Welfare Reform: Three States' Approaches Show Promise of Increasing Work
Participation (Chapter Report, 05/30/97, GAO/HEHS-97-80).

Pursuant to a congressional request, GAO reviewed states' experiences
under waivers with increasing the participation of welfare recipients in
work and work-related activities, focusing on: (1) the policies and
programs states initiated under waivers to increase participation in
work and work-related activities; and (2) whether states with statewide
waivers achieved participation rates comparable to those specified by
the Personal Responsibility and Work Opportunity Reconciliation Act.

GAO noted that: (1) the three states GAO reviewed, Massachusetts,
Michigan, and Utah, initiated policy changes designed to increase the
proportion of welfare recipients participating in activities intended to
move them toward self-sufficiency; (2) the three states changed their
requirements regarding who must participate and what constitutes
participation; (3) Michigan and Utah eliminated prior exemptions,
thereby requiring all recipients to participate in some activity, while
Massachusetts exempted a substantial portion of the caseload, mandating
only parents with school-age children to participate in a work-related
activity; (4) Utah permitted a broad range of activities to count as
participation, tailoring activity to client circumstances, while
Massachusetts and Michigan limited the activities for recipients
required to participate to job search, followed by a job or, in the case
of Massachusetts, by community service if a job is not found after 60
days; (5) all three states reported being able to provide the services
specified in their plans to support clients in their activities; (6)
they accomplished this in different ways: (a) by increasing available
services; (b) by limiting participation requirements to fit available
resources; or (c) by providing lower cost services; (7) the three states
had a common strategy of strengthening sanctions for noncompliance as a
tool to increase participation; (8) Massachusetts, Michigan, and Utah
were able to engage a substantial number of welfare recipients in work
and work-oriented activities; (9) most recipients who counted as
participating were working in unsubsidized jobs, which reflects the
emphasis placed on work in all three programs as well as changes to the
amount of income they can earn and continue to receive benefits; (10) on
the basis of GAO's analysis of state participation data, it is almost
certain that these three states will meet their all-families target
participation rate in the first year; but according to state officials,
future rates may prove more difficult to achieve; (11) furthermore,
Massachusetts and Michigan are concerned about their ability to meet the
higher two-parent families rates; (12) all three states are concerned
about meeting the future all-families rates, which are not only higher
but require more hours of participation; and (13) state officials
expressed concern that as employable recipients find jobs, the remainin*

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-97-80
     TITLE:  Welfare Reform: Three States' Approaches Show Promise of 
             Increasing Work Participation
      DATE:  05/30/97
   SUBJECT:  Welfare benefits
             Public assistance programs
             Welfare recipients
             Workfare
             State-administered programs
             Waivers
             Federal/state relations
             Statutory limitation
             Employment or training programs
IDENTIFIER:  Aid to Families with Dependent Children Program
             AFDC
             HHS Temporary Assistance for Needy Families Program
             Job Opportunities and Basic Skills Training Program
             Massachusetts 1995 Welfare Reform Program
             Michigan Family Independence Program
             Utah Single Parent Employment Demonstration Program
             Michigan Work First Program
             Massachusetts Full Employment Program
             JOBS Program
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


Cover
================================================================ COVER


Report to the Ranking Minority Member, Committee on Finance, U.S. 
Senate

May 1997

WELFARE REFORM - THREE STATES'
APPROACHES SHOW PROMISE OF
INCREASING WORK PARTICIPATION

GAO/HEHS-97-80

Welfare Work Participation Requirements

(106609)


Abbreviations
=============================================================== ABBREV

  AFDC - Aid to Families With Dependent Children
  AFDC-UP - Aid to Families With Dependent Children-Unemployed Parent
  FEP - Full Employment Program
  FIS - family independence specialist
  FSA - Family Support Act
  HHS - Department of Health and Human Services
  JOBS - Job Opportunities and Basic Skills Training
  SPED - Single Parent Employment Demonstration
  TANF - Temporary Assistance for Needy Families

Letter
=============================================================== LETTER


B-271515

May 30, 1997

The Honorable Daniel P.  Moynihan
Ranking Minority Member
Committee on Finance
United States Senate

Dear Senator Moynihan: 

This report, prepared at your request, reviews states' experiences
under waivers with increasing the participation of welfare recipients
in work and work-related activities.  The report examines the
policies and programs states initiated to increase participation in
such activities, determines what participation rates states have
achieved under their programs, and assesses whether states are likely
to meet the work participation rates specified in the new welfare
law. 

We are sending copies of this report to the Chairman, Committee on
Finance, U.S.  Senate; the Chairmen and Ranking Minority Members,
Committee on Ways and Means and its Subcommittee on Human Resources,
House of Representatives; the Secretary of Health and Human Services;
the Assistant Secretary for Children and Families; and other
interested parties.  We will also make copies available to others on
request. 

If you or your staff have any questions concerning this report,
please call me at 202-512-7215 or David P.  Bixler, Assistant
Director, at 202-512-7201.  Other GAO contacts and major contributors
to this report are listed in appendix II. 

Sincerely yours,

Mark V.  Nadel
Associate Director, Income Security Issues


EXECUTIVE SUMMARY
============================================================ Chapter 0


   PURPOSE
---------------------------------------------------------- Chapter 0:1

The Personal Responsibility and Work Opportunity Reconciliation Act
of 1996 (P.L.  104-193) ended the individual entitlement to welfare
benefits established by the Social Security Act in 1935.  Under the
new law, welfare benefits are time limited and recipients are
required to participate in work and work-related activities. 
Moreover, the new law requires states to have a minimum percentage of
their caseload participating in work or work-related activities to
avoid a financial penalty.  Even before passage of the act, many
states were experimenting with ways to increase participation in work
by reforming their welfare programs through waivers of Aid to
Families With Dependent Children (AFDC) program rules, which allowed
them to try innovative approaches beyond the rules. 

To get an early indication of likely state progress in meeting work
participation requirements as well as issues facing welfare
administrators under the new welfare law, the Ranking Minority Member
of the Senate Committee on Finance requested that GAO review (1) the
policies and programs states initiated under waivers to increase
participation in work and work-related activities and (2) whether
states with statewide waivers achieved participation rates comparable
to those specified by the new federal law. 


   BACKGROUND
---------------------------------------------------------- Chapter 0:2

AFDC provided assistance to economically needy families with children
who lacked support from one or both of their parents because of
death, absence, or incapacity.  Recipients typically were not
required to prepare for, or look for, work and needed only to
establish and maintain income eligibility to receive benefits.  The
federal and state governments shared the costs of AFDC.  The states
administered the program, and the Department of Health and Human
Services (HHS) had oversight responsibility. 

As welfare caseloads grew and more women with young children entered
the labor force, a major theme of welfare reform debates through the
years was the need to increase the employability of welfare
recipients to enable them to work their way off assistance.  In the
1970s, the Congress changed the AFDC program to require certain
recipients to participate in activities aimed at increasing their
employability.  In 1988, as part of the Family Support Act (FSA), it
expanded these requirements by creating the Job Opportunities and
Basic Skills Training (JOBS) program to provide education, training,
and employment services to welfare recipients and to make AFDC a
transitional path to self-sufficiency.  In an attempt to focus their
welfare programs more on work and increase participation further,
many states more recently requested waivers of federal law to reform
their programs.  As of June 1996, HHS had granted 33 states waiver
authority to test policies intended to increase the number of welfare
recipients participating in work and work-related activities. 

The new welfare reform law completely transforms AFDC by replacing
the individual entitlement to benefits with Temporary Assistance for
Needy Families (TANF) block grants to the states.  In addition, the
law includes several provisions to ensure cash assistance is
transitional and that those receiving aid work, or prepare for work,
in exchange for benefits.  While states have a great deal of
discretion over the design of their programs, the new law establishes
a 5-year lifetime limit to assistance and requires recipients to work
after 2 years of benefit receipt.  Furthermore, the law requires
states to achieve specified levels of recipient participation in work
activities to avoid financial penalty.  In fiscal year 1997, 25
percent of a state's entire caseload--the all-families rate-- must be
participating in work and work-related activities.  The rate
increases by 5 percent a year to 50 percent by fiscal year 2002.  A
separate and much higher minimum work participation rate is specified
for two-parent families:  75 percent in fiscal year 1997 and 90
percent in fiscal year 1999.  Each state's minimum participation
rates are reduced by an amount equal to the percentage that the
state's welfare caseload has declined since fiscal year 1995. 

GAO selected Massachusetts, Michigan, and Utah to review.  These
states had statewide waiver programs with work requirements that were
sufficiently mature to allow for analysis of participation data. 


   RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3

The three states GAO reviewed--Massachusetts, Michigan, and
Utah--initiated policy changes designed to increase the proportion of
welfare recipients participating in activities intended to move them
toward self-sufficiency.  These states increased the percentage of
those participating and, at least in the first year, are almost
certain to meet the all-families participation rate specified by the
new federal welfare law. 

The three states changed their requirements regarding who must
participate and what constitutes participation.  Michigan and Utah
eliminated prior exemptions, thereby requiring all recipients to
participate in some activity, while Massachusetts exempted a
substantial portion of the caseload, mandating only parents with
school-age children to participate in a work-related activity.  Utah
permitted a broad range of activities to count as participation,
tailoring activity to client circumstances, while Massachusetts and
Michigan limited the activities for recipients required to
participate to job search, followed by a job or, in the case of
Massachusetts, by community service if a job is not found after 60
days.  All three states reported being able to provide the services
specified in their plans to support clients in their activities. 
They accomplished this in different ways:  by increasing available
services, by limiting participation requirements to fit the available
resources, or by providing lower cost services.  The three states had
a common strategy of strengthening sanctions for noncompliance as a
tool to increase participation. 

Massachusetts, Michigan, and Utah were able to engage a substantial
number of welfare recipients in work and work-oriented activities. 
Most recipients who counted as participating were working in
unsubsidized jobs, which reflects the emphasis placed on work in all
three programs as well as changes to the amount of income they can
earn and continue to receive benefits.  Compared with prior
participation in JOBS program activities, the three states achieved
higher participation rates under their waiver programs. 

On the basis of GAO's analysis of state participation data, it is
almost certain that these three states will meet their all-families
target participation rate in the first year; but according to state
officials, future rates may prove more difficult to achieve. 
Furthermore, Massachusetts and Michigan are concerned about their
ability to meet the higher two-parent families rates; and all three
states are concerned about meeting the future all-families rates,
which not only are higher but require more hours of work
participation.  State officials expressed concern that as employable
recipients find jobs, the remaining caseload will consist of
individuals with substantial barriers to employment, making the
higher future target rates difficult to achieve. 


   PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4


      STATES TOOK DIFFERENT
      APPROACHES TO INCREASE
      PARTICIPATION
-------------------------------------------------------- Chapter 0:4.1

Under waivers, many states attempted to increase welfare recipients'
participation in work-related activities.  The three states GAO
reviewed--Massachusetts, Michigan, and Utah--took different
approaches to increasing the participation of welfare recipients in
work.  Michigan and Utah eliminated exemptions, expecting the entire
caseload to participate in some activity.  In contrast, Massachusetts
mandated that only parents of school-age children
participate--approximately 20 percent of its caseload.  Massachusetts
and Michigan emphasized work-focused activities, such as a short job
search/job readiness program or simple job search; while Utah allowed
a broad range of activities tailored to individual needs, including
education and training. 

To encourage employment, states used different strategies.  First,
they informed individuals early in the application process of their
obligation to participate in an activity that would lead to
employment.  Then they monitored and tracked client progress toward
that goal and offered needed supportive services, such as child care
and transportation assistance.  To further promote employment, the
states changed incentives in the AFDC program by allowing working
recipients to disregard more of their earned income in the
calculation of their benefits, thereby enabling them to earn more
before becoming ineligible.  To help recipients who earned their way
off welfare make the transition to self-sufficiency, some services
such as transitional Medicaid and child care were extended beyond the
12 months required by the JOBS program.  Utah diverted needy
job-ready individuals from the welfare rolls by offering them a
one-time payment to cover emergency expenses until they received a
paycheck.  Finally, these states used sanctions to enforce the
participation requirements by first reducing benefit amounts for
failure to participate in planned activities and, if the failure
persists, terminating benefits entirely. 

The three states' different approaches reflect the strategies that
they believe work to help welfare recipients in their states move
toward self-sufficiency; consequently, these states plan to continue
their programs, with minor changes, under the new law.  Michigan and
Utah, which attempted universal participation, are concerned about
the welfare recipients left behind after the job-ready clients become
employed and leave the rolls.  They believe this group, which in the
past generally was exempted or deferred from participation
requirements because of their multiple barriers to employment, will
require intensive services to be able to participate in the
activities that meet the restricted definition of work participation
under the new law. 

Through their waiver programs, Massachusetts, Michigan, and Utah were
able to increase the percentage of their AFDC recipients
participating in activities designed to move them toward
self-sufficiency.  Reflecting the underlying work focus of their
programs, the largest percentage of all participants in each of the
three states were engaged in unsubsidized employment.  Counting only
activities allowed by the JOBS program, GAO compared the
participation levels that these states achieved under their statewide
waiver programs in September/October 1996 to the levels that they
reported for their fiscal year 1995 JOBS program.  GAO's analysis
found that Massachusetts' participation increased from 19 to 36
percent of the adult caseload, Michigan's from 21 to 42 percent, and
Utah's from 42 to 57 percent. 


      STATES LIKELY TO ACHIEVE
      INITIAL ALL-FAMILIES
      PARTICIPATION RATE, BUT MAY
      BE CHALLENGED TO MEET OTHER
      RATES
-------------------------------------------------------- Chapter 0:4.2

GAO's analysis showed that all three states are almost certain to
meet the first year all-families TANF participation rate requirement. 
Taking credit for their caseload reductions since 1995, Massachusetts
and Michigan officials estimate that their actual target
participation rate for fiscal year 1997 will be about 13 percent,
while Utah officials believe their rate will be about 14 percent. 
Using the definition of work participation in the new law, all three
states were achieving participation rates higher than their estimated
target rates under their waiver programs.  GAO estimates that
Massachusetts was achieving an overall participation rate of 25.3
percent; Michigan, 32.6 percent; and Utah, 30.9 percent.  However,
officials in Massachusetts and Michigan reported they will have
difficulty meeting the separate two-parent families participation
rate, and officials in all three states expressed concern over their
ability to meet the higher future rates.  As those recipients who are
able become employed and move off the welfare rolls, officials are
concerned that the remaining caseload will consist of less employable
recipients, making it difficult to meet the higher future
participation rates.  Furthermore, since these programs were
implemented under favorable economic conditions, which many believe
contributed to the large caseload reductions, officials are concerned
that an economic downturn could limit future program success.  Also,
as states are left with a more difficult population to serve, they
are concerned about the cost and availability of services to meet
this population's multiple needs in moving toward work and
self-sufficiency. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5

GAO is making no recommendations in this report. 


   COMMENTS FROM HHS AND STATES
---------------------------------------------------------- Chapter 0:6

GAO received comments on a draft of this report from the three case
study states:  Massachusetts, Michigan, and Utah.  These states
provided technical clarifications about their programs and additional
interpretations about certain data, which were incorporated as
appropriate.  Comments were requested from HHS, but none were
received. 


INTRODUCTION
============================================================ Chapter 1

From fiscal years 1989 through 1994, average monthly Aid to Families
With Dependent Children (AFDC) caseloads rose sharply, from 3.8
million families to 5 million.  Caseloads began to decline in 1994
and in fiscal year 1996, the federal and state governments spent over
$20 billion on AFDC benefit payments for about 4.6 million families. 
Concerned about the growth in the number of recipients and believing
that the permanent guarantee of benefits contributed to welfare
dependency, the Congress fundamentally altered the nature of welfare
when it passed the Personal Responsibility and Work Opportunity
Reconciliation Act of 1996 (P.L.  104-193).  This major welfare
reform law ended the individual entitlement to assistance under AFDC
and replaced it with capped block grants to the states called
Temporary Assistance for Needy Families (TANF), making cash benefits
temporary and conditional upon work participation.  The new law
promotes self-sufficiency through employment and time limits,
imposing obligations on welfare recipients to engage in work-related
activities after 2 years of benefit receipt--or sooner if a state
determines the recipient is ready to work--and limiting the length of
time individuals may receive federal assistance to 5 years.  The law
specifies the types of activities that are considered work-related
and establishes minimum yearly requirements for the percentage of
each state's welfare caseload that must be participating in one of
these activities. 

Before enactment of the new law, many states were experimenting with
work requirements for welfare recipients through waivers of AFDC
program rules that the Department of Health and Human Services (HHS)
approved.  Section 1115 of the Social Security Act allows the
Secretary of HHS to waive provisions of the act for the purpose of
permitting states to innovate within the AFDC program.  As of June
1996, 33 states had received waivers to test policies aimed at
increasing the participation of welfare recipients in work and
work-related activities. 


   PAST LEGISLATIVE EFFORTS TO
   IMPOSE WORK REQUIREMENTS ON
   AFDC RECIPIENTS
---------------------------------------------------------- Chapter 1:1

AFDC provided assistance to economically needy families with children
who lack support from one or both of their parents because of death,
absence, or incapacity.  The federal and state governments shared the
costs of AFDC.  States had discretion over the amount they paid to
entitled individuals.  They administered the program, and HHS had
oversight responsibility. 

Since AFDC's beginning, a number of developments have generated
discontent with the program.  First, by the early 1990s, close to 90
percent of all AFDC families were headed by separated, divorced, or
never-married mothers, rather than the widows who constituted most of
the single-parent population when the law was passed in the 1930s. 
Second, employment rates for all women, including single parents and
women with very young children, have increased dramatically, making
it difficult to defend the equity of supporting poor mothers who are
unemployed.  Third, there was a rapid increase in the size of AFDC
caseloads, resulting in increased costs both to states and the
federal government.  Finally, concern that welfare fostered
dependency was fueled by research that identified a substantial
portion of the caseload who remain poor and receive assistance for
long periods of time. 

In the late 1960s, changes were made in AFDC eligibility requirements
so that certain groups of recipients were expected to take a job or
participate in activities designed to enhance their employability. 
In addition, attempts were made to encourage work by excluding a
portion of earnings from the benefit calculation--a provision known
as an earned income disregard.\1 These efforts were not successful in
stemming the increases in AFDC caseloads during the 1970s; and, in
1981, the Congress passed legislation that restricted the amount of
money working welfare recipients could disregard from their benefit
calculations, resulting in many working welfare recipients losing
their eligibility for benefits.\2

In 1988, the Congress passed the Family Support Act (FSA), which
established the Job Opportunities and Basic Skills Training (JOBS)
program to provide education, training, and employment-related
activities to help recipients support their families and avoid
long-term welfare dependency.  However, approximately one-half of the
adults receiving AFDC were exempt from the JOBS program, usually
because they were caring for a young child.\3 FSA also expanded
eligibility for AFDC to certain two-parent families.\4

The JOBS program established explicit and gradually increasing
participation rates that states had to achieve to receive the full
federal match of JOBS funds.  These participation rates allowed the
federal government to promote uniform goals among states with
different program models.  Minimum participation requirements
conveyed to both welfare program administrators and recipients that
welfare receipt was conditioned on recipients' involvement in
employment-promoting activities or in work itself.  The JOBS
participation rates started at 7 percent of the nonexempt caseload in
fiscal year 1991 and rose to 20 percent in fiscal year 1995.  While
most states met the minimum participation rates specified under JOBS,
the number of AFDC recipients actually participating was only about
13 percent of the entire adult AFDC caseload in fiscal year 1994.  In
addition, our previous work showed that the JOBS program was not well
focused on employment, despite its objective of preparing recipients
for work.\5

Since passage of FSA, many states have been using waivers to change
their AFDC programs to increase work participation and decrease
welfare dependency.  Waivers were designed to allow states to
experiment with programs and try out policies that were not allowed
under current law.  States were required to evaluate these waiver
programs to determine whether they were successful and, therefore,
worth continuing.  In the early 1990s, waiver authority was granted
almost routinely to states interested in reforming their welfare
programs.  States that reformed their welfare programs to focus more
on work and work-related activities made changes in their JOBS
program participation requirements, their sanctioning policies, and
their earned income disregards. 


--------------------
\1 Initially, if a person on AFDC earned a dollar, their AFDC benefit
would be reduced by that amount, imposing a 100-percent marginal tax
rate on earnings.  This was viewed as a strong work disincentive,
since working would not necessarily result in increased income. 

\2 In a 1985 report, An Evaluation of the 1981 AFDC Changes:  Final
Report (GAO/PEMD-85-4, July 2, 1985), we estimated that these changes
decreased the national monthly AFDC caseload by 442,000 cases. 

\3 Individuals aged 16 through 59 were exempt from the JOBS program
if they were ill or incapacitated; caring for a child under age 3 (or
age 1 at state option); attending high school; in the second or third
trimester of pregnancy; caring for an ill or incapacitated family
member; employed 30 hours or more per week; residing in an area where
the program was not available; or providing care to a child under 6
and child care could not be guaranteed.  Teenagers with children
under age 3 and who were not in school were not exempt from JOBS. 

\4 Under the original Social Security Act, two-parent families were
ineligible for AFDC unless one parent was incapacitated.  However, in
1961, states were given the option to extend coverage to two-parent
families in which one parent was unemployed.  In 1988, FSA required
all states to offer an unemployed parents provision to their AFDC
programs.  States that did not have an AFDC-Unemployed Parents
(AFDC-UP) program before FSA were allowed to place a time limit on
benefits for these families.  As of February 1996, 12 states had time
limits on eligibility, including Utah. 

\5 Welfare to Work:  Current AFDC Program Not Sufficiently Focused on
Employment (GAO/HEHS-95-28, Dec.  19, 1994). 


   TANF WORK PARTICIPATION
   REQUIREMENTS
---------------------------------------------------------- Chapter 1:2

While the new welfare law gives states broad discretion over program
design, states are required to meet minimum performance standards for
work participation to avoid financial penalty.  States are required
to have 25 percent of their entire caseload participating in work and
work-related activities in fiscal year 1997, a rate that will
increase by 5 percent each succeeding year until it reaches 50
percent by fiscal year 2002.  A second, substantially higher,
participation rate applies to two-parent families:  75 percent in
fiscal years 1997 and 1998, and 90 percent in fiscal year 1999 and
thereafter.  The required number of hours of participation each week
for single heads of household is 20 hours in fiscal years 1997 and
1998, 25 hours in fiscal year 1999, and 30 hours in fiscal year 2000
and thereafter.  However, if the single parent has a child under the
age of 6, the parent can meet the work requirement by participating
20 hours per week.  For two-parent families, one parent must
participate 35 hours and, in cases where a family receives federally
funded child care, the second parent must participate 20 hours. 
States' target work participation rates are lowered if they have had
caseload reductions since fiscal year 1995. 

Activities that count toward participation under the new law are
oriented toward work rather than education and training.  The
emphasis on activities directly related to work reflects the
objective of the law to involve welfare recipients in
employment-promoting activities or in work itself, so that they can
leave welfare and become self-sufficient.  Although actual movement
off welfare through employment is not counted in the participation
rate, to the extent that employment results in caseload reductions,
states receive credit for these terminations through reductions in
the target participation rates they are required to achieve.  HHS can
penalize states for failing to meet the federal participation
requirement.  The penalty starts at 5 percent of the state's block
grant amount and rises gradually to 21 percent of the grant if the
state continues failing to satisfy the minimum participation
requirements. 

TANF funding is capped at $16.4 billion per year through fiscal year
2002.  The amount of TANF funds that a state receives is based on
recent federal expenditures for that state's AFDC program. 
Consequently, states that spent more for AFDC in the past will
receive more in TANF funds.  However, states will not automatically
receive additional funds should their caseloads increase.  States are
required to maintain at least 75 percent of their 1994 level of AFDC
spending on the new program.  States that fail to meet required work
participation rates must maintain 80 percent of their past spending. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:3

The Ranking Minority Member of the Senate Committee on Finance
requested that we review states' early experiences with increasing
the participation of welfare recipients in work and work-related
activities under waivers.  This review would provide an early
indication of states' progress in meeting work participation
requirements as well as issues facing welfare administrators under
the new federal law.  Specifically, we were asked to examine (1) the
policies and programs states initiated under waivers to increase
participation in work and work-related activities and (2) whether
three states with statewide waiver programs achieved participation
rates comparable with those specified by the new federal law. 

To determine what programs states initiated under waivers to increase
the work participation of welfare recipients, we analyzed the terms
and conditions of waivers, as of June 30, 1996, for all states that
HHS indicated had waiver provisions related to work participation. 
We also reviewed three states in depth that had implemented their
waiver programs statewide.  The three states selected--Massachusetts,
Michigan, and Utah--had programs in operation long enough to have
generated data for preliminary analysis.  In addition, these states
took distinctly different approaches toward increasing participation. 
To review the strategies these states used to implement their welfare
reform, we analyzed program reports and documents and met with state
and local welfare officials, state employment agency officials,
service providers, welfare advocates in Michigan, and program
evaluators in Utah and Michigan.\6 In addition, we met with HHS
officials responsible for monitoring these states' waiver programs as
well as welfare policy experts familiar with these programs.  We also
reviewed the published work of research organizations that studied
state welfare reform efforts.  To determine why some recipients do
not participate, we conducted reviews of case records in Michigan and
Utah.  To estimate JOBS and TANF participation rates for the waiver
programs in Massachusetts and Utah, we relied on aggregate data
submitted to us by the states; program administrators in these states
helped us interpret their data and calculate their participation
rates.  For Michigan, we worked with Abt Associates, the contract
evaluator for the state, using data supplied by the state to
calculate the participation rates. 

We conducted our work between March 1996 and March 1997 in accordance
with generally accepted government auditing standards. 


--------------------
\6 Massachusetts has yet to contract with an evaluator to do their
waiver evaluation. 


STATES INCREASED PARTICIPATION
RATES THROUGH DIFFERENT APPROACHES
============================================================ Chapter 2

When the Personal Responsibility and Work Opportunity Reconciliation
Act was enacted, Massachusetts, Michigan, and Utah already had
implemented statewide programs designed to increase the work
participation rates of welfare recipients.  All three states had
changed their programs to be more work-focused, supporting
participation in activities designed to help recipients move off the
welfare rolls and into self-sufficiency.  These states had changed
the rules regarding the categories of recipients required to
participate and the activities in which they were allowed to
participate; actively encouraged participation by providing
incentives and support services as well as information emphasizing
the obligation to work; and enforced participation by establishing
stricter sanctions for nonparticipation in agreed-upon activities. 
These approaches were also found in other states' waiver programs. 
Through their redesigned programs, Massachusetts, Michigan, and Utah
increased the proportion of their caseloads participating in
activities designed to promote self-sufficiency. 


   STATES CHANGED PARTICIPATION
   REQUIREMENTS
---------------------------------------------------------- Chapter 2:1

The three states we reviewed changed their policies regarding who
must participate, as defined under their respective programs, and the
activities in which they must participate, in two cases eliminating
exemptions and requiring all adult AFDC recipients to engage in some
activity as a condition of benefit receipt.  Thus, while the JOBS
program exempted many recipients from participation, both Michigan
and Utah eliminated exemptions while Massachusetts strictly mandated
a smaller segment of the caseload to participate than under the JOBS
program.  Moreover, while JOBS broadly defined the activities states
could require to meet the participation requirement, Utah further
expanded allowable activities, while Massachusetts and Michigan
restricted activities to those more closely linked to work. 
Eliminating exemptions can enable states to achieve greater levels of
participation by including a larger portion of the caseload. 
Restricting activities to those closely linked to work can enable
states to more quickly move job-ready recipients into employment. 


      STATES CHANGED POLICIES
      REGARDING WHO MUST
      PARTICIPATE
-------------------------------------------------------- Chapter 2:1.1

The three states reviewed took different approaches to required
participation.  Michigan and Utah expanded participation requirements
to the entire caseload as a way to increase participation, while
Massachusetts required only a fraction of its caseload to engage in a
work-related activity but strictly enforced the requirement in an
effort to increase participation of those recipients mandated to
participate.  Table 2.1 and the following pages summarize the
work-related policy changes made under waivers in Massachusetts,
Michigan, and Utah. 



                                    Table 2.1
                     
                        Work-Related Policy Changes Under
                             Waivers in Three States

                    Massachusetts       Michigan            Utah
------------------  ------------------  ------------------  --------------------
Program name        Welfare Reform '95  Family              Single Parent
                                        Independence        Employment
                                        Program             Demonstration
                                                            Program

Participation       60 days of job      4 weeks of job      Activity based on
requirements        search followed by  search followed by  individualized self-
                    a job, subsidized   unsubsidized        sufficiency plan
                    employment, or      employment
                    community service

Activities that     --Unsubsidized      --Unsubsidized      --Unsubsidized
count as            employment          employment          employment
participation
                    --Community         --Community         --Job search
                    service             service
                                                            --Training
                    --Subsidized        --Job search
                    employment                              --Education

                    --Job search for                        --Mental health and
                    60 days only                            substance abuse
                                                            counseling, among
                                                            others

Exemptions          --Disabled or       --VISTA volunteer   Children under 16
                    caring for a
                    disabled person     --Dependent under
                                        16
                    --Third trimester
                    of pregnancy        (Others are
                                        deferred for
                    --Child in          reasons such as
                    assistance unit is  having a child
                    under mandatory     under 3 months old
                    full-time school-   or caring for an
                    age, usually age    incapacitated
                    6, or child not in  household member,
                    assistance unit     or being ill,
                    but living with     under age 16, or
                    recipient is under  over age 65;
                    3 months old        however, they are
                                        required to
                                        participate in a
                                        social contract
                                        activity)

Time limit          2 years in any 5-   No change           No change
                    year period for
                    families with no
                    children under age
                    2

Sanction            Eliminate           25-percent grant    Extensive
process             conciliation        reduction for 1     conciliation/case
                                        year, followed by   staffing/home visit;
                    Parent's benefit    termination of the  $100 grant reduction
                    removed first,      entire family's     for 2 months, then
                    then entire         benefit             termination of the
                    family's benefit                        entire family's
                    eliminated                              benefit

Amount of earned    First $30 of        First $200 of       First $100 of
income excluded     earnings + 1/2 of   earnings + 20       earnings + 50
from benefit        remainder excluded  percent of          percent of remainder
calculation         with no time limit  remainder excluded  excluded with no
                    for families with   with no time limit  time limit
                    no children under
                    age 2; for
                    families with
                    children under age
                    2 there is no
                    change in
                    disregard amount
                    but it is no
                    longer time
                    limited

Eliminate 100-      Yes                 Yes                 Yes
hour rule

Transitional child  Eliminate           No change           Eliminate
care (TCC) and      requirement that                        requirement that
transitional        recipient have                          recipient have
Medicaid (TM)       received welfare                        received welfare
                    benefits for "3 of                      benefits for "3 of
                    last 6 months" to                       last 6 months" to be
                    be eligible for                         eligible for TCC and
                    TCC and TM                              TM

                                                            Extend TCC until
                                                            family income
                                                            exceeds sliding fee
                                                            schedule; extend TM
                                                            for 24 months if
                                                            still income-
                                                            eligible

Unique features     One-stop career                         Postemployment
                    centers                                 services

                    AFDC benefit of                         One-time diversion
                    families with no                        from AFDC
                    children under age
                    2 reduced by 2.75%
--------------------------------------------------------------------------------
Note:  VISTA stands for Volunteers in Service to America. 

Utah, which has a history of work requirements for welfare
recipients, achieved higher JOBS participation rates than most other
states between fiscal years 1991 and 1995 and was attempting, under
its waiver program, to place 95 percent of its caseload in activities
that ultimately would lead to self-sufficiency.  Under its Single
Parent Employment Demonstration (SPED) program, implemented in part
of the state in January 1993, Utah required all recipients to
participate in an activity, even those with significant barriers such
as drug abuse and mental health problems.  Since this expansion meant
working with recipients who had been exempted or deferred from
participation in the past, Utah was exploring new territory in terms
of what is required to enable these recipients to move toward
self-sufficiency.  The SPED program director believes that the
greatest savings come from moving recipients with multiple barriers
into employment because these are the recipients most likely to
become long-term welfare recipients.  Like many other states, Utah
has experienced substantial caseload declines over the last several
years--attributable, according to officials, in part to its good
economy and low unemployment rates--freeing up resources for
recipients who may need more than a strong economy and the more
traditional JOBS services to help them exit welfare through
employment.  Utah does excuse some recipients from participation for
reasons such as lack of transportation or child care, or the
existence of a family medical condition or emergency; however, the
program's self-sufficiency worker and the recipient are expected to
work on resolving problems so that the recipient can ultimately
participate.  If recipients do not participate, they are subject to
sanctions and the possibility of benefit termination. 

Under Michigan's Family Independence Program, established in March
1996, all adult welfare recipients have an obligation to engage in an
activity leading to self-support, referred to as their social
contract.\7 In contrast to this universal participation requirement,
approximately 35 percent of Michigan's recipients were exempted from
JOBS participation requirements in fiscal year 1994.\8 Although the
goal is for all recipients to work, they can be deferred from job
search if they are not immediately employable, for example, if they
have children under 3 months of age, are under 16 years of age, are
ill, or are caring for an incapacitated household member.  They are,
however, still obligated to participate in some activity defined in
their social contract.  Recipients required to participate in work
activities are subject to sanctions and benefit termination for
failure to participate. 

In contrast to Michigan and Utah, Massachusetts mandates only a small
segment of its caseload to participate in a work-related activity as
a condition of eligibility for benefits.  Under Welfare Reform '95,
implemented starting in November 1995, recipients with children under
the age of 2--that is, approximately 46 percent of the caseload on
October 31, 1996--are not required to work in order to keep their
benefits.  Recipients with children aged 2 to 6--approximately 34
percent of the caseload--are expected to work but are not subject to
sanctions for failure to participate in a work-related activity.\9
Only recipients with school-aged children--approximately 20 percent
of the caseload--are strictly mandated to work.  If they fail to
comply with the work requirements, they are subject to a progressive
set of sanctions for continued noncompliance after which the entire
benefit is eliminated. 

Like Michigan and Utah, other states with work requirement waivers
changed exemptions, requiring a larger portion of the caseload to
participate than in the past.  Eleven states, including Michigan and
Utah, extended the participation requirement to include parents with
children under 1 year of age (see table I.1).\10 In contrast, seven
states, including Massachusetts, increased the age for the youngest
child exemption, thereby decreasing the number of adults required to
participate because of the age of their youngest child.  The age for
the youngest child that exempts a parent from participation in states
with work requirement waivers ranges from under 12 weeks in Indiana,
Maryland, Michigan, Nebraska, Oregon, Utah, and Vermont to under 6
years in Ohio and New Hampshire.\11


--------------------
\7 Michigan began its reform effort in 1992 with the demonstration
program To Strengthen Michigan Families.  Additional waivers granted
in 1994 eliminated exemptions from work participation and instituted
Work First, a mandatory job search program. 

\8 During the period that To Strengthen Michigan Families was in
effect, the state was required to report JOBS participation rates to
HHS. 

\9 Recipients with children between the ages of 2 and 6 who volunteer
to participate in the work program and complete an employment
development plan are subject to sanctions for failure to participate. 
Furthermore, all recipients with no children under the age of 2 are
subject to a need and benefit payment standard that is 2.75 percent
lower than that used for families with children under the age of 2. 

\10 Under the JOBS program states were to exempt parents with
children under the age of 3 with an option to exempt only parents of
children under the age of 1. 

\11 Some states have a separate rule for children conceived while the
mother is receiving AFDC.  For example, in Massachusetts, such
children are not eligible for a welfare benefit and can exempt the
mother from a work participation requirement only while they are
under the age of 3 months. 


      STATES CHANGED POLICIES
      REGARDING ALLOWABLE
      ACTIVITIES
-------------------------------------------------------- Chapter 2:1.2

In addition to changing who must participate, states changed the
activities allowed to meet the participation requirement.  Their
choices regarding which activities to allow reflect their approach
toward moving welfare recipients into self-sufficiency as well as
their fiscal situations.  Some states chose "labor force attachment"
models emphasizing rapid entry into the labor market, believing the
market to be the best indicator of who is employable.  Other states
chose a human capital development approach that provides education
and training to support employment at higher wages.  In general,
waiver programs have reflected states' move toward more work-focused
models.  Massachusetts and Michigan implemented programs in which
mandated individuals participate in short-term, work-focused
activities.  In contrast, Utah tailored activities to the individual,
reflecting the notion that there is no one route to self-sufficiency
appropriate for all.  Consequently, in Utah, a client's employment
plan may consist of immediate job search or it may consist of
education or training. 

Recognizing that not all welfare recipients are immediately ready to
enter the labor market and find a job, Utah permits participation in
a range of activities aimed at getting previously exempted or
deferred families with multiple barriers to participate in some
activity.  Consequently, Utah has expanded allowable activities to
include activities such as substance abuse and mental health
counseling in addition to the more traditional employment, job
search, training (including on-the-job training), and education
(including 2 years of college).  Believing that recipients should be
able to compete in the labor market for unsubsidized jobs, Utah does
not allow subsidized employment to count as participation.  The
number of hours of participation required is individualized.  For
example, a client with a mental health problem might be required to
see a therapist for 1 hour a week to meet the participation
requirement, while another client could be expected to conduct a job
search for 20 hours a week.  The emphasis of the program is on making
progress toward self-sufficiency.  For a person with a mental health
problem, dealing with that particular condition is viewed as an
essential first step toward self-sufficiency. 

Michigan, like Utah, attempts to include the entire adult caseload in
some activity; however, unlike Utah, Michigan requires all applicants
and recipients, unless deferred or considered not immediately
employable, to participate in Work First, a short mandatory job
readiness/job search program.  After 4 weeks in the program, Work
First participants who do not find a job are reassessed and engage in
further job search activity, but officials expect most recipients to
succeed in finding a job within the allotted period.  Recipients who
are deferred from Work First are required to engage in productive
activities leading to their personal growth or to their community's
enhancement.  The activity in which a deferred recipient agrees to
participate is established by a social contract between the client
and the welfare agency; actual participation in the agreed-upon
activity is self-reported and not closely monitored.  A final group
of recipients, including teen parents, for whom immediate employment
is not appropriate, participate in Michigan's JOBS program, where
they can, for example, continue their educations.  Originally, under
its waiver, Michigan did not provide on-the-job training or
subsidized employment; however, effective January 1997, on-the-job
training will be available to two-parent families.  Officials
indicated that when attempts to find a job have failed, a client may
be placed in community service for a maximum of 6 months, at which
point the client is assessed for further appropriate
employment-related activities. 

Massachusetts requires all nonexempt recipients with no children
under school-age to conduct a 60-day job search.\12 Those who do not
find employment at the end of the job search period must participate
in community service or one of two subsidized work programs, the Full
Employment Program (FEP) or supported work.  The community service
must be 20 hours or more per week as an uncompensated volunteer in a
public, quasi-public, or nonprofit organization.  For example, a
recipient could be a volunteer in his or her child's school or at a
hospital.  Under FEP, the Massachusetts Department of Transitional
Assistance provides a direct subsidy to employers who hire welfare
recipients.  Designed to benefit individuals with no prior work
experience, the subsidy is intended to allow the employer to provide
on-the-job training to the recipient.  Under the supported work
program, the Department pays contractors to work with recipients and
place them in jobs.  The contractor supplies a "shadow" to
continually supervise the participant.\13 Some officials believe that
subsidized work or supported work may be more successful than
community service at moving welfare recipients into paid employment. 

A total of 13 states expanded the definition of "work activity" under
waivers to include activities that did not meet the participation
requirement under the JOBS program, such as family counseling or drug
and alcohol abuse counseling.  Twelve of the 13 states added
parenting or family skills training as an option (see table I.2).  In
addition, 19 states waived the JOBS program limits on job search,
extending the time they could require individuals to look for
work.\14 Under the new law, job search is limited to a total of 6
weeks, which cannot be continuous for more than 4 weeks.  However, if
the state's unemployment rate is 50 percent greater than the U.S. 
rate, the allowable job search period is extended to 12 weeks. 
Eighteen states also increased their use of subsidized work programs
or community services jobs (see table I.3). 


--------------------
\12 In Massachusetts, aliens who are not allowed to work in the
United States are not subject to the 60-day job search requirement;
they are assigned directly to community service sites. 

\13 The Manpower Demonstration Research Corporation's evaluation of
the Supported Work Demonstration found this approach increased the
employment of long-term welfare recipients.  Sites included in the
study were Atlanta, Chicago, Hartford, Newark, New York City,
Oakland, and Fond du Lac and Winnebago counties in Wisconsin.  See
Stanley H.  Masters and Rebecca Maynard, The Impact of Supported Work
on Long-Term Recipients of AFDC Benefits (New York:  Manpower
Demonstration Research Corporation, Feb.  1981). 

\14 Under the JOBS program, states could require up to 8 weeks of job
search for applicants and recipients each year, resulting in a
possible 16 weeks of required job search in the first year, with 8
weeks per year thereafter.  Additional job search could not be
required unless combined with some other activity designed to improve
a participant's prospects for employment, and in no case could a
state require a recipient to participate in more than 3 weeks of job
search before conducting an employability assessment. 


   STATES IMPLEMENTED POLICIES TO
   ENCOURAGE PARTICIPATION IN WORK
   AND WORK-RELATED ACTIVITIES
---------------------------------------------------------- Chapter 2:2

In addition to redefining what constitutes participation under their
waiver programs, states implemented policies designed to encourage
recipients to meet the new requirements and thereby increase actual
participation in work and work-related activities.  Strategies
included providing emergency assistance to divert applicants from
becoming welfare recipients, delivering a strong work message to
clients early during the application process, continually reinforcing
the work message through close monitoring to ensure clients attend
work-related activities, and allowing recipients to keep more of
their earnings once they become employed.  Additionally, the states
structured their programs to provide the support services clients
needed to work, such as child care and transportation. 


      STATES PREVENT WELFARE
      RECEIPT BY DIVERTING
      POTENTIAL PARTICIPANTS WITH
      ONE-TIME PAYMENTS
-------------------------------------------------------- Chapter 2:2.1

One way some states encourage work is to divert job-ready individuals
from becoming welfare recipients.  In Utah, individuals applying for
assistance who have immediate job prospects or other sources of
income are offered job placement assistance and a one-time financial
payment to meet immediate needs, such as car repair or rent.  The
purpose of the diversion is to provide the family with financial
security until a paycheck arrives, thereby curtailing the family's
need for welfare.  Utah also makes these diverted families eligible
for transitional child care and Medicaid, benefits that usually are
only available to individuals who have earned their way off welfare. 
According to Utah's program director, about 20 percent of eligible
applicants are diverted from ever going on assistance in this manner. 
Five other states divert individuals from AFDC with lump sum
payments.  In most cases, if individuals do apply for assistance
before a specified period, the amount of the diversion is taken into
account in determining the amount of the family's benefit. 
Massachusetts and Michigan did not have a diversion component in
their programs.  (See
table I.4.)


      CLIENTS INFORMED DURING
      APPLICATION PROCESS OF
      PARTICIPATION REQUIREMENT
-------------------------------------------------------- Chapter 2:2.2

To emphasize their obligation to participate in activities that will
help them become employed, clients in the three states we visited are
told early in the application process that they must engage in a
work-related activity as a condition of benefit receipt.  In
Michigan, an assistance payment worker explains at the time of
application that all recipients must participate, even clients who
typically had been exempted or deferred in the past.  For example,
recipients with a child under 3 months of age who are deferred from
Work First might be expected to attend parenting classes or medical
appointments to meet their obligation to participate under the social
contract.  The obligation to participate is again discussed at
orientation to Work First.  To address the chronic problem of welfare
recipients failing to show up for orientation, beginning in October
1996, Michigan made attendance a condition of eligibility for
benefits. 

In Utah, applicants for assistance meet with a self-sufficiency
worker before income eligibility is determined.  They are shown a
special slide presentation, developed by the state, that emphasizes
the client's responsibility to participate and take immediate steps
toward achieving self-sufficiency.  The self-sufficiency worker
establishes a self-sufficiency plan with the individual.  If an
individual is determined to be eligible for cash benefits, the
benefit is viewed as a support enabling the individual to meet the
participation requirements spelled out in the self-sufficiency plan. 
Caseworkers commented that early participation in some activity is
crucial for program success.  Their experience shows that clients not
assigned to an activity within 2 months will conclude that the new
program is the same as the old program and disregard the mandate to
participate. 

Our review of state waivers found that other states used similar
strategies to inform individuals of their obligation to participate
in a work-related activity.  In Indiana, for example, the state
developed a script for staff to use at the initial eligibility
meeting with applicants that focuses the meeting on job placement and
assistance rather than entry into the welfare system.\15 In addition,
states such as Florida, Indiana, Virginia, and Wisconsin require
clients to sign personal responsibility agreements that set forth the
expectation that the client will participate in activities leading to
self-sufficiency. 


--------------------
\15 Welfare Waivers Implementation:  States Work to Change Welfare
Culture, Community Involvement, and Service Delivery
(GAO/HEHS-96-105, July 2, 1996).  We reviewed the changes made under
waiver programs to implement time limits and work requirements in
Florida, Indiana, Virginia, and Wisconsin. 


      CASEWORKERS RESPONSIBLE FOR
      MONITORING CLIENT ACTIVITIES
-------------------------------------------------------- Chapter 2:2.3

Caseworkers in the states we visited also encourage work by
monitoring client participation in work-related activities.  Under
the previous AFDC program, most of the caseworkers' time was spent
determining eligibility, and they had little ongoing contact with
recipients.  Utah has a self-sufficiency worker position whose
primary functions have changed over the years as the state's program
changed.  Currently, these workers negotiate a self-sufficiency plan
and monitor client progress with frequent follow-up calls to ensure
that clients participate in the activities agreed to in the
self-sufficiency plan.  The self-sufficiency workers have smaller
caseloads than income maintenance workers to enable them to make the
additional client contacts.  Utah officials have found that many of
the recipients they currently serve require a lot of time and
attention because they are the more difficult cases left behind after
the employable recipients have found jobs.  To become
self-sufficient, these recipients may need assistance in addressing a
number of problems, including drug or alcohol abuse, physical or
mental disabilities, family responsibilities, domestic violence, and
learning disabilities.  In the past, many of these recipients would
have been exempted or deferred from participation.  Officials
commented that working with such clients requires more skill and
education, such as a degree in social work. 

The Michigan Family Independence Agency is in the process of
converting the assistance payments worker position to the position of
family independence specialist (FIS).  The FIS workers will have a
target caseload of 65, compared with the average caseload of 200 for
the traditional assistance payments worker.  If a client fails to
attend a Work First appointment, the workers are responsible for
learning why and attempting to remove any barriers to participation. 
While Michigan is making an effort to monitor recipients in Work
First, the state does not closely monitor recipients assigned to
social contract activities, who self-report their participation. 

In Massachusetts, the caseworker role has been expanded to include
responsibility for making referrals to employment and training
services.  Caseworkers contact recipients by telephone or scheduled
office visit after 30 days and again after 45 days into their 60-day
job search period to discuss the clients' progress and to remind them
of the approaching deadline for finding a job.  Although caseworkers
inform clients of the job search services available, it is the
client's responsibility to request needed services and assistance in
their job search.  On day 60, any client who does not have a job is
given 10 days in which to select or be assigned to a community
service position.  Participation in community service is tracked
through automatically generated attendance sheets that the client
fills out and the supervisor at the community service site signs. 

Other states also encourage clients' participation through close
monitoring to ensure they are making progress toward
self-sufficiency.  Florida created a case manager position to
coordinate and broker a comprehensive set of services clients might
need to become employed.  Escambia County in Florida expanded the
responsibility of its existing staff to include both eligibility and
case management functions.  Former eligibility workers now are
responsible for overseeing client activity and coordinating support
services.  Other states also report an increased monitoring of
clients' participation.  For example, in Fond du Lac County,
Wisconsin, caseworkers monitor certain clients' participation in
activities by receiving daily attendance reports from the local
technical college. 


      STATES ALLOW PARTICIPANTS TO
      KEEP MORE OF THEIR EARNINGS
-------------------------------------------------------- Chapter 2:2.4

In an attempt to encourage recipients to work, the three states we
reviewed disregard a larger portion of their earnings in determining
eligibility and benefits, thereby enabling more recipients to combine
work and welfare and, thus, to be better off financially than when
only receiving welfare or earnings.  These states not only changed
the amount of the earnings disregard, but also allowed it to remain
in effect indefinitely.  Utah allows recipients to disregard the
first $100 of income plus 50 percent of the remainder, Michigan
allows $200 plus 20 percent of the remainder, and Massachusetts
allows $30 plus 50 percent of the remainder for working recipients
with no children under the age of 2.  Families with children under
age 2 receive the traditional $30 and one-third disregard with no
time limit.  Twenty-four other states with work requirements also
changed the amount of earned income disregarded in determining
eligibility and benefit amounts (see table I.5). 


      STATES ELIMINATE 100-HOUR
      RULE TO INCREASE TWO-PARENT
      FAMILIES' PARTICIPATION IN
      WORK
-------------------------------------------------------- Chapter 2:2.5

Some states waived the special rules that affected AFDC eligibility
for two-parent families.  Federal regulations specified that poor
two-parent families in which neither adult was disabled could receive
aid only if the principal wage earner was working less than 100 hours
per month and had a significant work history.  This restriction was
believed to have a negative effect on recipients' work effort since
families would lose eligibility if one parent worked more than 100
hours a month, regardless of how much they earned.  By abolishing the
100-hour rule, more two-parent families with earnings qualify for
assistance.  These are poor families that are already working part
time that would have been ineligible for AFDC because they were
working more than 100 hours a month, rather than because they had
made enough money to be ineligible for benefits.  The three states we
visited eliminated the 100-hour rule so that poor two-parent families
could work more hours and still receive welfare until they earned
enough to no longer be eligible for benefits.  Of all states with
expanded work requirements under waiver, 25 states eliminated the
100-hour rule for two-parent families. 


      STATES USE DIFFERENT MEANS
      TO PROVIDE NEEDED SUPPORT
      SERVICES
-------------------------------------------------------- Chapter 2:2.6

To make the transition from welfare to work, many recipients need
services such as training or job search and job readiness activities,
child care, and transportation assistance.  The three states we
reviewed reported they were able to provide the services their
programs were designed to offer and did not need to exempt clients
because resources were unavailable.  They achieved this outcome in
different ways as discussed below. 


         TAILORING SERVICES TO FIT
         AVAILABLE RESOURCES
------------------------------------------------------ Chapter 2:2.6.1

Both Massachusetts and Michigan stressed immediate movement into work
after a period of job search.  This labor force attachment approach
allows states to serve larger numbers of clients because recipients
generally participate in job search for shorter periods of time than
they do in education and training programs.  Manpower Demonstration
Research Corporation results show that labor force attachment
programs increase the number of recipients who leave the rolls
because of employment, which can result in welfare savings because
recipients leave the rolls sooner than they would have without the
program.  Savings can also occur when recipients' earnings reduce
AFDC benefits.\16

Massachusetts officials said the state has not experienced shortages
of resources needed to meet its participation goals.  Costs are
contained by providing services during the 60-day job search to only
those recipients who request them.  Furthermore, only parents of
school-aged children are required to participate and they are
expected to work or perform community service during school hours
whenever possible. 


--------------------
\16 Stephen Freedman and Daniel Friedlander, The JOBS Evaluation: 
Early Findings on Program Impacts in Three Sites (New York:  Manpower
Demonstration Research Corporation, Sept.  1995).  Study sites were
Atlanta, Georgia; Grand Rapids, Michigan; and Riverside, California. 
According to the study, most of the savings appear to be associated
with recipients' having left the rolls sooner rather than with lower
grants for recipients who did not earn enough to become ineligible
for welfare. 


         EXPANDING AND ENHANCING
         SERVICES FOR MORE
         RECIPIENTS
------------------------------------------------------ Chapter 2:2.6.2

In contrast to Massachusetts and Michigan, Utah's approach is
individualized.  The state provides a broader range of services
tailored to individual need, allowing recipients to engage in
longer-term activities, such as postsecondary education and training,
for a maximum of 2 years.  State officials believe that the cost of
Utah's range of services is offset by the savings resulting from
moving more difficult-to-serve clients into employment since they
tend to become long-term recipients.  Officials indicated that the
savings resulting from the caseload decline allow Utah to provide
more intensive services to its remaining caseload.  To be able to
serve everyone, Utah began its program at only three sites and
gradually phased in the remaining areas of the state.  Furthermore,
funds were added to pay for work activity programs and supportive
services.  At one of the initial sites in the Salt Lake City area,
the state hired 13 new workers for the program.\17

Child care is an important support in the welfare-to-work transition,
and Utah has used savings from caseload declines to ensure that child
care resources are adequate.  According to an Urban Institute report,
the overall reduction in AFDC expenditures 10 months after the Utah
demonstration was implemented greatly exceeded the increase in child
care expenditures.  As a result, total expenditures on child care and
AFDC payments actually declined by 20 percent.\18

During the first year of its welfare reform, Massachusetts increased
spending on child care for welfare recipients and transitional
assistance for former recipients by 13 percent.  Michigan's child
care expenditures increased by 26 percent from fiscal year 1995 to
fiscal year 1996 with the average monthly number of child care cases
increasing by over 3,000.  However, Michigan has not run out of child
care funds and officials report no increase in the use of child care
funds by recipients subject to participation requirements.  Other
states initiating welfare reform also increased available resources
for child care.  Wisconsin, for example, dramatically increased
expenditures for child care since implementation of its waiver
demonstration, Pay for Performance, from $12 million in 1987 to $52
million in 1995.\19 Under a new program, Wisconsin Works, the
exemption for age of youngest child will be reduced from 1 year to 12
weeks, and officials anticipate child care expenditures to triple to
$158 million.  To decrease demand for child care generated by its
welfare reform, Iowa lowered the income level qualifying a family for
a child care subsidy, thereby reducing the number of low-income
working families eligible for subsidies. 


--------------------
\17 Though the program added staff at the beginning, during its 1997
session, the state's legislature eliminated 40 positions in Utah's
Office of Family Support as a result of caseload declines. 

\18 LaDonna Pavetti and Amy-Ellen Duke, Increasing Participation in
Work and Work-Related Activities:  Lessons from Five Welfare Reform
Demonstration Projects (Washington, D.C.:  The Urban Institute, Sept. 
1995).  States studied were Colorado, Iowa, Michigan, Utah, and
Vermont. 

\19 We visited these two states as part of a companion assignment on
welfare benefit termination, Welfare Reform:  States' Early
Experiences With Benefit Termination (GAO/HEHS-97-74, May 15, 1997). 


         REORGANIZING SERVICES
------------------------------------------------------ Chapter 2:2.6.3

The states structured their delivery of services in different ways to
emphasize recipients' responsibility to become self-sufficient.  Utah
provided services through its social service agency, the Department
of Human Services.  Employment counselors and mental health/drug
counselors are co-located in the local social service agency.  This
client-centered approach may be most appropriate in states, such as
Utah, that have succeeded in assisting the most job-ready clients to
earn their way off welfare, leaving behind harder-to-serve
recipients.  Clients who receive all services in one location
coordinated by the same caseworker as they do in Utah may be less
likely to fall through the cracks than when they are referred to
other agencies for services.\20

Michigan's Family Independence Program provides services to clients
through two agencies.  The Michigan Jobs Commission, a newly created
agency responsible for all economic development programs in the
state, provides job search services, while the Family Independence
Agency oversees the case and tracks client compliance.  With its
large caseload, organizing the delivery system along functional lines
may allow Michigan to provide services more efficiently and at lower
cost per client served. 

Massachusetts recently added one-stop career centers as an available
source of assistance during the 60-day job search period for
recipients who are required to work.  The centers are funded through
demonstration grants from the U.S.  Department of Labor, and the
Private Industry Council/Regional Employment Boards have contracting
and oversight responsibility for the centers that are accessible to
all citizens.  The Department of Transitional Assistance negotiates
agreements with the centers to serve an expected number of welfare
recipients; however, the centers are reimbursed only for recipients
actually served.  When welfare recipients visit the center, they
usually attend an orientation and meet with a career counselor to
develop and sign an action plan for searching for a job.  The
Department of Transitional Assistance assists the client with health
insurance, child care, and housing; the career center develops job
search strategies for the client. 

Other states have made similar efforts to structure their programs to
deliver services differently to emphasize the responsibility to
become self-sufficient.  In addition to Utah, for example, Florida
and Wisconsin brought together a variety of staffs at one location
and formed case management teams to quickly help provide a
comprehensive set of services that clients need to participate. 
Indiana adopted a policy requiring that local public assistance
offices co-locate with workforce and job placement agencies as well
as other social service agencies.  Like Michigan, Iowa reorganized
its service delivery along functional lines.  Its Department of
Economic Development contracted with workforce development agencies
to provide job services such as assessment, job club, job search,
education and training, and support services, while its Department of
Human Services maintained responsibility for eligibility
determination. 


--------------------
\20 As part of its seven-site JOBS program evaluation, the Manpower
Demonstration Research Corporation evaluated the impact of different
case management approaches on client participation.  They found that
in Columbus, Ohio, an integrated case management approach in which
one case manager did both benefit determination and monitoring of
employment and training activities resulted in significantly higher
levels of client participation in orientation and JOBS activities
than the case management approach that separated income maintenance
and JOBS activities.  See Two Approaches to JOBS Case Management in
Columbus, Ohio:  Implementation, Participation, and Impact Findings
(New York:  Manpower Demonstration Research Corporation, Nov.  1996). 


      TRANSITIONAL SERVICES
      EXPANDED TO ENCOURAGE WORK
-------------------------------------------------------- Chapter 2:2.7

To aid the transition from welfare to self-sufficiency, Utah extended
supportive assistance to recipients who earned their way off welfare. 
This extra support is designed to help recipients stay off welfare. 
For example, the state extended Medicaid coverage to 2 years after a
recipient leaves welfare and transitional child care for as long as
the family is eligible based on income.  In addition, Utah created a
new caseworker position, called transitional caseworker, to continue
assisting individuals who have earned their way off financial
assistance.  These case management services are available to
individuals for 2 years after leaving welfare.  While Michigan and
Massachusetts provide transitional child care and Medicaid to
individuals for 1 year after they leave welfare, this does not
represent a change from JOBS, which required states to offer 1 year
of transitional child care and Medicaid. 

Of states with work requirement waivers, 11 states extended
transitional child care, most increasing the period from 12 months to
24 months after a client becomes ineligible for AFDC because of
earnings; and 7 states waived the former requirement that a recipient
have received AFDC for 3 of the last 6 months to be eligible for
transitional child care (see table I.6).  Ten states also extended
transitional Medicaid, most increasing the period to 24 months, and
10 states waived the "3 of the last 6 months" requirement for
Medicaid. 


   STATES IMPLEMENTED STRICTER
   SANCTIONS FOR NONCOMPLIANCE
   WITH PARTICIPATION REQUIREMENTS
---------------------------------------------------------- Chapter 2:3

Strict sanctions for failure to comply with work requirements are
intended to encourage reluctant recipients to participate to retain
their benefits.  The three states we visited implemented tougher
sanctions than they had in the past for failure to comply.  After
varying lengths of time under a progressive set of sanctions, the
states terminate the entire family's benefit, called a full-family
sanction, for not cooperating with the program requirements.  In
contrast, under the JOBS program, only the parent's portion of the
grant was removed as a sanction and the family continued to receive
the child's portion of the benefit until the parent complied.  For
the first noncompliance, the sanction lasted until the participant
complied.  For the second and subsequent cases of noncompliance, the
sanction lasted for 3 months or 6 months, respectively, or until
compliance, whichever was longer.  As a result of these rules, a
client could remain in sanction status with a reduced benefit
indefinitely. 

Officials from all three states viewed the full-family sanction as a
way to get a participant's attention and eliminate from the rolls
those who do not participate because they have other sources of
support.  Caseworkers report that clients often ignore all attempts
to contact them until they are notified their grant will be
terminated.  Furthermore, officials in all three states said a grant
reduction was not sufficient to get all recipients to comply and that
the ultimate consequence of benefit termination was useful in some
cases. 

As part of the sanction procedure in Michigan and Utah, a process
called conciliation is used to discover and help alleviate reasons
for nonparticipation previously not disclosed to caseworkers. 
Conciliation, a requirement of the JOBS program, protects clients
against unwarranted benefit terminations by providing an opportunity
to resolve misunderstandings or disagreements before they result in a
sanction.  Both Michigan and Utah do home visits to ensure that
clients understand their obligations prior to benefit termination for
failure to participate.  Massachusetts eliminated its conciliation
process, allowing a client to be sanctioned for failure to
participate with no formal procedure for dispute resolution. 
Massachusetts replaced conciliation with a set of progressive
sanctions prior to benefit termination. 

Utah requires an extensive conciliation process before terminating a
recipient's benefits.  A meeting called a case staffing is held to
discuss barriers to employment and offer additional assistance.  This
meeting to which the client is invited, is attended by the case
manager and other Department of Human Services and allied agency
representatives familiar with the case.  Home visits are made in some
cases.  Officials said the conciliation process was established to
ensure the client understands the choice that is being made:  to not
participate and to have benefits terminated.  They indicated the
process is expensive because it is labor-intensive.  In November
1995, Utah initiated the sanction policy of full benefit termination
for noncompliance with work requirements.  Before that time, clients
who did not participate received a $100 reduction in their benefit
amount indefinitely.  This could lead to recipients remaining in
sanction status for long periods; for example, one case we reviewed
had been in sanction status for 3 years.  From the time the first
family's benefits were terminated in December 1995 through December
1996, 180 families had their benefits terminated for failure to
comply with the new program requirements.\21

Michigan's conciliation process involves phone contact, scheduled
office visits, and home visits if necessary.  Clients can resume
receiving full benefits at any point if they agree to comply, subject
to a 5- or 10-day compliance test.  If the client still does not
comply after repeated attempts at contact, the grant is reduced 25
percent.  After 12 months of receiving a reduced grant, the case is
closed.  Michigan terminated the benefits of its first cases for
failure to participate in April 1996, and as of December 31, 1996,
765 cases had their benefits terminated for failure to comply with
the work requirements. 

Massachusetts imposes stringent sanctions for failure to comply on
the group of recipients mandated to work.  Clients are given an
opportunity to discuss any hardship or barriers to participation when
they are assessed, and agency workers are expected to review
legitimate reasons for not participating, called good cause, before
mandating employment or imposing sanctions.  However, once a client
fails to comply, there is no formal opportunity for conciliation and
the adult portion of the grant is removed after a 10-day notice is
sent.  For an average grant of $468 per month, this results in a
reduction of about $84.  If the noncompliance continues, a second
10-day notice is sent and the entire grant is terminated unless
participation occurs within the 10-day period.  No home visit is
required before benefits are terminated.  Clients can return to
assistance at any time if they comply with their participation
requirements for 2 weeks.  Furthermore, the sanction process stops if
the recipient demonstrates an exemption or good cause, or files a
timely appeal.  As of December 31, 1996, Massachusetts had terminated
the benefits of 1,322 families for an adult's failure to comply with
the work participation requirements.\22

Under waiver, many additional states increased the duration and
severity of sanctions imposed on mandatory participants who do not
comply.  Twenty-two states with expanded work requirements now allow
the entire benefit to be terminated for failure to comply with work
requirements (see table I.5). 


--------------------
\21 A small number of these cases--three as of June 30, 1996--had
their benefits terminated for failure to comply with child support
enforcement requirements. 

\22 An estimated 386 additional families had their benefits
terminated for noncompliance by a dependent child aged 16 to 18
required to participate in JOBS.  Benefits can be terminated for
reasons other than failure to comply with work requirements, such as
the failure to comply with requirements for teen living arrangements
or school attendance.  For detailed discussion of benefit termination
under waiver, see GAO/HEHS-97-74, May 15, 1997. 


   STATES ACHIEVE HIGHER
   PARTICIPATION RATES UNDER
   WAIVER PROGRAMS
---------------------------------------------------------- Chapter 2:4

By changing the rules and strengthening sanctions for noncompliance,
Massachusetts, Michigan, and Utah were able to involve a substantial
portion of their welfare recipients in some activity they believe
will move them toward self-sufficiency. 


      MOST PARTICIPANTS ARE IN
      WORK-RELATED ACTIVITIES
-------------------------------------------------------- Chapter 2:4.1

During September and October 1996, when all three states had their
waiver programs operational statewide, 11 percent of the caseload in
Massachusetts, 32 percent in Michigan, and 90 percent in Utah were
participating in activities that count as participation under the
requirements of the states' waiver programs.  (See table 2.2.) The
largest percentage of those participating in all three states were
engaged in unsubsidized employment, reflecting the work focus of
these programs. 



                               Table 2.2
                
                Participation in Allowable Activities in
                     Three States' Waiver Programs,
                         September/October 1996

                                                Massac
                                                husett  Michig
                                                     s      an    Utah
----------------------------------------------  ------  ------  ------
Average number participating                    8,491\  43,271   9,813
                                                     a
Average monthly adult caseload                  79,521  136,85  10,881
                                                             2
Percentage of adult caseload participating         11%     32%     90%
----------------------------------------------------------------------
Note:  Percentages are rounded to nearest whole number. 

\a In Massachusetts, only individuals who are strictly mandated to
participate and are participating for 20 hours are counted as
participating in the work program.  During this same period,
approximately 19,793 additional recipients not subject to the work
requirement or not participating for 20 hours a week were
participating in a work-related activity. 

Utah has the highest overall participation rate, reflecting the fact
that its program requires almost all recipients to participate and
more broadly defines participation, allowing a range of activities to
meet the participation requirement that are not counted in other
states.  Of those not participating, the largest number, over 5
percent of the caseload, were excused because of a family medical
condition.  Utah also has no minimum required number of hours of
participation for some activities, tailoring the required hours to
the recipient and the activity. 

Although Michigan also has a universal participation requirement, it
defers a substantial number of recipients from work participation
whose participation in social contract activities is neither counted
toward participation nor monitored.  In addition, Michigan requires
20 hours per week of participation in most work activities.  If
Michigan counted everyone who was participating in an activity for
any number of hours during a week, its participation rate would be
closer to 42 percent for the same period. 

Massachusetts had the lowest participation rate reflecting the fact
that only one-fifth of its caseload is strictly mandated to
participate and the state only counts the participation of that
mandated group as meeting the participation requirement for its work
program.  However, the participation for this group is substantial. 
In September and October 1996, for example, an average of 15,734
individuals were mandated to participate in a work activity.  During
that period, 8,491, approximately 54 percent, of these individuals
were participating in an activity meeting their participation
requirement.  Table 2.3 shows that of those participating, 58 percent
were employed, 23 percent were in community service, 13 percent were
in their 60-day job search period, and 6 percent were grandfathered
into educational activities.  Reflecting the different definitions of
participation in the three states, table 2.3 shows the percentage of
those participating in the different countable activities. 



                               Table 2.3
                
                 Distribution of Participant Activities
                   in Three States' Waiver Programs,
                         September/October 1996

                                                Massac
                                                husett  Michig
Activity                                             s      an    Utah
----------------------------------------------  ------  ------  ------
Unsubsidized employment                            56%     76%     34%
Subsidized employment                               2%      1%      NA
Community service/community work experience        23%    1%\a      2%
Job search/job readiness                           13%     13%     23%
Vocational education and job training               NA      8%      6%
Education                                           6%      2%   12%\b
Life skills training                                NA      NA    9%\c
Being assessed                                      NA      NA     14%
Average number participating                    8,491\  43,271   9,813
                                                     d
----------------------------------------------------------------------
Note:  Percentages are rounded to nearest whole number.  NA means
that either the state does not have this component or it is not
counted as participation. 

\a Only Michigan has a community work experience program. 

\b In Utah, 32 percent of the participants in this category are in
postsecondary education. 

\c Life skills includes such activities as parenting, mental health,
drug, and alcohol abuse counseling. 

\d In Massachusetts, only individuals who were mandated to
participate are included in the total. 

In each of the three states we reviewed, the largest percentage of
all participants were engaged in unsubsidized employment.  This
pattern reflects the work focus of these programs and the fact that
the income disregards were changed, allowing more recipients to
simultaneously work and receive welfare benefits.\23 Of those
participating in an activity, 56 percent in Massachusetts, 76 percent
in Michigan, and 34 percent in Utah were in unsubsidized employment. 
After unsubsidized employment, the percentage of recipients engaged
in other activities varied, reflecting the different program designs. 
In Michigan and Utah, the second largest group was engaged in job
search/job readiness activities; but in Massachusetts, a significant
group participated in community service.  While Massachusetts had 6
percent in education,\24 many of these were grandfathered in from the
previous JOBS program.  Participation in subsidized employment was
very low in Massachusetts and Michigan and nonexistent in Utah. 


--------------------
\23 Increasing the amount of the earned income disregard can
simultaneously increase recipients' total income while allowing them
to count toward states' work participation rate under the new law. 
However, in states with low benefit levels, small amounts of earned
income can lead to ineligibility for welfare benefits, leaving
smaller numbers of working welfare recipients to count toward the
work participation rate.  The time recipients receive benefits, even
while they are working, may count toward their time limit on benefit
receipt. 

\24 These recipients are in English as a second language, adult basic
education, and postsecondary education programs. 


   THREE STATES ALSO INCREASED THE
   PERCENTAGE OF THE CASELOAD
   PARTICIPATING IN JOBS
   ACTIVITIES UNDER WAIVERS
---------------------------------------------------------- Chapter 2:5

To determine whether states had actually increased participation
under their waiver programs over what they had been achieving in the
prior year under the JOBS program, we assessed participation rates
using the JOBS program definition of allowable activities.  In
addition to employment-related activities such as job search,
community work experience, on-the-job training, and job entry, JOBS
activities include high school or high school equivalency, remedial
education, English as a second language, higher education, jobs
skills training, and vocational training. 

In comparison to reported participation in their JOBS programs for
fiscal year 1995, all three states increased the percentage of their
AFDC recipients participating in JOBS activities under their
statewide waiver programs during September/October 1996.  Through
their waiver programs, Massachusetts increased the proportion of its
caseload participating in such activities for any number of hours
from 19 to 36 percent, Michigan from 21 to 42 percent, and Utah from
42 to 57 percent, as illustrated in figure 2.1. 

   Figure 2.1:  Average Monthly
   Participation Rates in JOBS
   Activities (FY 1995) and
   Comparable Activities in Three
   Statewide Waiver Programs
   (September and October 1996)

   (See figure in printed
   edition.)

We used the percentage of the caseload participating in a federally
defined JOBS activity for any number of hours as a consistent
benchmark.  To calculate comparable rates, we included in the
numerator any welfare recipient active in a JOBS activity for any
number of hours, and in the denominator we used total adult caseload
for the comparable period.  Consequently, although the absolute
number of recipients participating in Utah actually declined slightly
from fiscal year 1995 to September/October 1996, because of
substantial reductions in the overall caseload, the number
participating was a larger percentage of the caseload.  Both
Massachusetts and Michigan had large increases in the absolute
numbers of recipients participating as well as caseload declines. 


   STATES PLAN TO CONTINUE
   PROGRAMS UNDER TANF
---------------------------------------------------------- Chapter 2:6

Since the enactment of the new welfare reform law, states no longer
need to use waivers of AFDC rules as a vehicle for instituting
policies aimed at increasing the number of welfare recipients
participating in work-related activities.\25 The policies that
Massachusetts, Michigan, and Utah pursued under waivers provide an
indication of what strategies they believe are effective in attaining
greater participation.  First, to overcome recipients' frequent
failure to attend program activities, these states deliver a stronger
message to clients regarding their obligation to participate in
required activities and then monitor their participation.\26 All
three states also changed the disregard for earned income to remove
financial disincentives in the AFDC program.  In addition, officials
emphasize that these efforts must be coupled with the threat of
benefit termination for failure to comply with work participation
requirements.  All three states underscored the importance of
sanctions in achieving increased participation. 

Massachusetts, Michigan, and Utah plan to continue their welfare
programs relatively unchanged under TANF.  The major change that
Massachusetts will be making to its program is limiting benefit
receipt for all nonexempt recipients--two-parent families and single
heads of household with no children under the age of 2--to 24 months
in any 60-month period, a policy that had been denied in the state's
request for waivers.  Otherwise, Massachusetts will continue to
require that mandatory recipients work or perform community service
for 20 hours per week and not increase the hours to 30 as specified
in the new law.  In addition, Massachusetts will not limit the length
of job search to 6 weeks as required in the new law.  These
differences between the state's plan and the law require that waivers
stay in force until the scheduled end of the waiver demonstration, at
which time the state will be required to comply with requirements of
the new welfare law.  Utah had no time limit in its waiver program
but plans to implement a 36-month limit on benefit receipt, from
which a maximum of 20 percent of the caseload can be exempted for
hardship.  Eligible individuals who are employed at least 80 hours a
month in unsubsidized employment and had worked at least 80 hours per
month in 6 of the past 24 months can have their benefits extended on
a month-to-month basis beyond the 36th month up to the 60th month. 
Michigan added a 2-year work requirement and a 5-year limit on
benefit receipt, the same work requirements and time limits specified
in TANF, and is reducing from 12 months to 4 months the length of
time a client can be noncompliant before benefits are terminated. 
The state is considering using state funds to support certain
unemployed recipients beyond the end of the 5-year limit. 


--------------------
\25 TANF provides that if a state had a waiver in effect on the date
of enactment, it can continue the waiver until its expiration and
will not be required to comply with provisions of the act that are
inconsistent with the waiver.  If a state had a waiver pending on the
date of enactment and the waiver is approved on or before July 1,
1997, the state may not be required to comply with provisions of the
act that are inconsistent with the waiver though the state will be
subject to the work participation requirements of the act. 

\26 See Pavetti and Duke, pp.  43 and 46. 


STATES LIKELY TO ACHIEVE INITIAL
TANF ALL-FAMILIES PARTICIPATION
RATE BUT MAY BE CHALLENGED TO MEET
OTHER RATES
============================================================ Chapter 3

Massachusetts, Michigan, and Utah are likely to meet the TANF work
participation rate for all families in the short run.  However, state
officials in Michigan and Massachusetts are concerned about their
ability to meet the initial two-parent families' rate, and officials
in all three states are concerned about the higher future rates for
all families and two-parent families. 


   CALCULATING PARTICIPATION RATES
   UNDER TANF
---------------------------------------------------------- Chapter 3:1

Under the new law, states are required to meet a participation rate
for the entire caseload--the "all-families rate"--and a separate
participation rate for families with two parents--the "two-parent
families rate." For the all-families rate, the numerator includes the
number of families receiving assistance in which an adult or a minor
child head of household is engaged in work.  The denominator includes
the total number of families receiving assistance reduced by the
number of families who are being sanctioned for failure to
participate.  States have the option to exclude from the denominator
families with children under the age of 1.  For the two-parent
families rate, the numerator includes only families in which both
parents meet their participation obligation and the denominator
includes all two-parent families except those being sanctioned. 

To count as a participant for purposes of calculating the
all-families rate in the first year, a single head of household must
work at least 20 hours per week, and one adult in a two-parent family
must participate 35 hours a week, at least 30 hours of which must be
in one of the following activities: 

  unsubsidized employment,

  subsidized public or private sector employment,

  work experience,

  on-the-job training,

  community service,

  provision of child care to an individual participating in community
     service,

  vocational education, not to exceed 12 months for any individual,
     or

  job search and job readiness assistance, but not to exceed 6 weeks. 

If a two-parent family receives federally funded child care and an
adult in the family is not disabled or caring for a severely disabled
child, the second parent must participate for 20 hours for the family
to be counted as participating. 

The minimum participation rates specified in the law for each year
are lowered to the extent that state caseloads have declined since
1995.  Given that all but three states nationwide have experienced
declines in their caseloads beginning in 1994, the majority of states
should have target participation rates for fiscal year 1997 that are
lower than the minimum 25 percent specified in the law.  Figure 3.1
illustrates the caseload decline since 1994 in Massachusetts,
Michigan, and Utah. 

   Figure 3.1:  Caseload Declines
   in Three States, FY 1994-96

   (See figure in printed
   edition.)

Under the new law, caseload reductions not only provide states a
lower target work participation rate, they also contribute to states
having more money per recipient than in the past.  This results from
the fact that TANF block grant amounts are based on federal
expenditures in prior years, when most states had higher caseloads,
and remain fixed until fiscal year 2002.  Recent statistics compiled
by the U.S.  House of Representatives' Committee on Ways and Means
estimate the average benefit available per family in fiscal year 1998
will be $5,662 compared with $3,624 in fiscal year 1994. 


   THREE STATES LIKELY TO ACHIEVE
   THE ALL-FAMILIES PARTICIPATION
   RATE IN THE SHORT RUN
---------------------------------------------------------- Chapter 3:2

Using the new law's definition of work participation and reducing the
minimum participation rate by the caseload reduction credit, all
three of the states we reviewed should be able to meet the
all-families participation rate for fiscal year 1997.  Table 3.1
shows the target participation rates the three states will need to
achieve under the new law in fiscal year 1997, taking into account
the reductions in the caseloads these states have had from 1995 to
1996.\27 The table also shows the participation rates the three
states would have achieved if the TANF participation rules were
applied to their welfare programs in September/October 1996. 



                         Table 3.1
          
            Estimated FY 1997 All-Families TANF
            Target Participation Rates and TANF
            Rates Achieved Under Waiver in Three
              States in September/October 1996

                         (Percent)

                                        Participation rate
                      Estimated target      achieved using
                  rate with credit for  TANF definition of
State               caseload reduction       participation
----------------  --------------------  ------------------
Massachusetts                       13                25.3
Michigan                            13                32.6
Utah                                14                30.9
----------------------------------------------------------
Since Massachusetts has had a caseload decline of over 12 percent
from fiscal year 1995 to fiscal year 1996, its first-year TANF
participation rate is estimated to be about 13 percent.  Michigan's
caseload reduction over the same period is just under 12 percent, and
Utah's is just over 11 percent.  Therefore, Michigan's first-year
TANF target will be about 13 percent, and Utah's about 14 percent. 

Table 3.1 also shows that Michigan and Utah, the two states with a
universal participation requirement, would easily make the
all-families participation rate in the first year, even without the
adjustment for the caseload decline.  Massachusetts, which exempts a
substantial portion of the caseload from participation requirements
and is the only state of the three opting to exclude families with
children under age 1 from the TANF work participation requirement,
has a lower estimated work participation rate but would also meet the
minimum 25-percent all-families participation rate in the first year. 
Since Massachusetts' target rate is estimated to be 13 percent,
officials are confident about meeting this target.  Because the three
states are experiencing continued caseload declines in fiscal year
1997, their targets for fiscal year 1998 should also be lower than
those specified in the law. 


--------------------
\27 As of March 1997, HHS had not yet promulgated regulations telling
states how to calculate caseload reductions for purposes of TANF
participation rates.  However, state program administrators are using
the law and their own data to project what they think their
reductions will be.  According to the law, the minimum participation
rate for a year is reduced by the number of percentage points equal
to the number of percentage points by which the average monthly
number of families that received assistance during the immediately
preceding fiscal year is less than the number of families receiving
assistance in fiscal year 1995.  Therefore, the 1997 credit is equal
to the percentage change between the 1996 and 1995 average monthly
caseloads. 


   STATES MAY HAVE PROBLEMS
   MEETING THE TWO-PARENT
   PARTICIPATION RATE
---------------------------------------------------------- Chapter 3:3

The two-parent families participation rate under the new law is
higher than the two-parent rate under the JOBS program, both in terms
of the hours of participation required and the percentage of the
two-parent families required to participate.  Though Massachusetts,
Michigan, and Utah reported meeting the JOBS program's two-parent
participation rates in 1994 and 1995, many states failed to meet
them.\28 As shown in table 3.2, Michigan is unlikely to meet the
initial two-parent TANF work participation rate of 75 percent for
fiscal year 1997 without some program changes.  Massachusetts was not
able to provide us data to estimate a TANF two-parent family
participation rate. 



                         Table 3.2
          
           Estimated FY 1997 Two-Parent Families
          TANF Target Work Participation Rates and
            TANF Rates Achieved Under Waiver in
           Three States in September/October 1996

                         (Percent)

                                        Participation rate
                      Estimated target      achieved using
                  rate with credit for  TANF definition of
State               caseload reduction       participation
----------------  --------------------  ------------------
Massachusetts                       66                  NA
Michigan                            60                  35
Utah                                53                  87
----------------------------------------------------------
Note:  NA means data were not available. 

Michigan's concern over meeting the two-parent families participation
rate has led the state to develop an on-the-job training program
specifically for this population.  Of the three states we visited,
Utah is the most likely to meet the two-parent rate in the first
year.  The state has a very small two-parent family caseload,
approximately 110 cases, and has required this population to work in
exchange for benefits since 1983.  One parent is required to
participate 40 hours per week and the other parent 20 hours.  Also,
Utah limits the length of time the benefit can be received to 6
months in any 12-month period.  Utah officials estimate the state's
current work participation for two-parent families using the criteria
established by the new law would be 87 percent.  Consequently, Utah
officials are not concerned about meeting their estimated target rate
for two-parent families of 53 percent in the first year. 

Massachusetts plans to maintain its waiver provision requiring 20
hours a week of work participation for its mandatory work
participants, including two-parent families, rather than the higher
number of hours specified in the new law.  Consequently,
Massachusetts officials are not as concerned about their ability to
meet the initial two-parent participation rates as are Michigan
officials.  However, they are concerned about the future rate
increase to 90 percent and the fact that when their waiver authority
expires in 2005 they will be required not only to meet the minimum
participation rates but also the increased hours of participation
specified in the law. 


--------------------
\28 The JOBS participation rate target for AFDC-UP families was 40
percent in 1994 and 50 percent in 1995.  Forty-three states failed to
make the rate in 1994, and 28 states failed to meet the 1995 rate. 


   STATES MAY HAVE DIFFICULTY
   MEETING FUTURE PARTICIPATION
   RATES
---------------------------------------------------------- Chapter 3:4

Officials in the three states we visited are concerned about their
ability to meet higher future participation rates established under
the new law because of the requirement that larger numbers or
recipients work more hours, the possibility of a future recession
undermining employment opportunities, and the unique employment
problems in urban areas.  Finally, there is concern that as those who
are more employable get jobs and move off welfare, the remaining
caseload will be increasingly difficult to employ and participation
rates increasingly difficult to achieve. 


      INCREASES IN HOURS OF
      PARTICIPATION MAY BE
      DIFFICULT TO ACHIEVE
-------------------------------------------------------- Chapter 3:4.1

According to state officials, it may be easier for recipients to find
part-time employment than full-time employment because employers may
not want to pay the fringe benefits associated with full-time
employment.  Initially, states can count recipients engaged in
employment for 20 hours per week; however, as the required hours
increase to 30 hours in the year 2000, the officials said it may be
more difficult for recipients to find employment for at least this
many hours.  Moreover, in many states, especially those with low
benefit levels, working 30 hours or more is likely to make a family
ineligible for benefits.  Once recipients have earned their way off
the rolls, they will not count toward the participation rate.  In
Utah, during September and October 1996, only about 16 percent of the
caseload participated over 30 hours a week, whereas about 31 percent
participated over 20 hours in activities that would count toward the
TANF participation rate.  In Michigan, approximately 24 percent of
the caseload participated 30 hours per week compared with almost 33
percent participating 20 or more hours per week in activities that
meet the TANF participation requirements.  The concern may be
mitigated to some degree by the fact that TANF limits required work
participation to 20 hours a week for single parents with a child
under the age of 6.  Furthermore, since these three states did not
require more than 20 hours of participation under their waivers, the
low levels of participation for more hours may not reflect what can
be achieved in the long run. 


      CHANGES IN THE EMPLOYMENT
      OUTLOOK COULD AFFECT STATES'
      ABILITY TO MEET THE TANF
      PARTICIPATION RATES
-------------------------------------------------------- Chapter 3:4.2

Many observers believe that low unemployment rates have contributed
to the dramatic decline in welfare caseloads during the past several
years.  A recent study of caseload declines concluded that most of
the changes in the AFDC single-parent caseload since 1994 are
attributable to factors such as the unemployment rate and demographic
trends, including the number of children under age 18 in a state.\29
There are a number of factors beyond the control of policymakers and
program officials that could contribute to changes in the caseload. 
Should the economy deteriorate in the future and unemployment rates
go up, states may not receive the target rate reduction associated
with a caseload drop and they will be trying to achieve higher target
participation rates under less favorable economic conditions. 


--------------------
\29 David J.  Fein, Understanding Recent Declines in State AFDC
Caseloads:  An Analysis of Cross-State Variation (Cambridge, Mass.  : 
Abt Associates, Oct.  1996). 


      STATES WITH WELFARE
      POPULATIONS CONCENTRATED IN
      URBAN AREAS MAY HAVE
      DIFFICULTY MEETING THE
      PARTICIPATION RATES
-------------------------------------------------------- Chapter 3:4.3

Many argue that the urban poor on welfare tend to have significant
barriers to employment.  Frequently lacking in education and prior
work history, they are uniquely unsuited to the jobs available, which
increasingly demand more skills.  The estimated TANF work
participation rate in Wayne County, Michigan, which includes Detroit,
is 23 percent, approximately 19 percent lower than for the rest of
Michigan.\30 Furthermore, early results from Project Zero, a Michigan
initiative to have the entire caseload employed in demonstration
sites, suggests that a site in Detroit is experiencing greater
difficulty getting participants employed than are other Michigan
sites. 


--------------------
\30 For October 1996, the estimated TANF participation rate for Wayne
County was 23 percent compared with 42 percent for the remaining
Michigan counties and 33 percent for the entire state. 


      RECIPIENTS WHO DO NOT FIND
      JOBS TEND TO HAVE MULTIPLE
      BARRIERS TO EMPLOYMENT
-------------------------------------------------------- Chapter 3:4.4

Officials in the three states expressed concern that recipients who
do not find employment have multiple barriers and therefore are
considered hard to serve.  In Utah, where all welfare recipients are
required to participate, approximately 30 percent of the caseload is
described by the program director as being hard to serve.  Studies of
recipients in this group have found that, in addition to being less
likely to have prior work experience and being more likely to have
lower literacy levels, they have multiple problems that make program
participation difficult.  The problems include mental health issues,
such as depression, anxiety, and personality disorders; domestic
violence; substance abuse; behavior problems with children; and legal
problems.  Since the new law allows states to exempt 20 percent of
their caseload from the time limit, they may choose to exempt these
hard-to-serve individuals; however, they will be included in the
denominator of the participation rate calculation. 

While Utah's waiver program is the only one of the three state
programs we reviewed that allows participation in mental health or
substance abuse treatment programs to meet the work participation
requirement, Michigan will temporarily defer from work participation
a recipient who is receiving such treatment.  Officials reported
services were insufficient to meet client need in this area.  Even if
clients cannot participate because they need, but do not receive,
such services, they will remain in the denominator of the
participation rate calculation, potentially lowering a state's
participation rate. 


CONCLUSIONS AND HHS' AND STATES'
COMMENTS
============================================================ Chapter 4

One of the goals of the Personal Responsibility and Work Opportunity
Reconciliation Act of 1996 is to end the dependence of needy parents
on government benefits by making welfare a transitional program that
moves recipients into employment and off the rolls.  To this end, the
act encourages states to increase the proportion of those receiving
cash assistance in work and work-related activities, ultimately
requiring recipients to work as a condition of benefit receipt and
limiting the total time of benefit receipt.  The three states we
reviewed were attempting to accomplish this goal under waiver and,
although these states were taking different approaches, they all
appeared to have been able to increase participation.  While we
cannot conclude whether any of the specific policy changes were
linked to changes in participation, we did observe some common
elements across programs.  For example, although these states
differed on their participation requirements, they used similar
strategies to get recipients to meet the requirements.  The sanction
of benefit termination as a way to motivate client participation was
found in all three programs we reviewed and in 19 other states with
waivers to increase work participation.  Officials in the three
states viewed sanctions as a powerful tool to increase participation
rather than as a punishment for noncompliant behavior. 

While the new law limits the types of activities that count as work
participation, the experience of states, such as Utah and Michigan,
that are trying to engage the entire caseload in self-sufficiency
activities has been that some portion of the caseload requires more
than job search and assistance with child care and transportation to
be able to engage in a strictly defined work-related activity.  To
the extent that states move the more job-ready recipients into
employment, they will be challenged to provide the services necessary
to help clients with multiple barriers move into employment.  Since
these are the recipients who would have been exempted or deferred
from participation in the past, states have limited experience in
what is required to enable them to become employed, earn sufficient
wages, and sustain employment in order to exit welfare permanently. 

The experience of the three states we reviewed demonstrates that,
with the programs they had in place and with credit for caseload
reductions since fiscal year 1995, the initial all-families
participation rate should be attainable.  However, these states are
unsure of their capacity to reach higher rates in the future, in part
because they believe that their success in moving families off the
rolls will make it increasingly difficult to meet future
participation rates.  As states assist recipients who are employable
into jobs and off the rolls, they are concerned that they will be
left with recipients who are essentially unemployable or at least in
need of multiple supportive services.  On the other hand, the large
reductions in caseload that appear to be continuing not only will
lower states' target participation rates, but also will result in
more money per recipient under the block grant, providing an
opportunity for states to develop the range of services required to
help the least job-ready make the transition to employment. 

If a state chooses to provide the services necessary to enable
recipients with multiple employment barriers to participate in an
activity that will ultimately lead to self-sufficiency, the
particular activity may not meet the strict definition of work
participation under the new law.  Consequently, even though the state
may actually assist many of these recipients in getting jobs and
leaving the rolls, it may not appear to be successful in meeting the
participation requirements of the new law during the time it is
preparing the least employable to make the transition from welfare to
work.  To ensure that participation rates are promoting the desired
program outcome, they need to be carefully studied and monitored in
the future. 


   COMMENTS FROM HHS AND STATES
---------------------------------------------------------- Chapter 4:1

We obtained comments on a draft of this report from our three case
study states:  Massachusetts, Michigan, and Utah.  The states
provided technical clarifications about their programs and data
interpretation.  We incorporated their comments in the report as
appropriate.  We requested, but did not receive, comments from HHS. 


SELECTED CHARACTERISTICS OF
WELFARE REFORM PROGRAMS IN STATES
WITH WAIVERS CHANGING WORK
REQUIREMENTS
=========================================================== Appendix I



                         Table I.1
          
           Age of Youngest Child Exemptions Cited
            in Waivers for States Changing Work
                        Requirements

              Parent exempt from participating if age of
State         youngest child is...
------------  --------------------------------------------
Arizona       Under 1 year old

Colorado      No change requested in waiver

Connecticut   Under 1 year old

Delaware      Under 13 weeks old

Florida       Under 6 months old

Georgia       Under 5 years old

Hawaii        Under 6 months old

Illinois      Under 5 years old

Indiana       Under 12 weeks old

Iowa          Under 3 months old

Louisiana     No change requested in waiver

Maryland      Under 12 weeks old

Massachusett  Under mandatory school-age, usually 6 years
s             old, if child is included in the assistance
              unit, under 3 months old if the child was
              conceived while the parent was receiving
              AFDC

Michigan      Under 12 weeks old\a

Mississippi   No change requested in waiver

Missouri      No change requested in waiver

Montana       Under 1 year old

Nebraska      Under 12 weeks old

New           Under 6 years old if child is included in
Hampshire     the assistance unit, under 3 months old if
              the child was conceived while the parent was
              receiving AFDC

North         Under 5 years old
Carolina

North Dakota  Under 3 years old or at the state option,
              less than 3 but not less than 1 year old

Ohio          Under 6 years old

Oklahoma      No change requested in waiver

Oregon        Under 3 months old

South         Under 1 year old; no exemptions for parents
Carolina      under 25 years old who have not completed
              high school

South Dakota  No change requested in waiver

Texas         Under 5 years old; after September 1997,
              exemption changes to children under 4 years
              old

Utah          No exemptions

Vermont       Under 6 months old

Virginia      Under 18 months old; temporary 6-week
              extension granted for children conceived
              while the parent was receiving AFDC

West          No change requested in waiver
Virginia

Wisconsin     Under 1 year old for one parent

Wyoming       Under 1 year old; 3-month exemption for
              children conceived while the parent was
              receiving AFDC
----------------------------------------------------------
\a Originally, under Michigan's waiver, all exemptions were
eliminated; however, the state legislaure made single parents with
children under the age of 12 weeks exempt from the work requirement. 

Source:  The terms and conditions in the state waivers, HHS. 



                               Table I.2
                
                    Expanded Activities Meeting the
                  Participation Requirement in States
                       Changing Work Requirements

                        Parenting
            Life        or family               Drug and    Mental
            skills      skills      Family      alcohol     health
State       training    training    counseling  counseling  counseling
----------  ----------  ----------  ----------  ----------  ----------
Delaware                X

Florida                 X

Georgia                 X

Iowa                    X

Nebraska                X

North                   X           X           X
Dakota

Oregon                                          X           X

South                   X                       X
Carolina

Texas                   X

Utah                    X           X           X           X

Vermont                 X

Wisconsin   X           X
----------------------------------------------------------------------
Note:  Other states that changed their work requirements did not
count these activities, if provided, as meeting the participation
requirements under their waivers. 

Source:  The terms and conditions in the state waivers, HHS. 



                               Table I.3
                
                    Changes in Age of Youngest Child
                 Exemption and Work-Related Activities
                  Cited in Waivers for States Changing
                           Work Requirements

              Changed age
              of youngest   Extended job  Subsidized    Community
              child         search        work          service
State         exemption     component     component     component
------------  ------------  ------------  ------------  --------------
Arizona       X                           X

Colorado

Connecticut   X             X                           X

Delaware      X             X

Florida       X                           X

Georgia       X             X

Hawaii        X             X

Illinois      X             X             X

Indiana       X             X             X             X

Iowa          X             X

Louisiana

Maryland      X             X

Massachusett  X             X             X             X
s

Michigan      X             X

Mississippi                               X

Missouri                                                X

Montana       X                                         X

Nebraska      X             X

New           X             X             X             X
Hampshire

North         X             X             X
Carolina

North Dakota  X             X

Ohio          X                           X

Oklahoma

Oregon        X             X             X

South         X             X
Carolina

South Dakota                                            X

Texas         X             X                           X

Utah          X

Vermont       X             X             X             X

Virginia      X             X             X             X

West
Virginia

Wisconsin     X                           X             X

Wyoming       X
----------------------------------------------------------------------
Source:  The terms and conditions in the state waivers, HHS. 



                         Table I.4
          
           Diversions Cited in Waivers for States
                 Changing Work Requirements

            Diversion amount        Waiting period to
State       available               reapply
----------  ----------------------  ----------------------
Maryland    Not to exceed 3 times   Not eligible for the
            the monthly benefit     number of months
            amount unless           covered by the one-
            determined to be        time payment
            warranted by the local
            department, but in no
            case to exceed 12
            months' AFDC benefit
            amount

Montana     Not to exceed 3 times   Ineligible to receive
            the monthly benefit     2 months of future
            amount                  AFDC benefits for each
                                    month's AFDC benefit
                                    received as part of
                                    the one-time payment

North       Not to exceed 3 times   Required to repay
Carolina    the monthly benefit     these benefits
            amount                  following procedures
                                    used for AFDC
                                    overpayments

Texas       $1,000 regardless of    12 months
            family size

Utah        Not to exceed 3 times   3 months; eligible for
            the monthly benefit     prorated AFDC benefits
            amount                  if reapplies before 3-
                                    month period

Virginia    Not to exceed 120       160 days
            days' benefit amount
----------------------------------------------------------
Note:  Other states changing work requirements did not cite
diversions in their waivers. 

Source:  The terms and conditions in the state waivers, HHS. 



                               Table I.5
                
                  Changes to Sanctions and Incentives
                  Cited in Waivers for States Changing
                           Work Requirements

            Full-       Changed                 Changed     Allow
            family      earned      Changed     vehicle     individual
            sanctions\  income      asset       value       developmen
State       a           disregards  limit       limit       t accounts
----------  ----------  ----------  ----------  ----------  ----------
Arizona                 X           X                       X

Colorado                X           X           X

Connecticu  X           X           X           X
t

Delaware    X                       X           X           X

Florida                 X           X           X

Georgia                 X                       X

Hawaii                  X           X           X

Illinois    X           X           X

Indiana     X                       X

Iowa        X           X           X           X           X

Louisiana   X

Maryland    X           X           X           X

Massachuse  X           X           X           X           X
tts

Michigan    X           X                       X

Mississipp  X           X           X                       X
i

Missouri                X           X           X

Montana                 X           X           X

Nebraska    X           X           X           X

New         X           X           X           X
Hampshire

North                   X           X           X
Carolina

North       X           X           X           X
Dakota

Ohio        X           X                       X

Oklahoma                                        X

Oregon      X           X           X           X           X

South       X           X           X           X           X
Carolina

South       X           X           X           X
Dakota

Texas                   X           X           X           X

Utah        X           X           X           X

Vermont     X           X           X           X

Virginia    X           X           X           X           X

West        X
Virginia

Wisconsin   X           X           X           X           X

Wyoming                             X
----------------------------------------------------------------------
\a Table shows states with full-family sanctions for failure to
comply with work requirements.  Four other states, Idaho, Kansas,
Minnesota, and Tennessee, which did not change work requirements with
waivers, did initiate full-family sanctions for failure to comply
with existing work requirements.  Full-family sanctions can be
applied for other reasons, for example, failure to comply with teen
living arrangement requirements.  For a detailed discussion see
Welfare Reform:  States' Early Experiences With Benefit Termination
(GAO/HEHS-97-74, May 15, 1997). 

Source:  The terms and conditions in the state waivers, HHS. 



                               Table I.6
                
                 Changes to Transitional Child Care and
                 Transitional Medicaid Requirements for
                   States Changing Work Requirements


State         Child care    Medicaid      Child care    Medicaid
------------  ------------  ------------  ------------  --------------
Arizona                                   24 months     24 months

Colorado                    Eliminated

Connecticut   Eliminated    Eliminated    Extended      24 months
                                          until
                                          family's
                                          income
                                          exceeds 75%
                                          of state
                                          median
                                          income

Delaware                                  24 months     24 months

Florida                                   24 months

Illinois      Eliminated                  24 months     24 months

Iowa                                      24 months

Massachusett  Eliminated    Eliminated
s

Mississippi   Eliminated    Eliminated

Montana       Changed to 1  Eliminated
              month

Nebraska                                  24 months     24 months

New                         Eliminated
Hampshire

Ohio                                      18 months

South         Eliminated    Eliminated    24 months     24 months
Carolina

Texas                       Eliminated                  18 months

Utah          Eliminated    Eliminated    Extended      24 months
                                          until family
                                          income
                                          exceeds
                                          sliding fee
                                          scale

Vermont                                                 36 months as
                                                        long as income
                                                        does not
                                                        exceed 185% of
                                                        poverty line

Virginia                    Eliminated    24 months     24 months
----------------------------------------------------------------------
\a Other states that changed their work requirements through waiver
did not change their transitional benefits. 

Source:  The terms and conditions in the state waivers, HHS. 


GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
========================================================== Appendix II

GAO CONTACTS

David P.  Bixler, Assistant Director, (202) 512-7201
Margaret Boeckmann, Senior Social Science Analyst, (202) 512-6992

STAFF ACKNOWLEDGMENTS

In addition to those named above, the following individuals also made
important contributions to this report:  Ellen Soltow, Evaluator,
conducted fieldwork in Massachusetts, Michigan, and Utah and
contributed to the data analysis and report writing; Karen Brown,
Evaluator, conducted the analysis of the terms and conditions of
state waivers; and Rachel Rosenthal, Intern, assisted in the early
design phase of the assignment. 


BIBLIOGRAPHY
============================================================ Chapter 1

Bane, Mary Jo, and David T.  Elwood.  Welfare Realities:  From
Rhetoric to Reform (Cambridge, Mass.:  Harvard University Press,
1994). 

Brown, Amy.  Work First:  How to Implement an Employment-Focused
Approach to Welfare Reform (New York:  Manpower Demonstration
Research Corporation, Mar.  1997). 

Fein, David J.  Understanding Recent Caseload Declines in State AFDC
Caseloads:  An Analysis of Cross-State Variation (Cambridge, Mass.: 
Abt Associates, Oct.  1996). 

Freedman, Stephen, and Daniel Friedlander.  The JOBS Evaluation: 
Early Findings on Program Impacts in Three Sites (New York:  Manpower
Demonstration Research Corporation study for HHS, Sept.  1995). 

Friedlander, Daniel, and Gary Burtless.  Five Years After:  The
Long-Term Effects of Welfare-to-Work Programs (New York:  Russell
Sage Foundation, 1995). 

Gueron, Judith M., and Edward Pauly.  From Work To Welfare (New York: 
Russell Sage Foundation, 1991). 

Hamilton, Gayle.  The JOBS Evaluation:  Monthly Participation Rates
in Three Sites and Factors Affecting Participation Levels in Welfare
to Work Programs (New York:  Manpower Demonstration Research
Corporation study for HHS, Sept.  1995). 

Holzer, Harry J.  Employer Demand, AFDC Recipients, and Labor Market
Policy (Lansing:  Michigan State University, Sept.  1996). 

Masters, Stanley H., and Rebecca Maynard.  The Impact of Supported
Work on Long-Term Recipients of AFDC Benefits (New York:  Manpower
Demonstration Research Corporation study for HHS, Feb.  1981). 

Manpower Demonstration Research Corporation.  Two Approaches to JOBS
Case Management in Columbus, Ohio:  Implementation, Participation,
and Impact Findings (New York:  Manpower Demonstration Research
Corporation, Nov.  1996). 

Mead, Lawrence.  The New Politics of Poverty:  The Non-Working Poor
in America (New York:  Basic Books, 1992). 

Nightingale, Demetra Smith, and Robert H.  Haveman, eds.  The Work
Alternative:  Welfare Reform and the Realities of the Job Market
(Washington, D.C.:  The Urban Institute Press, 1995). 

Olson, Krista, and LaDonna Pavetti.  Personal and Family Challenges
to the Successful Transition from Welfare To Work (Washington, D.C.: 
Urban Institute report prepared for HHS, May 1996). 

Pavetti, LaDonna, and Amy-Ellen Duke.  Increasing Participation in
Work and Work-Related Activities:  Lessons From Five Welfare Reform
Demonstration Projects (Washington, D.C.:  Urban Institute report
prepared for HHS, Sept.  1995). 

Pavetti, LaDonna, Krista Olson, Nancy Pindus, and others.  Designing
Welfare-to-Work Programs for Families Facing Personal or Family
Challenges:  Lessons From the Field (Washington, D.C.:  Urban
Institute Report prepared for HHS, Dec.  1996). 

Wilson, William Julius.  When Work Disappears:  The World of the New
Urban Poor (New York:  Alfred A.  Knopf, 1996). 


RELATED GAO PRODUCTS
============================================================ Chapter 2

Welfare Reform:  States' Early Experiences With Benefit Termination
(GAO/HEHS-97-74, May 15, 1997). 

Welfare Waivers Implementation:  States Work to Change Welfare
Culture, Community Involvement, and Service Delivery
(GAO/HEHS-96-105, July 2, 1996). 

Employment Training:  Successful Projects Share Common Strategy
(GAO/HEHS-96-108, May 7, 1996). 

Welfare to Work:  Approaches That Help Teenage Mothers Complete High
School (GAO/HEHS/PEMD-95-202, Sept.  29, 1995). 

Welfare to Work:  Child Care Assistance Limited; Welfare Reform May
Expand Needs (GAO/HEHS-95-220, Sept.  21, 1995). 

Welfare to Work:  State Programs Have Tested Some of the Proposed
Reforms (GAO/PEMD-95-26, July 14, 1995). 

Welfare to Work:  Most AFDC Training Programs Not Emphasizing Job
Placement (GAO/HEHS-95-113, May 19, 1995). 

Welfare to Work:  Participants' Characteristics and Services Provided
in JOBS (GAO/HEHS-95-93, May 2, 1995). 

Welfare to Work:  Measuring Outcomes for JOBS Participants
(GAO/HEHS-95-86, Apr.  17, 1995). 

Welfare to Work:  Current AFDC Program Not Sufficiently Focused on
Employment (GAO/HEHS-95-28, Dec.  19, 1994). 

Child Care:  Current System Could Undermine Goals of Welfare Reform
(GAO/HEHS-94-238, Sept.  20, 1994). 

Families on Welfare:  Sharp Rise in Never-Married Women Reflects
Societal Trend (GAO/HEHS-94-92, May 31, 1994). 

Families on Welfare:  Teenage Mothers Least Likely to Become
Self-Sufficient (GAO/HEHS-94-115, May 31, 1994). 

Families on Welfare:  Focus on Teenage Mothers Could Enhance Welfare
Reform Efforts (GAO/HEHS-94-112, May 31, 1994). 


*** End of document. ***