
Federal rental assistance that would make rent affordable don't receive 
it because of limited funding, making SNAP even more vital.\39\ In 
addition, low-income individuals with poor health are oftentimes one 
injury or illness away from falling into homelessness. Without food 
assistance, people who were already struggling to pay rent may have to 
choose between paying rent or putting food on the table, increasing 
their risk of eviction and homelessness.
---------------------------------------------------------------------------
    \39\ ``Three Out of Four Low-Income At-Risk Renters Do Not Receive 
Federal Rental Assistance,'' Center on Budget and Policy Priorities, 
August 8, 2017, https://www.cbpp.org/three-out-of-four-low-income-at-
risk-renters-do-not-receive-federal-rental-assistance.
---------------------------------------------------------------------------
    Homelessness--which is primarily caused by the lack of affordable 
housing--can contribute to new health issues and worsen existing ones. 
For example, people who live outdoors or in a homeless shelter 
oftentimes don't have a secure place to store medication or prepare the 
foods needed to manage health conditions, such as diabetes. 
Furthermore, soup kitchens and shelters are often not conducive to 
maintaining a healthy diet since many offer meals with high sugar, 
salt, and starch content.\40\ Homelessness itself also creates 
additional barriers to employment, because individuals do not have easy 
access to computers and phones to apply for jobs, or showers and 
laundry facilities to maintain personal hygiene. Taking nutrition 
assistance away from people experiencing homelessness could therefore 
make it even more difficult to find and maintain employment and become 
stably housed.
---------------------------------------------------------------------------
    \40\ ``Homelessness and Health: What's the Connection?'' National 
Health Care for the Homeless Council, 2011, https://www.nhchc.org/wp-
content/uploads/2011/09/Hln_health_fact
sheet_Jan10.pdf.

    Restricting a state's waiver authority and use of individual 
exemptions will harm women who want to work but face challenges in 
---------------------------------------------------------------------------
receiving and reporting 20 hours of work each week.

    Women with low incomes may face particular barriers to consistently 
working (and reporting) 20 hours of work per week. Women are over-
represented in the low-wage workforce,\41\ which is plagued by unstable 
and unpredictable work schedules.\42\ Compared to women's 
representation in the overall workforce, women of virtually all races 
and ethnicities are over-represented in low-wage jobs (typically paying 
less than $11.50 per hour).\43\ Of the nearly 22.6 million people 
working in low-wage jobs, \2/3\ are women.\44\
---------------------------------------------------------------------------
    \41\ Nat'l Women's Law Ctr., Interactive Map: Women And Men In The 
Low-Wage Workforce (July 20, 2018), https://nwlc.org/resources/
interactive-map-women-and-men-low-wage-workforce/.
    \42\ See generally Nat'l Women's Law Ctr., Collateral Damage: 
Scheduling Challenges for Workers in Low-Wage Jobs and Their 
Consequences (Apr. 2017), available at https://nwlc-
ciw49tixgw5lbab.stackpathdns.com/wp-content/uploads/2017/04/Collateral-
Damage.pdf (hereinafter ``Collateral Damage'').
    \43\ Nat'l Women's Law Ctr. calculations based on IPUMS 2018, supra 
note 4.
    \44\ Id.
---------------------------------------------------------------------------
    The unpredictable and unstable work schedules that are common in 
retail, food service, and other low-wage jobs can prevent women from 
working 20 hours per week, every week. Many low-wage jobs lack paid 
leave,\45\ which presents another obstacle for women with caregiving 
responsibilities \46\ for people outside of the narrow time limit 
caregiving exemption. In addition, many low-wage jobs offer only part-
time work, despite many workers' need and desire for full-time 
hours.\47\ And the combination of insufficient hours and variable 
schedules can impede women from working more than one job to make ends 
meet.\48\
---------------------------------------------------------------------------
    \45\ Andrea Johnson, et al., Stepping Up: New Policies and 
Strategies Supporting Parents in Low-Wage Jobs and Their Children 22 
(Aug. 2018), available at https://nwlc-
ciw49tixgw5lbab.stackpathdns.com/wp-content/uploads/2018/08/
v2_final_nwlc_SteppingUp
KelloggReport.pdf.
    \46\ Women perform the majority of caregiving. See Bureau of Labor 
Statistics, U.S. Dep't of Labor, American Time Use Survey, Table A-1. 
Time spent in detailed primary activities and percent of the civilian 
population engaging in each activity, averages per day by sex, 2017 
annual averages (2018), available at https://www.bls.gov/tus/
a1_2017.pdf.
    \47\ Collateral Damage, supra note 19, at 2.
    \48\ Collateral Damage, supra note 19, at 1.
---------------------------------------------------------------------------
    In addition, many women, particularly those in low-wage jobs, face 
discrimination and harassment at work.\49\ Between 2012 and 2016, 36 
percent of women who filed sexual harassment charges also alleged 
retaliation, such as lost hours or job loss.\50\
---------------------------------------------------------------------------
    \49\ See, e.g., Rest. Opportunities Ctrs. United & Forward 
Together, The Glass Floor: Sexual Harassment in the Restaurant Industry 
5 (2014), available at http://rocunited.org/wp-content/uploads/2014/10/
REPORT_The-Glass-Floor-Sexual-Harassment-in-the-Restaurant-Industry
2.pdf; Human Rights Watch, Cultivating Fear: The Vulnerability of 
Immigrant Farmworkers in the U.S. to Sexual Violence and Sexual 
Harassment (May 2012), available at https://www.hrw.org/report/2012/05/
15/cultivating-fear/vulnerability-immigrant-farmworkers-us-sexual-
violence-and-sexual (documenting pervasive sexual harassment and 
violence among immigrant farmworker women); Irma Morales Waugh, 
Examining the Sexual Harassment Experiences of Mexican Immigrant 
Farmworking Women, 16 Violence Against Women 237, 241 (Jan. 2010), 
available at http://vaw.sagepub.com/content/16/3/237.abstract (80% of 
female farmworkers in California's Central Valley reported experiencing 
some form of sexual harassment); Unite Here Local 1, Hands Off, Pants 
On: Sexual Harassment In Chicago's Hospitality Industry (July 2016), 
available at https://www.handsoffpantson.org/wp-content/uploads/
HandsOffReportWeb.pdf (58% of hotel workers and 77% of casino workers 
surveyed reported being sexually harassed by a guest); Hart Res. 
Assoc., Key Findings From a Survey of Women Fast Food Workers (Oct. 5, 
2016), available at http://hartresearch.com/wp-content/uploads/2016/10/
Fast-Food-Worker-Survey-Memo-10-5-16.pdf (nationwide survey of workers 
in the fast food industry found nearly 40% of the women reported 
experiencing unwanted sexual behaviors on the job, and 21% of those 
workers reported that they suffered negative workplaces consequences 
after raising the harassment with their employer).
    \50\ Amanda Rossie, Jasmine Tucker & Kayla Patrick, Nat'l Women's 
Law Ctr., Out of the Shadows: An Analysis of Sexual Harassment Charges 
Filed By Working Women 8 (Aug. 2018), available at https://nwlc-
ciw49tixgw5lbab.stackpathdns.com/wp-content/uploads/2018/08/
SexualHarassmentReport.pdf.
---------------------------------------------------------------------------
    All of these factors can make it difficult for low-income working 
women to satisfy SNAP's 20 hour per week reporting requirement. As a 
result, women struggling with underemployment face a double risk: if 
their employer schedules them for fewer hours, then their wages 
decrease, and they are at risk of losing SNAP benefits. Even women who 
happen to receive enough hours to meet the time limit reporting 
requirement are still at risk of losing benefits if they are unable to 
meet burdensome administrative requirements to document their hours of 
work.\51\
---------------------------------------------------------------------------
    \51\ See Robin Fudowitz, MaryBeth Musumeci & Cornelia Hall, Kaiser 
Family Found., Year End Review: December State Data for Medicaid Work 
Requirements in Arkansas (Jan. 17, 2019), https://www.kff.org/medicaid/
issue-brief/state-data-for-medicaid-work-requirements-in-arkansas/ 
(noting that nearly all of the Arkansas Medicaid enrollees not exempt 
from the reporting requirement did not report any work activities, 
which could result from difficulty accessing the online reporting 
portal).
---------------------------------------------------------------------------
The Proposed Rule Is Likely to Have a Disparate Impact on People of 
        Color
    People of color face significant disparities in access to and 
utilization of care, and often fare worse than white people on measures 
of health status and health outcomes.\52\ In the Notice of Proposed 
Rulemaking, the Department concedes that the proposed changes ``have 
the potential for disparately impacting certain protected groups due to 
factors affecting rates of employment of these groups, [it] find[s] 
that implementation of mitigation strategies and monitoring by the 
Civil Rights Division of FNS will lessen these impacts.'' The 
Department is correct in noting that a consequence of restricting the 
ability of states to request waivers will disproportionately affect 
certain groups, because people of color have far higher unemployment 
rates than white adults.\53\ But no explanation of the mitigation 
strategies and monitoring is provided, so there is no opportunity for 
us to comment on whether the acknowledged disparate impact will in fact 
be mitigated. However, if the proposed rule results in higher rates of 
people of color losing SNAP benefits, this could exacerbate existing 
racial and ethnic disparities in health status.
---------------------------------------------------------------------------
    \52\ Samantha Artiga, et al., ``Key Facts on Health and Health care 
By Race and Ethnicity,'' Kaiser Family Found[a]tion, June 07, 2016, 
https://www.kff.org/disparities-policy/report/key-facts-on-health-and-
health-care-by-race-and-ethnicity/.
    \53\ Bureau of Labor Statistics, https://www.bls.gov/opub/ted/2017/
unemployment-rate-and-employment-population-ratio-vary-by-race-and-
ethnicity.htm.
---------------------------------------------------------------------------
Waivers Should Not Be Determined Predominantly by The Unemployment Rate
    The Department suggests that insufficient jobs are reflected in 
unemployment data, but that data excludes key evidence, such as 
unemployed persons who searched for work in the previous year but not 
in the past 4 weeks, and workers who are part-time for economic 
reasons. According to Bureau of Labor Statistics data, Blacks are twice 
as likely than Whites to have searched for work in the previous year 
but not in the past 4 weeks, and Latinos are 66 percent more likely 
than Whites to work part-time for economic reasons.\54\ These and other 
data points suggest that the proposed core standard for determining 
lack of sufficient jobs, unemployment data, disproportionately impacts 
protected classes.
---------------------------------------------------------------------------
    \54\ ``People in the Labor Force and Not in the Labor Force by 
Selected Characteristics, 2017 Annual Averages,'' U.S. Bureau of Labor 
Statistics, https://www.bls.gov/opub/reports/race-and-ethnicity/2017/
home.htm; ``Employed and Unemployed Full- and Part-time Workers by Age, 
[Sex], Race, and Hispanic or Latino Ethnicity,'' U.S. Bureau of Labor 
Statistics, December 2018, https://www.bls.gov/cps/cpsaat08.htm.
---------------------------------------------------------------------------
Conclusion
    The Administration proposed rule seeks to make changes explicitly 
not intended by Congress, which just concluded a review and 
reauthorization of SNAP in the 2018 Farm Bill and did not make the 
changes proposed.
    We strongly oppose any administrative action by USDA that would 
expose more people to this time limit policy that cuts people from the 
program, thereby cutting their benefits, and putting them at greater 
risk for food insecurity and the host of associated negative 
consequences. Under the law, states have the flexibility to waive areas 
within the state that have experienced elevated unemployment. The rules 
governing areas' eligibility for waivers have been in place for nearly 
20 years and the waiver rules have proven to be reasonable, 
transparent, and manageable for states to operationalize. Adding 
additional barriers to accessing nutritious food will make it even more 
difficult for individuals already facing economic inequity to find and 
maintain employment. By failing to consider existing disparities, the 
proposed policy will only exacerbate food insecurity in our county. Any 
change that would restrict, impede, or add uncertainty to states' 
current ability to waive areas with elevated unemployment should not be 
pursued.
    Feeding America strongly opposes the proposed rule that would 
expose even more people to the arbitrary SNAP food cutoff policy and 
harm our individuals facing food insecurity.
    The only action we encourage USDA to take with respect to this time 
limit rule that impacts Able-Bodied Adults Without Dependents is to 
propose its elimination. Restoring SNAP's ability to provide food 
assistance to impoverished unemployed people would be a powerful policy 
improvement that would reduce food insecurity among those seeking work.
            Sincerely,

Kate Leone,
Chief Government Relations Officer,
Feeding America.
                              attachment 1
Feeding America Food Banks Brace for Increased Need for Food Assistance 
        As Up to One Million Americans Lose Access to Food Stamps
April 8, 2016

    Feeding America, the nation's largest domestic hunger-relief 
organization, today warned that many food banks across the country will 
struggle to meet a significant increase in the need for emergency food 
assistance as between 500,000 and one million Americans are cut from 
the Supplement Nutrition Assistance Program (http://
www.feedingamerica.org/about-us%20press-room%20feeding-america-food-
banks.html) (SNAP, commonly known as food stamps) due to the return in 
many states of a harsh 3 month time limit on SNAP benefits for certain 
SNAP recipients.
    ``This is the equivalent of approximately $75-150 million in lost 
SNAP benefits per month on average, which equates to between 27 and 54 
million meals per month that SNAP recipients will lose,'' said Diana 
Aviv, CEO of Feeding America. ``These totally unnecessary cuts would 
increase demand on the nation's charitable food system at a time when 
food banks and other hunger-relief groups are stretched to meet 
sustained high need.''
    These cuts will affect unemployed adults aged 18-49 who are not 
disabled or raising minor children, also known as ``Able Bodied Adults 
Without Dependents'' ([ABAWDs]). [ABAWDs] are limited to 3 months of 
SNAP benefits in any 36 month period unless they are employed or 
participating in a training program for at least 20 hours a week. Even 
SNAP beneficiaries who are diligently looking for work and whose state 
does not offer them a slot in a work or training program are faced with 
losing their benefits.
    This year 22 states chose to, or were required to, re-impose time 
limits in all or part of the state for the first time since 2008.
    States are not required to offer SNAP recipients a place in a work 
or training program and only five states have pledged to offer a 
qualifying work slot to every individual subject to the 3 month time 
limit. Those impacted by the time limit face significant barriers to 
finding work or enrolling in training programs--25 percent do not have 
a high school degree, 33 percent face physical and mental limitations, 
and 38 percent were formerly incarcerated. They are also among the 
poorest SNAP recipients with an average income of about $2,000 per 
year.
    ``We are deeply concerned about the impact on some of the poorest 
and most vulnerable people in our communities. SNAP is often the only 
program providing benefits to unemployed adults without dependent 
children, and the loss of benefits will be catastrophic for those 
affected. The notion that we can readily make up for this unnecessary 
loss is just not realistic'' Aviv said.
    The Supplemental Nutrition Assistance Program (SNAP) helps millions 
of low-income Americans put food on the table, providing benefits that 
are timely, targeted and temporary. SNAP responds quickly to changes in 
need, growing in response to increases in poverty and unemployment and 
shrinking as need abates. The nutrition assistance program is targeted 
at our most vulnerable citizens, predominantly serving households with 
children, elderly and disabled members.
                              attachment 2
Feeding America Statement On Able-Bodied Adults Without Dependents 
        Proposed Rule
Attributed to Kate Leone, Chief Government Relations Officer
December 20, 2018

    ``Today, the Administration released a Notice of Proposed 
Rulemaking regarding Able-Bodied Adults Without Dependents (ABAWDs) and 
their receipt of Supplemental Nutrition Assistance Program (SNAP) 
benefits. Feeding America, the nation's largest domestic hunger-relief 
organization, is deeply dismayed by this proposal, which will force 
more adults into food insecurity by creating unreasonable restrictions 
on food assistance.
    ``According to recent reports, the United States Department of 
Agriculture (USDA) projects that the proposed rule would cut $15 
billion in benefits from the program over a decade, which Feeding 
America calculates would result in a loss of more than 8.5 billion 
meals from the tables of individuals facing hunger. USDA's most recent 
figures cite 40 million Americans across the United States facing food 
insecurity.
    ``Presently, unemployed or underemployed adults without dependents 
face strict time limits for receiving benefits if they are unable to 
find work. Specifically, adults ages 18 to 50 who do not receive 
disability benefits and do not have children are only able to receive 
SNAP benefits for 3 months, over the course of a 3 year period, unless 
they are working at least 20 hours a week or taking part in a 
comparable workforce program or training.
    ``Current law permits states to waive this rule temporarily in 
areas with elevated unemployment. Nearly every state has opted to use 
these waivers at some time. The proposed rule would effectively do away 
with state waivers by restricting the underlying criteria upon which 
waiver requests can be granted and expanding the grounds upon which 
they can be denied.
    ``By restricting access to ABAWD waivers, this rule would increase 
the risk of food insecurity for nearly one million people. In turn, 
that puts pressure on hunger-relief organizations and it is unlikely 
that our network of food banks can shoulder this burden. For each meal 
provided by Feeding America, SNAP provides 12 meals. Private charity 
simply cannot compensate for the breadth of the impact of cuts to the 
program.
    ``This rule is aimed at individuals who are most in need of our 
help--people who without resources who are unemployed. While 
participating in SNAP, the average income of an unemployed or 
underemployed adult without a family is just 18 percent of the poverty 
line, or about $2,171, per year, for a single-person household in 2018. 
On average, that person's SNAP benefit equates to $170 per month. It is 
inconceivable that we would deny food assistance to a person trying to 
live on just over $2,000 annually.
    ``Today, we anticipate the farm bill being signed into law. During 
the years of debate and negotiations to develop that legislation, there 
were many ideas similar to the Administration's proposed rule. Congress 
soundly rejected all of them, and the farm bill makes improvements to 
SNAP by increasing investments in job training and proven workforce 
management approaches.
    ``Once published in the Federal Register in the coming days, the 
public will have 60 days to generate comments on this proposal, in 
which Feeding America will actively participate. It is imperative that 
the Administration hear just how dangerous this proposal is to the 
health and well-being of many Americans. We encourage the 
Administration to rescind this rule.''
                              attachment 3
Feeding America Opposes Harmful SNAP Proposed Rule Released by USDA
Attributed to Kate Leone, Chief Government Relations Officer
February 1, 2019

    ``Feeding America is disappointed that the United States Department 
of Agriculture (USDA) continues to push a policy that will take 
billions of meals away from people struggling with hunger with the 
publication of their proposed rule to restrict states' ability to waive 
time limits on Supplemental Nutrition Assistance Program (SNAP) 
benefits in high unemployment areas. USDA projects that the proposed 
rule would cut $15 billion in benefits from the program over a decade. 
Feeding America, the nation's largest domestic hunger-relief 
organization, calculates that this would result in a loss of more than 
8.5 billion meals from the tables of people facing hunger. This rule 
would increase the risk of food insecurity for nearly one million 
people, which would put additional pressure on our network of 200 
member food banks. Private charity simply cannot compensate for the 
breadth of the impact of cuts to the program, as SNAP provides 12 meals 
for each meal provided by Feeding America.
    ``Presently, unemployed or underemployed adults without dependents 
face strict time limits for receiving benefits if they are unable to 
find work. Specifically, adults ages 18 to 50 who do not receive 
disability benefits and do not have children are only able to receive 
SNAP benefits for 3 months, over the course of a 3 year period, unless 
they are working at least 20 hours a week or taking part in a 
comparable workforce program or training.
    ``Current law permits states to waive this rule temporarily in 
areas with elevated unemployment. Nearly every state has opted to use 
these waivers at some time. The proposed rule would effectively do away 
with state waivers by restricting the underlying criteria upon which 
waiver requests can be granted and expanding the grounds upon which 
they can be denied.
    ``This rule is aimed at individuals who are in great need of our 
help--people without resources who are unemployed. While participating 
in SNAP, the average income of an unemployed or underemployed adult 
without dependents is just 18 percent of the poverty line or about 
$2,171 per year in 2018. On average, that person's SNAP benefit equates 
to $170 per month. It is inconceivable that we would deny food 
assistance to a person trying to live on just over $2,000 annually.
    ``The reality of low-wage employment is that individuals often face 
volatile job schedules and insufficient work hours, even if they are 
willing to work more. Over the past several weeks, the government 
shutdown provided a stark illustration of the impact one missed 
paycheck can make, and how little control workers have over their 
schedules. Ironically, this rule poses a threat to low-income workers, 
including Federal employees and contractors, who in the event of 
another extended shutdown could find their hours insufficient to meet 
program rules, threatening access to critical nutrition they need. SNAP 
by increasing investments in job training and proven workforce 
management approaches.
    ``Once published in the Federal Register in the coming days, the 
public will have 60 days to generate comments on this proposal, in 
which Feeding America will actively participate. It is imperative that 
the Administration hear just how dangerous this proposal is to the 
health and well-being of many Americans. We encourage the 
Administration to rescind this rule.''
                                 ______
                                 
 Submitted Comment Letter by Hon. Al Lawson, Jr., a Representative in 
  Congress from Florida; Authored by Center for Law and Social Policy
March 29, 2019

  Ms. Sasha Gerstan-Paal,
  Chief,
  Certification Policy Branch,
  Program Development Division,
  Food and Nutrition Service,
  Alexandria, Virginia

  Re: Proposed Rule: Supplemental Nutrition Assistance Program (SNAP): 
            Requirements for Able-Bodied Adults without Dependents RIN 
            0584-AE57

    Dear Ms. Gerstan-Paal:

    I am writing on behalf of the Center for Law and Social Policy 
(CLASP). CLASP is a national, nonpartisan, anti-poverty nonprofit 
advancing policy solutions for low-income people. We work at both 
Federal and state levels, supporting policy and practice that makes a 
difference in the lives of people living in conditions of poverty.
    CLASP submits the following comments in opposition to the U.S. 
Department of Agriculture's proposed regulation regarding the time 
limits within the Supplemental Nutrition Assistance Program (SNAP) that 
apply to working-age adults without minor children. We are deeply 
concerned by attempts to restrict food assistance to individuals for 
whom SNAP is essential to meeting their basic needs and providing a 
work support. While we strongly support the goal of helping SNAP 
recipients obtain and keep quality jobs that enable them to achieve 
economic security, we believe the proposed restrictions will not 
advance this goal. In fact, because the changes will result in more 
people losing their SNAP benefits, they will make it harder to achieve 
this goal.
    SNAP already has harsh time limits in place requiring states to 
limit food assistance to just 3 months out of every 3 years for most 
working-age adults without minor children, unless they have a 
documented disability or report 20 hours of work or related activities 
each week. This policy alone cuts off hundreds of thousands of 
unemployed people from food assistance when they need it most. The 
proposed rule would make the policy even harsher by taking away food 
from even more people struggling to find steady work. By the Trump 
Administration's own estimates, approximately 755,000 to 851,000 
individuals are at risk of losing food assistance through SNAP under 
the proposed rule.
    In the general comments that follow, we explain in more detail the 
reasons why the Department should immediately withdraw this proposed 
regulation. At the close of our general comments, we address major 
elements of the proposed rule section by section.
1. Background
    SNAP is our nation's most important anti-hunger program. It 
provides food assistance to youth, working families, people with 
disabilities, seniors, and many more. SNAP helps approximately 39 
million people in nearly 20 million households put food on the 
table.\1\ In 2015, SNAP lifted approximately 2.1 million Black people 
(including one million children) \2\ and an estimated 2.5 million 
Latinos (including 1.2 million children) out of poverty.\3\ More than 
ten percent of Asian American and Pacific Islander (AAPI) families 
receive SNAP benefits,\4\ while many more are likely eligible but 
unenrolled due to cultural stigma and insufficient program outreach to 
AAPI groups.\5\
---------------------------------------------------------------------------
    \1\ U.S. Department of Agriculture, ``SNAP Participation,'' Food 
and Nutrition Service, September 2018, https://fnsprod.azureedge.net/
sites/default/files/pd/34SNAPmonthly.pdf.
    \2\ SNAP Helps Millions of African Americans, Center on Budget and 
Policy Priorities, updated February 2018, https://www.cbpp.org/
research/food-assistance/snap-helps-millions-of-african-americans.
    \3\ SNAP Helps Millions of Latinos, Center on Budget and Policy 
Priorities, updated February 2018, https://www.cbpp.org/research/food-
assistance/snap-helps-millions-of-latinos.
    \4\ ``Congressional Tri-Caucus Denounces Cuts to the Supplemental 
Nutrition Assistance Program (SNAP),'' Congressional Asian Pacific 
American Caucus, September 2013, https://capac-chu.house.gov/press-
release/congressional-tri-caucus-denounces-cuts-supplemental-nutrition-
assistance-program-snap.
    \5\ Victoria Tran, ``Asian Americans are Falling Through the Cracks 
in Data Representation and Social Services,'' Urban Institute, June 
2018, https://www.urban.org/urban-wire/asian-americans-are-falling-
through-cracks-data-representation-and-social-services.
---------------------------------------------------------------------------
    In addition to fighting hunger, SNAP encourages work in several 
ways.\6\ First, SNAP's structure encourages work because as earnings 
rise, benefits phase out gradually. And because of the earned income 
disregard, earnings are treated more favorably than other income when 
benefits are calculated. Second, SNAP promotes employment by ensuring 
people have their basic needs met. Those working and seeking work on 
SNAP do not have to worry about when they will get their next meal. 
Instead, they can focus their energy on finding and keeping a job.
---------------------------------------------------------------------------
    \6\ Nune Phillips, SNAP and Work, Center for Law and Social Policy, 
January 2018, https://www.clasp.org/sites/default/files/publications/
2018/01/2018_snapandwork.pdf.
---------------------------------------------------------------------------
Access to SNAP Has Positive Effects on Individuals' Long-Term Economic 
        and Educational Attainment, Which in Turn Contribute to Self-
        Sufficiency
    The face of hunger in working-age adults is often hidden. It can 
look like a single mother denying herself her medication so she can buy 
groceries for her family, a college student unable to focus in his 
classes, a hungry young adult unsuccessfully trying to find a job in a 
competitive labor market without money for interview clothes, or even a 
veteran with Post-Traumatic Stress Syndrome choosing between rent, 
heating, and food after serving our country. Studies have shown that 
lack of access to food and proper nutrition exacerbates stress, 
anxiety, and depression,\7\ causes sleep disturbances and fatigue, and 
impairs cognitive functioning \8\--conditions that are a significant 
barrier to finding a job, keeping a job, or getting training to improve 
wages.
---------------------------------------------------------------------------
    \7\ Adrienne O'Neil, Shae E. Quirk, Siobhan Housden, et al., 
``Relationship Between Diet and Mental Health in Children and 
Adolescents: A Systematic Review,'' American Journal of Public Health 
vol. 104, 10 (2014): e31-42, https://www.ncbi.nlm.nih.gov/pmc/articles/
PMC4167
107/.
    \8\ Michael W. Green, Peter J. Rogers, Nicola A. Elliman, and Susan 
J. Gatenby, ``Impairment of Cognitive Performance Associated with 
Dieting and High Levels of Dietary Restraint,'' Physiology & Behavior 
55.3 (1994): 447-452, http://www.seven-health.com/wp-content/uploads/
2018/08/Impairment-of-Cognitive-Performance-Associated-with-Dieting-
and-High-Levels-of-Dietary-Restraint-.pdf.
---------------------------------------------------------------------------
    SNAP is the antidote that helps hungry people become more 
employable \9\ and increase wages.\10\ The SNAP program has also been 
shown to stimulate economic growth,\11\ improve academic outcomes,\12\ 
and improve health outcomes.\13\ SNAP benefits allow recipients to 
spend less money on food and be better able to afford other basic needs 
such as medicine and housing. Subjecting SNAP recipients to time limits 
makes it harder, not easier, for them to become self-sufficient.
---------------------------------------------------------------------------
    \9\ Dottie Rosenbaum, ``The Facts on SNAP, Part 2: SNAP Supports 
Work,'' Center on Budget and Policy Priorities, May 2013, https://
www.cbpp.org/blog/the-facts-on-snap-part-2-snap-supports-work.
    \10\ Ibid., 6.
    \11\ Nune Phillips, SNAP Contributes to a Strong Economy, Center 
for Law and Social Policy, August 2017, https://www.clasp.org/sites/
default/files/SNAP-Contributes-to-a-Strong-Economy.pdf.
    \12\ Nisha Beharie, Micaela Mercado, and Mary McKay, ``A Protective 
Association between SNAP Participation and Educational Outcomes Among 
Children of Economically Strained Households,'' Journal of Hunger & 
Environmental Nutrition vol. 12, 2 (2016): 181-192, https://
www.ncbi.nlm.nih.gov/pmc/articles/PMC5513186/.
    \13\ Steven Carlson and Brynne Keith-Jennings, SNAP Is Linked with 
Improved Nutritional Outcomes and Lower Health Care Costs, Center on 
Budget and Policy Priorities, January 2018, https://www.cbpp.org/
research/food-assistance/snap-is-linked-with-improved-nutritional-
outcomes-and-lower-health-care.
---------------------------------------------------------------------------
    Further, although pregnant woman are exempt from the time limit on 
SNAP benefits, the restrictions will still impact the health of 
pregnant women and babies, because they apply to women who do not yet 
know that they are pregnant, or who do not yet have medical 
documentation of their pregnancies. The U.S. Centers for Disease 
Control strongly recommends that even before conceiving, women achieve 
a healthy weight and nutritious diet in order to maximize their odds of 
a healthy pregnancy.\14\
---------------------------------------------------------------------------
    \14\ Centers for Disease Control and Prevention, ``Before 
Pregnancy: Women'' Centers for Disease Control and Prevention, https://
www.cdc.gov/preconception/women.html.
---------------------------------------------------------------------------
    Nutrition assistance has been documented to promote healthy birth 
outcomes as well as to have long-term benefits for the children of 
recipients. Researchers compared the long-term outcomes of individuals 
in different areas of the country when SNAP expanded nationwide in the 
1960s and early 1970s and found that mothers exposed to SNAP during 
pregnancy gave birth to fewer low-birth-weight babies.\15\ If women in 
early pregnancy are cut off from nutrition services, the negative 
outcomes would extend decades into the future, diminishing their 
children's opportunity to thrive in tangible and entirely preventable 
ways.\16\ Low-income women are already more likely to have poorer 
nutrition and greater stress, which can impair fetal brain development 
and health during pregnancy.\17\ Economic stressors, combined with 
inadequate prenatal care for low-income pregnant women, are associated 
with higher rates of pre-term births and infant mortality.\18\
---------------------------------------------------------------------------
    \15\ Douglas Almond, Hillary Hoynes, and Diane Schanzenbach, 
``Inside the War on Poverty: The Impact of Food Stamps on Birth 
Outcomes,'' The Review of Economics and Statistics, 93(2), May 2011, 
https://www.mitpressjournals.org/doi/pdfplus/10.1162/REST_a_00089; and 
Hilary Hoynes, Diane Whitmore Schanzenbach, and Douglas Almond, ``Long-
Run Impacts of Childhood Access to the Safety Net,'' American Economic 
Review, 106(4): 903-934, April 2016, https://pdfs.semanticscholar.org/
c94b/26c57bb565b566913d2af161e555edeb7f21.pdf.
    \16\ Sharon Parrot, et al., Trump ``Public Charge'' Rule Would 
Prove Particularly Harsh for Pregnant Women and Children, Center on 
Budget and Policy Priorities, (May 1, 2018), available at https://
www.cbpp.org/research/poverty-and-inequality/trump-public-charge-rule-
would-prove-particularly-harsh-for-pregnant.
    \17\ Tess Lefmann, Terri Combs-Orme, ``Prenatal Stress, Poverty, 
and Child Outcomes,'' Child and Adolescent Social Work Journal 31 
(2014), https://link.springer.com/article/10.1007/s10560-014-0340-x.
    \18\ Maternal and Child Health Bureau, Child Health USA 2014: 
Prenatal Care, Health Resources and Services Administration, 2014, 
https://mchb.hrsa.gov/chusa14/dl/chusa14.pdf; Maternal and Child Health 
Bureau, Child Health USA 2013: Barriers to Prenatal Care, Health 
Resources and Services Administration, 2014, https://mchb.hrsa.gov/
sites/default/files/mchb/Data/Chartbooks/childhealth2013.pdf; Centers 
for Disease Control and Prevention, Preterm Birth, 2016, https://
www.cdc.gov/reproductivehealth/MaternalInfantHealth/PretermBirth.htm; 
Child Trends, Preterm Births, 2015, https://www.childtrends.org/wp-
content/uploads/2015/06/indicator_1434209915.291.html.
---------------------------------------------------------------------------
SNAP Already Has Harsh Time Limits and Work Reporting Requirements in 
        Place
    Federal law currently limits adults ages 18-49 without dependent 
children or documented disabilities to just 3 months of SNAP in a 36 
month period unless they engage in work or work-related activities at 
least half time or participate in workfare.\19\ The current rule is 
harsh and unfair. When several states reinstated this time limit in 
2016 after suspending it due to the Great Recession, at least 500,000 
people lost SNAP benefits.\20\ For instance, in a recent Urban 
Institute study of the effects of Kentucky's reinstatement of time 
limits, researchers found that at least 13,000 adults without dependent 
children or documented disabilities lost SNAP benefits because they 
reached the 3 month time limit, representing 20 to 22 percent of the 
caseload subject to time limits.\21\ If the proposed rule goes into 
effect, many more geographic regions will now be required to reinstate 
the time limit. Time limits harm vulnerable people by denying them food 
benefits at a time when they most need it.
---------------------------------------------------------------------------
    \19\ Center on Budget and Policy Priorities, ``Unemployed Adults 
Without Children Who Need Help Buying Food Only Get SNAP For Three 
Months,'' https://www.cbpp.org/unemployed-adults-without-children-who-
need-help-buying-food-only-get-snap-for-three-months.
    \20\ Ed Bolen, Dottie Rosenbaum, Stacy Dean, et al., More Than 
500,000 Adults Will Lose SNAP Benefits in 2016 as Waivers Expire, 
Center on Budget and Policy Priorities, March 2016, https://
www.cbpp.org/research/food-assistance/more-than-500000-adults-will-
lose-snap-benefits-in-2016-as-waivers-expire.
    \21\ Elaine Waxman and Nathan Joo, Urban Institute, ``Reinstating 
SNAP Work-Related Time Limits,'' March 2019 https://www.urban.org/
research/publication/reinstating-snap-work-related-time-limits.
---------------------------------------------------------------------------
    People subject to the time limit are a demographically diverse 
population in terms of race, education, and geography. Nearly \1/2\ (47 
percent) of the individuals subject to the time limit are ages 18 to 
29. Approximately 85 percent have at most a high school diploma or 
equivalent. Approximately 45 percent of the people subject to the time 
limit are women and, among those who report race, an estimated 48 
percent are White, 35 percent are Black, and 13 percent are Latino.\22\ 
People subject to the time limit face particular employment challenges, 
including a lack of reliable transportation, unstable housing 
arrangements, engagement with the criminal justice system, unstable 
work histories, or undiagnosed physical or mental limitations.\23\ In 
particular, those who reside in states that have not expanded Medicaid 
are likely to have trouble getting access to a doctor to document their 
disability.
---------------------------------------------------------------------------
    \22\ Steven Carlson, Dorothy Rosenbaum, and Brynne Keith-Jennings, 
Who Are the Low-Income Childless Adults Facing the Loss of SNAP in 
2016?, Center on Budget and Policy Priorities, February 2016, https://
www.cbpp.org/research/food-assistance/who-are-the-low-income-childless-
adults-facing-the-loss-of-snap-in-2016.
    \23\ Ibid.
---------------------------------------------------------------------------
2. Proposal Does Not Encourage Employment and Would Weaken Economy as a 
        Whole
Time Limits and Work Reporting Requirements Do Not Support Employment
    Unlike work reporting requirements in most public assistance 
programs, SNAP time limit rules do not require states to offer options 
for meeting work reporting requirements before cutting people off 
benefits. Historically, most states have chosen not to help people 
subject to the time limit find qualifying work or training 
activities.\24\ Many individuals will lose SNAP if they cannot find a 
qualifying activity--which does not include job search--on their own.
---------------------------------------------------------------------------
    \24\ United States Government Accountability Office ``Supplemental 
Nutrition Assistance Program: More Complete and Accurate Information 
needed on Employment and Training Programs,'' November 2018, https://
www.gao.gov/assets/700/695632.pdf.
---------------------------------------------------------------------------
    Lessons learned from TANF, SNAP, and other programs demonstrate 
that work reporting requirements are not effective in connecting people 
to living-wage jobs.\25\ As laid out by the Center on Budget and Policy 
Priorities in a review of rigorous evaluations, research shows that 
employment increases among individuals subject to work reporting 
requirements were modest and faded over time. In nearly all of the 
approximately dozen programs evaluated, employment among recipients not 
subject to work reporting requirements was the same as or higher than 
employment among individuals subject to work reporting requirements 
within 5 years.\26\
---------------------------------------------------------------------------
    \25\ Ladonna Pavetti, Work Requirements Don't Cut Poverty, Evidence 
Shows, Center on Budget and Policy Priorities, June 2016, https://
www.cbpp.org/research/poverty-and-inequality/work-requirements-dont-
cut-poverty-evidence-shows.
    \26\ Ibid.
---------------------------------------------------------------------------
    Work reporting requirements are not only ineffectual but have 
opportunity costs: the time that a SNAP recipient loses in low-
intensity programs or low-wage jobs simply to meet requirements could 
have been spent obtaining skills and credentials, finding a quality 
job, and increasing their earnings. A much better focus for public 
policy is to invest in strategies that support people to develop skills 
and access training that prepares them for jobs that pay living wages 
and foster an economy that creates more quality jobs with fair wages.
Proposal Would Grow Government Bureaucracy
    Under the Work Support Strategies (WSS) project, CLASP worked 
closely with six states that sought to dramatically improve the 
delivery of key work support benefits to low-income families, including 
health coverage, nutrition benefits, and child care subsidies through 
more effective, streamlined, and integrated approaches. From this work, 
we learned that reducing unnecessary steps in the application and 
renewal process both reduced burden on caseworkers and made it easier 
for families to access and retain the full package of supports that 
they need to thrive in work and school. Conversely, additional steps 
are burdensome to both caseworkers and participants.
    In order to remain compliant with work reporting requirements, 
recipients must show proof of work. Failure to submit paperwork, even 
if the person meets work reporting requirements, can result in 
terminated SNAP benefits and increased caseload churn. For SNAP 
oversight agencies, tracking work hours, reviewing proof of work, and 
keeping track of who is and is not subject to the work reporting 
requirement every month is a considerable undertaking and is prone to 
caseworker error. Moreover, because the time limit rules are distinct 
from the work registration requirements, states may need to track 
compliance separately and provide participants with separate sets of 
notices informing them of the consequences for non-compliance, which 
further adds to the complexity of administration.
    The complexity of the processes and the ensuing churn will also 
impose administrative costs on social service offices. People who lose 
benefits may later re-apply, which consumes more staff time. One of the 
key lessons of the Work Support Strategies project is that every time 
that a client needs to bring in a verification or report a change adds 
to the administrative burden on caseworkers and increases the 
likelihood that clients will lose benefits due to failure to meet one 
of the requirements.\27\ The WSS states found that reducing 
administrative redundancies and barriers used caseworkers' time more 
efficiently and helped with Federal timeliness requirements.\28\ These 
administrative requirements in the proposed rule are unnecessarily 
burdensome to SNAP agencies.
---------------------------------------------------------------------------
    \27\ Julia B. Isaacs, Michael Katz, Ria Amin, Improving the 
Efficiency of Benefit Delivery Outcomes from the Work Support 
Strategies Evaluation, Urban institute, November 2016, https://
www.urban.org/research/publication/improving-efficiency-benefit-
delivery.
    \28\ Ibid.
---------------------------------------------------------------------------
    In particular, USDA's Office of Inspector General found that SNAP's 
provisions regarding time limit rules are difficult to implement. The 
report finds that states have difficulty implementing time limit rules 
because the requirements are very complex. As a result, implementation 
of time limit rules can be error prone. The report also quoted state 
officials as using terms like ``administrative nightmare'' and 
``operational nightmare'' in describing the time limit rules. State 
officials also expressed concerns regarding the amount of time and 
resources spent implementing time limit provisions.\29\ Many states 
have chosen to waive the maximum areas from the time limits in order to 
simplify program administration and preserve resources for meaningful 
services for participants.
---------------------------------------------------------------------------
    \29\ United States Department of Agriculture, Office of Inspector 
General ``FNS Controls Over SNAP Benefits For Able-Bodied Adults 
Without Dependents'' https://www.usda.gov/oig/webdocs/27601-0002-
31.pdf.
---------------------------------------------------------------------------
Proposal Would Undermine Efforts to Provide Meaningful Training Through 
        Voluntary E&T Programs
    Mandated work programs are harmful because they threaten to take 
away benefits from people who are unable to comply with arbitrary 
rules.\30\ Instead of spending time receiving necessary skills, 
resources, and education, recipients must spend time complying with 
regulations to keep food on their tables, and states must spend time 
and resources on government bureaucracy rather than serving clients 
with the programs needed to succeed. Furthermore, mandatory work 
programs encourage recipients to enter into the labor market sooner, 
with less necessary tools to be successful in finding a stable position 
with livable wages.
---------------------------------------------------------------------------
    \30\ Julia B. Isaacs, Michael Katz, Ria Amin, Improving the 
Efficiency of Benefit Delivery Outcomes from the Work Support 
Strategies Evaluation, Urban institute, November 2016, https://
www.urban.org/research/publication/improving-efficiency-benefit-
delivery.
---------------------------------------------------------------------------
    Over the past decade, some state and local leaders have worked hard 
to intentionally engage SNAP recipients in high-quality, voluntary 
programs that give participants the skills and credentials to achieve 
lasting economic security and develop partnerships for SNAP Employment 
& Training (E&T). The effort to expand high-quality SNAP E&T programs, 
still in early stages, require substantial resources and capacity to 
deliver outcomes. A recent Government Accountability Office (GAO) study 
notes that ``many states have reported to [FNS] that offering employer-
driven, skills-based, intensive employment and training services, such 
as vocational training or work experience, through voluntary programs 
yields more engaged participants with stronger outcomes.'' FNS 
explained to GAO that ``voluntary programs are less administratively 
burdensome than mandatory programs, as they allow states to focus on 
serving motivated participants rather than sanctioning non-compliant 
individuals.'' \31\ Under the proposed rules, this investment in 
quality, high-intensity programs will likely be reduced as some states 
will seek to spread limited SNAP E&T resources thinly to help more 
people meet SNAP time limit rules.
---------------------------------------------------------------------------
    \31\ Ibid.
---------------------------------------------------------------------------
    Instead of penalizing people for being poor and requiring 
assistance to put food on the table, USDA should consider ways to 
create a foundation for long-term economic success. Voluntary SNAP E&T 
programs, for instance, do not subject individuals to sanctions that 
increase food insecurity. In fact, research shows that voluntary 
programs can significantly increase employment, while mandatory SNAP 
E&T programs withhold basic assistance if individuals cannot meet 
participation requirements in a given month.\32\ To attract SNAP 
recipients to voluntary SNAP E&T programs, states can partner with 
trusted service providers that operate programs with a successful track 
record. Given these outcomes, in recent years, states have increasingly 
moved from mandatory to voluntary SNAP E&T programs.\33\
---------------------------------------------------------------------------
    \32\ Ibid., 9.
    \33\ United States Government Accountability Office ``Supplemental 
Nutrition Assistance Program: More Complete and Accurate Information 
needed on Employment and Training Programs,'' November 2018, https://
www.gao.gov/assets/700/695632.pdf.
---------------------------------------------------------------------------
Proposal Would Weaken the Economy as a Whole
    SNAP has historically served as an economic stabilizer in changing 
times. It helps to shorten recessions and dampen the effects of an 
economic cycle in downturn. Without the mitigating effects of SNAP, the 
impact of recessions can escalate. The proposed rule inhibits SNAP from 
rapidly responding to changing economic conditions, and the resulting 
impact on the economy will affect all job seekers. In addition, by the 
Administration's own calculations, the proposed rule would take food 
away from 755,000 to 851,000 low-income Americans, resulting in a loss 
of at least $15 billion in SNAP benefits over 10 years. These cuts will 
also have negative economic ripple effects, as SNAP benefits have been 
shown to have positive multiplier effects on state and local economies 
and to create new agricultural jobs.\34\
---------------------------------------------------------------------------
    \34\ Mark M. Zandi, Assessing the Macro Economic Impact of Fiscal 
Stimulus 2008, January 2008, https://www.economy.com/markzandi/
documents/Stimulus-Impact-2008.pdf; Kenneth Hanson, The Food Assistance 
National Input-Output Multiplier (FANIOM) Model and Stimulus Effects of 
SNAP, U.S. Department of Agriculture, October 2013, https://
www.ers.usda.gov/webdocs/publications/44748/7996_err103_1_.pdf?v=41056; 
``The Benefits of Increasing the Supplemental Nutrition Assistance 
Program Participation in Your State,'' U.S. Department of Agriculture, 
December 2011, https://www.fns.usda.gov/sites/default/files/
bc_facts.pdf; ``Chart Book: SNAP Helps Struggling Families Put Food on 
the Table,'' Center on Budget and Policy Priorities, March 2017, 
https://www.cbpp.org/research/food-assistance/chart-book-snap-helps-
struggling-families-put-food-on-the-table#part8.
---------------------------------------------------------------------------
3. Proposal Would Have a Disparate Impact on People Trying To Make Ends 
        Meet
    We strongly oppose the proposed rule due to its disproportionate 
impact on certain protected classes, including communities of color, 
immigrants, and people with disabilities. The Department acknowledges 
that the rule will have a disparate impact on some populations. It 
notes that the proposed changes ``have the potential for disparately 
impacting certain protected groups due to factors affecting rates of 
employment of these groups, [it] find[s] that implementation of 
mitigation strategies and monitoring by the Civil Rights Division of 
FNS will lessen these impacts.'' But no explanation of the mitigation 
strategies and monitoring is provided, and we do not believe that 
mitigation strategies can be significant enough to fully address the 
disproportionate impact of increased food insecurity and poverty on 
protected classes.
Harm to Communities of Color
    Many people of color face considerable employment challenges and, 
under the proposed rule, would be disadvantaged from accessing critical 
food assistance. Compared to the national average, rates of food 
insecurity are already higher for Black and Latino headed 
households.\35\ Work reporting requirements are also part of a long 
history of racially-motivated critiques of programs supporting basic 
needs, with direct harms to people of color. As discussed in more 
detail in the sections that follow, the proposed rule would 
disproportionately impact communities of color.
---------------------------------------------------------------------------
    \35\ Alisha Coleman-Jensen, Matthew P. Rabbitt, Christian A. 
Gregory, et al., Household Food Security in the United States in 2016, 
U.S. Department of Agriculture, September 2017, https://
www.ers.usda.gov/webdocs/publications/84973/err-237.pdf.
---------------------------------------------------------------------------
Racial Income Disparities Persist in the United States
    Due to persisting racial economic disparities and discrimination in 
hiring practices, average hourly wages for Black and Latino workers are 
substantially lower than their white counterparts.\36\ In 2017, for 
adults age 18-64, the poverty rate of the general population is 11%. 
That percentage is significantly higher for Latinos who have a poverty 
rate of 15% and even higher for Black Americans who have a poverty rate 
of 18%.\37\ This makes it more likely that Black and Latino individuals 
will benefit from programs that support work by helping them access 
nutritious food. The same is true for certain subgroups of Asian and 
Pacific Islanders that are particularly at risk of poverty, such as 
Marshallese (41% poverty rate), Burmese (38%), Hmong (26.1%) and 
Tongans (22.1%).\38\
---------------------------------------------------------------------------
    \36\ Eileen Patten, ``Racial, Gender Wage Gaps Persist in U.S. 
Despite Some Progress,'' Pew Research Center, July 2016, http://
www.pewresearch.org/fact-tank/2016/07/01/racial-gender-wage-gaps-
persist-in-u-s-despite-some-progress/.
    \37\ ``POV-01. Age and Sex of All People, Family Members and 
Unrelated Individuals Iterated by Income-to-Poverty Ratio and Race,'' 
U.S. Census Bureau, 2017, https://www.census.gov/data/tables/time-
series/demo/income-poverty/cps-pov/pov-01.html.
    \38\ ``American Community Survey 2015 Five Year Estimates, table 
DP03,'' U.S. Census Bureau, 2015, https://factfinder.census.gov/faces/
tableservices/jsf/pages/productview.xhtml?src=bkmk.
---------------------------------------------------------------------------
Employment Discrimination Limits Access to the Workforce for Many 
        Immigrants and People of Color
    Studies show that racial discrimination remains a key force in the 
labor market.\39\ In a 2004 study, researchers randomly assigned names 
and quality to resumes and sent them to over 1,300 employment 
advertisements. Their results revealed significant differences in the 
number of callbacks each resume received based on whether the name 
sounded stereotypically White or Black. More recent research indicates 
that this racial bias persists. A study from 2013 submitted fake 
resumes of nonexistent recent college graduates through online job 
applications for positions based in Atlanta, Baltimore, Portland, 
Oregon, Los Angeles, Boston, and Minneapolis. Black people were 16% 
less likely to get called in for an interview.\40\ Similarly, a 2017 
meta-analysis of field experiments on employment discrimination since 
1989 found that white Americans applying for jobs receive on average 
36% more callbacks than Black people and 24% more callbacks than 
Latinos.\41\
---------------------------------------------------------------------------
    \39\ Robert Manduca, Income Inequality and the Persistence of 
Racial Economic Disparities, Sociological Science, March 2018, https://
www.sociologicalscience.com/download/vol-5/march/
SocSci_v5_182to205.pdf.
    \40\ Brett Arends, ``In Hiring, Racial Bias is Still a Problem. But 
Not Always for Reasons You Think,'' Fortune, November 2014, http://
fortune.com/2014/11/04/hiring-racial-bias/.
    \41\ Lincoln Quillian, Devah Pager, Ole Hexel, et al., Meta-
Analysis of Field Experiments Shows No Change in Racial Discrimination 
in Hiring over Time, PNAS October 10, 2017 114 (41) 10870-10875, 
September 2017, https://doi.org/10.1073/pnas.1706255114.
---------------------------------------------------------------------------
Latino and Black Workers Have Been Hardest Hit by the Structural Shift 
        Toward Involuntary Part-Time Work
    Despite wanting to work more, many low-wage workers struggle to 
receive enough hours from their employer to make ends meet. A report 
from the Economic Policy Institute found that 6.1 million workers were 
involuntary part-time; they preferred to work full-time but were only 
offered part-time hours. According to the report, ``involuntary part-
time work is increasing almost five times faster than part-time work 
and about 18 times faster than all work.'' \42\
---------------------------------------------------------------------------
    \42\ Lonnie Golden, ``Still Falling Short on Hours and Pay,'' 
Economic Policy Institute, December 2016, http://www.epi.org/
publication/still-falling-short-on-hours-and-pay-part-time-work-
becoming-new-normal/.
---------------------------------------------------------------------------
    Latino and Black workers are much more likely to be involuntarily 
part-time (6.8 percent and 6.3 percent, respectively) than whites, of 
whom just 3.7 percent work part time involuntarily. And Black people 
and Latinos are a higher proportion of involuntary part-time workers, 
together representing 41.1 percent of all involuntary part-time 
workers. The greater amount of involuntary part-time employment among 
Black people and Latinos is primarily due to their having greater 
difficulty finding full-time work and more often facing work conditions 
in which hours are variable and can be reduced without notice.\43\ 
Historical racial bias and work conditions, in which hours are variable 
and can be reduced without notice, disparately impacts Black people and 
Latinos and increases their likelihood of experiencing involuntary 
part-time employment.\44\
---------------------------------------------------------------------------
    \43\ Ibid.
    \44\ Ibid.
---------------------------------------------------------------------------
People of Color Are More Likely to Live in Neighborhoods with Poor 
        Access to Jobs
    In recent years, majority-minority neighborhoods have experienced 
particularly pronounced declines in job proximity. Proximity to jobs 
can affect the employment outcomes of residents and studies show that 
people who live closer to jobs are more likely to work.\45\ They also 
face shorter job searches and fewer spells of joblessness.\46\ As 
residents from households with low-incomes and communities of color 
shifted toward suburbs in the 2000s, their proximity to jobs decreased. 
Between 2000 and 2012, the number of jobs near the typical Latino and 
Black resident in major metropolitan areas declined much more steeply 
than for white residents.\47\
---------------------------------------------------------------------------
    \45\ Scott W. Allard and Sheldon Danziger, Proximity and 
Opportunity: How Residence and Race Affect the Employment of Welfare 
Recipients, Housing Policy Debate, September 2000, https://
pdfs.semanticscholar.org/4936/dfd925b78d9e81f8d5d44b95b6a15f8ba0ab.pdf.
    \46\ Elizabeth Kneebone and Natalie Holmes, ``The Growing Distance 
Between People and Jobs in Metropolitan America,'' Brookings 
Institution, March 2015, https://www.brookings.edu/research/the-
growing-distance-between-people-and-jobs-in-metropolitan-america/.
    \47\ Ibid.
---------------------------------------------------------------------------
Due to Overcriminalization Of Neighborhoods of Color, People of Color 
        Are More Likely to Have Previous Histories of Incarceration, 
        Which in Turn Limits Their Job Opportunities
    People of color, particularly Black people and Latinos, are 
unfairly targeted by the police and face harsher prison sentences than 
their white counterparts.\48\ National data show that Black people and 
Latinos are three times more likely to be searched than whites \49\ and 
people of color are significantly over-represented in the U.S. prison 
population, making up more than 60 percent of people behind bars.\50\
---------------------------------------------------------------------------
    \48\ Jamal Hagler, ``8 Facts You Should Know About the Criminal 
Justice System and People of Color.'' Center for American Progress, May 
2015, https://www.americanprogress.org/issues/race/news/2015/05/28/
113436/8-facts-you-should-know-about-the-criminal-justice-system-and-
people-of-color/.
    \49\ Lynn Langton, Matthew Durose, et al., Police Behavior during 
Traffic and Street Stops, 2011, U.S. Department of Justice, October 
2016, https://www.bjs.gov/content/pub/pdf/pbtss11.pdf.
    \50\ ``United States profile,'' Prison Policy Initiative, https://
www.prisonpolicy.org/profiles/US.html#disparities.
---------------------------------------------------------------------------
    After release, formerly incarcerated individuals fare poorly in the 
labor market, with most experiencing difficulty finding a job after 
release. Research shows that roughly \1/2\ of people formerly 
incarcerated are still unemployed 1 year after release.\51\ For those 
who do find work, it's common to have annual earnings of less than 
$500.\52\ Further, during the time spent in prison, many lose work 
skills and are given little opportunity to gain useful work 
experience.\53\ People who have been involved in the justice system 
struggle to obtain a driver's license, own a reliable means of 
transportation, acquire relatively stable housing, and maintain proper 
identification documents. These obstacles often prevent formerly 
incarcerated persons from successfully re-entering the job market and 
are compounded by criminal background checks, which further limits 
their access to employment.\54\ A recent survey found that 96 percent 
of employers conduct background checks on job applicants that include a 
criminal history search.\55\
---------------------------------------------------------------------------
    \51\ Adam Looney and Nicholas Turner, Work and Opportunity Before 
and After Incarceration, The Brookings Institution, March 2018, https:/
/www.brookings.edu/research/work-and-opportunity-before-and-after-
incarceration/; Joan Petersilia, When Prisoners Come Home: Parole and 
Prisoner Reentry, Chicago, Ill: University of Chicago Press, 2003, 
https://www.amazon.com/When-Prisoners-Come-Home-Prisoner/dp/0195386124; 
Jeremy Travis, But They All Come Back: Facing the Challenges of 
Prisoner Reentry, Washington, D.C.: Urban Institute Press, 2005, 
https://www.amazon.com/But-They-All-Come-Back/dp/0877667500.
    \52\ Ibid., 40.
    \53\ Christy Visher, Sara Debus, and Jennifer Yahner, Employment 
after Prison: A Longitudinal Study of Releasees in Three States, The 
Urban Institute, October 2008, https://www.urban.org/sites/default/
files/publication/32106/411778-Employment-after-Prison-A-Longitudinal-
Study-of-Releasees-in-Three-States.PDF.
    \54\ Marina Duane, Nancy La Vigne, Mathew Lynch, et al., Criminal 
Background Checks: Impact on Employment and Recidivism, Urban 
Institute, March 2017, https://www.urban.org/sites/default/files/
publication/88621/2017.02.28_criminal_background_checks_report_final
ized_blue_dots_1.pdf.
    \55\ Thomas Ahearn, ``Survey Finds 96 Percent of Employers Conduct 
Background Screening,'' Employment Screening Resources, August 2017, 
http://www.esrcheck.com/wordpress/2017/08/03/survey-finds-96-percent-
of-employers-conduct-background-screening/.
---------------------------------------------------------------------------
People of Color May Be Less Likely to Receive Exemptions Based on 
        Health Conditions
    Research suggests that people of color, in particular Black people, 
may be negatively impacted by racial bias in pain assessment and 
treatment recommendations, which would affect their ability to receive 
exemptions based on health conditions. One study found individuals with 
at least some medical training hold false beliefs about race that 
inform medical judgements, which may contribute to racial disparities 
in pain assessment and inadequate treatment recommendations for Black 
patients' pain.\56\ Further, the Government Accountability Office (GAO) 
found in the early-1990s that Black people with serious ailments were 
much more likely than White people to be rejected for benefits under 
Social Security disability programs.\57\ While this particular analysis 
has not been repeated recently, there remains widespread evidence of 
disparities in medical treatment. These findings suggest that people of 
color may be less likely to receive exemptions based on health 
conditions, potentially subjecting more people to time limit rules than 
would otherwise be the case.
---------------------------------------------------------------------------
    \56\ Kelly M. Hoffman, et al., ``Racial bias in pain assessment and 
treatment recommendations, and false beliefs about biological 
differences between blacks and whites'' https://www.ncbi.nlm.nih.gov/
pmc/articles/PMC4843483/.
    \57\ Stephen Labaton, ``Benefits Are Refused More Often To Disabled 
Blacks, Study Finds'' New York Times, May 1992, https://
www.nytimes.com/1992/05/11/us/benefits-are-refused-more-often-to-
disabled-blacks-study-finds.html.
---------------------------------------------------------------------------
Work Reporting Requirements Are Part of a Long History of Racially-
        Motivated Critiques of Programs Supporting Basic Needs
    False race-based narratives have long surrounded people 
experiencing poverty, with direct harms to people of color. For decades 
these narratives have played a role in discussions around public 
assistance benefits--including SNAP--and have been employed to garner 
support from working-class White people.\58\ Below are a few examples 
of the relationship between poverty, racial bias, and access to basic 
needs programs.
---------------------------------------------------------------------------
    \58\ Josh Levin, ``The Welfare Queen,'' Slate, December 2013,http:/
/www.slate.com/articles/news_and_politics/history/2013/12/
linda_taylor_welfare_queen_ronald_reagan_made_her_a_
notorious_american_villain.html.

   When the ``Mother's Pension'' program was first implemented 
        in the early 1900s, it primarily served white women and allowed 
        mothers to meet their basic needs without working outside of 
        the home. Only when more African American women began to 
        participate were work reporting requirements implemented.\59\
---------------------------------------------------------------------------
    \59\ Rachel Black and Aleta Sprague, ``Republicans' Fixation on 
Work Requirements is Fueled by White Racial Resentment,'' Slate, June 
2018, https://slate.com/human-interest/2018/06/trump-administrations-
fixation-on-work-requirements-for-snap-benefits-is-part-of-a-long-
racist-policy-history.html.

   Between 1915 and 1970, over six million African American 
        people fled the south in the hope of a better life. As more 
        African Americans flowed north, northern states began to adopt 
        some of the work reporting requirements already prevalent in 
        assistance programs in the South.\60\
---------------------------------------------------------------------------
    \60\ Kali Grant, Funke Aderonmu, Sophie Khan, et al., Unworkable 
and Unwise: Conditioning Access to Programs that Ensure a Basic 
Foundation for Families on Work Requirements, Economic Security and 
Opportunity Initiative at Georgetown Law, February 2019, http://
www.georgetownpoverty.org/issues/tax-benefits/unworkable-unwise/.

   As civil rights struggles intensified, the media's portrayal 
        of poverty became increasingly racialized. In 1964, only 27 
        percent of the photos accompanying stories about poverty in 
        three of the country's top weekly news magazines featured Black 
        people; by 1967, 72 percent of photos accompanying stories 
        about poverty featured Black people.\61\
---------------------------------------------------------------------------
    \61\ Rachel Black and Aleta Sprague, ``The Rise and Reign of the 
Welfare Queen,'' New America, September 2016, https://
www.newamerica.org/weekly/edition-135/rise-and-reign-welfare-queen/.

   Many of Ronald Reagan's presidential campaign speech 
        anecdotes centered around a Black woman from Chicago who had 
        defrauded the government. These speeches further embedded the 
        idea of the Black ``welfare queen'' as a staple of dog whistle 
        politics, suggesting that people of color are unwilling to 
        work.\62\
---------------------------------------------------------------------------
    \62\ Gene Demby, ``The Truth Behind the Lies of the Original 
'Welfare Queen','' National Public Radio, December 2013, https://
www.npr.org/sections/codeswitch/2013/12/20/255819681/the-truth-behind-
the-lies-of-the-original-welfare-queen.

   In 2018, prominent sociologists released a study looking at 
        racial attitudes on welfare. They noted that white opposition 
        to public assistance programs has increased since 2008--the 
        year that Barack Obama was elected. The researchers also found 
        that showing white Americans data suggesting that white 
        privilege is diminishing led them to express more opposition to 
        spending on programs like SNAP. They concluded that the 
        ``relationship between racial resentment and welfare opposition 
        remains robust.'' \63\
---------------------------------------------------------------------------
    \63\ Rachel Wetts and Robb Willer, Privilege on the Precipice: 
Perceived Racial Status Threats Lead White Americans to Oppose Welfare 
Programs, Social Forces, Volume 97, Issue 2, 1 December 2018, Pages 
793-822, May 2018, https://academic.oup.com/sf/article/97/2/793/
5002999.
---------------------------------------------------------------------------
The Unemployment Rate Does Not Reflect Opportunities Available to 
        People of Color and, Because of Barriers to Employment, there 
        is a Disproportionate Rate of Employment for People of Color
    A reduction in time limit waivers and the resulting loss in SNAP 
benefits will disproportionately affect certain protected classes based 
on (a) an inadequate method for determining lack of sufficient jobs, a 
criterion for approving time limit waivers; and (b) the 
disproportionate rate of unemployment and underemployment for people of 
color.
    First, the Department suggests that insufficient jobs are reflected 
in unemployment data, but that data excludes key evidence, such as 
unemployed persons who searched for work in the previous year but not 
in the past 4 weeks, and workers who are part-time for economic 
reasons. According to Bureau of Labor Statistics data, Black people are 
twice as likely than White people to have searched for work in the 
previous year but not in the past 4 weeks, and Latinos are 66 percent 
more likely than White people to work part-time for economic 
reasons.\64\ These data points suggest that the proposed core standard 
for determining lack of sufficient jobs, unemployment data, 
disproportionately impacts protected classes.
---------------------------------------------------------------------------
    \64\ ``People in the Labor Force and Not in the Labor Force by 
Selected Characteristics, 2017 Annual Averages,'' U.S. Bureau of Labor 
Statistics, https://www.bls.gov/opub/reports/race-and-ethnicity/2017/
home.htm; ``Employed and Unemployed Full- and Part-time Workers by Age, 
[Sex], Race, and Hispanic or Latino Ethnicity,'' U.S. Bureau of Labor 
Statistics, December 2018, https://www.bls.gov/cps/cpsaat08.htm.
---------------------------------------------------------------------------
    Second, because of the systemic barriers to employment facing 
communities of color described in detail above, there is a 
disproportionate rate of employment for people of color. For instance, 
nationwide, the unemployment rate for Black people was 9.5 percent and 
6.0 percent for Latinos, compared to 4.5 percent for their White 
counterparts in 2017.\65\ Further, even within states, unemployment 
rates for Black people and Latinos are still relatively higher than 
their White counterparts. For example, in California--a state with a 
statewide time limit waiver in place--the unemployment rate was 5.9 
percent in 2017.\66\ However, the unemployment rate was considerably 
higher for Black people and Latinos in California in 2017; 10.7 percent 
for Black people and 6.7 percent for Latinos, compared to 5.5 percent 
for their White counterparts.\67\
---------------------------------------------------------------------------
    \65\ Census Bureau, American FactFinder, 2017 American Community 
Survey 1-Year Estimates, Table S0201 https://factfinder.census.gov/
faces/tableservices/jsf/pages/productview.xhtml?pid=
ACS_17_1YR_S0201&prodType=table.
    \66\ Ibid.
    \67\ Ibid.
---------------------------------------------------------------------------
Harm to Immigrants
Immigrant Eligibility for SNAP is Extremely Limited and Current SNAP 
        Participation Is Already Declining
    The Trump Administration's relentless anti-immigrant rhetoric and 
policies are driving low-income immigrant families away from SNAP.\68\ 
The requirements for eligibility in SNAP haven't changed recently but 
immigrant households legally eligible for SNAP benefits stopped 
participating in the program at a higher-than-average rate in 2018.\69\ 
Following welfare reform in 1996, a person must be a U.S. citizen or an 
eligible, lawfully-present non-citizen to qualify for SNAP 
benefits.\70\ Recent data presented at the 2018 American Public Health 
Association Annual Conference shows that after a decade of steady 
increases, enrollment nationwide among immigrant families eligible for 
SNAP has dropped by ten percent.\71\ The study's lead researcher said 
in a press release, ``We believe the drop in participation may be 
related to more nuanced changes in national immigration rhetoric and 
increased Federal action to deport and detain immigrants. These 
findings demonstrate that rhetoric and the threat of policy changes, 
even before changes are enacted, may be causing families to forego 
nutrition assistance.''
---------------------------------------------------------------------------
    \68\ Helena Bottemiller Evich, ``Immigrant Families Appear to be 
Dropping Out of Food Stamps,'' POLITICO, November [2018], https://
www.politico.com/story/2018/11/14/immigrant-families-dropping-out-food-
stamps-966256.
    \69\ Allison Bovell-Ammon, ``Trends in Food Insecurity and SNAP 
Participation Among Immigrant Families of U.S. Born Young Children,'' 
Children's HealthWatch, November 2018, http://childrenshealthwatch.org/
study-following-10-year-gains-snap-participation-among-immigrant-
families-dropped-in-2018/.
    \70\ Supplemental Nutrition Assistance Program: Guidance on Non-
Citizen Eligibility, U.S. Department of Agriculture, June 2011, https:/
/fns-prod.azureedge.net/sites/default/files/snap/Non-
Citizen_Guidance_063011.pdf.
    \71\ ``Study: Following 10-year gains, SNAP participation among 
immigrant families dropped in 2018,'' American Public Health 
Association, November 2018, https://www.apha.org/news-and-media/news-
releases/apha-news-releases/2018/annual-meeting-snap-participation.
---------------------------------------------------------------------------
    In addition, immigrants are often unaware of the SNAP program or 
are confused about their eligibility for benefits.\72\ Many immigrants 
in mixed-status families are not aware that some of their family 
members are eligible for SNAP, and immigrants face complicated 
administrative burdens due to caseworkers' lack of familiarity with 
foreign identity documents.\73\ In fact, most Federal agencies have 
been working to overcome the barriers immigrants face to enrolling in 
benefits rather than adopting policies such as this proposal, which 
will only exacerbate current disparities in immigrant access to the 
SNAP program.\74\ Given SNAP's record of alleviating poverty and food 
insecurity and improving health and employment outcomes, the USDA 
should be working to remove the barriers immigrant families face in 
accessing SNAP rather than further restricting access.
---------------------------------------------------------------------------
    \72\ Susan Bartlett, Nancy Burstein, William Hamilton, et al., Food 
Stamp Access Study: Final Report, U.S. Department of Agriculture, 
November 2004, https://naldc.nal.usda.gov/download/45671/PDF.
    \73\ Krista M. Perreira, Robert Crosnoe, Karina Fortuny, et al., 
Barriers to Immigrants' Access to Health and Human Services Programs, 
U.S. Department of Health and Human Services, May 2012, https://
aspe.hhs.gov/system/files/pdf/76471/rb.pdf.
    \74\ Robert Crosnoe, Juan Manuel Pedroza, Kelly Purtell, et al., 
Promising Practices for Increasing Immigrants' Access to Health and 
Human Services, U.S. Department of Health and Human Services, May 2012, 
https://aspe.hhs.gov/basic-report/promising-practices-increasing-
immigrants-access-health-and-human-services.
---------------------------------------------------------------------------
Harm to People with Disabilities
    People who are unable to work due to disability or illness are 
likely to lose food assistance under the proposal. Although the statute 
and regulations both provide for exemptions from the time limit for 
individuals with work limitations, the reality in practice is that many 
individuals with disabilities are not identified and granted 
exemptions. In many states, only individuals who are receiving 
government disability benefits are exempted from the time limit.
    Many individuals characterized as able-bodied adults have 
significant physical or mental barriers to employment. In a Franklin 
County, Ohio report, approximately \1/3\ of individuals characterized 
as able-bodied reported having a ``physical or mental limitation.'' 
\75\ Of those, 25 percent indicated that the condition limited their 
daily activities, and nearly 20 percent had filed for Disability/SSI 
within the previous 2 years.\76\ Although some conditions may not meet 
the stringent standard to qualify the individual for a Federal 
disability benefit, they still may have significant barriers to working 
20 hours or more per week. For instance, BLS reported that \1/2\ of 
working-age adults with a disability who were not working reported 
barriers to employment, including a lack of transportation and the need 
for accommodations in a workplace.\77\ Another BLS report shows that 
workers with disabilities are nearly twice as likely as workers with no 
disability to be employed part-time.\78\
---------------------------------------------------------------------------
    \75\ Franklin County Work Experience Program, Ohio Association of 
Foodbanks, 2015, http://admin.ohiofoodbanks.org/uploads/news/
ABAWD_Report_2014-2015-v3.pdf.
    \76\ Ibid.
    \77\ U.S. Department of Labor, ``Persons with a Disability: 
Barriers to Employment, Types of Assistance, and Other Labor-Related 
Issues--May 2012,'' Bureau of Labor Statistics, April 2013, https://
www.bls.gov/news.release/archives/dissup_04242013.pdf.
    \78\ U.S. Department of Labor, ``Persons with a Disability: Labor 
Force Characteristics--2016,'' Bureau of Labor Statistics, June 2017, 
https://www.bls.gov/news.release/pdf/disabl.pdf.
---------------------------------------------------------------------------
    Additionally, we know that many disabilities go undiagnosed either 
because they are difficult to diagnose or the person does not have the 
resources to seek out a diagnosis. Moreover, many people who are unable 
to work due to disability fail to receive an exemption because of the 
complexity of paperwork required for exemptions. A Kaiser Family 
Foundation study found that 36 percent of unemployed adults receiving 
Medicaid reported illness or disability as their primary reason for not 
working but were not receiving Disability/SSI.\79\ Because of the 
historic unemployment and underemployment of people with disabilities--
which workforce and SNAP employment systems are not adequately 
structured or funded to solve--a reduction in time limit waivers would 
result in the loss of crucial nutrition assistance for large numbers of 
low-income people with disabilities.
---------------------------------------------------------------------------
    \79\ Rachel Garfield, Robin Rudowitz, and Anthony Damico, 
Understanding the Intersection of Medicaid and Work, February 2017, 
http://kff.org/medicaid/issue-brief/understanding-the-intersection-of-
medicaid-and-work/.
---------------------------------------------------------------------------
Harm to College Students
Many Students Cannot Meet Requirements of Proposed Rule
    Students enrolled at least half-time are not subject to the time 
limit, and this will not change under the proposed rule. However, 
students enrolled less than half-time and not otherwise exempt will now 
be at increased risk of losing benefits under the proposed rule if they 
are unable to meet SNAP time limit rules. Many low-income students must 
work part-time to support themselves and their families, and therefore 
enroll in college less than half-time. However, an 80 hours per month 
requirement does not allow enough time for many students to be able to 
attend classes and complete their homework.
    Given these challenges, this rule stands in direct contradiction to 
its stated principle of `` . . . improv[ing] employment outcomes and 
economic independence.'' This rule will limit the ability of students 
with low incomes to successfully maintain SNAP and complete a post-
secondary education that can lead to quality employment with family-
sustaining wages \80\ and employer sponsored healthcare and retirement 
savings.\81\
---------------------------------------------------------------------------
    \80\ ``Measuring the Value of Education,'' U.S. Bureau of Labor 
Statistics, April 2018, https://www.bls.gov/careeroutlook/2018/data-on-
display/education-pays.htm.
    \81\ Teresa Kroeger and Elise Gould, The Class of 2017, Economic 
Policy Institute, May 2017, https://www.epi.org/publication/the-class-
of-2017/.
---------------------------------------------------------------------------
Proposal Would Exacerbate Confusion about Students' Eligibility for 
        SNAP
    SNAP has specific rules that determine which low-income students 
can receive food assistance. Low-income post-secondary students who are 
enrolled at least half-time and not otherwise exempt must meet all of 
the standard SNAP eligibility rules, as well as one of several 
additional qualifications, such as working at least 80 hours a month, 
participating in work-study, or participating in an employment and 
training program.\82\ According to the Government Accountability Office 
(GAO), post-secondary officials and students report being confused by 
these student rules. This leads to misinformation about the 
availability of SNAP on campus and low SNAP enrollment. A reported 57 
percent of potentially eligible students (those who have low incomes, 
and at least one additional risk factor for food insecurity) are not 
presently accessing SNAP.\83\ The proposed time limit rule will add to 
the confusion by imposing harsh restrictions on students who are 
enrolled less than half-time and trying to meet their basic need for 
food through SNAP. The proposed rule compounds the challenges of 
maintaining SNAP and undermines education activities that could lead to 
greater economic contributions and increased productivity.\84\
---------------------------------------------------------------------------
    \82\ ``Supplemental Nutrition Assistance Program (SNAP),'' Food and 
Nutrition Service, n.d., https://www.fns.usda.gov/snap/students; Carrie 
Welton, SNAP and Students: Food Assistance Can Support College Success, 
Center for Law and Social Policy, January 2019, https://www.clasp.org/
publications/fact-sheet/snap-and-students-food-assistance-can-support-
college-success.
    \83\ Food Insecurity: Better Information Could Help Eligible 
College Students Access Federal Food Assistance Benefits, U.S. 
Government Accountability Office, December 2018, https://www.gao.gov/
assets/700/696254.pdf.
    \84\ Noah Berger and Peter Fisher, A Well-Educated Workforce Is Key 
to State Prosperity, Economic Policy Institute, August 2013, https://
www.epi.org/publication/states-education-productivity-growth-
foundations/.
---------------------------------------------------------------------------
Proposed Rule Does Not Reflect Recent Changes in the Student Body
    Students who enroll full-time right after high school, receive help 
from their parents, and do not work during the school year are no 
longer the norm on college campuses.\85\ A recent report from the GAO 
demonstrated that 71 percent of undergraduate students now have at 
least one characteristic that complicates their ability to attend 
classes such as being financially independent from their parents. The 
additional financial strain of independence can contribute to lower 
retention and graduation rates as compared to their ``traditional'' 
counterparts, emphasizing the need for more robust and diverse 
supports.
---------------------------------------------------------------------------
    \85\ Ibid., 17.
---------------------------------------------------------------------------
    A reported 39 percent of all undergraduate students have a 
household income at or below 130 percent of the Federal poverty line. 
The GAO reported that the highest student risk of food insecurity is 
being low-income and the second is being a first-generation college 
student.\86\ In spite of the risk of food insecurity, low-income 
students are enrolling in college at rates that now exceed that of 
their middle-income peers.\87\ But this proposed rule would increase 
food insecurity and interfere with students' ability to attend and 
complete college.
---------------------------------------------------------------------------
    \86\ Ibid.
    \87\ Table 302.30. Percentage of Recent High School Completers 
Enrolled in College, by Income Level: 1975 through 2016, National 
Center for Education Statistics, July 2017, https://nces.ed.gov/
programs/digest/d17/tables/dt17_302.30.asp?current=yes.
---------------------------------------------------------------------------
Proposal Would Undermine Students' Completion of Post-Secondary 
        Education
    Analyses of the labor market over the past decade illustrate the 
considerable barriers to getting and maintaining employment without 
some form of post-secondary education. Research shows workers with a 
high school diploma or less lost 5.6 of the 7.2 million jobs wiped out 
in the Great Recession. These workers have recovered less than 80,000 
jobs in the decade since, while those with a bachelor's degree gained 
4.6 million jobs in the recovery. Ninety-nine percent of the jobs 
created since the Great Recession have gone to those with some form of 
post-secondary education.\88\ Workers with a post-secondary education 
also have the majority of jobs with livable wages and employer provided 
benefits.
---------------------------------------------------------------------------
    \88\ Anthony P. Carnevale, Tamara Jayasundera, and Artem Gulish, 
America's Divided Recovery College Haves and Have-Nots, Center on 
Education and the Workforce, 2016, https://cew.georgetown.edu/wp-
content/uploads/Americas-Divided-Recovery-web.pdf.
---------------------------------------------------------------------------
    Low-income individuals continue to enroll in post-secondary 
programs at increasing rates because they understand that post-
secondary education is the most reliable pathway to economic security. 
Without access to SNAP, low-income students who are food-insecure may 
struggle to persist in and successfully complete their post-secondary 
education.\89\ The proposed rule is therefore incredibly short-sighted 
in limiting student success in post-secondary education.\90\
---------------------------------------------------------------------------
    \89\ Duy Pham, Benefits Access for College Completion: Lessons 
Learned from a Community College Initiative to Help Low-Income 
Students, Center for Law and Social Policy, July 2016, https://
www.clasp.org/blog/benefits-access-college-completion-lessons-learned-
community-college-initiative-help-low-income.
    \90\ Ibid., 21.
---------------------------------------------------------------------------
Harm to Young Adults
    The proposed rule would have a disparate impact on youth and youth 
of color, given the considerable barriers they face in entering the 
labor market and maintaining employment. Nationwide, approximately 4.6 
million young adults ages 16 to 24 are out of school and 
unemployed.\91\ In 2018, the youth unemployment rate (9.2%) was more 
than double the overall unemployment rate of 3.9 percent.\92\
---------------------------------------------------------------------------
    \91\ Sarah Burd-Sharps and Kristen Lewis, More than a Million 
Reasons for Hope Youth Disconnection in America Today, Measure of 
America, March 2018, http://measureofamerica.org/youth-disconnection-
2018/.
    \92\ Table 2. Employment Status of the Civilian Noninstitutional 
Population 16 to 24 Years of Age by Sex, Race, and Hispanic or Latino 
Ethnicity, July 2015-2018, U.S. Bureau of Labor Statistics, August 
2018, https://www.bls.gov/news.release/youth.t02.htm; Household Data 
Annual Averages: 1. Employment Status of the Civilian Noninstitutional 
Population, 1948 to Date, U.S. Bureau of Labor Statistics, accessed 
March 2019, https://www.bls.gov/cps/cpsaat01.pdf.
---------------------------------------------------------------------------
    Among young adults, Black people (16.5 percent) and Latinos (10.8 
percent) have considerably higher rates of unemployment.\93\
---------------------------------------------------------------------------
    \93\ Ibid., 83a.
---------------------------------------------------------------------------
    Even when employed, young adults are more likely than older workers 
to have jobs with low wages and no benefits.\94\ Some struggle to 
receive enough hours from their employer to make ends meet. According 
to the Economic Policy Institute, young workers 16 to 24 years of age 
are more likely to be working part-time involuntarily among all age 
groups and account for approximately 28 percent of all involuntary part 
time workers, despite comprising 13 percent of those at work.\95\
---------------------------------------------------------------------------
    \94\ Alleviating Poverty for Opportunity Youth, JFF, December 2018, 
https://www.jff.org/resources/alleviating-poverty-opportunity-youth/.
    \95\ Ibid., 40.
---------------------------------------------------------------------------
    Furthermore, young adult workers are more likely to experience 
fluctuating work hours common to youth-hiring sectors such as retail, 
restaurants, agriculture, construction, and other services. For 
example, approximately 90 percent of young food service workers 
reported that their hours fluctuated in the last month by 68 percent, 
on average. In addition, \1/2\ of retail workers reported that they 
know their work schedule just 1 week or less in advance, and \1/2\ of 
janitors and housekeepers reported that their employer completely 
controls the timing of their work.\96\
---------------------------------------------------------------------------
    \96\ Susan J. Lambert, Peter J. Fugiel, and Julia R. Henly, 
Schedule Unpredictability Among Early Career Workers in the U.S. Labor 
Market: A National Snapshot, University of Chicago: Employment 
Instability, Family Well-being, and Social Policy Network, August 2014, 
http://www.academia.edu/21504026/
Schedule_Unpredictability_among_Early_Career_Workers_in_the_ 
US_Labor_Market_A_National_Snapshot.
---------------------------------------------------------------------------
    Young adults in these jobs use SNAP to help them cover basic needs, 
but many youths will lose SNAP under the proposed rule when their hours 
fall below 20 hours per week. The proposed rule penalizes young adults 
who struggle to find stable employment by increasing food insecurity.
4. Analysis of Major Elements of the Proposed Rule
    The majority of our comments to this point have addressed the 
harmful impact of the rule as a whole because different sections 
interact in ways that have a greater impact than any individual 
section. In order to ensure that our input is fully captured in the 
Department's analysis of the comments received, the following section 
addresses key elements of the proposed rule.
Conformance with the Agriculture Improvement Act of 2018 (Farm Bill)
    The just enacted Agriculture Improvement Act of 2018 maintains 
current law regarding SNAP time limit rules. The explanation given by 
the Act's Joint Explanatory Statement of the Committee of Conference 
is, ``the Managers . . . acknowledge that waivers from the ABAWD time 
limit are necessary in times of recession and in areas with labor 
surpluses or higher rates of unemployment.''
    While the NPRM states that applying SNAP time limit rules more 
broadly is in alignment with the House-passed Agriculture and Nutrition 
Act of 2018, H.R. 2, that bill did not ultimately become law. The final 
Agriculture Improvement Act of 2018 retains the SNAP time limit in 
current law and strikes the House bill modifications. In a letter to 
Secretary P[e]rdue requesting that the proposed rule be withdrawn, 
Senators Stabenow and Murkowski as well as 45 more Senators clarify 
Congressional intent. The letter from U.S. Senators states:

          In addition to being out line with Congressional intent 
        related to waivers, this rule also directly contradicts 
        Congressional direction related to waiver submissions and 
        carry-over exemptions included in the 2018 Farm Bill report. 
        This report, written by Chairman Pat Roberts, Ranking Member 
        Debbie Stabenow, Chairman Mike Conaway and Ranking Member 
        Collin Peterson and approved by the 369 Members of the House 
        and 87 Members of the Senate, explicitly directs the Department 
        not to make the changes made in this rule. This unilateral 
        Administrative action is in direct contradiction to the will of 
        Congress.\97\
---------------------------------------------------------------------------
    \97\ Senator Stabenow and Senator Murkowski, U.S. Senate, Letter to 
Secretary P[e]rdue, March 28, 2019, http://www.frac.org/wp-content/
uploads/19-03-28-Letter-to-Perdue-re-ABAWD-Rule.pdf.

    In contrast to the new law, the NPRM often cites the goal of 
ensuring that more people are subject to SNAP time limits and work 
reporting requirements as a justification for policy changes. For 
example, in describing options for a six percent floor, the NPRM states 
that, ``the Department is concerned that too many areas would qualify 
for a waiver of the ABAWD time limit and that too few individuals would 
be subject to the ABAWD work requirements.'' The Department's proposed 
floor of seven percent seems arbitrary and devised to produce the 
desired result of more individuals being subject to work reporting 
requirements, a goal that does not reflect the goals of Congress.
    Setting policy goals inconsistent with the intent of the final law 
is an over-reach of Departmental authority. The Department is expected 
to ensure that waivers for SNAP time limit rules are adequately 
responsive to nationwide recessions and relative areas of higher 
unemployment or labor surpluses. The three core standards proposed by 
the Department do not allow that role to be performed adequately.
Federalism Summary Impact Statement
    The proposed rule has federalism implications that contradict the 
intent of both the 2018 Farm Bill and Executive Order 13132. The Joint 
Explanatory Statement of the Committee of Conference of the farm bill 
states that ``the Managers intend to maintain the practice that bestows 
authority on the state agency responsible for administering SNAP to 
determine when and how waiver requests for ABAWDs are submitted.''
    Executive Order 13132 Section 7(b) states that, ``Each agency 
shall, to the extent practicable and permitted by law, consider any 
application by a state for a waiver of statutory or regulatory 
requirements in connection with any program administered by that agency 
with a general view toward increasing opportunities for utilizing 
flexible policy approaches at the state or local level in cases in 
which the proposed waiver is consistent with applicable Federal policy 
objectives and is otherwise appropriate.'' The Federal policy 
objectives stated above specifically maintain state agency 
responsibility for determining when and how waivers are submitted.
    State flexibility is critical to appropriate implementation of SNAP 
time limit rules. Consistent with the view of many researchers and 
agencies including the National Bureau of Economic Research, no single 
measure can truly identify economic downturns and a lack of sufficient 
jobs.\98\ Individual states have in-depth knowledge of their 
communities that allows them to identify qualitative data, Census 
Bureau data, Bureau of Labor Statistics employment-population data and 
U-6 measures, and other high-quality data to best make a case for the 
need for a waiver. The sections below highlight some of the strengths 
of these different measures.
---------------------------------------------------------------------------
    \98\ ``The NBER's Business Cycle Dating Committee,'' The National 
Bureau of Economic Research, September 2010, https://www.nber.org/
cycles/recessions.html.
---------------------------------------------------------------------------
Use of Bureau of Labor Statistics Data for Core Standards
    The United States Bureau of Labor Statistics produces two measures 
of labor under-utilization based on the Current Population Survey that 
will be discussed in this section.\99\
---------------------------------------------------------------------------
    \99\ The Current Population Survey universe is the non-
institutionalized civilian population at least 16 years of age.

  1.  U-3 measure: The U-3 measure is the official unemployment rate, 
            which is proposed by the Department as the basis for two of 
            three core standards: the fixed measure of unemployment 
            rate over ten percent, and the relative measure of 20 
            percent over the average unemployment rate over a 24 month 
            period. The U-3 calculates the unemployed as a percentage 
            of the labor force. The labor force includes employed as 
            well as unemployed, which is defined as those who have no 
            job and have made an attempt to look for work in the past 4 
---------------------------------------------------------------------------
            weeks.

  2.  U-6 measure: The U-6 is an alternative measure of labor 
            underutilization that captures:

      i.  the percentage of people who want and are available for full-
            time work but
                have had to settle for a part-time schedule for 
            economic reasons, such as
                their hours being cut back or being unable to find 
            full-time jobs (termed
                ``employed part-time for economic reasons'').

      ii.  the percentage of people who currently are neither working 
            nor looking
                for work but indicate that they want and are available 
            for a job and have
                looked for work sometime in the past 12 months (termed 
            ``marginally at-
                tached to the workforce.'') Discouraged workers, a 
            subset of the margin-
                ally attached, have given a job-market related reason 
            for not currently
                looking for work.

      iii.  The percentage of people who are unemployed, equivalent to 
            the U-3
                measure.

    These components of the U-6 measure are calculated as a percent of 
the labor force plus all persons marginally attached to the labor 
force.\100\
---------------------------------------------------------------------------
    \100\ Unlike the employment-population ratio (EPOP), the U-6 
measure depends on an individual accurately defining what wanting and 
being available for a job means to them. While EPOP is less subjective, 
the U-6 data more accurately reflects the ABAWD population. EPOP 
includes longer-term discouraged workers but does not distinguish them 
from retired persons or others who are not available to work.
---------------------------------------------------------------------------
    We oppose the proposed rule's heavy and exclusive reliance on the 
U-3 measure for two of its three core standards. The U-3 data tends to 
be biased downward as a reflection of available jobs, because it does 
not include those who are part-time due to a lack of available work or 
who are discouraged for a job-market related reason. Therefore, the U-3 
measure overstates the degree of recovery in the job market.
    The U-6 measure is distinguished from the U-3 by just two subsets 
of workers: workers who are part-time for economic reasons and workers 
who are marginally attached to the workforce. The impact of including 
these two subsets can be demonstrated by calculating a ratio of the U-6 
to U-3 measure: The U-6 measure of labor underutilization is 
significantly higher than the U-3 and the ratio varies geographically. 
As illustrated by graphic 1 in the appendix, some states have 
significantly lower numbers of workers who are part-time for economic 
reasons and/or marginally attached to the workforce than other states.
    Moreover, the U-3 measure does not accurately represent a large 
subset of individuals subject to the SNAP time limit. According to the 
USDA Food and Nutrition Service (FNS), more than 31 percent of 
nonelderly adult SNAP recipients were employed in an average month of 
2016. Many of these SNAP recipients were subject to the time limit who 
were employed part-time for economic reasons. A portion of the 
remaining 69 percent of non-elderly adult SNAP recipients were 
discouraged workers who had not looked for work in the past 4 weeks for 
job-market related reasons. Both subsets of individuals subject to the 
time limit are directly captured in the U-6 measure, but not in the U-
3.
    Exclusion of these subsets of individuals disproportionately 
impacts protected classes, who have higher rates of part-time 
employment for economic reasons and discouraged workers. Please see the 
Civil Rights Impact section below for a further discussion of this 
impact.
    If the Department bases waivers exclusively on U-3 unemployment 
rates, it will not count these individual subsets in a waiver review 
even though lack of sufficient jobs has impacted their employment 
status. If BLS data is to impact individuals' access to SNAP, it is 
imperative to address the U-3 measure's weaknesses.
Development of Core Standards and Other Data and Evidence in 
        Exceptional Circumstances
    While the 2018 Farm Bill requires waivers for SNAP time limit rules 
to be responsive to recessions and areas with labor surpluses and 
higher rates of unemployment, there are inherent challenges in defining 
these economic conditions, including weaknesses in existing data sets, 
complexities in defining recessions, and difficulty in using a single 
data set averaged across different categories of people, industries and 
geographic locations. As a result, many researchers use qualitative 
data to support an understanding of employment challenges. For example, 
the recognized agency for defining recessions, the National Bureau of 
Economic Research, does not use a single formula or data set for a 
definition of a recession.
    We oppose the proposed exclusion of additional data outside of the 
U-3. Additional data can support a picture of the strength of the labor 
market. For example, the BLS employment-population ratio, which 
measures employed persons as a percentage of the entire 
population,\101\ includes individuals who are employable but have not 
looked for a job in more than a year. In periods of severe and long-
term economic recessions, the number of individuals in this category 
will grow and the employment-population ratio will paint a clearer 
picture of the strength of the labor market than other measures.
---------------------------------------------------------------------------
    \101\ EPOP uses the relationship between the ratio of the monthly 
Current Employment Statistics Survey (CES) employment to the population 
and the ratio of the Current Population Survey (CPS) employment to the 
population. EPOP also includes trend and seasonal components to account 
for movements in the CPS not captured in the CES series. The seasonal 
component accounts for the seasonality in the CPS not explained by the 
CES (for example, agricultural employment movement), while the trend 
component adjusts for long-run systematic differences between the two 
series (for example, during expansions, the CES grows faster than the 
CPS).
---------------------------------------------------------------------------
    In addition, Census Bureau data should be an option for waiver 
applications, particularly for sub-state areas. BLS has ``concluded 
that data users often are better served by sub-state area data from the 
Census Bureau's American Community Survey (ACS). Data from the ACS 
provide more extensive geographic and demographic coverage, and have 
smaller sampling errors.\102\ The Census Bureau's ACS sample size is 30 
times larger than that of BLS, which accounts in large part for its 
increased accuracy.\103\ In 2013, the numbers of persons the ACS 
classified as `employed,' `unemployed,' and `not in the labor force' 
for the nation were all higher than the official CPS estimates. The ACS 
unemployment rate was 8.4 percent, compared to the CPS annual average 
of 7.4 percent.'' \104\ The variation in this one example reflects the 
challenge of standardizing the U-3 measure instead of allowing Census 
data to be used.
---------------------------------------------------------------------------
    \102\ More information is available at: ``Notes on Using Current 
Population Survey (CPS) Subnational Data,'' U.S. Bureau of Labor 
Statistics, June 2018, https://www.bls.gov/lau/notescps.htm.
    \103\ More information is available at: ``Fact Sheet: Differences 
Between the American Community Survey (ACS) and the Annual Social and 
Economic Supplement to the Current Population Survey (CPS ASEC),'' U.S. 
Census Bureau, May 2016, https://www.census.gov/topics/income-poverty/
poverty/guidance/data-sources/acs-vs-cps.html.
    \104\ More information is available at: ``American Community Survey 
(ACS) Questions and Answers,'' U.S. Bureau of Labor Statistics, April 
2017, https://www.bls.gov/lau/acsqa.htm.
---------------------------------------------------------------------------
    Finally, data on lack of jobs in declining occupations or 
industries is critical in assessing whether there are enough jobs, and 
should continue to be considered in waiver determinations. While a 
population may as a whole remain employed, a large subset may be 
significantly affected by declining occupations. This is expected to be 
the case, for example, when transportation evolves toward self-driving 
vehicles. While participation in WIOA's dislocated worker program meets 
SNAP time limit rules, there are inadequate opportunities for such 
participation in the United States, with only 400,000 people served 
nationwide in Federal Fiscal Year 2018.\105\
---------------------------------------------------------------------------
    \105\ A total of 406,407 were served from October 1, 2017 to 
September 30, 2018. More information is available, under quarterly 
reports, at: WIOA Performance Results, U.S. Department of Labor, 
February 2019, https://www.doleta.gov/performance/results/.
---------------------------------------------------------------------------
    The proposed rules would restrain SNAP from rapidly responding to 
changing economic conditions. Notably, according to the Brookings 
Institution, the most simulative type of spending during the Great 
Recession was a temporary increase in the SNAP maximum benefit, which 
was quicker to respond to deteriorating economic conditions than 
Congressional action and more effective dollar for dollar than 
increased spending on infrastructure and defense.\106\ Without the 
mitigating effects of SNAP, the impact of recessions can escalate. The 
USDA's Economic Research Service uses the Food Assistance National 
Input-Output Multiplier (FANIOM) model to estimate the multiplier 
effects from SNAP benefits at 1.79, which is a significant economic 
boost. We strongly oppose any changes that dilute the impact of SNAP 
benefits as an automatic stabilizer.
---------------------------------------------------------------------------
    \106\ Diane Whitmore Schanzenbach, Ryan Nunn, Lauren Bauer, et al., 
Nine Facts About the Great Recession and Tools for Fighting the Next 
Downturn, The Brookings Institution, May 2016, https://
www.brookings.edu/research/nine-facts-about-the-great-recession-and-
tools-for-fighting-the-next-downturn/.
---------------------------------------------------------------------------
    Economic downturns are not exceptional circumstances and should not 
be treated as such. The exceptional circumstances floor of ten percent 
is far too high to reflect the lack of sufficient jobs in a community, 
region, or country. On a national basis, the only time in the past 70 
years that the average unemployment rate was above ten percent was in 
1982-83. Yet from December 2007 through June 2009 the United States 
experienced the most severe recession in the post-war period, with over 
a four percent decline in gross domestic product (GDP).\107\ A floor 
above ten percent is therefore highly unresponsive to nationwide 
recessions and depressions. The 20 Percent Standard's use of a data 
over a 24 month period is also unresponsive--the period is more an 
indication of chronic economic depression than a new recession.
---------------------------------------------------------------------------
    \107\ More information from 1948 to present is available at: 
Databases, Tables & Calculators by Subject, U.S. Bureau of Labor 
Statistics, accessed March 2019, https://data.bls.gov/timeseries/
lns14000000. The unemployment rate reached but did not exceed ten 
percent in just 1 month of the Great Recession, October 2009. More 
information on GDP is available at: GDP & Personal Income, U.S. Bureau 
of Economic Analysis, accessed March 2019, https://apps.bea.gov/iTable/
index_nipa.cfm.
---------------------------------------------------------------------------
    This lack of responsiveness limits SNAP's ability to serve as an 
automatic stabilizer and is therefore inconsistent with the goals 
described in the Joint Explanatory Statement of the Committee of 
Conference to address times of recession. If a recession took effect 
tomorrow and the current unemployment rate of 3.9% (December 2018 BLS) 
doubled, the number of persons newly unemployed would be about five 
million across the United States. Many of these five million would be 
individuals unable to access SNAP beyond the time limit, and the loss 
of benefits would be detrimental to the economy.
Retaining the Extended Unemployment Benefits Qualification Standard
    The Unemployment Insurance Extended Benefits (EB) program extends 
individual unemployment compensation for an additional 13 weeks when a 
state's insured unemployment rate (IUR) or total unemployment rate 
(TUR) reaches at least 5% and is 120% of the average of the rates for 
the same 13 week period in each of the 2 previous years. There are two 
other optional thresholds that states may choose. EBs may be triggered 
if the state's IUR is at least 6%, or the TUR is at least 6.5% and is 
at least 110% of the state's average TUR for the same 13 weeks in 
either of the previous 2 years. An additional 20 weeks of benefits may 
be triggered if the TUR is at least 8% and is at least 110% of the 
state's average TUR for the same 13 weeks in either of the previous 2 
years.\108\
---------------------------------------------------------------------------
    \108\ Julie M. Whittaker and Katelin P. Isaacs, Extending 
Unemployment Benefits During a Recession, Congressional Research 
Service, May 2013, https://fas.org/sgp/crs/misc/RL34340.pdf.
---------------------------------------------------------------------------
    While these triggers are lower than the standard proposed for 
waivers, many researchers have found that EB triggers are set too high, 
which prevents many states from activating the program for extra weeks 
of benefits above and beyond the standard 26 weeks.\109\ Moreover, the 
trigger requires ever increasing unemployment rates in order to remain 
triggered, which means that many states cycle out of the system too 
early. Congress has regularly passed legislation to provide extended UI 
benefits in states that do not meet the EB criteria, or else extended 
benefits nationwide. Congress established temporary programs of 
extended UI benefits in 1958, 1961, 1971, 1974, 1982, 1991, 2002, and 
2008. In the Great Recession, Congress created a temporary program 
nationwide. The need for actions of Congress demonstrate that EB 
qualification as a core standard for approval is not adequate for 
states with high unemployment rates that are not rising rapidly.
---------------------------------------------------------------------------
    \109\ Unemployment Insurance Trigger 101: Is the Trigger System 
Working?, Center for American Progress, February 2010, https://
www.americanprogress.org/wp-content/uploads/issues/2010/02/pdf/
ui_101.pdf.
---------------------------------------------------------------------------
Establishing a Floor for Waivers Based on the 20 Percent Standard
    We strongly oppose the use of a floor for waivers. The 20 percent 
standard is an adequate relative measure that demonstrates that an area 
of the country is in a more difficult economic position than the rest 
of the country.
    The Department seeks to create a fixed floor at seven percent, well 
above the natural rate of unemployment. We believe a floor at any level 
above the natural rate of unemployment is unnecessary, arbitrary and 
needlessly disadvantages members of protected classes, as described in 
the Civil Rights Impact section below. In addition, it would subvert 
the intent of the Agriculture Improvement Act of 2018 to permit waivers 
in labor surplus areas.
    Further, there is significant disagreement amongst respected 
economists about the exact number for the natural rate of unemployment. 
While a five percent natural rate of unemployment seemed to be the norm 
at one point, the evidence has shifted in the past twenty years. If the 
natural rate is defined in part by the point at which unemployment 
leads to inflation, then the current unemployment rate of 3.9 percent 
is arguably above the natural unemployment rate, according to Jared 
Bernstein, President Clinton's former economic advisor. ``While 
inflation is picking up a bit, it has been very low for a very long 
time, unresponsive to falling unemployment, and no one is arguing that 
it is . . . spiraling up in response to a full-capacity economy.'' 
\110\ A selection of a five percent rate is arbitrary given the lack of 
consensus by experts.
---------------------------------------------------------------------------
    \110\ Jared Bernstein, ``I've Come to Believe That `Are We at Full 
Employment?' Is the Wrong Question,'' The Washington Post, May 2018, 
https://www.washingtonpost.com/news/posteverything/wp/2018/05/24/ive-
come-to-believe-that-are-we-at-full-employment-is-the-wrong-question/
?noredirect=on&utm_term=.c249a3003278.
---------------------------------------------------------------------------
Restricting Statewide Waivers and Combined Areas
    We oppose elimination of the option of statewide waivers as well as 
waivers for areas that are not economically tied together, due to the 
administrative complexities of implementing this change. There are 63 
counties or county equivalents on average per state; Texas has the most 
counties at 254. Some states have chosen to submit statewide waivers to 
avoid the administrative burden of creating dual systems for those SNAP 
recipients who are or are not subject to time limits. Under the 
proposed rules, states will need to collect data for each county or 
Labor Market Area in order to submit a waiver request. States will then 
have to set up dual systems and train caseworkers to treat SNAP 
recipients differently based on their county of residence. This adds to 
caseworker confusion and potential error. States will also need to 
train SNAP E&T service providers to treat SNAP recipients differently 
based on their residence.
    In addition, we oppose eliminating combined areas that fall outside 
of Labor Market Areas. The areas covered by Workforce Development 
Boards are not always consistent with Labor Market Areas; some include 
multiple counties, including some outside of Labor Market Areas, while 
others are smaller than a county. The proposed rules will make planning 
more difficult given the inability to group areas consistent with 
Workforce Development Boards.
    Finally, we oppose the elimination of waivers in sub-county areas. 
Many counties in the U.S. have extreme disparities in the labor market 
in different geographic areas. Traveling to jobs with an adequate labor 
supply may not be feasible for many low-income SNAP recipients. For 
example, traveling from rural Gorman to Long Beach in the most populous 
county in the United States, Los Angeles, takes about 2 hours by car 
and is not possible via public transportation. The proposed rules do 
not account for the immense variety in local conditions that can make 
finding a job nearly impossible for many people.
Ending ``Carryover'' Exemptions
    Current law allows unused exemptions to carry over and accumulate 
from one year to the next. Up until now, states understood these 
exemptions to be ``earned'' and made decisions about whether to use the 
exemptions in a given year based on a good faith assumption of FNS' 
continued allowance of carryover exemptions. Past and recent FNS 
Memoranda on the subject, including the most recent ``SNAP--FY 2018 
ABAWD 15 Percent Exemptions Totals, Adjusted for Carryover,'' clearly 
demonstrates FNS approval of carryover exemptions to date.
    We strongly oppose the use of a carryover formula in which the 
current year is adjusted based on the number of exemptions earned in 
the preceding fiscal year minus the number of exemptions used in the 
preceding fiscal year. The formula penalizes states for using carryover 
from the previous year by subtracting any used carryover amount from 
earned exemptions. We do not believe penalties for the use of carryover 
was the intent of the act. As can be seen in Example 2 (Varied 
Exemption Use) of the NPRM, the number of exemptions after adjustment 
becomes highly erratic as well as difficult to track under the proposed 
formula. The formula also incentivizes states to use their full amount 
in the year earned rather than prudently reserving exemptions for a 
downturn in the economy. Overall, we hold that the proposed method of 
calculating exemptions would create confusion, discourage the use of 
exemptions, and increase errors.
Civil Rights Impacts under the Civil Rights Impact Analysis
    The Department has stated that in accordance with the Department 
Regulation 4300-4 Civil Rights Impact Analysis, implementation of 
mitigation strategies and monitoring by the Civil Rights Division of 
FNS will lessen the disproportionate impacts of ``certain protected 
groups due to factors affecting rates of employment of members of these 
groups.''
    There are two main CRIA challenges to the NPRM. A reduction in 
waivers and the resulting loss in SNAP benefits will disproportionately 
affect certain protected classes based on (a) their disproportionate 
rate of unemployment and under-employment, as stated in the NPRM; and 
(b) an inadequate method for determining lack of sufficient jobs, a 
criterion for approving waivers.
Disproportionate Rate of Unemployment and Under-Employment
    We strongly oppose further restrictions to waivers due to their 
disproportionate impact on many protected classes including women, 
Black people, Latinos, and people with disabilities. We do not believe 
that mitigation strategies will be significant enough to address the 
impact of increased food insecurity and poverty on protected classes.
Inadequate Methodology
    The NPRM not only impacts protected classes disproportionately due 
to unemployment factors, but it further impacts protected classes due 
to the use of the U-3 measure, which excludes certain employment 
statuses that are more common amongst certain protected classes. We 
strongly oppose the NPRM for these reasons.
    Data from the Bureau of Labor Statistics illustrate the 
disproportionate impact of the data excluded from the U-3 measure on 
selected protected groups. For instance, as illustrated by BLS data in 
graphics 2 and 3 in the appendix, Black individuals are more than twice 
as likely than their White counterparts to have searched for work in 
the previous year but not in the past 4 weeks (see graphic 2), and 
Latinos are 66 percent more likely than Whites to work part-time for 
economic reasons (see graphic 3). Also, women are 38 percent more 
likely than men to work part-time for economic reasons.\111\
---------------------------------------------------------------------------
    \111\ Bureau of Labor Statistics, ``Labor Force Statistics from the 
Current Population Survey, A-18. Employed and Unemployed Full- and 
part-time workers by age, sex, and Hispanic or Latino ethnicity'' 
https://www.bls.gov/web/empsit/cpseea18.htm; and Bureau of Labor 
Statistics ``Labor Force Characteristics by Race and Ethnicity, 2017, 
Table 15. People in the Labor Force and Not in the Labor Force by 
Selected Characteristics, 2017 annual averages'' https://www.bls.gov/
opub/reports/race-and-ethnicity/2017/home.htm.
---------------------------------------------------------------------------
5. Conclusion
    In conclusion, we urge the Department to withdraw the proposed 
regulation in its entirety. As anti-poverty experts, we believe that 
the proposed changes will not incentivize or equip people with what 
they need to seek and maintain work, and will also have profound and 
damaging consequences for the well-being and long-term success of 
struggling workers and their families. We encourage the Department to 
dedicate its efforts to advancing policies that truly support economic 
security, self-sufficiency, and a stronger future for the United States 
by promoting--rather than undermining--the ability of unemployed and 
underemployed workers, their families, and their communities to thrive.
    Further, the proposed rule does not provide the analytical 
information needed to justify the policy change and to evaluate the 
proposed rule's likely impacts. Because of the deficiencies in 
reasoning and analysis, the proposed rule fails to answer basic 
questions related to the impact of the change and the people whom the 
proposed rule would affect. All in all, the proposed rule does not 
contain the information and data necessary to fully evaluate the 
proposal or to comment on key aspects on the Department's justification 
for the rule.
    Last, our comments include citations to supporting research and 
documents for the benefit of the Food and Nutrition Service in 
reviewing our comments. We direct FNS to each of the items cited and 
made available to the agency through active hyperlinks and as 
attachments, and we request that these, along with the full text of our 
comments, be considered part of the formal administrative record on 
this proposal.
    Thank you for the opportunity to submit these comments. Contact 
Elizabeth Lower-Basch ([email protected]) and Renato Rocha 
([email protected]) with any questions.
               appendix a: graphics referenced in comment
Graphic 1: Ratio of U-6 to U-3 Measures by State

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

          Powered by Bing.
          GeoNames, HERE, MSFT.
          Source: BLS, Alternative measures of labor under-utilization 
        by state, fourth quarter of 2017 through third quarter of 2018 
        averages.
Graphic 2: Data Excluded from U-3 Measure and Included in U-6 Measure: 
        Unemployed Who Searched for Work in Previous Year but Not in 
        Past 4 Week
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: BLS, People in the labor force and not in the labor 
        force by selected characteristics, 2017 annual averages.
Graphic 3: Data Excluded from U-3 Measure and Included in U-6 Measure: 
        Part-Time for Economic Reasons
        
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
        
          Source: BLS, Employed and unemployed full- and part-time 
        workers by age, sex, race, and Hispanic or Latino ethnicity, 
        December 2018.
           appendix b: contributors to clasp's public comment
Listed Alphabetically
    Kisha Bird is CLASP's director of youth policy. Ms. Bird works to 
expand access to education, employment, and support services for low-
income and opportunity youth. She is an expert in Federal youth policy 
and helps ensure national legislation has maximum impact for youth of 
color. Before joining CLASP, Ms. Bird was director of the Pennsylvania 
Statewide Afterschool/Youth Development Network, working to make 
quality education and afterschool programs accessible to young people. 
Prior to that, she was a program officer at the Philadelphia 
Foundation, where she helped develop and manage the Fund for Children, 
Youth Advisory Board, and discretionary grants process. She also has 
direct service experience, working in various community settings with 
children, youth and families. Ms. Bird holds a master of social service 
and master of law and social policy from Bryn Mawr College Graduate 
School of Social Work and Social Research. Additionally, she earned a 
bachelor's in sociology from Spelman College.
    Whitney Bunts is a policy analyst with CLASP's youth policy team, 
with a focus on juvenile justice, mental health, racial equity, and 
Opportunity Youth. Whitney has a competence in education policy, 
opportunity, at-risk, LGBTQ+ youth, racial equity, system dynamics, and 
evaluation. Preceding her career at CLASP, Whitney was a graduate 
student at Washington University in St. Louis. During her time in grad 
school, she interned for the 22nd District Circuits Attorney Office as 
a Victim Services Advocate. In this role, she educated victims on the 
criminal and juvenile justice process, while partnering with 
prosecutors to advocate for their rights. Additionally, she served as 
Policy Associate Intern at Covenant House Missouri. As an intern she 
had the opportunity to update, revise, and align their policies with 
the Housing First Federal policy. Additionally, she facilitated 
workshops to build racial equity and inclusion within the organization 
using System Dynamic tools. As a student, Whitney was a Graduate Policy 
Scholar, served on the Student Coordinating Council and Graduate 
Professional Council, and was selected for the ``Excellence Award for 
Activism.'' Prior to attending graduate School, Whitney served as a 
City Year AmeriCorps member where she tutored and mentored hundreds of 
high school students in reading and writing. She holds a Master of 
Social Work, with a concentration in Children, Youth & Families, and a 
specialization in Policy and System Dynamics from Washington University 
in St. Louis. Additionally, she has a double bachelors in political 
science and psychology from Georgia State University.
    Aimee Chitayat is a consultant for CLASP's Income and Work Supports 
team. As Principal of AC Strategic Solutions, she has led efforts to 
expand SNAP Employment and Training (E&T) since 2007. Aimee developed 
the first SNAP E&T third-party partner programs implemented in 
California--the Fresh Success intermediary model and the county-based 
Cal Success model--and designed innovative policies, procedures, tools, 
and templates for their implementation. She provides intensive training 
and technical assistance on SNAP E&T to community colleges, community-
based organizations, social enterprises, counties and statewide 
intermediaries throughout the country. She supported New Jersey in 
drafting successful legislation for a SNAP E&T program and provided 
oversight to the USDA's SNAP E&T Pilot Project in Fresno County as a 
consultant to the California Department of Social Services. She 
developed, supported, or commented on numerous state and national bills 
and policy clarifications on SNAP and SNAP E&T. She earned her Master 
of Social Welfare from the University of California Berkeley and her 
undergraduate degree from Brown University.
    Parker Gilkesson is a policy analyst with CLASP's Income and Work 
Support team. She works with low-income and work support programs with 
a focus on the Supplemental Nutrition Assistance Program (SNAP). Parker 
is a subject matter expert in social policy, benefit eligibility, human 
services delivery, racial equity, and state and local policy regarding 
SNAP, TANF, and Medicaid. Prior to joining CLASP, Parker began her 
career as a Human Services Specialist in Mecklenburg County, Charlotte, 
NC. In this role, she worked directly with recipients receiving 
Medicaid, TANF, and SNAP to determine their eligibility for low-income 
and work support programs. She has other experience including TANF 
policy research, cancer research, public health, public service, and 
nonprofits. Parker holds a Master of Public Policy, with a 
concentration in Public Administration from Liberty University and a 
Bachelors in Health Education, Maternal and Child Health from Howard 
University. Furthermore, Parker believes in the importance of bridging 
the gap between policy analysis and policy effectiveness. She is very 
passionate about social change taking place within our communities, 
therefore, Parker aspires to be a part of the equation to solve poverty 
and inequities in health and social welfare among citizens of the 
United States.
    Madison Hardee is a senior policy analyst/attorney at CLASP, where 
she focuses on issues affecting access to health care and public 
benefits for immigrants and mixed-status families. Ms. Hardee coleads 
the Protecting Immigrant Families, Advancing Our Future Campaign in 
collaboration with the National Immigration Law Center. Prior to 
joining CLASP, Ms. Hardee spent 5 years as an attorney with Charlotte 
Center for Legal Advocacy, where she provided direct legal 
representation to low-income clients across public benefit programs and 
saw first-hand how programs like Medicaid, SNAP and SSI reduce economic 
hardship, improve health, and increase stability. She successfully 
challenged state agency decisions and identified several areas for 
systemic advocacy. Working together with partner organizations, Ms. 
Hardee negotiated significant changes to Medicaid and ACA eligibility 
policies, providing access to health care for tens of thousands of low-
income immigrants. Ms. Hardee holds a Juris Doctor from Tulane Law 
School and a bachelor's degree in public health from George Washington 
University. In 2016, she was presented with the New Leader in Advocacy 
Award by the National Legal Aid and Defender Association.
    Elizabeth Lower-Basch is director of CLASP's income and work 
supports team. Her expertise is Federal and state welfare (TANF) 
policy, other supports for low-income working families (such as 
refundable tax credits), systems integration, and job quality. From 
1996 to 2006, Ms. Lower-Basch worked for the Office of the Assistant 
Secretary for Planning and Evaluation at the U.S. Department of Health 
and Human Services. In this position, she was a lead welfare policy 
analyst, supporting legislative and regulatory processes and managing 
research projects. She received a Master of Public Policy from Harvard 
University's Kennedy School of Government.
    Judy Mortrude is a senior policy analyst with CLASP's Center for 
Postsecondary and Economic Success. Ms. Mortrude has more than 30 
years' experience developing, delivering, and evaluating workforce 
education, particularly with low-literacy and high-barrier populations. 
She has been a classroom teacher, school administrator, and state 
agency staff. Currently, Ms. Mortrude supports cross-agency state teams 
as they scale and sustain integrated education and training career 
pathway policies and practices; focus attention on racial and economic 
equity; and build two-generational strategies. Additionally, she 
analyzes Federal adult and post-secondary education policy and supports 
organizations like the National Coalition for Literacy and the Open 
Door Collective.
    Renato Rocha is a policy analyst within CLASP's Income and Work 
Supports team. He focuses on issues regarding work reporting 
requirements across benefit programs as well as access to public 
benefits for immigrant families. Prior to CLASP, Renato was an economic 
policy analyst at UnidosUS (formerly National Council of La Raza), 
where he conducted analysis of consumer protection, budget, tax, 
disaster relief, and labor issues that impact the well-being of Latino 
and immigrant communities. In graduate school, he also had the 
opportunity to work at the National Immigration Law Center, where he 
analyzed policy issues affecting deferred action recipients. Renato 
holds a Master in Public Affairs from Princeton University's Woodrow 
Wilson School of Public and International Affairs and a B.A. in 
Politics from Occidental College. In 2013, Renato served as a Fulbright 
Public Policy Initiative Fellow to Mexico.
    Darrel Thompson is a research assistant with CLASP's Income and 
Work Supports team. He provides research support and analysis on 
various low-income and work support programs. Prior to joining CLASP, 
Darrel interned at the Center on Budget and Policy Priorities and the 
Lou Frey Institute of Politics and Government. He holds a bachelor's 
degree in political science from the University of Central Florida.
    Isha Weerasinghe is a senior policy analyst focused on mental 
health and sits in CLASP's youth team. She works on how CLASP's issue 
areas impact individuals' mental health, with a specific focus on 
youth, young adults, and mothers. Ms. Weerasinghe previously worked as 
the Director of Policy and Advocacy at the Association for Asian 
Pacific Community Health Organizations (AAPCHO), where she focused on 
the intersections of how Asian Americans, Native Hawaiians, and Pacific 
Islanders (AA&NHPIs) can better access linguistically concordant and 
culturally appropriate care. She also did a great deal of coalition 
building and provided policy guidance nationally, for AA&NHPI-serving 
community health centers and AA&NHPI-serving organizations, in health 
access and equity. Ms. Weerasinghe has done community based 
participatory research, as well as local and state policy advocacy in 
her work at New York University's Center for the Study of Asian 
American Health (CSAAH), working within New York City and New York 
state. Over the past 8 years, she has done extensive coalition work and 
policy advocacy on the impacts of hepatitis B in the United States. 
Isha has a bachelor's in arts degree in biology from Bryn Mawr College, 
and a master's in science degree in health policy and demography from 
the London School of Economics and Political Science.
    Carrie Welton is a policy analyst on the income and work supports 
team. Her work focuses on advocating for policy reforms that improve 
the lives of people with low income and communities of color using a 
racial equity lens. This includes improving access to public benefit 
programs for post-secondary students and student parents to advance 
their academic success. She also advocates for policy reforms that 
strengthen the Earned Income Tax Credit (EITC) and the Child Tax Credit 
(CTC). Previously, Ms. Welton spent 3 years at the W.K. Kellogg 
Foundation on the national Education and Learning team focused on early 
childhood systems alignment. In addition, she spent 4 years at the 
Kellogg Company conducting research and providing strategic direction 
to inform the organization's government relations and lobbying efforts. 
Ms. Welton also served on the state board of the American Civil 
Liberties Union (ACLU) of Michigan, furthering the civil liberties and 
civil rights of residents. As a member of the executive committee, she 
provided fiduciary, strategic, and generative leadership to the 
organization. She earned her Master of Public Administration from the 
Gerald R. Ford School of Public Policy at the University of Michigan 
and her undergraduate degree in Public Law from Western Michigan 
University.
                                 ______
                                 
  Submitted Comment Letter by Hon. Jimmy Panetta, a Representative in 
 Congress from California; Authored by Abby J. Leibman, President and 
      Chief Executive Officer, MAZON: A Jewish Response to Hunger
March 19, 2019

  Chief, Certification Policy Branch,
  SNAP Program Development, USDA Food & Nutrition Services,
  Alexandria, VA 22302

  Re: Proposed Rule: Supplemental Nutrition Assistance Program (SNAP): 
            Requirement for Able-Bodied Adults without Dependents RIN 
            0584-AE57

    To Whom It May Concern:

    On behalf of MAZON: A Jewish Response to Hunger, I am pleased to 
submit these comments in opposition to public notice FR Doc. 2018-
28059. Based on our organization's many years of expert involvement in 
anti-hunger related issues, we submit for your consideration these 
comments focused on whether USDA should reconsider certain rules that 
govern--and restrict--the current waiver standards for able-bodied 
adults without dependents (ABAWDs) who participate in the Supplemental 
Nutrition Assistance Program (SNAP).

    We unequivocally oppose the proposed rule change, which would 
restrict states' flexibility to provide vital nutrition support to 
people who struggle to feed themselves and their families.

    Inspired by Jewish values and ideals, MAZON is a national advocacy 
organization working to end hunger among people of all faiths and 
backgrounds in the United States and Israel. For more than 30 years, 
MAZON has been committed to ensuring that vulnerable people have access 
to the resources they need to be able to put food on the table. MAZON 
is a leading voice throughout the country on anti-hunger issues, 
especially those that involve populations or problems that have been 
previously overlooked or ignored--this includes food insecurity among 
veterans, currently-serving military families, seniors, rural and 
Native American communities, and college students. In fact, MAZON has 
already commented on similar draconian proposals, as evidenced by the 
attached letter we submitted on April 4, 2018 regarding RIN 0584-AE57.
    It is with this experience and focus that we address the proposed 
rule.
    The Supplemental Nutrition Assistance Program (SNAP) is the 
cornerstone of our nation's nutrition safety net, and most SNAP 
recipients who are able to work do, in fact, actually work. Under 
current law, childless adults ages 18 to 49 are restricted to only 90 
days of SNAP benefits in 3 years unless they can prove they are working 
or participating in an employment and training program for 80 hours per 
month. States currently have flexibility to request waivers from this 
harsh and arbitrary time limit for communities that face high 
unemployment or insufficient job opportun[i]ties. USDA's proposed rule 
change severely limits this critically important and common-sense 
flexibility that is utilized by the majority of states.
    Harsher limitations on accessing SNAP ignore the complex realities 
of low-income families. This decision to restrict waivers will 
exacerbate already difficult circumstances, not alleviate them.

Needless and Egregious Harm to the American People
    This proposed rule change will have a devastating impact on 
Americans of all walks of life. Working-age adults without minor 
children are by no means a monolithic population. Approximately 45% are 
female, and of them, nearly \1/3\ are over the age of 40. Roughly \1/2\ 
of ABAWDs are Caucasian, \1/3\ African American, and 10% Latina.\1\
---------------------------------------------------------------------------
    \1\ https://www.cbpp.org/research/food-assistance/who-are-the-low-
income-childless-adults-facing-the-loss-of-snap-in-2016.
---------------------------------------------------------------------------
Rural Americans
    Rural areas often face unique barriers to achieving food security 
including a lack of public transportation, scarcity of childcare 
services, lower educational attainment, fewer economic opportunities 
and higher unemployment rates than urban areas.\2\ With the largest 
proportion of SNAP participants, rural counties and small metropolitan 
areas are more dependent on SNAP than urban counties. Of the top 100 
counties that participate in SNAP, 85 are rural.\3\ There is stunning 
evidence that this proposed rule change would cause extraordinary harm 
to rural communities in southern states.\4\
---------------------------------------------------------------------------
    \2\ https://www.ers.usda.gov/webdocs/publications/90556/eib-
200.pdf.
    \3\ https://www.dailyyonder.com/geography-food-stamps/2018/12/31/
25422/.
    \4\ https://www.mathematica-mpr.com/our-publications-and-findings/
publications/proposed-changes-to-the-supplemental-nutrition-assistance-
program-waivers-to-work-related-time.
---------------------------------------------------------------------------
    In a recent speech about the need for economic development in high-
poverty rural communities, Chairman of the Federal Reserve Bank Jerome 
Powell explained that rural areas ``generally lack diverse industries 
and employment options and often have suffered from decline in a 
traditional industry.'' \5\ While current data show a strong economy 
nationally, this is not true for rural America where poverty remains a 
persistent challenge.\6\
---------------------------------------------------------------------------
    \5\ https://www.reuters.com/article/us-usa-fed-powell-analysis/in-
rural-mississippi-still-waiting-on-recovery-idUSKCN1Q30JH.
    \6\ https://www.federalreserve.gov/newsevents/speech/
powell20190212a.htm.
---------------------------------------------------------------------------
The Working Poor
    We also know that the majority of SNAP recipients who can work do 
work. Among those who would be harmed by this proposed rule, roughly 
75% worked the year before and/or the year after receiving SNAP. Many 
of these people continuously experience periods of work and 
unemployment, stuck in a devastating cycle of inconsistent low-skill, 
low-wage jobs that are unable to lift anyone out of poverty.\7\ The 
individuals most at risk of losing SNAP benefits under the proposed 
rule are workers who experience normal labor market fluctuations and 
those who should be eligible for exemptions but often do not receive 
them.\8\
---------------------------------------------------------------------------
    \7\ https://www.cbpp.org/research/food-assistance/who-are-the-low-
income-childless-adults-facing-the-loss-of-snap-in-2016.
    \8\ http://www.hamiltonproject.org/blog/
workers_could_lose_snap_benefits_under_trumps_
proposed_rule.
---------------------------------------------------------------------------
    Ample evidence suggests that harsh SNAP time limits fail to 
``increase self-sufficiency, well-being, and economic mobility'' as 
intended.\9\ In fact, we know that the vast majority of people 
subjected to these time limits remained poor or even became poorer.\10\ 
Even among conservative policy experts who support the principle of 
work requirements, poorly-designed policies like this proposed rule 
raise concerns and are considered to be unreasonable, unrealistic, 
untested, and clearly designed to cut caseloads and costs--not provide 
needed assistance and a pathway to self-improvement for those who are 
struggling.\11\
---------------------------------------------------------------------------
    \9\ https://www.whitehouse.gov/presidential-actions/executive-
order-reducing-poverty-america-promoting-opportunity-economic-mobility/
 
    \10\ https://www.cbpp.org/research/poverty-and-inequality/work-
requirements-dont-cut-poverty-evidence-shows.
    \11\ https://mlwiseman.com/wp-content/uploads/2019/01/
Farmbill.120118.pdf.
---------------------------------------------------------------------------
U.S. Veterans
    We are deeply concerned by the evidence that this proposed rule 
change would severely impact veterans who often face unique challenges 
in securing full-time work and may require more than 3 months to secure 
employment.
    An estimated 1.4 million veterans live in households that 
participate in SNAP.\12\ Evidence suggests that veteran households 
participate in SNAP at lower rates than non-veteran households, 
indicating that there are thousands who qualify but have not applied 
for this essential lifeline.\13\ Post-9/11 veterans have nearly double 
the average rate of food insecurity \14\ and recent scholarship has 
raised concerns about the high rate of food insecurity and resultant 
health impacts for women veterans.\15\ We know that many veterans 
return from combat with disabilities, sometimes undiagnosed or not 
fully recognized, that make it more difficult to maintain gainful 
employment and provide food for themselves and those who rely on them, 
even if they do not meet the definition of ``dependent.'' Households 
with a disabled veteran are nearly twice as likely to be food-insecure 
as households that do not have someone with a disability.\16\
---------------------------------------------------------------------------
    \12\ https://www.cbpp.org/research/food-assistance/snap-helps-
almost-15-million-low-income-veterans-including-thousands-in.
    \13\ https://www.cbpp.org/research/food-assistance/snap-helps-
almost-15-million-low-income-veterans-including-thousands-in.
    \14\ https://www.ncbi.nlm.nih.gov/pubmed/24806818.
    \15\ https://www.whijournal.com/article/S1049-3867(17)30419-X/
abstract.
    \16\ https://www.cbpp.org/research/food-assistance/snap-helps-
almost-15-million-low-income-veterans-including-thousands-in.
---------------------------------------------------------------------------
    The Blue Star Families 2018 Military Family Lifestyle Survey--the 
largest and most comprehensive survey of active duty service members, 
veterans, and their families--found employment to be one of the top 
three issues of primary concern among veterans.\17\ Veterans often 
struggle to find jobs that match their skills, especially if they have 
little work experience beyond military service. They might also face 
discrimination from employers, particularly if they have a mental or 
physical disability. Furthermore, many recently transitioning veterans 
take temporary jobs but struggle to find full-time sustained work that 
is a good fit for their skills and experience--these veterans will not 
be able to regularly report 20 hours of work per week in order to 
receive SNAP benefits.
---------------------------------------------------------------------------
    \17\ https://bluestarfam.org/survey/.
---------------------------------------------------------------------------
    In addition to employment concerns, veterans who are awaiting a 
disability determination face enormous challenges in making claims 
through the U.S. Department of Veterans Affairs['] (VA) daunting claims 
process, where delays and multiple appeals are commonplace. During this 
waiting period, many veterans who cannot work are also unable, or 
limited in their ability, to access Federal assistance.
    The State of Maine offers a deeply concerning example of the 
harmful impacts of this proposed rule change on veterans. In 2014, 
Governor Paul LePage chose not to request a SNAP waiver for working-age 
adults without minor children, for which the State of Maine was 
eligible. As a result of this action, many thousands of Maine residents 
were stripped of access to needed nutrition assistance from SNAP, 
including an estimated 2,800 veterans affected by these harsh time 
limits, many of whom continue to face unemployment and must turn to the 
charitable food sector to meet their basic needs.
    We urge USDA to consider the story of Tim Keefe, a veteran living 
in Maine whose story provides a personal and painful glimpse of the 
impact of this proposed rule change. When Governor Le Page decided not 
to seek a waiver for the ABAWD SNAP time limit, Tim lost his access to 
SNAP--one of the only supports that helped him get by as he was 
desperately trying to find employment. He became homeless and reported 
feeling ``like a cave man.'' \18\ Tim resorted to eating squirrels that 
he caught to survive the brutal Maine winter, taking a great toll on 
his health and well-being. He eventually was able to qualify again for 
SNAP assistance when he turned 50. The SNAP benefits he receives now 
are a lifeline for Tim and enable him to regularly put food on the 
table once again.
---------------------------------------------------------------------------
    \18\ https://bangordailynews.com/2017/06/03/politics/i-felt-like-a-
caveman-how-work-requirements-for-state-benefits-hurt-one-maine-man/.
---------------------------------------------------------------------------
    Sadly, Tim's story is not unique to the veteran experience in 
America. Veterans regularly need temporary supplemental nutrition 
assistance precisely because they frequently find themselves in periods 
of transition. It does not matter whether they are recently returning 
from service or have already long contributed to our workforce. Nobody 
deserves to be destabilized by hunger while trying to get back on their 
feet. Ensuring that all veterans have access to adequate and nutritious 
food is critical, and providing such access to veterans is the least 
this nation owes to its returning servicemembers who have made such 
great sacrifices in service to our country.
Exacerbated Hunger Among Native Americans
    As the first non-Native member of the Native Farm Bill Coalition, 
MAZON is deeply concerned about the profound harm this proposed rule 
change will have on American Indian and Alaska Native individuals. We 
know that one in four Native Americans is food-insecure (double the 
national average of one in eight), and this assault on SNAP eligibility 
clearly will exacerbate hunger and poverty in this particularly 
vulnerable and frequently overlooked population.\19\
---------------------------------------------------------------------------
    \19\ http://www.nativepartnership.org/site/DocServer/2017-PWNA-
NPRA-Food-Insecurity-Project-Grow.pdf?docID=7106.
---------------------------------------------------------------------------
    Despite reports of high employment on a national scale, 
unemployment rates on reservations remain dangerously high, in some 
cases as high as 21%, and in some communities much higher.\20\ For 
these communities, waivers for SNAP time limits literally save lives, 
especially considering the geographic isolation and the impact still 
felt today by historic violation of treaties with multiple Tribes, 
generations of discrimination, forced attempts of assimilation, and 
state-sponsored genocide.
---------------------------------------------------------------------------
    \20\ https://www.bloomberg.com/news/articles/2018-04-05/where-u-s-
unemployment-is-still-sky-high-indian-reservations.
---------------------------------------------------------------------------
    Denying states the ability to apply for waivers will further strain 
the Food Distribution Program on Indian Reservations (FDPIR) which 
serves American Indians and Alaska Natives living on reservations or in 
designated Tribal areas. In 2018, FDPIR served an average of 87,216 
participants--mostly low-income individuals and families, working 
adults, children, people with disabilities, and seniors.\21\ FDPIR was 
designed as an alternative to SNAP and serves some overlapping 
populations on Tribal reservations, so changes to SNAP eligibility 
policies will impact FDPIR.\22\ Because FDPIR's funding is capped at a 
fixed dollar amount, there is a real concern about the exhaustion of 
FDPIR funds in the event of a spike in participation caused by 
individuals cut off from SNAP benefits due to the proposed rule 
change.\23\
---------------------------------------------------------------------------
    \21\ https://fns-prod.azureedge.net/sites/default/files/pd/
fdpart.pdf.
    \22\ https://fns-prod.azureedge.net/sites/default/files/ops/
StudyofFDPIR.pdf.
    \23\ https://www.phi.org/uploads/files/FDPIR%20Module%20-
%20CCRWF%20Nutrition%20
Primer.pdf.
---------------------------------------------------------------------------
    Mary Greene Trottier, a member of the Spirit Lake Sioux Nation and 
President of the National Association of FDPIR, recently testified 
before the U.S. House Committee on Natural Resources about the impacts 
of the recent partial government shutdown on Indian Country and the 
importance of FDPIR to Tribal members. In a compelling portion of her 
testimony, Ms. Trottier recounted how FDPIR was impacted by changes to 
SNAP benefits in 2013:

          We know from experience that any time SNAP benefits are 
        reduced or taken away, our program [FDPIR] sees an immediate 
        rise in applications as people seek to feed themselves and 
        their families. In some cases there is a 25 percent increase in 
        participation [ . . . ] when SNAP benefits are reduced. We saw 
        this in October 2013, when the American Recovery and 
        Reinvestment Act (ARRA) expired and SNAP benefits were reduced. 
        In the month after ARRA's expiration, we saw an immediate rise 
        in participation across FDPIR sites in all our regions. 
        Unfortunately, this rise in participation does not come with 
        increased funding. We must try to do more with less.\24\
---------------------------------------------------------------------------
    \24\ https://naturalresources.house.gov/imo/media/doc/
Mary%20Greene%20Trottier-Testimony.pdf.

    We cannot count on FDPIR to meet the needs of every food-insecure 
Native American. Of the 573 Tribes recognized by the Federal 
Government, FDPIR operates among only 276 Tribes. SNAP is the only 
option available to alleviate food insecurity in 297 Tribal 
communities.\25\ Furthermore, since FDPIR exclusively applies to Tribes 
recognized by the Federal Government, the hundreds of Tribes recognized 
by states alone and not by the Federal Government are already unable to 
utilize the limited amount of funding available that exists for FDPIR 
to supplement loss of access to SNAP.
---------------------------------------------------------------------------
    \25\ https://fns-prod.azureedge.net/sites/default/files/fdpir/pfs-
fdpir.pdf.
---------------------------------------------------------------------------
    Finally, the Federal Government must adequately and appropriately 
consult with all federally-recognized Tribes to ensure meaningful and 
timely input on legislative proposals, policy matters and regulatory 
changes that have Tribal implications. Consultations and related 
efforts to improve operation and administration of Federal nutrition 
programs operating in Indian Country stem from a recognition that the 
U.S. has a solemn obligation to support Tribal sovereignty and protect 
the well-being of these communities at a level comparable to non-
Natives. USDA's proposed rule change will have a substantial direct and 
disproportionate impact on Native communities, on and off reservations. 
Accordingly, before moving forward with this rulemaking proposal, our 
government has a duty to consult with Tribal sovereigns about this 
proposed rule change and to consider their concerns and recommendations 
about how to mitigate potential negative impacts on Native 
communities.\26\
---------------------------------------------------------------------------
    \26\ https://www.federalregister.gov/documents/2000/11/09/00-29003/
consultation-and-coordination-with-indian-tribal-governments.
---------------------------------------------------------------------------
College Students
    The proposed rule change denies SNAP access to certain post-
secondary students without dependents, as well as noncustodial student-
parents who are enrolled less than part-time. When these students are 
denied the ability to document hours of countable work-related 
activities while otherwise not being exempt, the Federal Government 
will be harming one of the greatest sources of our workforce 
development by making students food-insecure and decreasing their 
ability to complete coursework.
    A new report from the Government Accountability Office (GAO) found 
that a shocking 39% of all undergraduate students in the country--
almost 7.3 million--are at risk of hunger because of low household 
income.\27\ Unfortunately food insecurity often prevents students from 
completing degrees and credentials because they are too hungry to 
learn. Stable part-time work remains elusive to this student 
population, many of whom participate in SNAP to ensure that they can 
cover basic needs because of inconsistent schedules, low wages, and 
lack of benefits.\28\
---------------------------------------------------------------------------
    \27\ https://www.gao.gov/products/GA0-19-95.
    \28\ https://www.jff.org/resources/alleviating-poverty-opportunity-
youth/.
---------------------------------------------------------------------------
    There is already confusion about SNAP eligibility for students, and 
this proposed rule change will only worsen the situation. This 
confusion will also increase difficulty for higher education 
administrators and state regulators in identifying clear eligibility 
determinations for students.
Subversion of Democracy
    Not only would this proposed rule cause unprecedented harm to 
already struggling populations in America, it is an unprecedented 
undermining of our democracy itself.
    At a time of unprecedented political polarization, it is notable 
that the Agriculture Improvement Act of 2018, commonly referred to as 
the farm bill, was reauthorized with historic bipartisan margins of 
support by votes of 369-47 in the House of Representatives \29\ and 87-
13 in the Senate.\30\ As a result of thoughtful and engaged debate and 
deliberation, Congress agreed that significant changes to the SNAP 
ABAWD waivers were unwarranted and unwise--the bill instead strengthens 
ten pilot programs that are currently examining best practices for SNAP 
employment and training. In stark contrast, this arbitrary new proposed 
rule change was announced on the same day that President Trump signed 
the farm bill into law. Designed to curtail SNAP participation, the 
Administration's proposal contradicts express Congressional intent and 
is a callous and calculated attempt to circumvent the democratic 
process as evidenced clearly in the carefully-negotiated final farm 
bill.
---------------------------------------------------------------------------
    \29\ http://clerk.house.gov/evs/2018/roll434.xml.
    \30\ https://www.senate.gov/legislative/LIS/roll_call_lists/
roll_call_vote_cfm.cfm?congress=115
&session=2&vote=00259.
---------------------------------------------------------------------------
    The proposed rule change could not be more out of touch with the 
reality of struggling American workers and families. USDA should focus 
on implementing the 2018 Farm Bill provisions that will help Americans 
get back to work, not resort to rulemaking that is a slap in the face 
to Democracy and jeopardizes critical nutrition assistance for those 
who need help to put food on the table.
    By USDA's own estimate, the proposed rule change would result in 
755,000 people losing access to life-saving nutrition benefits.\31\ The 
proposal completely ignores the realities of people who are willing to 
work but face inconsistent work hours, lack access to reliable 
transportation, live in areas where the economy has been slow to 
recover from the Great Recession, or are unable to access employment 
and training programs--all of whom could fail to meet the burdernsome 
work reporting requirements imposed on SNAP recipients.
---------------------------------------------------------------------------
    \31\ https://www.washingtonpost.com/business/economy/trump-
administration-aims-to-toughen-work-requirements-for-food-stamps-
recipients/2018/12/20/cf687136-03e6-11e9-b6a9-0aa5c2
fcc9e4_story.html
---------------------------------------------------------------------------
    The Administration's stated goal of subjecting more working-age 
adults without minor children to time limits for SNAP benefits is a 
tactic designed to cause more hardship to the very people USDA claims 
to help. The current SNAP eligibility restrictions are already punitive 
as is, with waivers intended for parts of the country where jobs and 
training opportunities are not readily available. Restricting states' 
ability to issue waivers will unrealistically penalize people and 
increase hunger--the very opposite of SNAP's intended purpose.
    Importantly, the 2018 Farm Bill lowered the number of people that 
states can exempt from SNAP time limits. The new law limits states to 
exempting only up to 12% (down from 15%) of adults subject to current 
SNAP time limits, which clearly marks the intent by Congress for policy 
adjustments concerning state waivers for SNAP time limits. However, the 
USDA proposed rule change would prohibit states from carrying over any 
unused percentages from year to year, which could result in penalizing 
those states in years when their economies take a downward turn and 
more families struggle to put food on the table.
State Flexibility
    The proposed rule change directly assaults states' flexibility and 
ability to devise meaningful workforce development programs that 
actually empower SNAP recipients to find and sustain stable work.\32\ 
Waivers help states provide reprieve for communities with high 
unemployment and limited capacity for civil society to support and 
empower SNAP recipients.\33\
---------------------------------------------------------------------------
    \32\ https://www.cbpp.org/research/food-assistance/waivers-add-key-
state-flexibility-to-snaps-three-month-time-limit.
    \33\ https://www.cbpp.org/research/food-assistance/waivers-add-key-
state-flexibility-to-snaps-three-month-time-limit.
---------------------------------------------------------------------------
    Every state except Delaware has at some point requested to waive 
time limits on SNAP since the adoption of the 1996 welfare reform law--
this fact demonstrates that states need a certain amount of flexibility 
in order to ensure that individuals and families can try to stave off 
hunger when they fall on hard times. In fact, 33 states, the District 
of Columbia, Guam, and the Virgin Islands are currently approved for 
statewide or partial time limit waivers, again affirming that there is 
a genuine need for states to have the flexibility to address their 
unique economic circumstances.\34\
---------------------------------------------------------------------------
    \34\ https://www.fns.usda.gov/snap/abawd-waivers.
---------------------------------------------------------------------------
Charity Alone Cannot End Hunger
    The charitable food sector invests a mighty $5 billion per year to 
meet emergency hunger needs, however the Federal Government--mostly 
through SNAP--provides the vast majority of all food assistance in this 
country. Additionally, SNAP and similar programs pump $5.8 billion per 
month, or $70 billion per year, into the U.S. economy.
    Stripping people of critical SNAP benefits will directly impact the 
charitable food system, which is already strained. The recent 
unprecedented government shutdown revealed that many Americans are 
living paycheck to paycheck, with limited savings in the event of 
economic hardship. With thousands of people--including Federal 
workers--turning to the charitable food sector to meet their basic 
needs, the shutdown also illuminated the vital importance of our 
Federal nutrition safety net. What happens when 755,000 low-income 
people are kicked off the meager support SNAP currently provides? To 
expect an extraordinarily generous philanthropic sector to increase its 
expenditures by even a fraction of the twenty-fold difference it holds 
with our government's ability to be a solution is woefully unrealistic 
and dangerous.\35\
---------------------------------------------------------------------------
    \35\ https://www.seattletimes.com/nation-world/private-charity-no-
match-for-federal-poverty-aid-experts-say/.
---------------------------------------------------------------------------
    Charities cannot make up the difference.
SNAP Strengthens Our Economy
    We know that SNAP fuels economic growth in this country. Retailers 
of all sizes benefitted from the $63 billion redeemed in 2017 through 
SNAP funds.\36\ Based on the most recent data available, 10% of all 
food consumption dollars comes from SNAP.\37\ Firms that accept SNAP 
experienced an average 4% increase in business between 2013-2017.\38\
---------------------------------------------------------------------------
    \36\ https://www.cbpp.org/research/food-assistance/snap-boosts-
retailers-and-local-economies.
    \37\ https://www.cbpp.org/research/food-assistance/snap-boosts-
retailers-and-local-economies.
    \38\ https://www.cbpp.org/research/food-assistance/snap-boosts-
retailers-and-local-economies.
---------------------------------------------------------------------------
    Without SNAP, our economy would lose between 0.53% and 1.03% of 
GDP.\39\ In fact, every SNAP dollar spent expands the economy by 
$1.70.\40\ By removing 755,000 people from SNAP, USDA's proposed rule 
change would result in a self-inflicted wound on our economy that would 
be felt in every state and the District of Columbia.
---------------------------------------------------------------------------
    \39\ https://www.marketwatch.com/story/prolonged-shutdown-could-
slash-gdp-along-with-food-stamps-economist-says-2019-01-07.
    \40\ https://www.cbpp.org/research/food-assistance/snap-boosts-
retailers-and-local-economies#ftn6.
---------------------------------------------------------------------------
Invest in Evidence-Based Solutions
    SNAP is first and foremost a food security program, not a catalyst 
for workforce development. It remains unclear how restricting SNAP 
benefits would help people find and sustain gainful employment. Placing 
more stringent restrictions on struggling Americans will not help 
anyone find gainful employment. A more meaningful way to encourage work 
among SNAP recipients would be to invest in effective job training 
programs with robust case management to help individuals successfully 
overcome barriers to employment--especially for people in rural areas, 
on or near Indian reservations, and in economically-distressed 
communities.
    USDA has already invested in pilot employment and training programs 
in ten socioeconomically and geographically diverse states.\41\ Among 
the most successful of these programs is the partnership with the 
Washington State Department of Social and Health Services Resources to 
Initiate Successful Employment (RISE) project. RISE empowers and serves 
individuals receiving SNAP who face significant barriers to 
employment--this includes veterans, people experiencing homelessness, 
individuals with limited English proficiency, and non-custodial parents 
with child support obligations. Case managers employed by community 
colleges and community-based organizations help lower barriers to 
employment by leveraging housing resources, working with the Division 
of Child Support for clients who are delinquent in child support 
payments, and creating accelerated training strategies and job 
placements within in-demand or high growth industries.\42\
---------------------------------------------------------------------------
    \41\ https://fns-prod.azureedge.net/sites/default/files/snap/SNAP-
ET-Pilot-Summaries.pdf.
    \42\ https://fns-prod.azureedge.net/sites/default/files/snap/SNAP-
ET-Pilot-Summaries.pdf.
---------------------------------------------------------------------------
    Washington's RISE program played a critical role in the state's 
workforce development efforts, empowering over 40,000 people with 
employment, training, and support services in the Greater Seattle 
Area.\43\ Of the individuals enrolled in the program between 2009 and 
2011, 71% were employed with a median hourly wage of $11 per hour and 
over $33 million was generated for the community-based organizations 
and community colleges to deliver the program training.\44\
---------------------------------------------------------------------------
    \43\ https://www.nationalskillscoalition.org/resources/
publications/file/Washington-SNAP-brief-web_FINAL.pdf
    \44\ https://www.nationalskillscoalition.org/resources/
publications/file/Washington-SNAP-brief-web_FINAL.pdf
---------------------------------------------------------------------------
    The U.S. would be far better served by replicating this program's 
success across the country instead of pulling the rug out from under 
vulnerable Americans while they are already experiencing hardship.
Conclusion
    MAZON is deeply concerned that the Administration is going down a 
dangerous path of proposed rulemaking that seems intended to discourage 
SNAP use without any meaningful alternatives to economic empowerment. 
If adopted, this proposed rule change would hurt hundreds of thousands 
of SNAP recipients--veterans, Native Americans, college students, and 
people living in rural and remote communities, and other vulnerable 
sectors of our nation--all of whom are vital to our collective strength 
and success. MAZON urges USDA to rescind this proposed rule change and 
instead dedicate resources toward strengthening employment and training 
opportunities to help people find pathways to sustainable and 
meaningful employment.
    If the purported goal of this proposed rule change is to ``increase 
self-sufficiency, well-being, and economic mobility,'' this 
Administration's actions are misguided and its priorities 
troubling.\45\ I would refer you to the comments submitted on behalf of 
MAZON on April 4, 2018 in response to the Advance Notice of Proposed 
Rulemaking regarding SNAP and requirements and services for ABAWDs 
(included with these comments), which included recommendations about 
how to improve SNAP and better help those who struggle with food 
insecurity in this country.
---------------------------------------------------------------------------
    \45\ https://www.whitehouse.gov/presidential-actions/executive-
order-reducing-poverty-america-promoting-opportunity-economic-mobility/
 
---------------------------------------------------------------------------
    We remain unwavering in our opposition to this proposed rule 
change, which makes an end-run around Congressional intent and would 
severely curtail states' flexibility to provide life-saving nutrition 
support to their residents who struggle to feed themselves and their 
loved ones.
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Abby J. Leibman, President and Chief Executive Officer.
                               attachment
April 4, 2018

  Ms. Sasha Gersten-Paal,
  Chief, Certification Policy Branch,
  SNAP Program Development, USDA Food & Nutrition Service,
  Alexandria, VA

  Re: Advanced Notice of Proposed Rulemaking: Supplemental Nutrition 
            Assistance Program: Requirements and Services for Able-
            Bodied Adults Without Dependents RIN 0584-AE57

    Dear Ms. Gersten-Paal,

    On behalf of MAZON: A Jewish Response to Hunger, I am pleased to 
submit these comments regarding input requested about whether USDA 
should reconsider certain rules that govern the 3 month time limit on 
childless adults for the Supplemental Nutrition Assistance Program 
(SNAP).
    Inspired by Jewish values and ideals, MAZON is a national advocacy 
organization working to end hunger among people of all faiths and 
backgrounds in the United States and Israel. For more than 30 years, 
MAZON has been committed to ensuring that vulnerable people have access 
to the resources they need to be able to put food on the table. MAZON 
is a leading voice throughout the country on anti-hunger issues, 
especially those that involve populations or problems that have been 
previously overlooked or ignored--this includes food insecurity facing 
veterans, currently serving military families, seniors, Native 
Americans, and college students.
    The Jewish community has a rich tradition of asking questions and 
wrestling with different perspectives in order to discover the meaning 
and truth about an issue. However, asking the right questions is 
critical in the pursuit of understanding. We are concerned that the 
focus for USDA's solicited input to inform the proposed rulemaking is 
misdirected, and the questions being asked will not advance the stated 
goals of addressing food insecurity and helping people move out of 
poverty. In short, they are not the right questions.
    SNAP is first and foremost a food security program, not a catalyst 
for workforce development. The framing of this request for public 
comments that links the goal of addressing food insecurity with 
``helping able-bodied SNAP recipients obtain and maintain employment 
and aligning program regulations with the President's Budget proposals 
related to ABAWDs'' misconstrues the purpose of the SNAP program and 
approaches the issue using an incorrect premise, ensuring that 
recommendations will not be responsive to the actual goals of this 
long-standing program. This request for comments does not inspire us to 
provide suggestions, input, and answers; instead, it has only provoked 
more questions:

   How would placing more stringent time limits on able-bodied 
        adults without dependents (ABAWDs) receiving SNAP actually help 
        people to find and sustain gainful employment? What is the 
        evidence and data to support this notion? What data exists to 
        show how severe time-limits have contributed to employment for 
        recipients of SNAP and other Federal assistance programs?

   Why is the focus for USDA's inquiry on processes and 
        procedures affecting ABAWDs and not on additional investments 
        to ensure access to employment and training opportunities to 
        help move them towards self-sufficiency? Why does this inquiry 
        not address the importance of case management and other 
        supports that take into consideration the real circumstances 
        and challenges faced by unemployed individuals that have not 
        only demonstrated efficacy but are designed to make the 
        transition to employment much more effective?

   Why do states have the option to offer Employment and 
        Training (E&T) on a voluntary basis to certain or all SNAP 
        participants but are not mandated to do so, and why does this 
        request for comments not seek insights about whether this 
        policy should be reconsidered?

   Does USDA recognize that ABAWDs are by no means a monolithic 
        population, including veterans, college students, those 
        suffering from mental health challenges, and individuals 
        formerly in the foster care system? How does USDA envision 
        ensuring the employment of this diverse and complex population 
        in 3 months or fewer? What considerations do USDA and states 
        currently make to account for the diverse circumstances and 
        challenges by unemployed ABAWDs trying to find employment? What 
        more could USDA and states do to recognize and address the 
        needs of this diverse population?

   What is USDA doing to increase employment, training, and 
        workforce participation among ABAWDs in collaboration with 
        other Federal agencies that have more experience and expertise 
        in these areas? Why is so much burden to increase employment 
        being placed on SNAP, which is a nutrition assistance program 
        not an employment training program?

   Why is USDA seeking input on these questions now, when the 
        results of the SNAP E&T pilot programs called for in the last 
        farm bill are not yet available? Would it not be more prudent 
        and helpful to consider the data, best practices, and lessons 
        learned from those pilot programs before seeking to make policy 
        and programmatic changes? Would making such changes without the 
        benefit of the results of these pilot programs be considered 
        irresponsible and a disservice to the SNAP program and the 
        millions of Americans who receive vital assistance from it?

   Does USDA truly believe that the receipt of SNAP benefits 
        prevents unemployed ABAWDs from seeking and securing gainful 
        employment? Does USDA believe that revoking SNAP benefits after 
        a 3 month time period will notably increase employment rates 
        for ABAWDs? What is the evidence to support these beliefs?

   Does USDA believe that hunger is the best motivator for 
        self-improvement? Is government-imposed hunger--forcing people 
        off needed benefits without adequate training or 
        opportunities--an appropriate function of the Federal 
        Government? What are the moral justifications for this policy?

    While we welcome the chance to address opportunities for 
improvement of the SNAP program, we fail to understand the singular 
focus on ABAWDs, which completely ignores the complex realities of 
their lives, the economic circumstances of the diverse regions of this 
country, and the lack of data to support cutting them off from 
nutrition assistance. The rhetorical framing of these proposals, and 
the proposals themselves, seek to punish struggling Americans with 
harsh penalties for conditions beyond their control. MAZON hopes that 
USDA can answer our questions to help provide greater clarity and 
vision to the proposed changes to SNAP. We hope that USDA can take a 
step back to ask the right questions about how to improve SNAP and 
better help those who struggle with food insecurity in this country.
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Abby J. Leibman, President and Chief Executive Officer.
                                 ______
                                 
  Submitted Comment Letter by Hon. Jimmy Panetta, a Representative in 
  Congress from California; Authored by Keith Carson, Vice President, 
Board of Supervisors, District 5; Chair, Personnel, Administration and 
                      Legislation (PAL) Committee
March 29, 2019

  Certification Policy Branch,
  SNAP Program Development Division,
  Food and Nutrition Service, USDA,
  Alexandria, Virginia

  RE: Proposed Rule: Supplemental Nutrition Assistance Program (SNAP): 
            Requirements for Able-Bodied Adults without Dependents RIN 
            0584-AE57

    Dear Certification Policy Branch:

    The County of Alameda, California (``Alameda County'' or the 
``County'') submits these comments in response to the February 1, 2019 
Notice of Proposed Rulemaking (``NPRM'' or ``proposed rule'') from the 
Department of Agriculture's Food and Nutrition Service (FNS) on the 
Supplemental Nutrition Assistance Program that proposes changes to the 
able-bodied adults without dependents (ABAWD) state waivers. This 
includes rejecting the proposed seven percent unemployment floor for 
the 20 percent standard, limitation of applicable data to only that 
provided by the Bureau of Labor Statistics (BLS), limitation of waiver 
applicability to only the fiscal year in which it is implemented, 
restriction on regions that can be combined for a waiver application, 
the elimination of historical seasonal unemployment as a criterion for 
approval, and restricted use of the 15 percent exemptions. The County 
of Alameda strongly opposes these proposed changes and urges FNS to 
immediately withdraw the proposed rule.
    The Alameda County Social Services Agency (ACSSA) administers 
eligibility for CalFresh (the Supplemental Nutrition Assistance 
Program, or SNAP) and the SNAP Employment and Training (SNAP E&T) 
programs. SNAP plays a critical role in addressing hunger and food 
insecurity in our community. It is the first line of defense against 
hunger for low-income residents. SNAP drives over $11 billion in total 
economic activity annually in California and the proposed rule would 
harm our local economies, retailers and agricultural producers by 
reducing the amount of SNAP dollars people have to spend on food. 
Approximately 99,180 individuals receive nutritional support through 
SNAP in Alameda County.
    Under the current rules, an Able-Bodied Adult Without Dependents 
(ABAWD) is a non-assistance CalFresh/SNAP recipient age 18 to 49 who is 
able-bodied without dependent children. ABAWD eligibility for CalFresh/
SNAP is time limited to any 3 full months of benefits in a 36 month 
period unless the individual: (1) satisfies the ABAWD work requirement; 
(2) is exempt from the ABAWD time limit; (3) qualifies for an 
additional 3 consecutive month period of eligibility; (4) receives a 15 
percent exemption; or (5) lives in a county or area with a waiver of 
the ABAWD time limit. Many ABAWDs subject to the time limit face 
significant barriers and struggle to find and maintain employment. As a 
result, the work requirement functions more as a time limitation, and 
individuals categorized as ABAWDs simply lose nutritional support at 
the end of 3 months.
    A county, multi-county region, area within a county, or an entire 
state can be approved for a waiver of the ABAWD time limit if it meets 
federally established criteria regarding high unemployment or lack of 
sufficient jobs. Alameda Comity has had a waiver for ABAWD eligibility 
since May 2008, when a statewide waiver was approved at the start of 
the recession. The County previously operated under a waiver from 
December 1996 through August 1997. While the State of California no 
longer holds a statewide waiver, Alameda County is operating CalFresh/
SNAP with a regional ABAWD waiver until August 31, 2019. The waiver has 
permitted ABAWD CalFresh/SNAP recipients to continue receiving 
nutritional support during times when jobs have been scarce, the 
unemployment rate [is] higher than the national average, and the job-
to-employment ratio [is] historically low. The County also operated 
under a 15 percent exemption for ABAWDs for Fiscal Years 2000-2007.
    California has structured the use of exemptions such that 
exemptions can be used to encourage individuals to engage in employment 
and training activities. For example, exemptions could be used for 
individuals living in rural areas who may require additional time to 
engage in job search activities, or for those individuals who are 
engaged in employment and training but may happen to not meet the hours 
needed during a given month, for example if an individual falls ill for 
a day and therefore falls short of meeting the hourly requirement in 
the month.
Establishing a Lower Unemployment Floor
    One of the proposals is to set a floor of seven percent 
unemployment for which the 20 percent standard for regional waivers 
would be considered. The NPRM requests comments on setting the floor at 
six, seven, or ten percent. Given that the ``natural rate of 
unemployment'' cited in the NPRM is five percent, Alameda County is of 
the position that the proposed floors are higher than what is 
reasonable to meet the waiver's intention: to provide relief for those 
areas where unemployment rates limit one's ability to find a job and 
meet ABAWD work requirements. The theory of a natural rate of 
unemployment is debated by economists, and is impacted by the larger 
conditions of the economy--meaning it does not stand at a steady rate. 
There is not one agreed-upon steady figure for the current natural rate 
of unemployment, and of recent estimates, five percent is on the higher 
side. If the natural rate were actually closer to four percent, as some 
economists and Federal Reserve Banks have argued, then a seven percent 
floor for the 20 percent standard would require an unemployment rate 75 
percent higher than the natural rate for waiver eligibility. This means 
that in an economic downturn, regions would need to lose more jobs than 
the market provides--leaving more workers out of work--before they may 
be potentially eligible for an ABAWD waiver. Because the natural rate 
of unemployment is one that fluctuates with the larger economy, Alameda 
County believes setting a constant floor in relation to a proposed 
constant estimate of the ``natural rate'' for the consideration of 
ABAWD waivers is inappropriate. It is potentially due to these 
fluctuations that the original ABAWD guidance following the 1996 
Personal Responsibility and Work Opportunity Reconciliation Act 
(PRWORA) did not set a ``floor''; rather, the regulations set regional 
waiver eligibility in relation to the national unemployment average. 
This relationship between a state or region and the national average 
better estimates the actual job losses or gains that impact individuals 
within the region as they seek employment. Alameda County believes the 
20 percent standard should maintain the relationship between the state/
region unemployment rate and the national unemployment rate.
    The NPRM notes that with the current relational metric for the 20 
percent standard, 44 percent of ABAWDs live in waived areas. A seven 
percent standard would reduce this to 11 percent of ABAWDs, six percent 
standard to 24 percent, and a ten percent standard to only two percent 
of ABAWDs. All proposed floors would result in a dramatic loss of 
nutritional support for individuals who are eligible for SNAP. 
According to the Bureau of Labor Statistics (BLS), the Oakland-Hayward-
Berkeley division, which encompasses much of Alameda County, had a 
civilian labor force of 1,449,098 in November 2018. To achieve a seven 
percent unemployment rate, more than 60,190 jobs would need to be lost. 
To achieve a ten percent unemployment rate, more than 103,660 jobs 
would need to be lost. These would be significant changes to the local 
economy, causing considerable harm to the most vulnerable as the 
economy loses jobs, and due to strict waiver requirements if the NPRM 
becomes finalized, ABAWDs would be ineligible for CalFresh/SNAP 
benefits appropriate to local and individual needs. If the intention is 
to remove beneficiaries from the program, then setting one of these 
floors would meet that intention. However, the goal stated in the Food 
and Nutrition Act of 2008 is ``the policy of Congress, in order to 
promote the general welfare, [is] to safeguard the health and well-
being of the Nation's population by raising levels of nutrition among 
low-income households.'' A proposal to restrict state ABAWD waivers, 
such as proposed by this NPRM, was rejected by Congress during the 
recent reauthorization of the farm bill in fall 2018; therefore, this 
NPRM exceeds Congressional intent.
Limiting State Waiver Areas To Those Designated As a Labor Market Area 
        (LMA)
    The NPRM includes a proposal to limit the areas states could group 
together for a state waiver to those that are [designated] as a Labor 
Market Area (LMA) by the Federal Government. The NPRM justifies this 
proposal by stating that the limitation ensures that only areas that 
are economically tied together are grouped together. Alameda County 
believes that this proposal is significantly too narrow. For example, 
the LMA that includes much of Alameda County is the San Francisco-
Oakland-Hayward Metropolitan Statistical Area. Due to a jobs/housing 
imbalance throughout much of the San Francisco Bay Area, County to 
County Commute Estimates compiled by the Employment Development 
Department illustrate that the MSA is insufficient. These estimates 
show that 235,311 workers commute to jobs in Alameda County from--
Contra Costa (92,797), Santa Clara (38,339), San Joaquin (26,121), San 
Francisco (22,009), San Mateo (13,417), Stanislaus (8,198), and nearly 
a dozen other counties in the vicinity. Similarly, 227,757 workers who 
reside in Alameda County commute to jobs outside the County to--San 
Francisco (71,861), Santa Clara (64,696), Contra Costa (39,883), San 
Mateo (34,369) and nearly a dozen other counties in the vicinity.\1\ 
The Bay Area is significantly larger than the traditional nine county 
Bay Area, as ``super-commuters'' from regional counties travel to 
polycentric urban centers throughout the region to reach their place of 
employment. Restricting ABAWD waivers to LMAs would significantly 
underestimate the impact of labor market shifts in Alameda County's 
multi-county relationships. Job losses in any of the neighboring LMAs 
would directly impact workers who live in Alameda County. With the 
proposal in the NPRM, these workers would not have their job losses 
included in ABAWD waiver consideration.
---------------------------------------------------------------------------
    \1\ State of California Employment Development Department, ``County 
to County Commute Patterns,'' https://www.labormarketinfo.edd.ca.gov/
data/county-to-county-commute-patterns.html.
---------------------------------------------------------------------------
Limiting the Duration of Waivers
    Alameda County has concerns about the NPRM proposal to limit ABAWD 
waivers to the fiscal year in which they are implemented. Processing 
time for waiver applications could limit the applicability of the 
waiver. For example, the current ABAWD waiver under which Alameda 
County operates was submitted September 18, 2017. FNS action did not 
occur until July 2, 2018--nearly 10 months later, and into a new fiscal 
year--with the waiver applying to September 1, 2018 through August 31, 
2019. If the NPRM were finalized as written, this waiver may have only 
been applicable for 1 month: through September 2018. We are concerned 
that with lengthy processing timelines, ABAWD waivers may be approved 
only for a matter of months per application, rather than providing 
relief for a year, as has been regular practice. County processes would 
be considerably complicated by this move, as a new waiver or end of a 
waiver changes ABAWD eligibility practices for workers throughout the 
agency. Significant staff time would be used to update systems and 
practices for each time-restricted waiver. Eliminating statewide 
waivers would result in a significant administrative burden in 
California which will not help save or reduce costs. The elimination of 
the waiver would require additional staff time and training to engage 
the ABAWD caseload to encourage participation in employment and 
training activities.
Harm to Individuals, Families and Children, and Impact to Local 
        Economies
    The goal for SNAP recipients to achieve self-sufficiency is 
predicated on the ability of local economies to provide adequate 
opportunities for gainful employment. Though the national unemployment 
rate has recovered from the Great Recession, many individuals in a wide 
variety of regions continue to struggle finding steady work that makes 
ends meet. The employment-to-population ratio, as tracked by BLS, has 
not yet recovered to pre-recession levels. The ABAWD waivers provide 
relief for these individuals, so that their ability to purchase 
adequate food necessary for their well-being is not negatively impacted 
by factors outside of their control. Nationwide, USDA data show that 
the individuals impacted by ABAWD waivers have an average monthly 
income of approximately 17 percent of the poverty line.\2\ These 
individuals typically qualify for no other income support. This rule is 
harsh and unfair. It harms vulnerable people by denying them food 
benefits at a time when they most need it and, yet does not result in 
increased employment and earnings. By time-limiting food assistance to 
this group, Federal law has shifted the burden of providing food to 
these unemployed individuals from SNAP to states, cities, and local 
charities. Removing nutrition support does not enable these individuals 
to secure employment, or work more hours. It does not increase self-
sufficiency; instead, the proposed policy change only causes harm to 
those most in need.
---------------------------------------------------------------------------
    \2\ Center on Budget and Policy Priorities, ``More than 500,000 
Adults Will Lose SNAP Benefits in 2016 as Waivers Expire,'' March 2016. 
https://www.cbpp.org/research/food-assistance/more-than-500000-adults-
will-lose-snap-benefits-in-2016-as-waiversexpire#_ftn3.
---------------------------------------------------------------------------
    In Alameda County, there are approximately 167,000 food-insecure-
households.\3\ While a two-working-parent family with two children 
needs $97,000/year to make ends meet,\4\ the annual income of 40% of 
the households served by the Alameda County Community Food Bank is 
$10,000/year or less.\5\ It has been demonstrated that individual, 
household, and societal costs of food insecurity are high. Food-
insecure households are sometimes unable to afford balanced, healthy 
meals, putting them at increased risk of diet-related disease such as 
diabetes and obesity. Further, food-insecure children are at higher 
risk of being hospitalized, iron deficient, obese, have lower tests 
scores than their peers, have greater difficulty getting along with 
other children, and may have impaired social development. Food 
insecurity also places a burden on our health care and safety net 
systems--systems in which millions of households run short of money for 
food at the end of each month. Being food-insecure is correlated to a 
nearly 50% increased likelihood of being in the top 5% of health care 
users, and it is estimated that food insecurity costs the U.S. health 
system $160 billion per year in poor health outcomes and additional 
health care.\6\
---------------------------------------------------------------------------
    \3\ California Food Policy Advocates, ``Nutrition & Food Insecurity 
Profile--Alameda County,'' July 2016. https://cfpa.net/county-profiles/
 
    \4\ California Budget and Policy Center, ``Making Ends Meet Fact 
Sheet: Alameda County Monthly Family Budget,'' 2017. https://
www.calbudgetcenter.org/wp-content/uploads/Fact-Sheet_Making-Ends-Meet-
2017_AlamedaCty.pdf.
    \5\ Alameda County Community Food Bank, ``2014 Hunger Alameda 
County Uncovered,'' 2014. https://www.accfb.org/wp-content/uploads/
2017/08/ACCFB-HungerStudy2014-smaller.pdf.
    \6\ Bread for the World Institute, ``The 2016 Hunger Report by the 
Numbers,'' 2016. http://hungerreport.org/2016/wp-content/uploads/2015/
12/HR2016-Fact-Sheet.pdf.
---------------------------------------------------------------------------
    Because SNAP is so important for low-income and food-insecure 
children, children under the age of 18 and the adults who live with 
them are technically exempt from the 3 month time limit for SNAP. 
However, though current rules around the SNAP time-limit explicitly 
exempt adults who have a dependent child under the age of 18 or live in 
a household with children under 18, this definition may not allow for 
the complex financial arrangements that low-income families utilize to 
put food on the table. Alameda County represents the interests of 
vulnerable children who as a result of this rule will experience a 
reduction in important resources that help meet their basic needs, even 
though FNS does not account for this in its cost-benefit analysis. This 
includes:
    Children with non-custodial parents: Poverty is a troubling reality 
for custodial and noncustodial parents. The most recent available data 
from 2015 suggests that 3.5 million custodial parents live below the 
poverty line, making access to food assistance all the more important 
for them and their children.\7\ Thus, some 4.5 million poor and low-
income custodial parents who rely on child support payments from non-
custodial parents (NCPs) also utilize SNAP to put food on the table for 
their children.\8\ Yet NCPs are often themselves low-income, with 2.1 
million living below the poverty line in 2015, and 1.5 million 
accessing SNAP to supplement their resources to afford child support 
payments.\9\ Because NCPs are not exempt from the ABAWD time-limit, the 
proposed rule not only threatens them, but their children. An under-
employed or unemployed NCP who loses SNAP may need to divert his or her 
income from child support payments in order to stay afloat financially, 
which would be particularly devastating given that child support 
represents more than \1/2\ of the income of the families in poverty who 
receive it.\10\
---------------------------------------------------------------------------
    \7\ U.S. Department of Health and Human Services Office of the 
Assistant Secretary for Planning and Evaluation, ``How Many Families 
Might be Newly Reached By Child Support Cooperation Requirements in 
SNAP and Subsidized Child care, and What Are Their Characteristics?'', 
July 2018, https://aspe.hhs.gov/pdf-report/how-many-families-might-be-
newly-reached-child-support-cooperation-requirements-snap-and-
subsidized-child-care-and-what-are-their-characteristics.
    \8\ U.S. Census Bureau, ``Custodial Mothers and Fathers and Their 
Child Support: 2015 Current Population Survey,'' April 2016, Table 4, 
https://www2.census.gov/programssurveys/demo/tables/families/2015/
chldsu15.pdf.
    \9\ Ibid. at 7.
    \10\ Heather Hahn, ``Navigating Work Requirements in Safety Net 
Programs: Potential Pathways for Parents,'' The Urban Institute, 
January 2019, https://www.urban.org/sites/default/files/publication/
99479/navigating_work_requirements_in_safety_net_programs_0.pdf.
---------------------------------------------------------------------------
    Children whose extended family members provide financial support: 
Some low-income children may rely on food, financial assistance, or 
free childcare from extended family members, family friends, or a 
parent's significant other who do not live with them but use SNAP to 
supplement their income. Households that are the most financially 
precarious are the most likely to rely on such transfers to make ends 
meet. Considering that financially precarious households are often 
embedded together within the same networks, they likely received money 
or assistance from others who are also struggling economically.\11\ If 
ABAWDs in these networks lose SNAP benefits due to tightened state 
waiver rules, it would disrupt their ability to lend that crucial 
assistance to low-income children.
---------------------------------------------------------------------------
    \11\ The Pew Charitable Trusts, ``Extended Family Support and 
Household Balance Sheets: Getting by with a little help from friends 
and relatives,'' March 2016, https://www.pewtrusts.org/-/media/assets/
2016/03/fsm_kinshipbrief.pdf.
---------------------------------------------------------------------------
    Youth aging out of foster care and unaccompanied homeless youth: 
Youth in foster care and unaccompanied homeless youth 
disproportionately experience significant barriers to obtaining a high 
school diploma, entering college, obtaining a driver's license, 
accessing health insurance, maintaining housing stability, and 
obtaining steady employment. SNAP plays a significant role in the 
health and well-being of youth aging out of care and unaccompanied 
homeless youth with no support systems. Former foster youth often 
experience poor nutrition and food insecurity, and SNAP benefits help 
to address this problem and increase the likelihood of healthy adult 
outcomes.\12\ However, because former foster youth and unaccompanied 
homeless youth often meet the definition of an Able-Bodied Adult 
Without Dependents, they face obstacles accessing this critical 
assistance and would likely disproportionately suffer under tightened 
state waiver requirements.
---------------------------------------------------------------------------
    \12\ Megan Martin, Shadi Houshyar, Alexndra Citrin, DeQuendre 
Neeley-Bertrand, DeQuendre and Raquan Wedderburn, ``Supporting Youth 
Aging Out of Poster Care through SNAP,'' The Center for the Study of 
Social Policy, 2014, https://www.cssp.org/policy/2016/supporting-youth-
aging-out-of-foster-care-through-SNAP.pdf.
---------------------------------------------------------------------------
    The Department provides little analysis to explain its conclusions 
about the impacts the changes would have on individuals and population 
groups nor of realistic plans to avert harm from those changes. USDA 
merely asserts its expectation that \2/3\ of those individuals made 
newly subject to the time limit ``would not meet the requirements for 
failure to engage meaningfully in work or work training.'' By the 
Administration's own calculations, the proposed rule would take food 
away from 755,000 low-income Americans, cutting food benefits by $15 
billion over 10 years. The Administration does not estimate any 
improvements in health or employment among the affected population. 
Moreover, while the Department concedes that the proposed changes 
``have the potential for disparately impacting certain protected groups 
due to factors affecting rates of employment of these groups, [it] 
find[s] that implementation of mitigation strategies and monitoring by 
the Civil Rights Division of FNS will lessen these impacts.'' But no 
explanation of the mitigation strategies and monitoring is provided, 
leaving no opportunity for the public to comment on whether the 
acknowledged disparate impact will in fact be mitigated.
    The economic impact of such a drastic change in ABAWD rules has an 
enormous economic impact not only on California as a state, but on 
local California communities and counties as well. Based on USDA 
Economic Research Service analysis, it is estimated that each $1 in 
Federal SNAP benefits generates $1.79 in economic activity. Those 
dollars help many food retailers operating on thin margins to remain in 
business; something that improves food access for all residents.
Conclusion
    For all of these reasons, Alameda County strongly opposes the 
proposed rule that would expose even more people to the arbitrary food 
cutoff policy by limiting state flexibility regarding area waivers and 
individual exemptions. Furthermore, we urge you to immediately withdraw 
the proposed rule.
            Sincerely,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Keith Carson,
Vice President, Board of Supervisors, District 5;
Chair, Personnel, Administration and Legislation (PAL) Committee.

CC:

California Congressional Delegation;
Members, Board of Supervisors;
Donna Ziegler, Alameda County Counsel;
Lori Cox, Director, Social Services Agency;
Colleen Chawla, Director, Health Care Services Agency;
C.J. Lake, LLC;
County Welfare Directors Association of California.
                                 ______
                                 
   Submitted Comment Letter by Hon. Ted S. Yoho, a Representative in 
                         Congress from Florida
February 19, 2019

  Hon. Sonny Perdue,
  Secretary of Agriculture,
  U.S. Department of Agriculture,
  Washington, D.C.

    Dear Secretary Perdue,

    We write to thank you for the United States Department of 
Agriculture's (USDA) recent notice of proposed rulemaking regarding 
work requirements for able-bodied adults without dependents (ABAWD) 
recipients of the Supplemental Nutrition Assistance Program (SNAP), RIN 
0584-AE57, and to urge our support for promulgation of the rule as 
proposed. At a time when our nation is seeing historic economic growth, 
including generationally-low unemployment rates, this proposed rule 
will allow our country to continue to thrive by restoring integrity to 
SNAP and by moving the American people toward complete self-
sufficiency, thereby saving American taxpayers billions of dollars.
    As you may know, SNAP was originally intended to give hard-working 
Americans a second chance should they encounter a difficult stretch in 
life--it was never intended to become one's livelihood, or their so-
called ``way of life.'' In spite of this, since our last welfare reform 
legislation in 1996, the program has repeatedly shifted from these 
first intentions, and has continually been weakened by increased 
administrative flexibility.
    This flexibility has allowed state governments to abuse their power 
and evade the reasonable work requirements that SNAP utilizes to ensure 
that recipients don't take advantage of the current system. These 
requirements obligate ABAWDs, who are non-disabled and between the ages 
of 18 and 49, to work or participate in an employment program for at 
least 20 hours a week to continue to receive benefits for more than 3 
months over a 36 month period.
    However, under the current law, state governments may waive these 
requirements in areas where the unemployment rate is above the national 
average. Given our nation's strong economy, this can include areas with 
unemployment rates under five percent--a rate that is conventionally 
considered as full employment. Additionally, states are allowed to 
grant partial state waivers by grouping together areas with similar 
labor markets, which allows a state to gerrymander areas for waiver 
purposes, thus potentially authorizing waivers for the entire state.
    Also, states may exempt up to 15 percent of their ABAWDs. However, 
should the states not use these exemptions, they are able to hoard them 
for use in future years, which has resulted in certain states 
accumulating hundreds of thousands of exemptions. All of this has led 
to an abdication of SNAP's original purpose and has disincentivized 
self-sufficiency, which resulted in 3.8 million individual ABAWDs on 
SNAP in 2016, of which 2.8 million were not working at all.
    The USDA's proposed rule would help to fix this significant problem 
by implementing several common-sense reforms to the current work 
requirement waiver laws. These include raising the necessary 
unemployment threshold for local area work requirement waivers to seven 
percent unemployment, ending the states' ability to gerrymander waiver 
districts by only granting partial state waivers for areas that are 
``economically tied,'' ending the states' ability to accumulate and 
carryover work requirement exemptions for more than 1 year, and 
increasing SNAP administrative efficiency by setting clearer standards 
for allowable waivers.
    These reforms would save hard-working American taxpayers $15 
billion over a 10 year period and would help to reestablish the true 
goal of the SNAP program, to help hard-working Americans in their 
attempts to gain self-sufficiency. As such, we support the USDA's 
proposed rule and urge you to promulgate this rule in its current 
proposed version, to ensure our nation's continued success.
            Thank you,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            

 
 
 
Hon. Kevin Hern,
Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]


 
 
 
Hon. Tom Cole,                       Hon. Markwayne Mullin,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Trent Kelly,                    Hon. Mark Meadows,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Kevin Brady,                    Hon. Mike Johnson,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Brian Babin,                    Hon. Gary J. Palmer,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Dusty Johnson,                  Hon. Van Taylor,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Jim Jordan,                     Hon. Matt Gaetz,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Andy Harris,                    Hon. John Joyce,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Michael C. Burgess,             Hon. Bob Gibbs,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. John Ratcliffe,                 Hon. Kelly Armstrong,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Ralph Norman,                   Hon. Ron Wright,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Scott Perry,                    Hon. Roger W. Marshall,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Mo Brooks,                      Hon. Alexander X. Mooney,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Jeff Duncan,                    Hon. Mike Bost,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. David Schweikert,               Hon. Jason Smith,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Tim Burchett,                   Hon. Andy Barr,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Roger Williams,                 Hon. Michael Guest,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Bradley Byrne,                  Hon. Randy K. Weber, Sr.,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Steve King,                     Hon. Paul A. Gosar,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Bruce Westerman,                Hon. Robert B. Aderholt,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Ted S. Yoho,                    Hon. Lance Gooden,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Glenn Grothman,                 Hon. Francis Rooney,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. W. Gregory Steube,              Hon. Daniel Meuser,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Bill Johnson,                   Hon. Barry Loudermilk,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Ron Estes,                      Hon. Mark E. Green,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Jackie Walorski,                Hon. Vicky Hartzler,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Mark Walker,                    Hon. John R. Moolenaar,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Jack Bergman,                   Hon. Carol D. Miller,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. David Rouzer,                   Hon. Jim Hagedorn,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Ben Cline,                      Hon. Warren Davidson,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Steve Chabot,                   Hon. Anthony Gonzalez,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Paul Mitchell,                  Hon. Garret Graves,
Member of Congress;                  Member of Congress;
 

[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
                                     

 
 
 
Hon. Guy Reschenthaler,              Hon. Michael Cloud,
Member of Congress;                  Member of Congress;
 

                                 ______
                                 
                          Submitted Questions
Questions Submitted by Hon. Dusty Johnson, a Representative in Congress 
        from South Dakota
Response from Karen Cunnyngham, Associate Director, Mathematica Policy 
        Research
    Question 1. Ms. Cunnyngham, does your analysis investigate the 
potential positive income, reduction in poverty, and overall quality of 
life improvement for individuals who newly obtain employment because of 
the rule change? If so, is this available as a supplement to your 
previously released analysis?
    Answer. My analysis relies on objective, rigorously derived 
estimates of the effects of changes to SNAP, and unfortunately, such 
data pertaining to people who newly gain employment as a result of 
changing ABAWD work requirements is not available. As such, my analysis 
does not address this issue directly. But these are important 
questions, and research on quality-of-life improvement using rigorously 
collected evidence would be an important contribution to the policy 
discussion.

    Question 2. Ms. Cunnyngham, your written testimony mentioned that 
access to a good SNAP employment and training program could help SNAP 
participants meet the work requirements. The 2018 Farm Bill clarified 
the programs that can count for meeting SNAP E&T. What estimates have 
you (or others) done on SNAP participants' access to all types of E&T 
programs, and on the overall ability of those programs to help 
participants comply with any changes in the SNAP waiver rules?
    Answer. Under a contract for the U.S. Department of Agriculture, 
Mathematica is currently testing innovative strategies to increase 
employment and earnings among SNAP participants and reduce their 
dependence on SNAP and other public assistance programs. This study was 
mandated in the 2014 Farm Bill, which authorized grants for up to ten 
pilots (or demonstration projects). The ten pilots offer diverse 
services and target different groups of SNAP participants in various 
geographic locations. Mathematica's random assignment impact analysis 
will give policymakers insight into effective strategies for increasing 
employment and earnings; decreasing public assistance; and positively 
influencing other outcomes of interest, including food security, 
health, and housing. The impact analysis will also help us understand 
whether any of these outcomes vary for different types of SNAP 
participants. In addition, the evaluation includes the following: (1) 
an implementation analysis that will document the operation of each 
pilot and provide context for interpreting and understanding observed 
impact both within and across pilots; (2) a participation analysis that 
will examine the characteristics and service paths of pilot 
participants and control group members and determine whether the 
pilots, the services they offer, and the requirements they impose 
affect individuals' decisions about whether to apply for SNAP; and (3) 
a cost-benefit analysis that will estimate the return on each dollar 
invested in providing E&T services. Interim study findings are set to 
be released in late 2019, and we would be happy to provide your office 
a copy of our findings.
    On the question of what access participants have to SNAP E&T 
services, Mathematica studied the characteristics of SNAP E&T several 
years ago. This was a nationally representative study of E&T 
participants and providers that documented the services available and 
received. The full report, with an executive summary, can is available 
at https://fns-prod.azureedge.net/sites/default/files/ops/
SNAPEandTCharacteristics.pdf.

    Question 3. Ms. Cunnyngham, did the estimates of the number of 
individuals affected by the proposed ABAWD rule changes or any of your 
estimates attempt to differentiate the reasons an individual might be 
impacted?
    Answer. Drilling down on individual motivations, behavioral 
insights, or other reasons individuals might or might not be impacted 
by the proposed rule change is not possible with the data we used for 
our analysis. Currently, we can only speak to the circumstances, not 
the ``why'' behind the circumstances. Additional data, however, 
possibly provided at the state and local levels, could help explain 
why.
Response from Sam Adolphsen, Vice President of Executive Affairs, 
        Foundation for Government Accountability
    Question 1. Mr. Adolphsen, how is one declared able-bodied? Can you 
provide various state examples of the eligibility process that 
determines such a classification?
    Answer. The term ``able-bodied'' in relation to the Supplemental 
Nutrition Assistance Program (SNAP), or food stamps, is a term found in 
Federal Food Stamp regulations at 7 CFR, Subchapter C.\1\ The relevant 
portion of food stamp law, U.S. Code Title 7, Chapter 51, and Section 
2015, contains a work requirement for able-bodied adults on the program 
and defines those individuals as between ages 15 and 60 who are 
``physically and mentally fit.'' \2\
---------------------------------------------------------------------------
    \1\ https://www.law.cornell.edu/cfr/text/7/subtitle-B/chapter-II/
subchapter-C.
    \2\ https://www.law.cornell.edu/uscode/text/7/2015.
---------------------------------------------------------------------------
    During the eligibility process for receipt of the food stamp 
benefit, a state worker from the administering agency will use the 
application process to determine if the applicant is in fact an able-
bodied adult subject to the work requirement in Section 2015. They will 
further determine if they are an Able-Bodied Adult Without Dependents 
(ABAWD) subject to a separate requirement in Section 2015(o). In either 
case, the applicant or participant can demonstrate they are unable to 
meet the requirement through a note from a doctor or social worker at 
any point. Recent guidance from Food and Nutrition Services has also 
allowed eligibility workers to make this determination of unfitness in 
certain extreme cases.

    Question 2. Florida is one of many states focused on work-oriented 
reforms, and a new report shows the incredible impact they're having on 
the state. Since the state implemented a food stamp work requirement in 
2016, nearly 94 percent of able-bodied, childless adults have left 
Florida's food stamp program. Those who left the program found work in 
over 1,104 industries--in nearly every corner of Florida's economy. 
Even better, 70 percent of those who initially found work in the fast 
food industry or at a temp agency left those industries within 1 year. 
Work requirements aren't just helping people climb out of dependency in 
Florida--they're helping them climb the economic ladder to better 
opportunities and bigger paychecks. Mr. Adolphsen, how can we replicate 
this in other states who may find it difficult or not cost-efficient to 
enforce work requirements and engage recipients?
    Answer. The results of work requirements for ABAWDs in the food 
stamp program in Florida were outstanding, as noted in the question. 
However, we know this is impact is not isolated only to Florida. Both 
Kansas and Maine implemented the same work requirement in 2014, and in 
subsequent studies, each of those states found results similar to 
Florida--incomes of those former participants in the program more than 
doubled in just 1 year. Furthermore, Kansas and Maine did not report 
any additional administrative burden or cost as a result of 
implementing the work requirement. In all three states, enrollment in 
the program dropped significantly as a result of the reform, and since 
most administrative costs are driven by caseloads, there is strong 
evidence to believe this type of reform can actually drive down 
administration burden.
    In fact, California, which has not implemented the work requirement 
for ABAWDs in almost the entire state, has the second highest 
administrative cost per case in the country at $68.52 per case, per 
month. That is more than double the national average of $29.98. 
Meanwhile Florida's cost per case, per month is the lowest in the 
nation.\3\
---------------------------------------------------------------------------
    \3\ https://fns-prod.azureedge.net/sites/default/files/snap/FY16-
State-Activity-Report.pdf.
---------------------------------------------------------------------------
    Additionally, most states' systems are already set up to handle the 
ABAWD work requirement, since it has been a part of food stamp law 
since 1996, and until the mid-2000s, most states were enforcing the 
requirement in at least part of their state. This means that most 
states simply have to turn that feature back on within their computer 
systems. While there are significant Employment and Training funds 
available to states to assist ABAWDs with work activities or training, 
there is no required number of ``slots'' that must be funded or 
staffed, so states have flexibility in this area, ensuring that it is 
not a legitimate obstacle to re-instating the requirement.

    Question 3. Mr. Adolphsen, we know the Foundation for Government 
Accountability has written on these waiver abuses and the proposed 
rulemaking. Your written testimony stated this proposed rule would help 
address waiver abuse. Do you have additional recommendations to the 
Department's solution?
    Answer. The proposed rulemaking is essential to ensuring that 
waivers are no longer abused. However, the rulemaking leaves in place 
some key loopholes that could easily be exploited by states that are 
more focused on keeping people on food stamps than helping them get 
back to work. To ensure that this gaming of the system cannot take 
place, the rule should be adjusted in several key ways.
    The proposed rule will still allow for the combining of certain 
areas for the purposes of applying for a waiver. This could mean that 
some counties, for example, with low unemployment, can still be grouped 
with higher-unemployment counties in a way that would result in both 
receiving a waiver, even though the lower unemployment county would not 
qualify on its own. The rule should require each individual area, 
county, or city, to qualify on its own without combination with a 
separate area.
    Making sure that areas that are within commuting distance of 
available jobs do not qualify for a waiver could strengthen the rule 
further. The USDA has defined certain ``commuting zones'' that more 
accurately capture the commuting patterns within the country. The rule 
should require that, even if a county qualifies itself for a waiver, if 
the area is within commuting distance to available jobs in another 
county, it should not receive a waiver.
    The rule could also be strengthened by using a ten percent 
unemployment rate as the floor for any waiver. Congress very clearly 
set ten percent unemployment as the chosen mark for approval of the 
waivers in the original work requirement law, and it would return the 
waivers to their original intent of being employed only in truly 
economically disadvantaged areas.
    There are three final changes that could improve the rule:

   First, current regulations exempt from the work requirement 
        ABAWDs who reside in a household with a sibling under 18. The 
        intent of the law is to exempt caretakers or parents of 
        children under 18, not anyone who resides in the household. 
        This loophole should be closed in this rule.

   Second, regulations currently exempt 50 year olds from the 
        work requirement. The law exempts those individuals ``over 50 
        years old'' but regulations exempt anyone over 49 years old. 
        The Office of Inspector General has recommended this be fixed, 
        and it should be changed in this rule.

   Third, regulation includes language that ``requires'' the 
        Secretary of USDA to approve ABAWD work requirement waivers 
        when certain criteria are met. However, the law states only 
        that the Secretary ``may'' approve waivers in these cases. The 
        rule should return this discretion to the Secretary as Congress 
        intended.

    Question 4. The 2018 Farm Bill allows states to exempt up to 12% of 
their ABAWDs, per month, for myriad reasons. How can states more 
effectively use this allowance to capture the appropriate exceptions, 
as opposed to waiving larger areas and more people from the work 
requirement?
    Answer. The use of this particular exemption is often ignored in 
the discussion about the larger geographic waivers. Although some 
proponents of the current waiver structure treat the new rule as if it 
will eliminate waivers altogether, it will not. The rule would simply 
move the standards for qualifying back towards Congressional intent of 
approving them only in economically-disadvantaged areas.
    Further, even taken to the extreme and assuming there are No 
geographic waivers, states still have the option to exempt 12 percent 
of their ABAWD caseload. This presents states with a much more targeted 
opportunity to exempt individuals from the work requirement when there 
is a true need to do so. A prime advantage of this for states is that 
there is no requirement to complete the administrative process or 
management of a geographic waiver. The state can simply deploy these 
exemptions as they deem necessary.
    A good example of how these 12 percent exemptions might be used is 
in the case of major and sudden job loss in the state. For example, if 
a large employer decides to suddenly leave town, the state may use 
these exemptions for the specific people impacted by layoffs. This 
allows for rapid response and individual targeting, instead of the 
heavy handed and administratively burdensome geographic waiver 
approach.

    Question 5. The House-passed farm bill included language that would 
have mandated the collection of integral data related to SNAP 
households. Data collection had long been a focus of the hearing 
series, with many bipartisan witnesses stating that there is a reality 
in SNAP: limited data. This data collection would have debunked claims 
about recipient households and would have informed future policy based 
on facts rather than (tiny) sample-based assumptions. As an example, 
most assumptions are based on a survey of \1/4\ of one percent of SNAP 
households (52,000 households of 20.8 million). Per The Center on 
Budget and Policy Priorities, this provision was seen as one that 
``unnecessarily puts personal information for tens of millions of 
Americans at risk.'' Mr. Adolphsen, did FGA support this data 
collection? Would it have mitigated the concerns that there is no data 
on ABAWDs?
    Answer. FGA does support the collection of better data to ensure 
proper evaluation and management of the program. For example, several 
witnesses at the recent hearing on ABAWDs were relying heavily on 
extrapolations of survey data to promote their viewpoint. Our research 
has tracked the actual experiences of more than 500,000 individuals, 
following their progress from welfare to work, and looking at each 
person's actual increase in income.
    While there could be better data collection, the assertion put 
forth that we ``do not know who these ABAWDs are'' is also not entirely 
accurate. We have enough data available to know that ABAWDs are mostly 
male, mostly young, and most are not working at all. We know for sure 
that they are not part-time or full-time students, and we know that 
they are not working in the seven million available jobs. We know this 
in particular because if an ABAWD works even in a minimum wage job 
full-time, they will be out of poverty and off of welfare. Improved 
data will help all interested parties better analyze this important 
program.
Response from Lisa Hamler-Fugitt, Executive Director, Ohio Association 
        of Foodbanks
    Dear Honorable Dusty Johnson, Ranking Minority Member:

    Thank you for your time and interest in the Ohio Association of 
Foodbanks SNAP Work Experience Program that serves only work-mandated 
recipients who are considered unemployed and underemployed Able-Bodied 
Adults without Dependents in Franklin County, Ohio. It is my pleasure 
to provide you with our responses to the questions I received from 
Jennifer Yezak, via email on April 17, 2019.
    Question 1. Ms. Hamler-Fugitt, your written testimony mentions the 
myriad ``self-reported'' issues individuals categorized as ``able-
bodied'' disclose. At what point in this assessment do you engage with 
medical service providers to either support or dispute such claims? 
What is the process once you engage with those medical service 
providers?
    Answer. When a SNAP work-mandated recipient states during their 
assessment that they have a disability and/or limitation, and the 
County SNAP Caseworker did not document the disability, the 
Association's Work Assessment Specialist directs the SNAP recipient to 
go back to their County SNAP Caseworker to secure the ``Franklin County 
ABAWD Employability Form'' (attached). The SNAP recipient is then 
required to locate a doctor or medical professional to complete the 
form on their behalf.
    The Association's Work Assessment Specialist encourages the SNAP 
recipient to apply for Medicaid, if they are without health care 
coverage, and provides them with a list of free medical clinics if they 
don't currently have a doctor. It is the responsibility of the SNAP 
recipient to locate and secure the medical service provider that will 
assess and document their medical condition and who is willing to 
complete the form, documenting the recipient's inability to work. It is 
becoming more difficult to locate medical professionals who are willing 
to complete these assessments and complete the paper work for free. We 
have had a number of SNAP recipients report that the local medical 
providers are charging $20 to complete the form in addition to the cost 
of an office visit.
    These costs are prohibitive for an unemployed SNAP recipient that 
has no income and is not covered by health insurance.

    Question 2. Ms. Hamler-Fugitt, your testimony presents a variety of 
circumstances that should void ABAWDs from work, and four reasons to 
withdraw the proposed rule. USDA made it a point--multiple times--to 
ask for additional feedback in both the ANPRM and the NPRM. Do you have 
suggestions that can be handled via rulemaking as to how to hold states 
accountable, engage ABAWDs in the labor force, and work to move 
individuals to personal autonomy?
    Answer. USDA does not mandate that states provide SNAP work-
mandated recipients with education, training, Work Force Investment 
Act, SNAP Employment and Training or any workforce supports. The 
Franklin County Department of Job and Family Services (FC[D]JFS) 
contracts with the Association to provide these services to SNAP work-
mandated recipients who reside in Franklin County, Ohio.
    The Ohio Association of Foodbanks supports work by providing 
assignments and placements at over 50 nonprofit and faith-based 
organizations to help recipients meet their work requirements when they 
are unable to secure paid employment or a slot to participate in a 
qualifying education and training program.
    In our experience there is significant evidence that restricting 
time limit waivers increases the direct cost to the state and county 
agencies, places an additional administrative burden on the SNAP 
recipients, and likely decreases the number of families who are able to 
get the food assistance they need. In fact, without additional funding 
from Congress, this rule is a significant unfunded mandate on state and 
local governments. Programs with lower administrative burdens reach 
more eligible people than programs that make participants jump through 
hoops to get help meeting their basic needs.
    The proposed rule change would make it harder for states to obtain 
and implement area waivers by dropping statewide waivers except when a 
state triggers extended benefits under unemployment insurance. It would 
unduly limit the economic factors considered in assessing an area's 
eligibility for a waiver (e.g., by no longer allowing employment to 
population ratios that demonstrate economic weakness to qualify areas 
for waivers). It would undermine efficient state implementation of area 
waivers by limiting their duration to 12 months and delaying their 
start dates until after USDA processes the request. In addition, the 
proposed rule would remove states' ability to use exemptions 
accumulated prior to the rule's implementation as well as limit the 
time states have to use exemptions they receive in the future.
    USDA provides little analysis to explain its conclusions about the 
impacts the changes would have on individuals and population groups nor 
of realistic plans to avert harm from those changes. Instead it merely 
asserts its expectation that \2/3\ of those individuals made newly 
subject to the time limit ``would not meet the requirements for failure 
to engage meaningfully in work or work training.'' Moreover, while USDA 
concedes that the proposed changes ``have the potential for disparately 
impacting certain protected groups due to factors affecting rates of 
employment of these groups, it finds that implementation of mitigation 
strategies and monitoring by the Civil Rights Division of FNS will 
lessen these impacts.'' But no explanation of the mitigation strategies 
and monitoring is provided, so there is no opportunity for us to 
comment on whether the acknowledged disparate impact will in fact be 
mitigated.
    The time limit proposal goes around Ohio's Governor and lawmakers 
and the rest of Congress, which just concluded a review and 
reauthorization of SNAP in the 2018 Farm Bill and did not make the 
changes proposed. The rules governing areas' eligibility for waivers 
and individual exemptions have been in place for nearly 20 years. In 
that time, the waiver rules have proven to be reasonable, transparent, 
and manageable for states to operationalize.
    We strongly oppose the proposed rule that would expose even more 
Ohioans to the SNAP food cutoff policy, increase hunger and food 
insecurity and harm our state and community.

    Question 3. Per your website, Feeding America's mission is to feed 
America's hungry through a nationwide network of food banks and to 
engage the United States in the fight to end hunger. Feeding America 
also ``works hard to protect and promote government programs that help 
families facing hunger.'' SNAP is one of those programs, which also 
includes an emphasis on work and self-sufficiency. Feeding America has 
over $2.8 billion in revenues, distributes over $2.7B of that in grants 
and other ``assistance;'' what is Feeding America doing to promote 
work, personal goals, self-sufficiency, and a lifetime of independence?
    Answer. The Ohio Association of Foodbanks a separate 501(c)(3) and 
is a partner state Association of Feeding America. The Association is 
unable to respond to this question and we have referred it on to 
Feeding America.
    For additional information about the Ohio Association of Foodbanks, 
the following link is to our 2018 Annual Report, which provides an in-
depth account of Ohio's hunger relief network and the programs and 
partners that drive our mission to provide food and other resources to 
people in need: http://ohiofoodbanks.org/docs/publications/
SFY2018_annual_report.pdf.*
---------------------------------------------------------------------------
    * The document entitled, Ohio Association of Foodbanks: Annual 
Report 2018, is retained in Committee file.

    Question 4. Feeding America's website states that over 46 million 
people benefit from their services. That's over 14% of our nation. We 
have spent decades and trillions of dollars fighting poverty. It is 
obvious that some, if not all, of the over 80 social welfare programs 
are not working. What, in your opinion, has worked? What is not 
working? What are the efficiencies that can be most easily achieved to 
improve the program in the near term?
    Answer. The Ohio Association of Foodbanks a separate 501(c)(3) and 
is a partner state Association of Feeding America. Our efforts in Ohio 
fed one in six hunger Ohioans last year.
    I appreciate the question of whether there are ways to improve 
efficiencies in government programs. It is, and has always been, 
crucial that each tax dollar is spent wisely and prudently. While I 
have thoughts to share, my responses only focus on Federal nutrition 
programs.
    Programs with overlapping purposes or services do not necessarily 
mean that families are receiving more help than they need. Too often, 
the programs and supports available to families and seniors do not 
reach them or fail to address the enormous problems so many families 
and individuals face in this economy. This is typically because the 
programs are difficult to access due to confusing rules and 
requirements and because many key services, such as housing, child 
care, and job training, have fixed funding and cannot respond to 
increased need, particularly during a weak economy. When key programs 
do reach Ohioans, they are a powerful weapon against poverty and 
hardship. We must do more to improve how programs and agencies work 
together to ensure that these programs are accessible to the families 
and individuals they are intended to serve.
    However, there are some real-world realities that I want to 
highlight. For one, it is crucial to recognize that program overlap 
does not always mean duplication. People who struggle to afford an 
adequate diet may be better served by different types of programs and 
rarely are the value of benefits provided by any one given program 
sufficient to support the food needs of an individual or family.
    For example, SNAP works, yet the benefits fall far short of what a 
family requires and do not last the whole month. Thus, families that 
receive SNAP may also turn to food banks to fill in the gaps. A recent 
study found that in Ohio, low-income families continued to be food-
insecure despite accessing several food assistance programs. For 
example, 42% of the households standing in our food lines receiving 
SNAP benefits in the study were at risk of hunger and another 42% had 
cut back on the number or size of their meals. Additionally, among 
households with children ages 0-3 years, 58% participate in the Special 
Supplemental Nutrition Program for Women, Infants, and Children (WIC). 
Among households with school-age children, 63% and 55%, respectively, 
participate in the Federal school lunch and school breakfast programs 
and 14% participate in the summer food program.
    These benefits are provided to families that ask for help and these 
families have to exert a significant amount of effort in order to 
access and maintain these benefits. So, it is crucial that we recognize 
that the food assistance programs that we already have do not provide 
enough to meet all the nutritional needs of families.
    A simple recommendation is that Congress increase the SNAP benefit 
to the low-cost and moderate food program. In addition, automatically 
enroll low-income seniors, persons with disabilities, and families with 
children into the SNAP program.
    The fact that food insecurity (or risk of hunger) remains at a high 
level confirms that too few, not too many, resources are being made 
available for the families that need them.
    That said, I recommend that Congress enact policies that simplify 
and streamline eligibility rules so that families who participate in 
one program can be easily enrolled into all the other low-income 
programs for which they are eligible. For example, states and school 
districts have been working to cut red tape and streamline enrollment 
into the school lunch program by automatically enrolling children from 
families receiving SNAP and TANF through a process known as direct 
certification. Children in households receiving SNAP benefits are 
eligible for free school meals and school districts are required to 
work with the SNAP administrators to enroll them automatically, using 
the direct certification process noted above. Parents who have already 
completed a lengthy and detailed SNAP application should not have to 
complete another application; schools should not have to process 
unnecessary paperwork.
    This same method of simplifying and streamlining eligibility should 
apply to low-income seniors on Social Security who are receiving 
assistance and extra help with Medicaid part D. A hungry senior is not 
a healthy senior and I recommend that Congress enact policies that 
would allow states to auto-enroll seniors on low, fixed incomes into 
the SNAP and Commodity Supplemental food Programs, without requiring 
these seniors to complete multiple applications for critical food 
assistance.
    Recommendations:

   Congress should facilitate the ability for states to 
        integrate streamlined enrollment of eligible people across 
        several key programs: SSI/SSDI, Medicaid, Medicare and Domestic 
        Food Programs. This could be supported by allowing for data 
        sharing and interconnected data systems and providing the 
        Social Security Administration with the ability and mandate to 
        directly enroll low-income seniors and person with disabilities 
        in all Federal nutrition programs for which they are eligible.

   Recipients of Unemployment Compensation should be notified 
        about Federal employment and training programs that are 
        available to them for additional education to enhance their 
        skills.

   The United States Department of Agriculture should require 
        states to provide a comprehensive plan for how they will ensure 
        that eligible individuals will be enrolled in all domestic food 
        programs rather than individual, program-specific efforts. In 
        support of that effort, USDA could provide information to 
        states and localities about how best to cross-leverage their 
        outreach efforts.

   The Federal Government should measure states' success with 
        enrolling eligible individuals in a core package of programs, 
        such as SNAP, Medicaid, and school meals, rather than assess 
        participation by individual programs. Agencies could establish 
        a national standard for multi-benefit application assistance 
        programs and create incentives and funding to maximize 
        participation by eligible low-income families and individuals 
        in income-enhancing programs. States and local governments that 
        create seamless enrollment systems to connect those most in 
        need to available supports should be recognized and rewarded. 
        Their best practices could be promoted in other locations.

   Expand efforts and resolve to strengthen and create 
        interdepartmental coordination, universal application, single 
        agency consolidation and facilitate streamlining of public 
        benefits, eligibility and applications processes. These efforts 
        will free up administrative funding that can be directed to 
        increasing benefits levels and support expansion of eligibility 
        standards.

    Question 5. USDA provided Regional Directors with a memorandum on 
November 19, 2015 that explicitly speaks to how states can better 
assess an individual's fitness for work ``methodically and 
comprehensively.'' The memorandum goes on to say that Federal rules 
(CFR 273.24) allow states this flexibility to prevent placing 
unnecessary burden on individuals who are clearly unfit for employment. 
Ms. Hamler-Fugitt, where do you see the breakdown in this process?
    Answer. CFR 273.24 pertains to General eligibility guidelines to 
states, and in my experience, states err on the side of more rules, not 
less when defining their policies. In addition, there is no 
Congressional oversight, no Federal funding, no administrative 
oversight, no national standards and no penalties on states that refuse 
and fail to comply with this Federal rule. Might I recommend that 
Congress request a GAO report be completed on how states have or have 
not implemented CFR 273.24. Another Congressional action that may be 
warranted is a review and report by state on the number of work, 
training, and work fair slots and programs are operating in each county 
in every state. This would provide Congress with baseline information 
about the capacity or lack thereof to provide these program services to 
SNAP recipients that are subject to the time limits.

    In closing, food insecurity remains at very high levels, which 
confirms that too few, not too many resources are being made available 
to struggling families who desperately need them. SNAP is one of the 
MOST efficient ``welfare'' programs out there. Tens of millions of the 
people served in emergency hunger relief are not eligible for ``social 
welfare programs'' and not considered to be ``living in poverty'' under 
Federal guidelines, yet nonetheless they can't afford to put food on 
the table. These are merely nutrition programs. SNAP should continue to 
exist because proper nutrition is critical to the healthy livelihoods 
of every person. Other programs should address employability 
separately. SNAP is not a jobs program. It is not funded as a jobs 
program.
    Thank you again for the opportunity to testify and I would be 
pleased to respond to any additional questions that you might have. I 
can be reached either my email at Redacted or my telephone at Redacted.
            Best regards,
            
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
            
Lisa Hamler-Fugitt,
Executive Director.
                               attachment
                               
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Response from Jay C. Shambaugh, Ph.D., Director, The Hamilton Project, 
        and Senior Fellow, Economic Studies, Brookings Institution; 
        Professor of Economics, George Washington University
    Dear Representative Johnson,

    Thank you for the questions regarding my testimony. I appreciate 
your interest in this important topic and welcome the opportunity to 
share thoughts on these issues. I answer the questions below.
            Best regards,

Jay Shambaugh.

    Question 1. Mr. Shambaugh, if the Administration were loosening the 
ability of states to receive waivers, or relaxing the data criteria 
that should be used, would you have concerns about whether it is within 
the purview of the Administration to make changes to ABAWD waiver 
regulations? Is there any evidence that this Administration's 
interpretation of the phrasing of the Food and Nutrition Act of 2008 of 
``does not have a sufficient number of jobs to provide employment for 
the individuals'' is inappropriate and not within an allowable range of 
possible regulatory interpretations?
    Answer. Given that SNAP is a highly effective automatic stabilizer, 
proposals that change the conditions by which economically distressed 
places become eligible for work requirement waivers should be held to 
the highest evidentiary standards. The concerns expressed in my 
testimony did not relate to USDA's jurisdiction or purview to implement 
regulatory changes, as such matters span beyond my research and 
academic expertise. What my testimony did, however, discuss in detail 
was the lack of evidence supporting the change. The U.S. Department of 
Agriculture's (USDA) Proposed Rule does not meet an evidentiary 
standard and would weaken SNAP's responsiveness to an economic downturn 
without increasing labor force participation rates. The USDA and its 
Regulatory Impact Analysis (RIA) fail to fully consider the costs and 
benefits of the proposed rule, including the costs and benefits under 
alternative economic conditions.
    The USDA and the RIA do not consider the benefits to program 
participation for individuals nor SNAP's role as an automatic 
stabilizer when weighing proposed changes. The rule is likely to push a 
considerable number of current beneficiaries who are either in the 
labor market or unable to work off the SNAP rolls while failing to 
expand for newly eligible participants at the onset of a recession. It 
does so absent evidence that labor force attachment among ABAWDs would 
increase as a result of this proposal even in a strong economy and 
without consideration to the costs both for individuals and the economy 
in any circumstance. Given the importance of SNAP in a recession, it 
seems impossible to change how waivers work without considering the 
effects of the policy change under a range of economic circumstances. 
The proposed rule and the RIA do not factor in such considerations.

    Question 2. Mr. Shambaugh, your written testimony states that the 
vast majority of ABAWDs are in the labor force. The recently released 
Characteristics of SNAP Households shows that just over 31% of ABAWDs 
have any countable earned income, meaning less than 32% work. Why the 
discrepancy? Even if the majority of ABAWDs work at some point, if the 
majority are not working while on SNAP than at any given point in time, 
it seems that the majority are not working, correct?
    Answer. Because states are not required to provide precise numbers 
on ABAWDs, the administrative data reports are also not precise. The 
Characteristics of SNAP Households Table 3.3 from which the 31 percent 
number is taken is for ``Adults age 18 to 49 without disabilities in 
childless households.'' ABAWDs are a subset of this larger group, as 
some in this group have been identified as not being an ABAWD by the 
eligibility caseworker or Quality Control reviewer. When President 
Trump's White House Council of Economic Advisors used the Quality 
Control data to look at the average monthly employment rate, which is 
lower than the rate of those in the labor force because it does not 
consider those seeking work, they found that 38 percent were working in 
a given month in 2017 (Economic Report of the President, March 2019, p. 
464).
    More importantly, these monthly snapshots do not capture the full 
picture of labor force engagement of those on SNAP. As our work has 
shown, the monthly snapshot provides an incomplete picture due to the 
volatility of the low wage labor market.
    In fact, when looking at an extended period, we have found that 75 
percent of 18-49 year old adults without dependents and without 
disability income who are SNAP beneficiaries are in the labor force at 
some point over a 2 year period. Using the same method with a slightly 
different sample (18 to 64 year olds as opposed to 18 to 49 year olds), 
The 2019 Economic Report of the President found that about 70 percent 
of those who were non-disabled working-age SNAP recipients in December 
2013 worked at some point from January 2013 to December 2014 (p. 463).
    As noted in our research, the majority of those not working in a 
given month who do work at other times are not working because of 
``work-related reasons'' such as problems finding hours, the loss of a 
job, closure of an employer, etc. These people are trying to work. They 
may not have income every month, but they are actively engaged in the 
labor force.

    Question 3. According to the American Enterprise Institute, many 
scholars and public policy professionals are moving away from using 
survey data to inform policy. AEI continues to say that these 
individuals cite a lack of reliability and the importance of comparing 
self-reported data to administrative data. Mr. Shambaugh, what do you 
envision as the best data to capture the information discussed at the 
hearing?
    Answer. There is value in both administrative data and surveys. 
Surveys are a critical tool to complement administrative data sources 
and are vital to providing valid evidence for the policy debate. In 
particular, surveys can provide evidence about why people are behaving 
as they are, what keeps them from working, what their health status is, 
among many other crucial pieces of information. As we have said, the 
data that we have used to analyze policy issues regarding work 
requirements have also been employed by the White House Council of 
Economic Advisors in its most recent Economic Report to the President.
    While data is a critical input, research design is more so. 
Research conducted this year by Jeehoon Han (University of Chicago) and 
Timothy Harris (Illinois State University for the W.E. Upjohn Institute 
for Employment Research) used administrative data, survey data, and a 
quasi-experimental research design to identify the effect of SNAP work 
requirement waivers on caseload levels and employment outcomes. Han 
finds no negative effect of place-based work requirement waivers on 
employment and no positive effect of the reimposition of work 
requirements on employment. These findings are consistent with work 
requirements having no effects on employment. Harris typically finds an 
extremely small (less than one percent) but statistically significant 
impact of work requirements on employment, but even the most generous 
specifications find the number incentivized to work is far smaller than 
those removed from the program. Crucially, they compare those facing 
work requirements to those who do not, giving a valid comparison of 
treatment and control groups.
    By contrast, we urge policymakers to be cautious in their review of 
published research using administrative data that does not employ a 
research design for causal inference. These studies, including one 
referenced repeatedly during the Subcommittee's April 3 hearing, do not 
identify treatment and control groups, incorrectly sample the study 
population, and do not account for factors that would otherwise explain 
their findings like work history before and while on the program. 
Because they fail to design their studies correctly, the administrative 
data employed cannot answer the research question asked.
    In [addition] to using administrative data and survey data in our 
research, The Hamilton Project has produced evidence in support of the 
Federal statistical agencies role in collecting data that supports 
economic growth and evidence-based policy. In a March 2017 report 
published jointly by the American Enterprise Institute and The Hamilton 
Project, the organizations supported government-collected data, 
including surveys. In particular, this report notes that research based 
on surveys is used not only by researchers, but by private businesses 
(see Chapter 1: Businesses).
    The report can be found at: http://www.aei.org/publication/in-
order-that-they-might-rest-their-arguments-on-facts-the-vital-role-of-
government-collected-data/.*
---------------------------------------------------------------------------
    * The document entitled, ``In Order That They Might Rest Their 
Arguments on Facts'': The Vital Role of Government-Collected Data, is 
retained in Committee file.

    Question 4. Mr. Shambaugh, as an economist you understand the 
difficulties in compiling longitudinal data on ABAWDs that would 
provide improved information on the status and situation that 
contribute to ABAWDs enrolling in and staying enrolled in SNAP? What 
can USDA do right now that would provide policymakers, states, and 
nonprofits serving the ABAWD population with even better information on 
what this group needs to receive the services they need and obtain 
employment if possible?
    Answer. I agree that in order to better understand the needs of the 
ABAWD population and to improve their labor market outcomes, 
longitudinal data that includes demographic, casework, and employment 
information on ABAWDs would provide improved descriptive information 
for oversight and policymaking. For example, identifying volatility as 
a key feature of the low-wage labor market and the longer-term outcomes 
for those who failed to meet a work requirement can only be done with 
longitudinal data.
    At present, using available information to better describe who 
ABAWDs are and addressing screening procedures to limit exposure to 
work requirements only to those who the law feels should face a work 
requirement are the best first steps. There should be improved 
screening mechanisms for disability and individual barriers to work. 
States have used individual exemptions to shield a variety of 
vulnerable populations from work requirements, such as victims of 
domestic abuse, veterans, and those who have aged out of the foster 
care system. This could be done more systematically with point-in-time 
information.

                                  [all]