Welfare Reform: States Are Restructuring Programs To Reduce Welfare
Dependence (Chapter Report, 06/17/98, GAO/HEHS-98-109).

Pursuant to a congressional request, GAO reviewed states' implementation
of the Temporary Assistance for Needy Families (TANF) block grant,
focusing on: (1) states' efforts to require and encourage welfare
recipients and potential recipients to assume greater personal
responsibility; (2) how states are providing services to support the
objectives of TANF; and (3) early reported data to assess states'
progress in achieving program objectives.

GAO noted that: (1) consistent with the thrust of the federal welfare
reform law, states are moving away from a welfare system focused on
entitlement to assistance to one that emphasizes finding employment as
quickly as possible and becoming more self-sufficient; (2) in the seven
states GAO visited, welfare offices are generally being transformed into
job placement centers; (3) adults with mental and physical impairments
and those caring for small children are less likely than before to be
exempt from participating in work activities; (4) in the states GAO
reviewed, the average proportion of adult recipients required to
participate in work activities increased; (5) to reinforce the
expectation that welfare is temporary, states have established time
limits on receiving cash assistance and have modified various policies
to help make welfare recipients financially better off if they obtain
jobs than if they do not; (6) states also have devised strategies to
reduce the need for monthly cash assistance; (7) states also have
modified their programs to better support welfare recipients in becoming
more self-sufficient; (8) in their efforts to change the culture of
welfare offices, states are expanding welfare workers' roles by shifting
their priorities from determining eligibility and cash assistance levels
to helping recipients obtain work and become more self-sufficient; (9)
at the same time, states plan to use some of the additional budgetary
resources available under the welfare reform law to enhance support
services; (10) moreover, some states have given local administrative
entities greater flexibility to design welfare-to-work programs tailored
to the needs of their recipients; (11) implementing all these changes
has not been quick or easy: among the most challenging and widespread
implementation issues reported by the states have been training staff to
perform their new roles and finding ways to involve recipients with
multiple barriers to participation in work activities; (12) it is too
early to draw definitive conclusions about the success of states'
programs because it is uncertain how states' programs will perform as
more of the most job-ready recipients leave welfare and states face
increasing proportions of recipients with multiple problems, or if the
current strong economy undergoes a major downturn; (13) moreover, little
is known about program impacts; and (14) future monitoring of states'
programs will need to focus on areas such as job retention and earnings
progression, children's welfare, and family stability.

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

 REPORTNUM:  HEHS-98-109
     TITLE:  Welfare Reform: States Are Restructuring Programs To Reduce 
             Welfare Dependence
      DATE:  06/17/98
   SUBJECT:  Block grants
             Disadvantaged persons
             Public assistance programs
             Single parents
             State-administered programs
             Welfare benefits
             Welfare recipients
             Eligibility determinations
             Workfare
IDENTIFIER:  HHS Temporary Assistance for Needy Families Program
             Aid to Families with Dependent Children Program
             AFDC
             HHS Child Care and Development Fund
             Job Opportunities and Basic Skills Training 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.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


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


Report to the Chairmen, Committee on Finance, U.S.  Senate, and
Subcommittee on Human Resources, Committee on Ways and Means, House
of Representatives

June 1998

WELFARE REFORM - STATES ARE
RESTRUCTURING PROGRAMS TO REDUCE
WELFARE DEPENDENCE

GAO/HEHS-98-109

Restructuring State Welfare Programs

(106614)


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

  AFDC - Aid to Families With Dependent Children
  APWA - American Public Welfare Association
  CCDF - Child Care and Development Fund
  CRS - Congressional Research Service
  DOL - Department of Labor
  HHS - Department of Health and Human Services
  HUD - Department of Housing and Urban Development
  JOBS - Job Opportunities and Basic Skills Training
  TANF - Temporary Assistance for Needy Families
  TIES - Texas Integrated Eligibility Services

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


B-276981

June 17, 1998

The Honorable William V.  Roth, Jr.
Chairman, Committee on Finance
United States Senate

The Honorable E.  Clay Shaw, Jr.
Chairman, Subcommittee on Human Resources
Committee on Ways and Means
House of Representatives

This report, prepared at your request, is the first of two reports
that examine states' implementation of the Temporary Assistance for
Needy Families (TANF) block grant, which was authorized by Public Law
104-193.  This report focuses on how states are restructuring their
welfare programs to meet the objectives of TANF specified in the law. 
The second report, to be issued later this year, will examine states'
fiscal decisions for their TANF grants to determine whether states
are taking steps to prepare for the effects of future economic
downturns on their welfare programs. 

We are sending copies of this report to the Ranking Minority Member,
Committee on Finance, U.S.  Senate; the Chairman and Ranking Minority
Member, Committee on Ways and Means, and the Ranking Minority Member
of 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 Gale C.  Harris, Assistant
Director, at (202) 512-7235.  Other GAO contacts and staff
acknowledgments for this report are listed in appendix VI. 

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) made sweeping changes to the nation's cash
assistance program for needy families with children.  Title I of the
law replaced the Aid to Families With Dependent Children (AFDC)
program with fixed block grants to the states to provide Temporary
Assistance for Needy Families (TANF) and ended the entitlement of
families to assistance.  In fiscal year 1996, AFDC paid benefits of
over $20 billion in combined state and federal funds to a nationwide
caseload that averaged about 4.6 million families a month.  As
specified by the new law, the goals of TANF include ending welfare
dependence by promoting job preparation, work, and marriage;
preventing and reducing the incidence of out-of-wedlock pregnancies;
encouraging the formation and maintenance of two-parent families; and
providing states increased flexibility to help them achieve these
goals.  Among other provisions, the law requires that, to avoid
financial penalties, states must impose work requirements for adults,
meet steadily rising requirements for the percentage of adults that
must participate in work activities, and enforce a 5-year lifetime
limit on receiving federal assistance. 

At the request of the Chairmen of the Senate Committee on Finance and
the House Committee on Ways and Means' Subcommittee on Human
Resources, this report (1) describes states' efforts to require and
encourage welfare recipients and potential recipients to assume
greater personal responsibility, (2) examines how states are
providing services to support the objectives of TANF, and (3) reviews
early reported data to assess states' progress in achieving program
objectives. 


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

Welfare reform gives states flexibility to design their own programs
and strategies for achieving program goals, including how to help
welfare recipients move into the workforce.  At the same time, states
must meet federal requirements that emphasize the importance of work
for those receiving assistance.  To avoid federal financial
penalties, in fiscal year 1997 states must ensure that adult
recipients in 25 percent of all their TANF families and 75 percent of
their TANF two-parent families are engaged in work activities.  These
rates increase in subsequent years.  To be counted in states'
participation rates, adults must participate a specified minimum
numbers of hours per week in work activities, such as unsubsidized
employment, on-the-job training, job search and job readiness
assistance, community service, and vocational educational training. 
If adults fail to participate as required, states must reduce their
cash assistance and may terminate assistance for the entire family. 
The Department of Health and Human Services (HHS) is the primary
federal agency providing oversight of states' welfare programs.  HHS'
responsibilities include developing regulations and assessing
penalties for noncompliance with the law. 

Because many states had already begun experimenting with changing
their AFDC programs through waivers of federal law, states were at
different stages of implementing their reform efforts when the
federal legislation was enacted.  For example, Oregon implemented its
welfare reform program statewide under waivers in mid-1996, while
California did not enact welfare reform legislation until mid-1997. 

To obtain information for this request, GAO selected seven states for
in-depth tracking on the basis of various program indicators and
demographic characteristics:  California, Connecticut, Louisiana,
Maryland, Oregon, Texas, and Wisconsin.  In these states, GAO
collected and analyzed program data as well as interviewed state and
local officials and members of community advisory committees and
advocacy groups.  In addition, to provide information on other
states--and on all 50 states, when data were available--GAO analyzed
data collected by HHS and other organizations.  In developing the
methodology for this study, GAO consulted with its Welfare Reform
Advisory Committee, which is composed of 11 experts in this field
(see app.  I for a list of members). 


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

Consistent with the thrust of the federal welfare reform law, states
are moving away from a welfare system focused on entitlement to
assistance to one that emphasizes finding employment as quickly as
possible and becoming more self-sufficient.  In the seven states GAO
visited, welfare offices are generally being transformed into job
placement centers, and in some instances applicants are expected to
engage in job search activities as soon as they apply for assistance. 
Adults with mental and physical impairments and those caring for
small children are less likely than before to be exempt from
participating in work activities, and adults who fail to participate
as required are more likely to have their family's assistance
terminated.  In the states GAO reviewed, the average proportion of
adult recipients required to participate in work activities increased
from 44 percent in 1994 to 65 percent in 1997.  In addition, to
reinforce the expectation that welfare is temporary, states have
established time limits on receiving cash assistance--in some cases
shorter than 5 years--and have modified various policies to help make
welfare recipients financially better off if they obtain jobs than if
they do not.  States also have devised strategies to reduce the need
for monthly cash assistance, such as providing one-time, lump-sum
payments in lieu of monthly payments and enhancing their efforts to
reduce the number of out-of-wedlock pregnancies. 

States also have modified their programs to better support welfare
recipients in becoming more self-sufficient.  In their efforts to
change the culture of welfare offices, states are expanding welfare
workers' roles by shifting their priorities from determining
eligibility and cash assistance levels to helping recipients obtain
work and become more self-sufficient.  At the same time, states are
using some of the additional budgetary resources available under the
welfare reform law to enhance support services, such as
transportation and child care, for recipients participating in work
activities and poor families who have found jobs and left the welfare
rolls.  In addition, states are working to enhance their capacity to
treat physical and mental health problems.  In Oregon, state
officials estimated that about 50 percent of the welfare caseload
requires drug or alcohol treatment.  Moreover, some states have given
local administrative entities greater flexibility to design
welfare-to-work programs tailored to the needs of their recipients. 
Implementing all these changes has not been quick or easy:  among the
most challenging and widespread implementation issues reported by the
states have been training staff to perform their new roles and
finding ways to involve recipients with multiple barriers to
participation, such as mental and physical health problems and low
literacy levels, in work activities. 

Nationwide, welfare dependence has decreased.  Welfare caseloads
decreased by 30 percent between January 1994 and September 1997--and
decreased by a larger percentage each year during this period.  In
addition, GAO's analysis showed that the seven states reviewed have
generally increased their job placement rates.  While these results
are promising, it is too early to draw definitive conclusions about
the success of states' programs because it is uncertain how states'
programs will perform as more recipients leave welfare for work and
states face increasing proportions of recipients with multiple
problems, or if the current strong economy undergoes a major
downturn.  Moreover, little is known about program impacts, such as
the effect the programs have had on the well-being of children and
families.  Future monitoring of states' programs will need to focus
on areas such as job retention and earnings progression, children's
welfare, and family stability. 


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


      STATES' POLICIES ARE
      SHIFTING EMPHASIS FROM
      ENTITLEMENT TO
      SELF-SUFFICIENCY
-------------------------------------------------------- Chapter 0:4.1

States have modified their policies to require and encourage welfare
recipients and potential recipients to seek work and become more
self-sufficient.  For example, in all seven of the states GAO
visited, the proportion of recipients assigned to job placement
activities--as opposed to education or training activities--was
substantially higher in 1997 than in 1994, and the proportion
assigned to job placement activities more than quadrupled in
Connecticut and Louisiana during this time.  All seven states now
require nonexempt recipients to participate in work activities
immediately upon applying for assistance or as soon as possible
thereafter.  In addition, five of the seven states have strengthened
their sanctions by adopting provisions for terminating the assistance
of the entire family for noncompliance with work requirements. 

As part of an effort to "make work pay," 42 states have changed their
policies relating to the treatment of earned income from those
previously in effect under AFDC to permit recipients to keep more of
their monthly cash assistance payments or retain them for longer
periods once they begin working.  Nearly all states have increased
the amount of assets or the value of a vehicle that recipients can
own and still remain eligible for cash assistance.  The asset and
vehicle limits under the AFDC program were widely considered to be
too low, creating barriers to families' efforts to become more
self-sufficient. 

States also have adopted varying time limits on the receipt of cash
assistance.  Nineteen states have established policies to terminate
assistance for some families sooner than the 5 years specified by the
federal law, but these states generally have also adopted policies to
extend assistance beyond these limits in certain circumstances.  For
example, recipients in Connecticut can receive 6-month extensions to
the state's 21-month time limit if they have made a good faith effort
to comply with work requirements but have been unable to find
employment.  Tracking the time families receive cash assistance poses
significant challenges for many states, given their need to upgrade
their automated information systems to collect all of the data
required for such an undertaking.  While states have set time limits
on eligibility for cash assistance, they generally have not used
their flexibility under TANF to reduce cash assistance levels or deny
eligibility to specific groups of people, except for convicted drug
felons. 

In addition, states are pursuing various strategies to reduce the
need for welfare.  As of November 1997, 30 states had reported that
they were using "diversion," a major new strategy that seeks to
divert some applicants from monthly cash assistance by providing
other forms of assistance, such as one-time, lump-sum payments;
support services such as child care or Medicaid; and assistance with
job search.  To enhance the collection of child support, the welfare
law requires states to reduce the amount of families' cash assistance
by at least 25 percent for noncooperation with child support
enforcement requirements.  Sixteen states have adopted stronger
provisions that call for terminating families' entire cash assistance
payments on this basis.  Finally, recognizing the strong link between
teenage childbearing and welfare dependence, states are enhancing
existing pregnancy prevention programs, especially for teens. 
Strategies being used by our case study states include abstinence
education, stronger enforcement of statutory rape laws, and male
involvement programs. 


      STATES ARE ENHANCING SUPPORT
      SERVICES FOR RECIPIENTS
-------------------------------------------------------- Chapter 0:4.2

As states seek to expand the number of adults participating in work
activities, they have generally expanded the roles of welfare workers
to better support the work focus of their programs.  Workers' new
responsibilities vary but include such tasks as motivating clients to
seek work, exploring the potential for diversions, and collecting
more information about applicants and recipients to determine what
they need to facilitate self-sufficiency.  Training workers to
perform these broader responsibilities has been especially
challenging because of the need to help workers change their
perspectives and help them cope with workload pressures.  For
example, Oregon responded to such challenges, in part, by
streamlining paperwork for determining eligibility to allow staff
more time for new responsibilities. 

As a result of their large, recent caseload declines, most states
have more budgetary resources for their welfare programs under the
TANF funding formula than they would have had under prior law, under
which funding was tied to caseload size.  The seven states GAO
reviewed are using some of the additional available budgetary
resources to provide services to help families address barriers to
employment.  For example, these states are using a range of
approaches to help recipients obtain reliable transportation, such as
providing funding for rural transportation systems, enlisting
volunteers to provide transportation for recipients, and providing
funds for vehicle repairs.  Some states have enhanced services by
providing mentors or making case management available to those who
have left welfare for jobs.  As the most readily employable
recipients leave welfare, states and localities are concerned that
they will face a more difficult to serve population.  In response to
such concerns, Baltimore added a social service component to its
welfare reform program, and Oregon placed counselors on site to
provide mental health and substance abuse services.  In addition, the
seven states GAO visited have used federal and state funding to
increase overall expenditures for their fiscal year 1997 child care
subsidy programs for TANF and other low-income families, with
increases over fiscal year 1996 expenditures ranging from about 2
percent in Maryland to 62 percent in Louisiana. 

As welfare reform has provided states greater flexibility, some of
GAO's case study states, in turn, have given local administrative
entities greater flexibility to design programs tailored to the needs
of their recipients.  While policies regarding eligibility and cash
assistance levels in these states continue to be set at the state
level, local administrative entities now have more flexibility to
customize their policies for moving recipients from welfare to work. 
To promote local accountability, these states are using methods such
as creating financial incentives and establishing performance
measures that focus more on desired outcomes.  For example,
California's welfare reform law stipulates that counties are to
receive financial bonuses on the basis of their cost savings from
recipients leaving welfare because of employment that lasted at least
6 months, increased earnings by recipients because of employment, and
diversion of applicants from welfare for at least 6 months. 


      WELFARE DEPENDENCE HAS
      DECREASED, BUT LITTLE IS
      KNOWN ABOUT IMPACTS ON
      FAMILIES
-------------------------------------------------------- Chapter 0:4.3

While the number of families receiving cash assistance nationwide
decreased 30 percent between January 1994 and September 1997, more
than two-thirds of this decrease has occurred since January 1996. 
GAO estimated that three of the case study states more than doubled
their job placement rates from 1995 to 1997, and two of them
increased their rates by more than 70 percent.  In addition, the
seven states generally have increased the percentages of families
participating in welfare-to-work programs under TANF compared with
their prior welfare-to-work programs.  While all seven states
reported that they would meet their required TANF all-families
participation rates for fiscal year 1997, two states reported that
they would not meet their required rates for two-parent families. 

In many states, favorable economic conditions appear to have
facilitated implementation of "work first" approaches.  It is not yet
known, however, how states' welfare reform programs will perform
under weaker economic conditions.  In addition, as indicated by
states that have experienced large caseload reductions, many of the
remaining recipients have multiple problems that interfere with their
ability to work.  So far, little is known about how effective states
will be in helping these families become more self-supporting. 

Despite early indications of progress toward key goals of welfare
reform, much remains unknown about how families fare after leaving
welfare with respect to economic stability and child and family
well-being.  However, some states have efforts under way to obtain
information on such topics.  For example, Maryland is tracking a
random sample of families that have exited welfare to provide
information on topics including employment and earnings, welfare
recidivism, and receipt of foster care.  In addition, concerned about
the importance of having national data on the impacts of states'
welfare reforms, the Congress included provisions in the welfare
reform law that direct HHS to conduct research on the benefits,
costs, and effects of state programs funded under TANF, as well as
mandate that the Bureau of the Census expand a national survey of
families to permit an evaluation of the law's impacts. 


   COMMENTS FROM HHS AND THE
   STATES
---------------------------------------------------------- Chapter 0:5

GAO obtained comments on a draft of this report from HHS and the
seven case study states.  HHS and the states generally agreed with
the report's findings and provided additional technical information
that GAO incorporated in the report as appropriate.  (Ch.  5 contains
additional information about HHS and the states' comments and GAO's
responses.)


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

The Personal Responsibility and Work Opportunity Reconciliation Act
(P.L.  104-193), enacted in August 1996, replaced AFDC with a block
grant to the states, entitled Temporary Assistance for Needy
Families.  As specified in the law, the objectives of TANF are to

  -- provide assistance to needy families so that children may be
     cared for in their own homes or in the homes of relatives;

  -- end the dependence of needy parents on government benefits by
     promoting job preparation, work, and marriage;

  -- prevent and reduce the incidence of out-of-wedlock pregnancies
     and establish annual numerical goals for preventing and reducing
     the incidence of these pregnancies; and

  -- encourage the formation and maintenance of two-parent families. 

All these goals are to be accomplished by providing states increased
flexibility to design their own programs, thus allowing states to
build on the initiatives they began experimenting with before federal
reform.  However, the law sets some parameters:  to avoid financial
penalties, states must impose work requirements for adults, meet
steadily rising rates of caseload participation in work activities,
and enforce a 5-year lifetime limit on the receipt of federally
funded assistance, along with several other provisions.  To enable
HHS to monitor states' progress in meeting key program objectives,
the law also requires states to meet new data reporting requirements. 
On November 20, 1997, HHS issued proposed regulations for
implementing TANF in accordance with the new federal law; however, as
of May 1998, these regulations had not yet been finalized.\1


--------------------
\1 HHS officials noted that while states were permitted to make their
own reasonable interpretations of the 1996 welfare reform law before
the publication of final rules for the TANF program, some of the
state interpretations of the statute reflected by our report may not
be sustained when HHS issues final rules. 


   THE WELFARE SYSTEM BEFORE
   REFORM
---------------------------------------------------------- Chapter 1:1

Under AFDC, states were required to provide benefits to all
economically needy families with children who applied and were
eligible under federal law and whose income and assets were within
state-prescribed limits.  In fiscal year 1996, over $20 billion in
combined federal and state funds were paid in cash assistance to a
nationwide caseload that averaged about 4.6 million families a month. 
Within AFDC, states were required to establish welfare-to-work
programs, called Job Opportunities and Basic Skills Training (JOBS),
to ensure that AFDC families obtained the education, training, and
employment that would help them avoid long-term welfare dependence. 
Federal matching funds for JOBS were available as a capped
entitlement set at $1 billion in fiscal year 1996.  Historically,
many families were exempt from participating in JOBS, and less than a
third of those required to participate actually were able to
participate because of limited state funding.\2 In addition, within
AFDC, unlimited federal matching funds were available for states
opting to provide emergency assistance for families to avoid the
destitution of a child or to provide for a child's living
arrangements.  Emergency assistance could be authorized for up to 30
days in any 12-month period, and eligibility criteria often differed
from those of the regular AFDC program.  As of 1995, all states but
Alaska and Mississippi had implemented emergency assistance programs,
spending about $3.2 billion on these programs nationwide in fiscal
year 1996.  To ensure states adhered to the goals and requirements of
the law governing these programs, HHS was given oversight
responsibility to be exercised through its approval, review, and
audit functions. 


--------------------
\2 See Welfare to Work:  Current AFDC Program Not Sufficiently
Focused on Employment (GAO/HEHS-95-28, Dec.  19, 1994), pp.  5-8. 


   MOST STATES' REFORM EFFORTS
   PRECEDED FEDERAL REFORM
---------------------------------------------------------- Chapter 1:2

Before passage of the new federal welfare reform law, HHS had
authority to waive certain statutory program requirements for AFDC,
and most states had begun experimenting with various reforms intended
to move more recipients from welfare to work.  Between January 1987
and the passage of welfare reform in August 1996, 46 states\3 had
received approval to implement waivers affecting their AFDC and JOBS
programs--in some cases, statewide; in others, only in selected
sites.  These waiver initiatives included time limits, strengthened
work requirements, and teen-parent requirements related to school
attendance and living in supervised settings--provisions similar to
those subsequently embodied in the new federal law.  (See app.  II
for a summary of states' waiver provisions approved before passage of
the federal welfare reform act.)

Because these waivers had been approved and implemented at different
times, states were in different stages of implementing their reform
efforts when the federal reform law was enacted.  Some states
continued their waiver programs implemented before federal reform,
while other states enacted new legislation to redesign their programs
after passage of the new federal law.\4 Oregon, for example, enacted
legislation in 1995 and implemented its welfare reform program
statewide in July 1996.  In contrast, California did not pass welfare
reform legislation until mid-1997, with statewide implementation of
its new program effective in January 1998. 


--------------------
\3 The term "state" includes the District of Columbia in this report. 

\4 Under the new federal reform law, states' waiver provisions that
are inconsistent with the new law may continue for the duration of
the waiver if they were already in effect before the new law was
enacted.  However, all states--regardless of any waivers--are subject
to certain requirements, such as the mandatory work requirements and
the calculation of participation rates, discussed later in this
chapter. 


   KEY PROVISIONS OF FEDERAL
   REFORM
---------------------------------------------------------- Chapter 1:3

The 1996 federal welfare reform law made sweeping changes to the
nation's cash assistance program for needy families with children. 
Title I of the law ended the entitlement of families to welfare
benefits and replaced the AFDC program, including JOBS and emergency
assistance, with block grants to the states under TANF.  The fixed
amounts of states' grants under the new law are based on the amount
of their grants received in specified fiscal years under prior law,
supplemented for population increases under certain circumstances.\5
For fiscal year 1997, federal grants available to the states ranged
from $21.8 million in Wyoming to over $3.7 billion in California,
totaling $16.7 billion nationwide.  With respect to state funding,
the federal reform law included a "maintenance-of-effort" provision
requiring states to provide 75 to 80 percent of their historic level
of funding.\6 Subsequent legislation provided additional federal
funds totaling $3 billion over 2 years for Welfare-to-Work Grants to
be allocated to states for increased support of activities helping to
place and keep individuals in unsubsidized jobs.\7

Federal law also makes federal funding available to states for child
care subsidies for low-income families, authorizing $2.9 billion for
fiscal year 1997 and up to $3.7 billion in the year 2002.  Under the
new reform law, states are required to ensure that a significant
percentage of these funds are used to provide child care assistance
to current or potential TANF recipients.\8

The federal welfare reform law also made significant changes to
Medicaid--a federal/state-funded program that provides medical
assistance to low-income families.  Before welfare reform, AFDC
recipients were automatically enrolled in Medicaid on the basis of
their eligibility for cash assistance under AFDC.  The new law
severed the connection between eligibility standards for Medicaid and
cash assistance, allowing states to set their own eligibility
standards for Medicaid within certain parameters.  However, most
families who would have been eligible for Medicaid before welfare
reform continue to qualify for services.\9


--------------------
\5 Under the new welfare reform law, states' grants are based on the
greatest of three options for determining the amount of their grants: 
(1) the average amount of grants received for fiscal years 1992,
1993, and 1994; (2) the amount of grants received for fiscal year
1994 (with some adjustments for states with high expenditures for
emergency assistance in fiscal year 1995); or (3) a formula based on
grants received for fiscal year 1995. 

\6 Under the new welfare reform law, states that meet federally
mandated minimum participation rates must provide at least 75 percent
of their historic level of funding; states that fail to meet mandated
rates must provide at least 80 percent.  (Participation rates are
discussed later in this chapter.)

\7 The Balanced Budget Act of 1997 authorized funds for
Welfare-to-Work Grants for fiscal years 1998 and 1999.  After certain
set-asides, 75 percent of the funds are to be made available in
formula grants and 25 percent in competitive grants.  To qualify,
states must fund their programs above their required TANF
maintenance-of-effort level.  For more details, Mark Greenberg,
Welfare-to-Work Grants and Other TANF-Related Provisions in the
Balanced Budget Act of 1997 (Washington, D.C.:  Center for Law and
Social Policy, Aug.  1997). 

\8 These funds are provided through the Child Care and Development
Fund, which combines the funding streams of what had previously been
four different programs and includes both discretionary and
entitlement funds (for more details, see ch.  3).  At least 70
percent of the entitlement funds must be used for current or
potential TANF families.  In fiscal year 1997, entitlement funds
equaled almost $2 billion, or about two-thirds of the total.  See
Welfare Reform:  States' Efforts to Expand Child Care Programs
(GAO/HEHS-98-27, Jan.  13, 1998). 

\9 To ensure continued Medicaid coverage for low-income families, the
law generally set Medicaid eligibility standards at AFDC levels in
effect on July 16, 1996.  The law provides the option to deny medical
assistance to individuals who fail to meet the work requirements but
does not permit states to deny medical assistance to minor children
who are not the head of a household.  For information on the
Medicaid-related actions states have taken since welfare reform, see
Medicaid:  Early Implications of Welfare Reform for Beneficiaries and
States (GAO/HEHS-98-62, Feb.  24, 1998). 


      MOVING FROM DEPENDENCY TO
      SELF-RELIANCE
-------------------------------------------------------- Chapter 1:3.1

Because of congressional concern that welfare had become a way of
life for some recipients, a key purpose of the new law was to promote
work over welfare and self-reliance over dependency.  In support of
this goal, the law provides that states must require able-bodied
recipients to participate in work or work-related activities and must
impose a 5-year lifetime limit on federal assistance. 

States must require adults in families receiving TANF-funded
assistance to participate in work or work-related activities after
receiving assistance for 24 months, or sooner, as defined by the
state.  If recipients fail to participate as required, states must at
least reduce the families' grant and may opt to terminate the grant
entirely. 

To avoid financial penalties, states must ensure that a certain
specified minimum percentage of their caseloads are participating in
work or work-related activities each year.  These percentages are
referred to as "minimum mandated participation rates." To count
toward states' mandated rates, adult recipients in families must
participate a certain minimum number of hours in work or a
work-related activity as prescribed in the law--such as job readiness
workshops; on-the-job training; and, under certain circumstances,
education.  The required number of hours of participation and the
percentage of a state's caseload that must participate to meet
mandated rates increase over time, as shown in table 1.1. 



                         Table 1.1
          
                Federal Law Sets Increasing
            Participation Requirements for One-
               Parent and Two-Parent Families

                             Fiscal year
            ----------------------------------------------
              1997    1998    1999    2000    2001    2002
----------  ------  ------  ------  ------  ------  ------
Minimum weekly average participation requirement (hours)
----------------------------------------------------------
One-            20      20      25      30      30      30
 parent
 families
Two-            35      35      35      35      35      35
 parent
 families\
 a

Minimum mandated participation rates (percentage)
----------------------------------------------------------
All             25      30      35      40      45      50
 families
Two-            75      75      90      90      90      90
 parent
 families
----------------------------------------------------------
\a To receive federally funded child care assistance, two-parent
families must participate for a combined total of at least 55 hours a
week. 

Source:  42 U.S.C.  sec.  607. 

Finally, to help ensure the temporary nature of assistance and to
provide further impetus for moving recipients to self-reliance, the
law prohibits the use of TANF funds to provide assistance for
families with adults who have received assistance for more than 5
years, cumulative over their lifetimes.\10 Families with no adult
receiving assistance (commonly referred to as "child-only" cases) are
not subject to this limit, and up to 20 percent of a state's average
monthly caseload may be exempt on the basis of hardship or having
been subjected to domestic violence.\11 States with time-limit
waivers inconsistent with these provisions may delay implementing the
5-year lifetime limit until their waivers expire.  Also, states may
opt to continue to provide assistance beyond the 5-year limit using
state funds. 


--------------------
\10 Months that the adult may have received assistance as a minor
dependent (not head-of-household) do not count toward this lifetime
limit. 

\11 The federal reform law allows states to exempt a family from the
time limit by reason of hardship or if the family includes an
individual who has been battered or subjected to extreme cruelty and
to elect as the base the average monthly caseload for either the
current fiscal year or the preceding fiscal year. 


      REDUCING OUT-OF-WEDLOCK
      PREGNANCIES
-------------------------------------------------------- Chapter 1:3.2

To address concerns about the growing number of children born to
unwed mothers and the impact on welfare caseloads, another key
purpose of the new law was to reduce out-of-wedlock pregnancies and
encourage the formation and maintenance of two-parent families.\12 To
this end, the law included provisions requiring HHS and the states to
establish goals and take action to reduce the incidence of
out-of-wedlock pregnancies, especially among teens, and to provide
annual progress reports to the Congress beginning this month.\13

In addition, citing studies that indicate that the increase of
teenage pregnancies among the youngest girls has been particularly
severe and linked to predatory sexual practices of men who were
significantly older, the law also requires states to conduct programs
to provide education and training on the problem of statutory rape. 
To expand their reach to males, these programs are to be provided
through law enforcement agencies, the education system, and other
relevant counseling services. 


--------------------
\12 From 1976 to 1992, the proportion of single women receiving AFDC
who had never been married increased from about 21 percent to about
52 percent.  See Families on Welfare:  Sharp Rise in Never-Married
Women Reflects Societal Trend (GAO/HEHS-94-92, May 31, 1994). 

\13 We have another study under way that will address the issues
around teen pregnancy in more detail. 


      PROVIDING INCREASED STATE
      FLEXIBILITY AND
      ACCOUNTABILITY
-------------------------------------------------------- Chapter 1:3.3

Instead of prescribing in detail how programs are to be structured,
the new law authorizes states to use their block grants in any manner
reasonably calculated to accomplish the purposes of TANF.  For
example, states are allowed to set forth their own criteria for
defining who will be eligible and what assistance and services will
be available, provided states ensure fair and equitable treatment. 
States may opt to deny assistance altogether for noncitizens, drug
felons, minor teen parents, or those determined to be able to work. 
Alternatively, these groups could be provided a different array of
assistance and services funded by TANF or a separate state-funded
program.  States may also choose when to require adults to
participate in work activities, what types of activities are allowed,
whom to grant good cause for failure to participate, and whether or
not to terminate grants to entire families for noncompliance.\14

The law balanced this increase in state flexibility, however, with an
increase in state accountability for working toward the goals of
reform and HHS' responsibilities for tracking state performance and
family outcomes.  To enable HHS to rank the states annually on
performance and make decisions concerning bonuses and penalties,
states are required to submit a detailed quarterly data report. 
Using the state-reported data, HHS is required to rank the states
annually according to the most and least successful work programs,
taking into account (1) placements in long-term private sector jobs,
(2) overall caseload reductions, (3) diversion of individuals from
applying for and receiving assistance, (4) the number of children
living in poverty, and (5) the amount of federal assistance provided
to the state.  HHS is also required to rank the states annually on
the basis of the percentage of out-of-wedlock births in families
receiving assistance and the reductions in the percentage of
out-of-wedlock births from the prior year. 

These state-reported data also provide the basis for awarding
financial bonuses to the states to promote key program objectives. 
The federal reform law provides $200 million per year for 5 years for
bonuses to reward states with high performance in achieving the goals
of TANF.  States may be awarded bonuses in amounts up to 5 percent of
their grants in fiscal years 1999 through 2003.\15 The law also
provides $100 million per year for 4 years for bonuses to reward
states that demonstrate net decreases in the number of out-of-wedlock
births.  Up to five states may be awarded these bonuses of $20
million in fiscal years 1999 through 2002.  (If fewer than five
states qualify, bonuses are increased to $25 million.) The law
specifies that the reduction in the number of out-of-wedlock births
is to be determined for the most recent 2-year period for which such
information is available and that only those states with no increase
in abortion rates since 1995 are eligible for the bonuses.\16

Data from the quarterly report are also to be used to determine
whether or not the state will be assessed a penalty.  The law, as
amended, specifies 14 grounds for penalties, including failure to
meet the maintenance-of-effort state funding requirement, failure to
satisfy the minimum mandated participation rates, failure to
implement the 5-year lifetime time limit, and failure to submit a
quarterly data report.  The total penalty in a single year can range
up to 25 percent of a state's grant.  (See app.  III for a
description of all 14 penalties.) Under some circumstances, HHS may
determine that a state has "reasonable cause" for failing to meet
some requirement and will not impose a penalty.  Proposed regulations
indicate, however, that if a state diverts cases to separate state
programs or is continuing with inconsistent provisions under waiver,
it may not be able to avoid a penalty through a reasonable cause
exception.\17

In addition to the quarterly report, states are required to submit
their child poverty rates annually, along with a corrective action
plan if the rate increases by 5 percent or more from the prior year. 
And beginning in 1999, HHS is required to submit annual reports on
the circumstances of families reaching their time limits and families
headed by teen parents. 

To ensure evaluation of states' programs under welfare reform, HHS is
mandated to conduct research on the costs and benefits of operating
different state programs, including effects on welfare dependency,
illegitimacy, teen pregnancy, employment rates, child well-being, and
any other area deemed appropriate.  HHS must also evaluate any
innovative approaches it assists states in developing and may help
fund states' evaluation efforts.  Finally, HHS is required to work
with the states to study and analyze outcome measures for evaluating
the success of states' efforts to move families from welfare to work
as alternatives to the minimum mandated participation rates. 


--------------------
\14 However, the law defines the types of activities that may count
toward the state's mandated participation rate, as well as the cases
that must be included in the calculation.  The law also stipulates
that if a one-parent family with a child under age 6 is unable to
obtain needed child care, the state may not sanction the family for
noncompliance with the work requirement, and failure to maintain
assistance to such families is grounds for a penalty of up to 5
percent of the state's grant. 

\15 The formula for measuring high performance was to be defined by
HHS in consultation with the National Governors' Association and the
American Public Welfare Association no later than August 22, 1997. 
In March 1998, HHS published the formula for awarding
high-performance bonuses in fiscal year 1999 and noted that a notice
of proposed rule-making was being drafted that would address the
applicable formula in future years.  (See ch.  4 for further
discussion of the high performance bonus.)

\16 In March 1998, HHS issued proposed regulations defining the data
to be used in these calculations.  (See ch.  2 for further discussion
of the out-of-wedlock bonus.)

\17 For example, under provisions of the proposed rule, exceptions
will not be made for failure to meet minimum participation rates in
states that divert cases to separate state programs to avoid the work
participation requirements, or that are continuing to use alternative
work requirements adopted under waivers, such as allowing a broader
range of activities to count toward their mandated rates.  However,
the proposed rule is not binding, and the final rule may be
different. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:4

The Chairmen of the Senate Committee on Finance and the Subcommittee
on Human Resources, House Committee on Ways and Means, asked us to
monitor and report on states' efforts to implement programs to meet
the stated objectives of title I of the new federal welfare reform
law.  Specifically, this report (1) describes states' efforts to
require and encourage welfare recipients and potential recipients to
assume greater personal responsibility, (2) examines how states are
providing services to support the objectives of TANF, and (3) reviews
early reported data to assess states' progress in achieving program
objectives. 

To obtain information for this request, we judgmentally selected for
in-depth tracking seven states that reflect a diversity of
circumstances based on various program indicators and demographics: 
California, Connecticut, Louisiana, Maryland, Oregon, Texas, and
Wisconsin.  Key factors considered in our selection of states
included previous AFDC grant levels, AFDC funds per child in poverty,
unemployment rates, child poverty rates, urban/rural population,
presence of large cities, geographic region, experience with waivers,
and whether programs are administered at the state level or by
counties.  Table 1.2 provides a summary profile of each state
selected. 



                                         Table 1.2
                          
                          Summary Profile of the Seven Case Study
                                           States

                                                                 AFDC
                                                                funds
                        Maximum                         Child     per
                           AFDC   Average              povert   child            Date
                          grant   monthly              y rate      in   Percent  statewide
                      level for    number  Unemployme  (perce  povert     urban  reform
          Type of      a family   of AFDC     nt rate     nt,       y  resident  program
          administra   of three  families   (percent,  1996)\  (1996)         s  implement
State     tion\a       (1996)\b  (1996)\c     1996)\d       e      \f  (1990)\g  ed\h
--------  ----------  ---------  --------  ----------  ------  ------  --------  ---------
Calif.    County           $596   895,960         7.2    25.5  $3,135      92.6  Jan. 1998

Conn.     State             543    58,117         5.7    22.7   2,587      79.1  Jan. 1996

La.       State             190    70,581         6.7    31.8     545      68.1  May 1997

Md.       State             373    74,106         4.6    16.6   2,113      81.3  Oct. 1996

Oreg.     State             460    33,444         5.9    20.1   1,851      70.5  July
                                                                                 1996\i

Tex.      State\j           188   254,953         5.6    24.4     634      80.3  Jan.
                                                                                 1997\k

Wis.      County\l          517    60,058         3.5    12.5   2,593      65.7  Sept.
                                                                                 1997\m

U.S.                       $394  4,553,30         5.4    20.5  $2,120      75.2
average                                 8
------------------------------------------------------------------------------------------
\a In a state-administered program, the local welfare office (if any)
is a unit of the state agency.  In a county-administered program, the
local welfare office is a unit of the local government. 

\b Maximum grant levels are amounts provided as of July 1996 for a
family of three with no income (see L.  Jerome Gallagher and others,
One Year After Federal Welfare Reform:  A Description of State
Temporary Assistance for Needy Families (TANF) Decisions as of
October 1997 [Washington, D.C.:  The Urban Institute, May 1998], pp. 
VI-1 through VI-3). 

\c Data on number of families receiving AFDC are from HHS,
Administration for Children and Families, Characteristics and
Financial Circumstances of AFDC Recipients, Fiscal Year 1996
(Washington, D.C.:  HHS, Apr.  1997). 

\d Data represent 1996 annual average unemployment rates from the
Bureau of Labor Statistics. 

\e Data on child poverty rates are from the U.S.  Bureau of the
Census, 1997 Current Population Survey (Washington, D.C.:  Bureau of
the Census, 1997). 

\f Calculations are based on HHS data on total AFDC and related
program expenditures for fiscal year 1996. 

\g "Urban residents" are defined as living in (1) a city, village, or
other place with 2,500 or more inhabitants or (2) an area with at
least 50,000 inhabitants, comprising one or more places and the
surrounding territory.  Data from the U.S.  Bureau of the Census,
Statistical Abstract of the United States--1997 (Washington, D.C.: 
Bureau of the Census, 1997), table 44. 

\h All states were required to implement TANF programs as of July 1,
1997, unless they opted to continue statewide programs previously
implemented under waivers.  Some states, such as California and
Wisconsin, had made changes to conform to the TANF requirements by
this date but subsequently implemented more fundamental changes to
the structure of their programs. 

\i Before implementing this reform program, Oregon was granted
waivers in 1992 and began implementing provisions to require more
clients to participate in JOBS; teens to finish high school; and JOBS
participants to be in drug, alcohol, or mental health treatment if
needed. 

\j In Texas, eligibility services are administered by the state,
whereas employment and child care services are being transferred to
local workforce development boards. 

\k This program was implemented statewide in all counties with JOBS
programs, covering about 90 percent of the state's welfare
recipients.  It was initially implemented as a pilot in one county in
June 1996 and was expanded to five additional counties in September
1996 before going statewide in January 1997. 

\l However, as a result of Wisconsin's privatization initiative, the
program is administered by private organizations in several counties. 

\m Before implementing this reform program, Wisconsin implemented a
statewide waiver in 1987 requiring teens to finish high school, a
limited waiver in four counties in 1994 encouraging teens to
participate in case management activities promoting family
responsibility, and a statewide waiver in March 1996 requiring
applicants to participate in 60 hours of job-search activities before
qualifying for monthly cash assistance and also requiring recipients
to engage in work activities (including job search) full-time or face
a prorated reduction of cash assistance up to the full amount of the
family's grant. 

To describe states' efforts to design new programs and provide
services, and to assess states' progress in achieving program
objectives, we obtained and analyzed available program documents and
data in each of the case study states.  We also interviewed state and
local program officials and members of welfare advocacy groups in the
states we visited.  We relied on the states for descriptions of their
applicable laws.  Table 1.3 lists the sites visited in each case
study state. 



                         Table 1.3
          
            States, Counties, and Cities Visited

                                              Largest city
                                                population
                                                     (1996
State       County          Largest city       estimate)\a
----------  --------------  --------------  --------------
Calif.      Fresno          Fresno                 396,011
            San             San                    183,474
            Bernardino      Bernardino             838,744
            Santa Clara     San Jose

Conn.       Hartford        Hartford               133,086
            Litchfield      Torrington              34,529

La.         Orleans         New Orleans            476,625
            Evangeline      Ville Platte                \b

Md.         Baltimore       Baltimore              675,401
            (city)\c        Hagerstown              34,633
            Washington

Oreg.       Multnomah       Portland               480,824
            Linn            Albany                  37,919

Tex.        Dallas          Dallas               1,053,292
            Navarro         Corsicana               23,320

Wis.        Fond du Lac     Fond du Lac             39,658
            Milwaukee       Milwaukee              590,503
----------------------------------------------------------
\a Data are from Population Estimates Program, Population Division,
U.S.  Bureau of the Census, Nov.  1997. 

\b Population less than 10,000. 

\c Independent city (not part of a county). 

In addition, we analyzed data collected by HHS and other
organizations to provide information on other states--and on all 50
states when available.  For example, we analyzed HHS data on AFDC
family characteristics and participation in the JOBS program to
better identify states' practices before reform.  We also reviewed,
and included where appropriate, nationwide data on states' TANF
programs gathered and summarized by HHS, the Congressional Research
Service (CRS), the Urban Institute, the Rockefeller Institute, the
American Public Welfare Association (APWA), and the National
Governors' Association. 

To help us formulate the methodology for our study, we consulted with
GAO's Welfare Advisory Committee, comprising 11 nationally recognized
experts in the welfare area.  (See app.  I for a list of members.) To
provide quantitative data on the programs in the seven states we
visited, we relied on data used by these states to manage their
programs.  In addition, we obtained comments on a draft of this
report from the seven states and HHS.  We conducted our work between
February 1997 and April 1998 in accordance with generally accepted
government auditing standards. 


STATES' POLICIES ARE SHIFTING
EMPHASIS FROM ENTITLEMENT TO
SELF-SUFFICIENCY
============================================================ Chapter 2

Consistent with the thrust of the federal welfare reform law of 1996,
states are shifting away from a welfare system that focuses on a
family's entitlement and eligibility determination to one focused on
moving recipients--and potential recipients--to self-sufficiency. 
This shift is reflected in states' efforts to strengthen the work
focus of their programs, modify policies concerning eligibility and
grant levels to encourage self-reliance, and implement strategies to
help reduce the need for welfare. 


   STATES HAVE STRENGTHENED THEIR
   FOCUS ON WORK
---------------------------------------------------------- Chapter 2:1

States have generally strengthened their focus on work as a central
feature of their reform efforts.  Under prior law, limited numbers of
AFDC recipients were referred to the JOBS program to prepare for and
accept employment.  Beginning under waivers and culminating with the
enactment of TANF, the AFDC and JOBS programs have been replaced by a
single, integrated program.  Now, in many states, when people apply
for welfare, they are simultaneously enrolling in a welfare-to-work
program, and welfare offices across the country are being transformed
into job placement centers.  In an effort to place individuals as
quickly as possible into jobs, these new work-focused welfare
programs generally require participation in activities leading more
directly to employment, require a higher percentage of recipients to
participate, require participation sooner, and impose more stringent
sanctions for failure to participate. 

In implementing this transformation of welfare programs to a more
work-focused approach, states have encountered numerous challenges. 
For example, increasing the percentage of recipients required to
participate in work-related activities has required states to change
caseworker roles from emphasizing eligibility determination to
helping recipients address barriers to work by providing more and
different services to those previously exempted from work
requirements.  How states have changed their programs to meet these
challenges is discussed further in chapter 3. 


      RECIPIENTS ARE BEING
      REQUIRED TO PARTICIPATE IN
      ACTIVITIES TIED TO
      EMPLOYMENT
-------------------------------------------------------- Chapter 2:1.1

Under most states' welfare reform programs, recipients are expected
to participate in work or work-related activities leading more
directly to employment, such as job readiness and job search.  In
response to TANF's restrictions on the types of activities that count
toward minimum mandated participation rates, some states are shifting
participants from education and training activities to community
service.  And in a few states, the number of hours recipients are
required to participate has increased to better reflect hours
comparable to full-time employment. 


         "WORK FIRST" APPROACH IS
         BEING ADOPTED IN MANY
         STATES
------------------------------------------------------ Chapter 2:1.1.1

The previous JOBS program provided states with broad discretion
regarding the type of activities assigned recipients; many states
emphasized education and training as much as, if not more than,
employment.\18 The typical sequence of activities involved initial
assessment; development of an "employability plan"; then assignment
to an activity such as education, job skills training, job readiness,
or job development and placement.  Under states' reform programs, and
consistent with the new federal law, greater emphasis is being placed
on employment.  Many states now require recipients to "test the job
market" for a specified length of time before investing in costly
assessments or vocational training programs--an approach commonly
referred to as "work first." Under these "work first" programs,
recipients typically are provided an orientation and assistance in
searching for a job; they may also receive some readiness training. 
Only those unable to find a job after several weeks of job search are
then assessed for placement in other activities, such as remedial
education or vocational training.\19

Among our case study states, all but Louisiana have adopted a "work
first" approach.  In Oregon, officials told us that among those
participating in a 4-week job readiness and job search program, 40 to
50 percent report finding jobs by the end of the 4 weeks.  Even in
sites such as Santa Clara County, California, that had traditionally
placed a strong emphasis on education and training under JOBS,
officials told us they have found that approach unsuccessful in
moving recipients into work and self-sufficiency and that they, too,
have embraced the "work first" approach.  While encouraging
recipients to find jobs as quickly as possible, however, Santa Clara
officials said they have continued to emphasize the importance of
education.  They estimated that about 25 percent of the county's
recipients are able to find jobs with a livable wage, but as many as
75 percent will need to combine work with further training to obtain
the skills necessary to move toward self-sufficiency.  In Louisiana,
officials told us that they have not adopted a "work first" approach
because a large percentage of their caseload has very low education
and skill levels and is not job-ready.\20


--------------------
\18 See Welfare to Work:  Most AFDC Training Programs Not Emphasizing
Job Placement (GAO/HEHS-95-113, May 19, 1995) and Welfare Reform: 
Three States' Approaches Show Promise of Increasing Work
Participation (GAO/HEHS-97-80, May 30, 1997). 

\19 For further discussion of the characteristics of a "work first"
approach, see Amy Brown, Work First:  How to Implement an
Employment-Focused Approach to Welfare Reform (New York:  Manpower
Demonstration Research Corporation, Mar.  1997). 

\20 As part of a multiyear evaluation of the effectiveness of
different welfare-to-work strategies, the federal government recently
issued a study of three sites that simultaneously operated two
different programs:  a "labor force attachment" program, which
emphasized placing people in jobs quickly, and a "human capital
development" program, which emphasized education and training as a
precursor to employment.  The study found that while both programs
increased individuals' 2-year cumulative employment and earnings, the
labor force attachment program had larger impacts within this time
period.  See U.S.  Departments of Health and Human Services and
Education, Evaluating Two Welfare-to-Work Program Approaches: 
Two-Year Findings on the Labor Force Attachment and Human Capital
Development Programs in Three Sites, National Evaluation of
Welfare-to-Work Strategies (Washington, D.C.:  U.S.  Departments of
Health and Human Services and Education, Dec.  1997). 


         ACTIVITIES THAT COUNT
         TOWARD PARTICIPATION HAVE
         BEEN RESTRICTED
------------------------------------------------------ Chapter 2:1.1.2

The increased emphasis on employment is reflected in the new federal
law's list of activities that count toward minimum mandated
participation rates--a more restricted list than that allowed
previously under JOBS (see fig.  2.1).  For example, under JOBS,
families could participate in job readiness and education-oriented
activities indefinitely, with few restrictions.  Under TANF, however,
job readiness activities may count for no more than 6 weeks per
recipient; vocational educational training cannot count for more than
12 months for any individual; and education other than high school or
its equivalent counts only if it is directly related to employment. 
In addition, not more than 30 percent of the number of individuals
counted as engaged in work may be in vocational education or
attending school as teen parents. 

   Figure 2.1:  Work Activities
   Under JOBS and Under TANF

   (See figure in printed
   edition.)

\a Although states were required to provide job development and job
placement services, participation in these activities did not count
toward states' participation rates.  Also, unsubsidized employment
was generally not considered part of the JOBS program, which exempted
those working 30 hours or more a week from participating. 

\b "Work supplementation" under JOBS provided for subsidized
employment activities. 

Source:  Family Support Act of 1988 (P.L.  100-485, sec.  482[d]
through [g]) and 42 U.S.C.  sec.  607. 

Federal law also specifies the minimum number of hours families must
participate to count toward the mandated rates.  The adult recipient
in a single-parent family must participate in countable activities an
average of at least 20 hours a week (escalating to 25 hours in 1999,
and to 30 hours in 2000 and thereafter).  Adult recipients in a
two-parent family must participate for a combined total of at least
35 hours a week.\21 Under JOBS, participation was based on the
average monthly number of individuals whose combined and averaged
hours exceeded 20 hours per week.  (For preliminary rates achieved
using the new federal criteria for participation, see ch.  4.)


--------------------
\21 Generally, to receive federally funded child care assistance
under TANF, two-parent families must participate for a combined total
of at least 55 hours a week. 


         ASSIGNMENT TO JOB
         PLACEMENT ACTIVITIES HAS
         INCREASED
------------------------------------------------------ Chapter 2:1.1.3

Consistent with the increased emphasis on employment in the new
federal law, most states have adopted restrictions on the criteria
for assigning recipients to training and education activities and the
length of time such activities can be assigned.  Available 1994 data
from the previous JOBS program compared with more recent data under
TANF indicate that assignment to job placement activities has
increased while assignment to education and training activities,
including vocational education, has decreased in most states.  While
these data reflect various months and may not be comparable across
states, they provide a rough indication of the shift taking place
within each state.  Among our seven case study states, this shift is
best exemplified by Connecticut and Louisiana, where the percentage
assigned to job placement activities increased more than fourfold
between 1994 and 1997, not including those who found unsubsidized
jobs (see fig.  2.2).  The shift is least apparent in Oregon, which
had begun implementing reforms in 1992 and already had a large
percentage of recipients assigned to job placement activities in the
base year, and in California, which had not yet implemented its
reform program but nevertheless doubled the percentage assigned to
job placement by 1997. 

   Figure 2.2:  Percentage of
   Active Participants Assigned to
   Job Placement Activities in
   Seven States Before and After
   Federal Reform

   (See figure in printed
   edition.)

Note:  "Job placement" activities include job search, job readiness,
job development, community service, work experience, subsidized
employment, and on-the-job training but not unsubsidized employment. 
Data on participation cover various time periods and may not be
comparable across states but can provide a rough indication of the
changes over time within each state. 

Sources:  The 1994 data are from Committee on Ways and Means, U.S. 
House of Representatives, 1996 Green Book (Washington, D.C.:  U.S. 
Government Printing Office, Nov.  4, 1996), Table 8-7:  Average
Monthly Percentage of JOBS Participants by State and Component,
Fiscal Year 1994, pp.  420-21.  The 1997 data, obtained from the
states, are for the following time periods:  California (June 1997),
Connecticut (Sept.  1997), Louisiana (Aug.  1997), Maryland (Apr.  to
Sept.  1997), Oregon (July to Sept.  1997), Texas (Sept.  1996 to
Aug.  1997), and Wisconsin (Jan.  to Dec.  1997). 


         TYPES OF JOB PLACEMENT
         ACTIVITIES VARY AMONG
         STATES
------------------------------------------------------ Chapter 2:1.1.4

All states had increased the percentage of active participants
assigned to job placement activities as compared with education and
training activities, although the types of work-related activities
assigned those not working in unsubsidized jobs varied greatly among
our case study states (see table 2.1).  For example, in 1997, all
states except Louisiana had assigned more than a quarter of their
active participants to job search and job readiness activities.  In
Maryland, Oregon, Texas, and Wisconsin, over half the active
participants not in unsubsidized jobs had been assigned to such
activities. 



                         Table 2.1
          
             Percentage of Active Participants
              Assigned to Various Work-Related
           Activities in Seven States Before and
                    After Federal Reform

                        Conn              Oreg
Year            Calif.     .   La.   Md.     .  Tex.  Wis.
--------------  ------  ----  ----  ----  ----  ----  ----
Education and training activities\a
----------------------------------------------------------
1994              76.7  85.0  87.8  65.1  44.4  75.3  60.4
1997              53.3  31.7  48.6  10.5  27.5  36.1  12.5

Job placement activities\b
----------------------------------------------------------

Job search, readiness, and development
----------------------------------------------------------
1994              15.6  11.0   7.8  33.8  55.2  24.2  29.0
1997              28.6  45.4  11.5  56.3  59.0  50.4  61.5

Community service and work experience\c
----------------------------------------------------------
1994               6.3   1.1   4.3     0   0\d     0   6.7
1997              15.4  22.7  39.3  31.3   5.3  12.8  25.0

Subsidized employment
----------------------------------------------------------
1994                 0     0     0   0.7   0.2     0   3.2
1997               1.4     0   0.5   1.9   7.9     0   0.8

On-the-job training
----------------------------------------------------------
1994               1.4   2.9   0.1   0.3   0.2   0.5   0.7
1997               1.3   0.3   0.1     0   0.3   0.7   0.3
----------------------------------------------------------
\a Education and training activities include high school or its
equivalent, English as a second language, higher education, and
vocational and job skills training. 

\b Job placement activities do not include program entry, assessment,
counseling, job entry, or unsubsidized employment. 

\c Under JOBS, states could operate a community work experience
program to provide work experience and training to recipients in
projects serving a useful public purpose in fields such as health,
education, and public safety.  Under TANF, both "work experience, if
sufficient private sector employment is not available" and "community
service programs" are included in the list of countable work
activities.  Since no further definitions of these activities have
been provided and the distinction is unclear, we combined these
categories. 

\d While this sample of the 1994 data reported to HHS included no
recipients participating in community work experience, Oregon
officials estimated that 2 to 5 percent of active participants had
participated in community work experience sometime in 1994. 

Sources:  Data for 1994 are based on state reports to HHS on the
average monthly number of participants, as summarized in Committee on
Ways and Means, 1996 Green Book, Nov.  4, 1996.  Data for 1997 were
provided by states and represent various reporting periods, ranging
from 1 month (California, Connecticut, and Louisiana), to 3 months
(Oregon), to 6 months (Maryland), to an entire year (Texas and
Wisconsin).  Shorter periods were preferable to avoid doublecounting
across months.  Data on participation cover various time periods and
may not be comparable across states but can provide a rough
indication of the shift taking place within each state.  Data
represent those actively participating, but they may not be
participating to the extent necessary to count toward the states'
minimum mandated participation rates.  See ch.  4 for data on
preliminary rates achieved using the new federal criteria for
participation. 

In addition, Connecticut, Louisiana, and Maryland had assigned a
significant proportion of participants to community service and work
experience placements.  According to a Louisiana official, many
participants were being shifted from education and training
activities to community work experience because of TANF provisions
that limit the extent to which recipient participation in education
and training activities can be counted toward meeting the state's
mandated participation rate.  Louisiana officials told us that many
of their recipients need remedial training before they can be
assigned to vocational education programs, however, and voiced
concerns about the limitations TANF places on counting recipients
assigned to these activities.  In contrast, other states placing
fewer recipients in community service activities told us they viewed
such placements as a last resort, useful only to provide recipients
with little or no work history the necessary experience to obtain an
unsubsidized job, or to comply with a state requirement to use such
placements once a certain length of time had elapsed without finding
a job.\22

Offering subsidies to employers who hire and pay wages to recipients
or who provide on-the-job training is one option states can use to
help create job opportunities and move recipients into work when more
training is needed.\23 In the past, however, such programs have been
criticized for subsidizing jobs for participants who could have been
hired into unsubsidized jobs.  While 37 states indicated in their
TANF plans as of November 1997 that they would provide employer
subsidies, we found such subsidized employment placements were not
widely used.  Among our five case study states with subsidized
employment programs (California, Maryland, Oregon, Texas, and
Wisconsin),\24 state officials told us widespread use of such
placements is generally not encouraged because they are expensive and
often difficult to negotiate with employers.  Maryland had
implemented only a small pilot project using subsidized employment in
the city of Baltimore, while Texas had implemented a project only in
the Corpus Christi area.  According to Texas officials, an evaluation
of the Corpus Christi project will be completed before the project is
expanded to other areas.  Oregon had used subsidized employment
placements most frequently but, even so, had assigned less than 10
percent of its participants to this activity in 1997. 

To the extent that such placements are used, job developers generally
screen recipients carefully and attempt to secure positions that are
likely to result in the recipient's being hired to an unsubsidized
position.  For example, Oregon's "JOBS Plus" program subsidizes
employers to hire recipients who lack job skills and experience, but
it is a relatively small part of Oregon's welfare-to-work program. 
Job developers prescreen recipients before referring them for
specific openings; nevertheless, only about half the placements
result in unsubsidized jobs, and few big companies are willing to
participate in the program.  Similarly, Wisconsin's "Trial Jobs"
component is designed for certain clients who lack sufficient work
experience to be able to move directly to unsubsidized employment. 
Trial Jobs reimburses employers up to $300 per month to offset some
of the initial costs of new employee training and supervision.  Such
on-the-job training placements are made for up to 3 months, with a
possible 3-month extension in some circumstances, and are expected to
result in unsubsidized employment. 


--------------------
\22 Use of community service placements as a last resort is also
reflected by the fact that most states chose to opt out of a TANF
provision allowing states to assign nonexempt adult recipients to
community service after receiving 2 months of cash assistance if they
were not otherwise participating in work activities.  As of November
1997, 49 of the 54 states and territories had decided not to
implement such a provision, including all of our case study states. 
Only Massachusetts, Michigan, New Mexico, South Dakota, and Guam had
opted to implement the provision.  (See National Governors'
Association Center for Best Practices, Summary of Selected Elements
of State Plans for Temporary Assistance for Needy Families, as of
November 20, 1997 [Washington, D.C.:  National Governors'
Association, Nov.  20, 1997], p.  3.) However, some states have
implemented their own requirements.  For example, in California, once
new applicants have received assistance for 18 months--or, once
ongoing recipients have received assistance for 24 months after
entering the welfare-to-work program--and have not found unsubsidized
employment, they must be assigned to community service. 

\23 Subsidized employment programs use TANF cash assistance payments,
and sometimes cashed-out food stamps, to develop and subsidize jobs
as an alternative to providing monthly cash assistance payments to
participants.  On-the-job training programs reimburse employers for
providing training and additional supervision to participants. 

\24 See National Governors' Association Center for Best Practices,
Summary of Selected Elements of State Plans for TANF, Nov.  20, 1997. 


         FEW STATES ARE OPTING TO
         EXCEED MINIMUM HOURLY
         REQUIREMENTS
------------------------------------------------------ Chapter 2:1.1.5

At least initially, most states adopted the federal minimum standard
of 20 hours of participation per week for most recipients in
single-parent families, and 35 hours of participation for two-parent
families.  However, a few states have used the flexibility under TANF
to increase the number of hours recipients in single-parent families
are required to participate to better reflect hours comparable to
full-time employment.  For example, in Wisconsin, program
requirements call for adult recipients to be assigned up to 40 hours
of activities per week, and families receive grant amounts that are
prorated by applying a sanction for hours missed without good cause
multiplied by the federal minimum wage.  In Oregon, the goal is to
require 40 hours per week for every adult recipient, but those in
certain situations are sometimes allowed to do less.  California
requires its counties to increase the number of hours of
participation per week from 20 to 32 by July 1999, or sooner at
county option.  Among the three counties we visited, only Fresno had
opted to require 32 hours immediately.  As discussed further in
chapter 4, some state officials noted that tracking actual hours of
participation in the various activities has created a huge increase
in workload for staff. 


      MORE RECIPIENTS ARE BEING
      REQUIRED TO PARTICIPATE
-------------------------------------------------------- Chapter 2:1.2

Under states' welfare reform programs, a higher percentage of adult
recipients are being required to participate in work activities and
find jobs.  Before reform, parents or caretaker recipients caring for
a child under age 3 (or under age 1, at state option), along with
those living in remote areas or with mental or physical impairments,
were generally not required to participate in JOBS.  Now, many
states' reform programs have lowered the age-of-youngest-child
exemption to 1 year or even younger\25 and require other
traditionally exempt recipients to address their barriers, engage in
work activities, and look for jobs. 

The new federal reform law generally allows states to exempt whomever
they choose from participating in their welfare-to-work programs.\26
However, virtually all adult recipients except for single custodial
parents caring for a child under 1 year of age are to be included in
the calculation of the state's participation rate.\27 As a result,
the extent to which states exempt recipients for other reasons could
affect states' ability to meet prescribed minimum participation rates
and avoid financial penalties.  Nevertheless, nearly all states are
continuing to provide exemptions from participation requirements
under some circumstances.  In a shift away from blanket exemptions of
broad categories of recipients, however, about half the states are
making such determinations on a case-by-case basis or describing
exemptions as deferrals, delays, or temporary exclusions that will
end with resolution of the problem. 

Among our seven case study states, the percentage of adult recipients
required to participate in work activities increased from an average
of 44.2 percent in 1994 to 64.8 percent in 1997 (see fig.  2.3). 
Percentages were highest in states such as Connecticut, Maryland, and
Oregon, where expanded participation requirements had been
implemented previously under waivers.  Percentages were lowest in
states such as California, where the state's welfare reform program
had not yet been implemented, and Texas, where the
age-of-youngest-child exemption is higher than in most other
states.\28

   Figure 2.3:  Percentage of
   Adult Recipients Required to
   Participate in Seven States
   Before and After Federal Reform

   (See figure in printed
   edition.)

Note:  Data represent those required to participate but not
necessarily actively participating or participating to the extent
necessary to count toward the states' minimum mandated participation
rates.  For data on preliminary rates achieved using the new federal
criteria for participation, see ch.  4. 

\a Connecticut provided estimated data for 1997 that are based on the
percentage of cases subject to time limits. 

\b In March 1996, Wisconsin began implementing a waiver that required
increased participation and provided cash assistance that was based
on hours of participation.  In June 1997, the JOBS exemption policy
was expanded to "universal participation," requiring referral of all
adult recipients to Wisconsin's welfare-to-work program under TANF. 
However, those caring for a child under 12 weeks old are referred
only as volunteers, and good cause exemptions continue to be granted
on a case-by-case basis.  No 1997 data on the percentage of adult
recipients exempted on these bases were available. 

\c 1994 data are based on the percentage of total adult recipients
designated "JOBS Mandatory" (Committee on Ways and Means, 1996 Green
Book, Nov.  4, 1996, Table 8-9:  JOBS Participation by State, Fiscal
Year 1994, pp.  425-27). 

\d 1997 data were obtained from the states and are based on the
percentage of total adult recipients required to participate (or
designated "nonexempt").  Data are for the following time periods in
1997:  April to June (California), September (Texas), October
(Connecticut, Louisiana, and Oregon), and December (Maryland). 

Some case study states are requiring recipients with significant
barriers to participate but are allowing a broader range of
activities and fewer hours of participation than specified under
TANF.  For example, in California, Oregon, and Wisconsin, program
rules allow recipients with a mental health or substance abuse
problem to be assigned initially to treatment or counseling programs,
and those with very young children to be assigned fewer than 20 hours
of participation per week.  In California, as long as such recipients
comply with their assigned requirements, they are eligible to
continue to receive assistance under TANF up to the 5-year time
limit, but they are not counted when determining the state's
participation rate.  In Oregon, proposed rules would allow such
recipients to receive assistance under TANF as well as count toward
the participation rate because of the state's waiver, which will not
expire until the year 2002.  During the last quarter of fiscal year
1997, Oregon reported 22 percent of its participants had been
assigned to drug, alcohol, mental health, or other counseling
programs. 


--------------------
\25 Before federal reform, 11 states had adopted waivers requiring at
least part-time participation before the youngest child reached the
age of 1 year (see HHS, Setting the Baseline:  A Report on State
Welfare Waivers [Washington, D.C.:  HHS, June 1997]).  Under TANF, 19
states have adopted such provisions, with 5 of these states allowing
no exemption based on the age of the youngest child (see L.  Jerome
Gallagher and others, One Year After Federal Welfare Reform, May
1998, p.  V-3). 

\26 TANF specifies only one mandatory basis for exemption:  a state
may not reduce or terminate assistance on the basis of a refusal of
an individual to work if the individual is a single custodial parent
caring for a child who is under 6 years old and needed child care
within a reasonable distance by a suitable informal, or affordable
formal, provider is unavailable. 

\27 Also not included in the calculation of the state's participation
rate are families with adult recipients being sanctioned for
nonparticipation for up to 3 months within a 12-month period. 
However, single custodial parents caring for a child under 1 year of
age may only be omitted from the calculation of a state's
participation rate for a cumulative lifetime total of 12 months. 
Thus, parents who have a subsequent child while on assistance may be
included in the calculation even though they may be exempt from the
state's participation requirements.  (See L.  Jerome Gallagher and
others, One Year After Federal Welfare Reform, May 1998, pp.  V-1 and
V-2.) In addition, proposed regulations issued in November 1997
indicate that states may be required to include cases that are exempt
because of waiver provisions in the calculation of their
participation rates, except research cases being studied as part of
an evaluation. 

\28 Before September 1997, Texas would exempt adult recipients caring
for a child under age 5; since September 1997, the
age-of-youngest-child exemption has been lowered to age 4 but is
still among the highest in the nation. 


      PARTICIPATION IS EXPECTED
      SOONER
-------------------------------------------------------- Chapter 2:1.3

Under states' welfare reform programs, participation requirements are
being imposed sooner than under JOBS, with many states requiring
participation in job search activities immediately upon application
for assistance.  Before reform, recipients could wait months--or even
years--before being required to participate, and many never were
required to participate because of the lack of sufficient services
and staff.\29 Now, in many states, new program rules call for
applicants to be assigned to job search and placement activities as
soon as they walk through the door and are initially screened as
potential recipients.  As a result, some find jobs and are diverted
from receiving monthly cash assistance, as discussed later in this
chapter. 

The new federal reform law specifies that states must require
recipients to engage in work activities either when the state
determines they are ready to do so or after 24 months of assistance,
whichever occurs earlier.  According to TANF plans submitted
nationwide, as of November 1997, 33 states had adopted the 24-month
federal maximum, while 21 indicated that they would require
recipients to engage in work activities sooner--including 11 states
that require recipients to participate immediately upon
application.\30

Two of our seven case study states have adopted the 24-month federal
maximum (Louisiana and Maryland), and five require recipients to
participate immediately if they are not exempt (California,
Connecticut, Oregon, Texas, and Wisconsin).  For example, Texas
implemented a policy in December 1997 requiring most applicants to
attend workforce orientation sessions as a condition of completing
the application process.  Nonexempt recipients are required to
participate in follow-up employment sessions and job readiness and
job search activities.  California's plan also calls for all
nonexempt recipients to be assigned immediately to job search
activities. 

Even in the two case study states that do not impose immediate
requirements according to their state plans, program officials
indicated the states are requiring nonexempt recipients to engage in
work activities as soon as possible.  For example, Maryland state
officials told us their program requires individuals to engage in
work activities from the moment they come in contact with the local
welfare departments.  In Louisiana, officials told us that case
managers assess recipients' skills and assign them to a work activity
as soon as possible. 


--------------------
\29 See GAO/HEHS-95-28, Dec.  19, 1994, pp.  5-8. 

\30 See National Governors' Association Center for Best Practices,
Summary of Selected Elements of State Plans for TANF, Nov.  20, 1997,
p.  3. 


      STRONGER SANCTIONS FOR
      NONPARTICIPATION ARE BEING
      IMPOSED
-------------------------------------------------------- Chapter 2:1.4

Many states' welfare reform programs call for imposing stronger
sanctions on recipients who fail to participate in their assigned
activities than the sanctions that were allowed before reform.  Under
JOBS, if a recipient failed to comply with work requirements without
good cause, grants could be reduced only by the amount attributable
to the noncomplying recipient, and conciliation processes were
required to try to secure cooperation before the sanction could be
imposed.\31 Now, in many states, program rules call for terminating
the entire family's grant, and conciliation processes have been
streamlined or, in some cases, eliminated. 

The new federal reform law stipulates that if an adult recipient
refuses to engage in work activities as assigned, the state must, at
a minimum, reduce the family's grant by an amount prorated on the
basis of the individual's failure to participate; the state may opt
to reduce the grant more, up to and including terminating the entire
family's grant.\32

Under waivers prior to federal reform, 26 states had implemented
provisions to terminate the entire family's grant for failure to
participate, commonly referred to as "full family sanctions."\33
According to a summary of state plans compiled by the Urban
Institute, a total of 36 states have adopted such provisions under
TANF.\34 However, many states have maintained their conciliation
processes even though they are no longer required. 

Five of our seven case study states have adopted full family
sanctions:  Connecticut, Louisiana, Maryland, Oregon, and Wisconsin. 
Because Louisiana had only implemented its policy in March 1998,
calling for a partial sanction for 3 months followed by a full-family
sanction for continued failure to comply, no Louisiana families had
received a full-family sanction at the time of our review.  In the
remaining four states, full-family sanctions had been adopted
previously under waivers and had been in effect for at least a year. 
In these states, we found the number of cases in which full-family
sanctions were imposed for failure to comply with work requirements
varied widely but that the number never exceeded 1 percent of each
state's total caseload per month (see table 2.2).  Nationwide data on
full-family sanctions imposed under TANF have not yet been issued by
HHS and may differ significantly from the experiences of these
states. 



                         Table 2.2
          
             Full-Family Sanctions Imposed for
          Failure to Comply With Work Requirements
                       in Four States

              Full-family sanctions imposed during first 9
                        months of implementation
              --------------------------------------------
                             Full-family
                            sanctions as
                   Average    percentage
                    number    of average
               imposed per       monthly
State                month    caseload\a  Months covered
------------  ------------  ------------  ----------------
Connecticut              9         < 0.1  Jan. 1996-Sept.
                                           1996
Maryland               174           0.3  Oct. 1996-June
                                           1997
Oregon                  13         < 0.1  July 1996-Mar.
                                           1997
Wisconsin            550\b           0.9  Mar. 1996-Nov.
                                           1996
----------------------------------------------------------
\a Data calculated using fiscal year 1996 average monthly caseloads,
as reported in HHS, Characteristics and Financial Circumstances of
AFDC Recipients, FY 1996, Apr.  1997, or using caseload data provided
by the state. 

\b Based on cases receiving no cash assistance for at least 1 month
and not open as of December 31, 1996. 

The structure of states' sanction and conciliation policies varies. 
For example, Wisconsin's policy requires providing cash assistance
prorated by applying a sanction for hours missed without good cause,
multiplied by the federal minimum wage, and the conciliation process
has been eliminated.  When initially implemented in March 1996, the
policy also required providing no cash assistance if families'
participation fell below 25 percent of scheduled hours.  After
initial implementation, a relatively high number of Wisconsin
families received no cash assistance as a result of sanctions, in
part, because no effort was made to contact the families and resolve
any barriers to participation before denying assistance.  Effective
June 1997, Wisconsin eliminated the portion of its policy that
required providing no cash assistance to families when participation
fell below 25 percent of scheduled hours without good cause, and the
number of full-family sanctions imposed has dropped significantly. 
In October 1997, 220 Wisconsin families received no cash assistance
as a result of sanctions, compared with an average of about 550 per
month during initial implementation.  Wisconsin officials told us
they believe their policy of reducing grants on a prorated basis for
work hours assigned and not performed and eliminating the
conciliation process before imposing sanctions work well as
motivators that mirror the real world. 

In contrast, county officials in Maryland told us their rigorous
conciliation processes resulted in fewer terminations of families'
grants, and Oregon officials told us they had terminated very few
grants for the same reason.  While the threat of grant termination
served as a motivator to some extent, welfare office workers in these
states said that conciliation processes take the teeth out of
sanctions if they provide families an opportunity to keep their
grants simply by agreeing to cooperate.  Workers said they hoped
their conciliation processes would be changed to shorten the process
and to prevent recipients from "gaming the system" by agreeing to
cooperate and then failing to do so. 

In Connecticut, relatively few families had their grants terminated
after initial implementation, primarily because of the state's
program structure of graduated sanctions.  While only 83 families
received full-family sanctions during the first 9 months of
implementation, 377 families received full-family sanctions over the
next 3 months.  However, more recent sanction data indicate that the
number of sanctions may be subsiding now that the program has been
fully implemented.  In May 1997, only three cases received
full-family sanctions, compared with an average of as many as 126 a
month earlier in the implementation process.  The early peak in the
number of sanctions most likely reflects the conversion of the
state's existing caseload to the new program rules, as well as
initial doubts among both recipients and caseworkers that grants
actually would be cut off. 

In imposing full-family sanctions, on the one hand, state officials
maintain that if families are experiencing hardship caused by the
loss of cash assistance, all they have to do is cooperate to have
their grant reinstated.  On the other hand, states have usually taken
steps to address concerns raised by welfare advocacy groups about the
well-being of children in families whose entire grant has been
terminated for noncooperation.  For example, states have generally
adopted safeguards to ensure that families fully understand the
implications of their actions, that services are available to help
them participate, and that exemptions may be provided under certain
circumstances.  To ensure the children are not at risk of placement
in foster care if grants are terminated, some states, such as Oregon,
require a home visit before terminating a family's grant for
noncooperation and may continue to pay rent and utilities directly on
behalf of a family to prevent destitution. 

In other states, including three of our case study states, such
concerns have resulted in continuing with only a partial reduction in
grant amounts for noncooperation instead of imposing full-family
sanctions.  All three of these states nevertheless strengthened their
sanction processes in other ways.  For example, both Louisiana and
Texas eliminated conciliation.  California shortened its conciliation
process and required assistance to be provided as vouchers for rent
and utilities for those families sanctioned for more than 3 months. 
In California's San Bernardino County, families accepting a reduced
grant rather than attempting to participate were automatically
referred to the fraud unit for investigation of possible unreported
sources of income.\35 Despite these changes, officials we spoke with
were skeptical about whether reduced grants alone would be a
sufficient motivator for recipients to participate and were concerned
that their inability to terminate such cases might affect their
state's ability to meet minimum mandated participation rates over
time.  Because of the way participation rates are calculated, states
can achieve higher rates by closing cases of families who fail to
comply with work requirements for more than 3 months.\36


--------------------
\31 Under JOBS, states were required to establish a conciliation
procedure for the resolution of disputes involving an individual's
participation in the program, including an opportunity for a hearing,
before taking any action to suspend, reduce, discontinue, or
terminate grants.  This is different from a state's administrative or
appeal process, which is still required under the new law and which
takes place after the administrative action, or sanction, is imposed. 

\32 The law also allows states to terminate Medicaid coverage for an
adult head-of-household for refusing to work; however, the law
requires a Medicaid sanction to be lifted once the recipient begins
complying with the state's rules and does not permit states to
terminate Medicaid benefits for pregnant women, infants, or children
who are not a head-of-household.  Further, the law allows states to
reduce a household's food stamp allotment by up to 25 percent for
failure to comply with program requirements under TANF, or more at
state option, consistent with the state's TANF provisions. 

\33 See Welfare Reform:  States' Early Experiences With Benefit
Termination (GAO/HEHS-97-74, May 15, 1997), p.  29. 

\34 See L.  Jerome Gallagher and others, One Year After Federal
Welfare Reform, May 1998, pp.  V-7 and V-8.



\35 Between January and March 1996, 60 such cases were referred for
investigation on this basis.  Investigators found fraud in 13, or
about 22 percent, according to the head investigator.  In all but one
case, the finding of fraud was based on an inability to locate the
families.  The families no longer lived at their reported addresses
but were using the addresses as mail drops and continuing to cash the
checks. 

\36 Under TANF, after a family has been sanctioned for 3 months, if
it continues to receive partial grants, it must be counted in the
denominator in the calculation of the state's participation rate.  As
the number of such families accumulates, it becomes increasingly
difficult for the state to meet its mandated participation rate and
avoid financial penalties.  Thus, if such cases cannot be terminated,
caseworkers are under pressure to find ways to motivate these
families to participate. 


   STATES HAVE MODIFIED POLICIES
   TO ENCOURAGE SELF-SUFFICIENCY
---------------------------------------------------------- Chapter 2:2

States' reform programs have generally included various policy
changes to reinforce the expectation that recipients will move from
welfare to work and that welfare should be temporary, not a way of
life.  By changing policies on the calculation of cash assistance
payments and the provision of support services, many states are
attempting to ensure that "work pays"--that is, that recipients who
find jobs are better off financially than recipients who do not find
jobs.  In addition, states have taken steps to limit the time cash
assistance can be received--most states have chosen limits consistent
with the new law, but some have shorter limits than the federally
mandated 5-year lifetime limit.  Many states have also adopted
various extension and exemption policies as a safety net for families
reaching these limits, at least under some circumstances.  Finally,
while making policy changes to help make work pay and cash assistance
temporary, few states have opted to use their increased flexibility
under TANF to restrict eligibility for certain groups or to lower
grant levels.  In fact, several states have expanded eligibility by
simplifying income standards and liberalizing the criteria for
two-parent families. 


      INCENTIVES TO MAKE WORK PAY
      ARE BEING INCREASED
-------------------------------------------------------- Chapter 2:2.1

Making work pay involves the interaction of many factors, including
grant levels, the impact of earned income on cash assistance, the
availability of support services, the earned income tax credit, and
child support.  Under the prior system, these factors could interact
in such a way that families were financially better off on welfare
than working at a low-wage job, and families already on welfare had
few incentives to work.\37 As a result, reforms to increase
incentives for families to choose work over welfare were among the
first reforms that states began making under waivers.  Under TANF,
states are provided broad discretion in defining policies to help
make work pay, such as determining how to calculate payment amounts
and grant levels and whether to provide services to those not
receiving monthly cash assistance. 

To be eligible for assistance, families must have income and
resources below certain limits defined by the states and referred to
as income standards.\38 In determining a family's eligibility and
calculating the amount of the cash assistance payment, allowances can
be made for certain assets, such as a home or a car, and if a family
member is employed, a portion of earned income can be excluded, or
"disregarded," in the calculation.\39

Under AFDC, the limits on a family's earnings, savings, and other
assets were widely viewed as too low, creating a barrier to families'
attempts to move toward self-reliance.  For working families with
incomes low enough to still qualify for assistance, after 12 months
the AFDC grant would be reduced dollar for dollar for earnings,
except for a $90 monthly allowance for work expenses.  As a result,
to help families prepare for a future without monthly assistance
payments, and as part of a general move to "make work pay," many
states adopted waivers to increase their asset limits and change
their "earnings disregard" policies to allow families to keep more of
their grants when working, or to keep them longer.\40 According to
states' TANF plans as of October 1997, 39 states had increased their
countable asset limits, 48 states had increased their vehicle
allowances, and 42 states had changed their earnings disregard
policies, as compared with what was allowed under AFDC.\41 All our
case study states have implemented changes to their asset limits,
vehicle allowances, or earnings disregard policies to help make work
pay (see table 2.3).\42



                         Table 2.3
          
            Asset Limits and Earnings Disregard
                  Policies in Seven States

                Vehicle
      Countabl  value     Individual
Stat   e asset  allowanc  development   Monthly earned
e        limit  e         accounts      income disregard\a
----  --------  --------  ------------  ------------------
Cali  $2,000\c  $4,650    Yes           $225 plus 50
f.\b                                    percent of the
                                        remainder up to
                                        the minimum basic
                                        standard of
                                        adequate care\d

Conn     3,000  One       No            All earnings up to
.               vehicle                 the poverty level
                                        until time limit
                                        reached

La.\     2,000  10,000    Yes\          $120; up to $900
e                                       for the first 6
                                        months of
                                        employment

Md.    2,000\f  One       No            26 percent up to
                vehicle                 the countable
                                        income limit

Oreg    10,000  10,000    Yes           50 percent up to
.           or                          the countable
       2,500\g                          income limit

Tex.   2,000\i  4,650     No            Unchanged
\h

Wis.     2,500  10,000    No            All income (except
                                        federally funded
                                        disregards)
                                        subject to the
                                        gross income limit
                                        of 115 percent of
                                        the poverty level,
                                        but disregarded in
                                        the calculation of
                                        the grants\j

Prio     1,000  1,500     No            $90; $30 plus 1/3
r                                       for the first 4
limi                                    months, then $30
ts                                      for the next 8
unde                                    months
r
AFDC
----------------------------------------------------------
\a Disregards for child care expenses are not included. 

\b California's policies are provided under its new welfare reform
law, enacted August 1997 and effective January 1998. 

\c The limit is $3,000 for cases with a member aged 60 or older. 

\d California's minimum basic standard of adequate care is determined
on the basis of the number of eligible people in the family, as
adjusted to reflect regional variations in housing costs and changes
in the cost of living. 

\e These policies were implemented in early 1998 as follows,
according to Louisiana state officials:  increase in asset limits and
vehicle allowance effective March 1998; adoption of individual
development accounts effective February 1998; and change to the
earnings disregard policy effective January 1998. 

\f Acording to state officials, Maryland also allows each dependent
child to save up to $2,000 from earnings without counting it against
the family's $2,000 asset limit. 

\g $10,000 is the limit for those participating satisfactorily in
Oregon's welfare-to-work program; $2,500 is the limit for those not
participating. 

\h These policies became effective in Texas in October 1997,
according to a state official.  Texas' welfare waiver includes
provisions for individual development accounts and "fill-the-gap"
budgeting in four pilot counties but was not yet implemented at the
time of our review.  "Fill-the-gap" budgeting is a method of paying
grants that allows working families to keep a greater portion of
their grant as earnings increase up to the maximum grant level or, in
some states, up to the need standard.  Additionally, 1997 state
legislation provided for the Texas Workforce Commission to establish
individual development accounts for TANF recipients. 

\i The limit is $3,000 for households with an elderly or disabled
person, regardless of whether that person is in the assistance unit. 

\j Wisconsin state officials provided this information. 

Source:  L.  Jerome Gallagher and others, One Year After Federal
Welfare Reform, May 1998, pp.  III-2, III-3, VI-5, and VI-6, except
as otherwise noted. 

Under the new federal welfare reform law, states can also opt to
allow recipients to establish individual development accounts to
accumulate assets for further education, purchasing a home, or
starting a business and not have these assets count in determining
their eligibility for assistance.  According to TANF plans as of
October 1997, 22 states had opted to allow recipients to accumulate
assets under restricted savings accounts of some type, including 3 of
our 7 case study states:  California, Louisiana, and Oregon (see
table 2.3).\43 Texas also had plans to implement individual
development accounts as authorized by 1997 state legislation. 

In addition, by making support services available to families who are
not receiving monthly cash assistance; by encouraging clients to take
advantage of the earned-income tax credit; and, in a few instances,
by adjusting grant levels, states are more able to ensure that the
working poor are better off than those families relying on welfare
and not working.  For example, Oregon and Wisconsin provide child
care services primarily on the basis of income, regardless of an
individual's status as a welfare recipient or nonrecipient, and
Oregon uses state funds to make income-based medical coverage
available to low-wage workers who do not meet Medicaid eligibility
criteria.\44 In states with policies to provide services on the basis
of income rather than welfare status, families may no longer have an
incentive to stay on welfare simply to gain better access to such
services, and some families may be diverted from receiving monthly
cash assistance, as discussed later in this chapter. 

Also, when clients attend orientation sessions in some states, they
are presented with charts to illustrate how they could be better off
working for wages than they are receiving welfare grants--financially
as well as psychologically.  For example, in both Wisconsin and
Oregon, a single mother with two children could expect to nearly
double her monthly income if working at an unsubsidized job for 40
hours per week, as compared with receiving a cash assistance grant
and not working (see fig.  2.4).  According to a California welfare
advocacy spokesperson, grant levels and income disregards were
lowered, in part, to help ensure that the nonwelfare working poor
would be better off than those on welfare. 

   Figure 2.4:  Making Work Pay in
   Oregon

   (See figure in printed
   edition.)

\a This figure includes the federal earned income tax credit, plus
the state earned income, working family, and child care tax credits. 

\b This figure represents monthly gross wages for 40 hours of work
per week at the Oregon minimum wage ($6 an hour), minus taxes and the
client's share of child care expenses. 


--------------------
\37 See discussion in Mary Jo Bane and David T.  Ellwood, Welfare
Realities:  From Rhetoric to Reform (Cambridge, Mass.:  Harvard
University Press, 1994), pp.  144-54. 

\38 Under prior law, families were required to pass both a gross
income test and a net income test.  Gross income could not exceed 185
percent of the state's need standard, which was based on the amount
of income each state determined as essential for a minimal standard
of living for a particular family size.  Net income could not exceed
100 percent of the state's need standard or the state's maximum grant
level (which was generally less than the need standard).  Under
states' reform programs, several states have changed or eliminated
the gross income test used to determine eligibility.  For example,
two of our case study states--Connecticut and Maryland--no longer
employ any gross income test for determining eligibility.  Other
states, such as California, have increased the needs standard.  In
Wisconsin, families must meet a gross income limit of 115 percent of
the poverty level to be eligible for assistance; if eligible on the
basis of this gross income test, grant levels are not reduced as a
result of receiving other income. 

\39 In addition, some states have changed how income is counted
toward this limit.  For example, Maryland has opted to no longer
count the income of a step-parent that falls below 50 percent of the
poverty level in determining the eligibility of a two-parent family. 

\40 Under waivers before federal reform, 25 states increased their
countable asset limits, 32 states increased their vehicle allowances,
and 31 states changed their AFDC earnings disregard policies, with
some states disregarding all earned income up to the poverty line. 
See HHS, Setting the Baseline, June 1997. 

\41 See L.  Jerome Gallagher and others, One Year After Federal
Welfare Reform, May 1998, pp.  III-1 and VI-4. 

\42 Changes in "earnings disregard" policies, such as increases in
the amount of earnings disregarded in calculating the amount of cash
assistance or extensions to the length of time such earnings are
disregarded, raise new issues in the context of time-limited cash
assistance.  For example, while they may encourage some recipients to
find jobs, they also make it possible for more families to continue
to receive cash assistance after finding work, thereby slowing their
exit from welfare.  Moreover, some of these families may be receiving
very small cash assistance payments yet continuing to use up some of
their remaining months of eligibility for cash assistance.  For
further discussion, see Dan Bloom, After AFDC:  Welfare-to-Work
Choices and Challenges for States (New York:  Manpower Demonstration
Research Corporation, 1997), ch.  5. 

\43 See L.  Jerome Gallagher and others, One Year After Federal
Welfare Reform, May 1998, pp.  III-1 through III-3. 

\44 Other states are also moving to establish income-based programs,
making child care available to families regardless of welfare status. 
See GAO/HEHS-98-27, Jan.  13, 1998, pp.  13-14. 


      VARIOUS TIME LIMITS ON
      ASSISTANCE HAVE BEEN ADOPTED
-------------------------------------------------------- Chapter 2:2.2

To promote the temporary nature of cash assistance and encourage
families to move toward self-reliance, the new federal reform law
prohibits states from using federal TANF funds to support families
for more than 5 years.  This time limit is cumulative, for life, for
all adult recipients.  However, states may continue to use federal
funds to provide assistance for up to 20 percent of their caseloads
beyond this 5-year limit if families are facing hardship and may
continue to provide assistance for additional families using only
state funds.\45 States' welfare reform programs generally include
provisions to place time limits on assistance, but they often count
months differently and provide different time frames and consequences
for those families reaching the limits, compared with those required
for claiming federal eligibility under the new law.  Officials told
us that tracking both state and federal time limits, especially
across state lines, poses new and significant challenges for many
states. 


--------------------
\45 States with time-limit waiver provisions inconsistent with the
new federal law may continue to use TANF funds to support families
consistent with their waiver provisions beyond the 5-year limit until
their waivers expire.  Under the proposed regulations, however, such
states would not then be eligible for a reasonable cause exception to
a penalty for failure to meet minimum work participation rate or time
limit requirements. 


         MANY STATES' TIME LIMITS
         INCLUDE SAFETY NETS
------------------------------------------------------ Chapter 2:2.2.1

States adopting time limits shorter than the federal 5-year maximum
generally have also adopted policies to extend assistance under
certain circumstances.  In addition, some states provide for using
state funds to extend assistance beyond the federal 5-year limit for
more than the 20 percent of their caseload that can be exempted on
the basis of hardship.  As of October 1997, 30 states' TANF plans
included a 5-year time limit on assistance for families, consistent
with the federal law.  However, nineteen states, including three of
our case study states--Connecticut, Louisiana, and Oregon--had opted
to impose shorter grant termination time limits on some families
instead of (or, in the case of five states, in addition to) the
5-year limit for families.\46 As indicated in table 2.4, Connecticut,
Louisiana and Oregon have adopted implementation policies to allow
assistance to continue for families reaching the limits under certain
circumstances. 



                                        Table 2.4
                         
                           Time-Limit Policies in Seven States

          Date of
          implementat  Length of     Effect on
State     ion          time limit    grants        Implementation policies
--------  -----------  ------------  ------------  --------------------------------------
Calif.    1/98         60 months     Grant         Months do not count against the time
                                     reduction     limit if recipient is exempt.

Conn.     1/96         21 months     Grant         6-month extensions are granted if
          (under                     termination   recipient demonstrates "good faith" in
          waiver)                                  complying with all work requirements
                                                   but is unable to find employment.

La.       1/97         24 months     Grant         Months do not count if recipient is
                       during a 60-  termination   disabled or incapacitated. Extensions
                       month period                may be granted if recipient is
                                                   actively seeking employment, requires
                                                   up to 1 year for completion of
                                                   training that will lead to employment,
                                                   or loses a job not as a result of
                                                   performance or if factors relating to
                                                   job availability are unfavorable.
                                                   Hardship exemptions are also granted
                                                   on the basis of individual
                                                   circumstances.

Md.       1/97         60 months     Grant         Extensions are granted to families
                                     reduction\a   that include an individual who has
                                                   been battered or subject to extreme
                                                   cruelty, and to other populations to
                                                   be determined.

Oreg.     7/95         24 months     Grant         Months do not count against the time
          (under       during an     termination   limit if recipient is judged not
          waiver)      84-month                    employable, has not been given the
                       period                      opportunity to participate, is
                                                   participating satisfactorily, or is
                                                   caring for a sick family member (for
                                                   up to 3 of the 24 months).

Tex.      1/97         12 months     Grant         Months do not count against the time
          (under       during a 72-  reduction     limit if JOBS employment services are
          waiver)\b    month                       not available or if recipient has good
                       period;                     cause for not participating.
                       24 months                   Extensions are granted for severe
                       during an                   personal hardship, lack of support
                       84-month                    services, or living in area of high
                       period; 36                  unemployment.
                       months
                       during a 96-
                       month period

Wis.      9/97\c       60 months     Grant         Months do not count against the time
                                     termination   limit for custodial parents of infants
                                                   12 weeks old or younger.\d Extensions
                                                   are granted on a case-by-case basis to
                                                   families in unusual circumstances,
                                                   such as when a member is unable to
                                                   work because of personal disability or
                                                   other significant limitations to
                                                   employment, or when the local labor
                                                   market precludes reasonable
                                                   opportunity.
-----------------------------------------------------------------------------------------
\a In Maryland, upon reaching the time limit, children continue to
receive state-funded assistance in the form of vouchers. 

\b Time limit policies were implemented statewide in all Texas
counties with JOBS programs, covering about 90 percent of the state's
welfare recipients.  These policies were initially implemented as a
pilot in one county in June 1996 and expanded to five additional
counties in September 1996, before going statewide in January 1997. 
Texas also has a 60-month grant termination time limit, with no
extensions. 

\c While Wisconsin's time limit policy was implemented in September
1997, it provides for counting months retroactively to October 1996,
excluding those months the recipient was an active JOBS participant. 

\d Months do count if the 12-week-old or younger infant was born more
than 10 months after the parent was first determined to be eligible
for assistance and the child was not the result of sexual assault or
incest. 

Sources:  National Governors' Association Center for Best Practices,
Summary of Selected Elements of State Plans for TANF, Nov.  20, 1997;
HHS, Setting the Baseline, June 1997; and documents provided by the
states. 

Among the three case study states with shorter grant termination time
limits, as of January 1998, families had begun reaching their time
limits only in Connecticut, where program provisions call for
terminating assistance after 21 months.  Among families reaching
their time limits, however, extensions were often granted (see fig. 
2.5). 

   Figure 2.5:  Reaching Time
   Limits in Connecticut

   (See figure in printed
   edition.)

\a This limit does not apply to those who are exempt from the time
limit on the basis of age, disability, caring for a disabled
relative, caring for a child under age 1, or being determined by the
state to be unemployable. 

Texas imposes both a 5-year grant termination time limit and shorter
limits on some recipients, resulting in grant redutions.  The shorter
limits were implemented under waiver and call for imposing time
limits of 12, 24, or 36 months, depending on the extent of the
recipients' education and work experience.  When these time limits
are reached, the amount of assistance is reduced by the adult's
portion of the grant, but assistance for the children continues.  As
of November 1997, no Texas families had reached their 5-year
termination time limit, but 129 families were receiving reduced
grants because they had reached the shorter limits. 

Two of our case study states--California and Maryland--plan to
provide state-funded assistance if the number of families reaching
the federal 5-year time limit exceeds the 20 percent of state
caseloads that can be exempted from the federal limit on the basis of
hardship.  In California, families reaching the federal 5-year limit,
but not yet reaching the state's 5-year limit (which started later
and counts months differently), will continue to receive full grants
with state funds; families reaching the state's 5-year limit will
continue to receive reduced grants for their minor children--again,
with state funds.  In Maryland, families reaching the federal 5-year
limit will continue to receive state-funded assistance for their
minor children through a third party or through vouchers. 


--------------------
\46 Among the remaining states, Iowa adopted individualized time
limits; Michigan and Vermont have no time limit under state law; and
four states plan to reduce, but not terminate, grants when families
reach the 5-year time limit.  See L.  Jerome Gallagher and others,
One Year After Federal Welfare Reform, May 1998, p.  IV-3. 


         TRACKING TIME LIMITS
         POSES NEW CHALLENGES
------------------------------------------------------ Chapter 2:2.2.2

To enforce federal lifetime time limits and determine federal
eligibility for a case, states are required to track the monthly
receipt of assistance for individual cases indefinitely, including
assistance that may have been received in other states.  In addition,
many states have adopted state-specific time limits differing from
the federal limit regarding when the clock started, when a month
counts toward the limit, and when the limit is reached.  As a result,
those states must track receipt of assistance against both a federal
and a state clock, often using different criteria.  States' automated
information systems generally were not collecting all the data
required to track cases against these federal and state limits, and
upgrading their systems to do so poses significant new challenges for
most states. 

HHS reported in December l997 that nearly two-thirds of states either
are doing nothing to track cases against the federal 5-year limit or
are still planning how to do so.\47 APWA reported that, while states
are making progress toward automating the tracking for the federal
5-year time limit internally, this tracking requires states to make
substantial modifications or replacements to information systems
software and communications networks, and to increase data storage
capacity, develop new applications, and improve transaction
processing speed. 

Among our case study states, Louisiana, Maryland, and Wisconsin had
systems in place to track both federal and state time limits.  In
California, Connecticut, and Texas, systems were still under
development.  California officials said that programming their
computer system to track both federal and state time limits will be
challenging because of the different rules to be applied.  The state
is trying to implement a "statewide automated welfare system," but
this system is not expected to be up and running before 2002. 
Meanwhile, California is using data from its Medicaid information
system to track which months, if any, a case receives TANF
assistance.  As of January 1998, Texas did not have the capability to
track cases against the federal 5-year time limit and was awaiting
guidance from HHS concerning implementation of inconsistent waiver
provisions before developing a system.  HHS had informed Texas that
until the year 2002, when its waiver expires, only those recipients
in counties that have been participating in the state's JOBS program
are subject to the federal time limit.  But state officials said this
gives recipients in non-JOBS counties no incentive to find work, and
officials were uncertain about providing guidance to local offices on
how time limits should be enforced pending final HHS regulations. 
Oregon's data system has the capacity to track months on aid within
the state, which it is doing for target populations under the new
program; however, no effort is being made to apply this tracking to
the time limit issue because, as a result of Oregon's waiver, no
cases are currently subject to the federal limit, and no cases are
expected to reach the state limit because of the state's sanction
policy.\48

States face even greater challenges in developing systems that will
allow tracking receipt of assistance across state lines.  Currently,
there is no national computer system that tracks the receipt of
assistance under TANF.  While HHS has proposed various options, many
issues about how best to design a system that can share or exchange
information nationwide remain unresolved.  Officials in some states,
such as Maryland and Wisconsin, told us they are attempting to
contact other states for information.  Maryland workers are relying
on telephone calls to other states to obtain this information, which
is time-consuming and lacks any assurance of accuracy, while
Wisconsin recently updated an out-of-state inquiry form requesting
the number of months recipients were subject to time limitations. 
Officials in Texas told us that they will track time limits on
assistance received only within the state, on the basis of criteria
under the state's waiver. 


--------------------
\47 The report, HHS, Report on Data Processing (Washington, D.C.: 
HHS, Dec.  1997), was based on a survey of states conducted by the
National Governors' Association, APWA, National Association of State
Information Resource Executives, and National Conference of State
Legislatures.  Although the survey was conducted in November and
December 1996, HHS believes the survey's results continue to reflect
the status of states' progress. 

\48 In Oregon, months count toward the time limit only if the family
fails to cooperate, and the state has graduated sanctions resulting
in a full-family sanction for failure to participate.  Officials told
us they do not expect any families to ever reach the state time
limits in Oregon because, if families are cooperating, they can
continue to receive cash assistance indefinitely (funded by the state
after the waiver expires in the year 2002); if families are not
cooperating, their grants will be terminated long before the time
limit is reached. 


      ELIGIBILITY FOR SPECIFIC
      GROUPS AND GRANT LEVELS
      REMAIN GENERALLY UNCHANGED
-------------------------------------------------------- Chapter 2:2.3

Under federal reform, states are allowed to establish their own
eligibility policies for specific groups and to determine grant
levels, but most states have not made significant changes.  Changes
to eligibility policies generally have been limited to simplifying
the criteria for two-parent families and denying assistance to
convicted drug felons.  Further, at least for the present, states
generally have maintained grant levels, dispelling concerns that they
would immediately engage in a "race to the bottom" by lowering the
amount of grants and redirecting those savings to other programs. 


         STATES ARE MAKING FEW
         CHANGES TO ELIGIBILITY
         CRITERIA FOR SPECIFIC
         GROUPS
------------------------------------------------------ Chapter 2:2.3.1

Although allowed under the new federal welfare law to deny assistance
to specific groups, states are continuing to provide assistance to
most groups that were eligible for AFDC, and some states are
expanding eligibility for two-parent families.  Most states, however,
have opted to deny assistance to convicted drug felons. 

Two-parent families continue to be eligible to receive assistance in
all states, and in some states, the eligibility criteria have been
simplified and expanded.\49 Liberalization of eligibility for
two-parent families has been promoted as a way to mitigate
disincentives to the formation and maintenance of two-parent families
that were perceived to have existed under prior law.  To be eligible
for AFDC, the law required that one parent in a two-parent family be
unemployed or working fewer than 100 hours a month, and have a
history of attachment to the workforce.  Beginning under waivers and
continuing under TANF, many states have taken steps to eliminate some
or all of these rules.  For example, 32 states received waivers to
eliminate the 100-hour rule, and 24 states received waivers for the
work history rule.  As of October 1997, TANF plans in 35 states
called for using the same eligibility criteria for two-parent
families as are used for one-parent families.  All but one of our
case study states had adopted such provisions.  In California, the
one exception, two-parent applicant families are still subject to the
100-hour rule.\50

Families headed by minor teen parents will also continue to be
eligible for assistance in nearly all states, but with new
restrictions.  For example, in Wisconsin, teen parents under age 18
are regarded as the financial responsibility of their parents and are
ineligible for cash assistance payments.  They may, however, receive
case management services and other types of assistance such as child
care and food stamps.\51 Similarly, in Maryland, minor teen parents
are not allowed to receive cash assistance directly but must receive
grants through a responsible third-party payee.  For other states
continuing to provide cash assistance to minor teen parent families,
the new federal law makes eligibility for these families contingent
upon compliance with requirements to attend school regularly and live
in an adult-supervised setting.  Before federal reform, 15 states had
adopted such requirements under waivers.  Now, all states providing
assistance to teen parent families must adopt such requirements or
provide assistance with state funds.\52 According to one Texas
official, these requirements will be challenging to implement because
the state has limited alternative adult-supervised settings for teen
parents who cannot live at home. 

Although the new federal law allows states to deny TANF assistance to
certain "qualified alien" families residing in the United States at
the time welfare reform was enacted\53 --August 22, 1996--all states
except Alabama indicated they will continue to provide assistance to
these families, according to state plans as of November 1997.  In
addition, although qualified alien families arriving after August 22,
1996, are prohibited from receiving TANF assistance for 5 years under
the new federal law, several states, including California and
Maryland, indicated they will use state funds to provide aid to these
families as well.\54 California officials estimated the state cost
for providing assistance for those no longer federally eligible will
be minimal, at least for the next 6 months, but they noted that they
may experience difficulty in distinguishing between those who are
federally eligible and those who are not because the date of entry
into the country has not been routinely gathered in the past. 

People convicted of drug felonies compose the one group to which most
states have opted to deny eligibility.  Under the new federal law, an
individual convicted of a felony involving the possession, use, or
distribution of a controlled substance after August 22, 1996, shall
not be eligible for assistance, unless the state opts out of this
provision by enacting a law specifically addressing this point. 
However, children in families that include adults who are convicted
drug felons continue to be eligible to receive assistance. 
Nationwide, 37 states have adopted this provision to deny cash
assistance to convicted drug felons, although some have narrowed the
types of felonies to which this prohibition applies.  The 17 states
passing legislation to opt out of this provision (including 2 of our
7 case study states:  Connecticut and Oregon) have generally made
continuing assistance to convicted drug felons contingent upon
participation in a drug treatment program.\55


--------------------
\49 AFDC's two-parent program, called the "Unemployed Parent"
program, was initiated in 1961 and was mandated nationwide under the
Family Support Act of 1988 (effective Oct.  1, 1990), enabling
two-parent families to qualify for assistance. 

\50 See L.  Jerome Gallagher and others, One Year After Federal
Welfare Reform, May 1998, pp.  III-13 through 15. 

\51 Minor teen parents in Wisconsin who cannot be supported by their
parents are referred to child welfare services for appropriate living
arrangements. 

\52 States that had implemented a waiver with inconsistent teen
parent provisions before federal reform was enacted are exempt from
this requirement.  For example, California is exempt from these
provisions because of its "Cal-Learn" waiver, discussed later in this
chapter. 

\53 Title IV of the Personal Responsibility and Work Opportunity
Reconciliation Act defines qualified aliens as legal permanent
residents, aliens paroled into the United States for at least 1 year,
refugees, and aliens granted asylum or certain similar relief. 

\54 GAO has another study under way examining the restrictions the
new welfare reform law has placed on the eligibility of legal
immigrants. 

\55 See National Governors' Association Center for Best Practices,
Summary of Selected Elements of State Plans for TANF, Nov.  20, 1997. 


         GRANT LEVELS REMAIN
         UNCHANGED IN MOST STATES
------------------------------------------------------ Chapter 2:2.3.2

Most states have not changed the grant levels paid to families that
were in effect when the federal reform law was passed, thus
dispelling for now concerns that federal block grants might result in
widespread reductions in grant levels.  In its summary of state
plans, the Urban Institute reported that only 11 states made
adjustments to their grant levels between 1996 and l997:  5 states
increased their grant amounts, 4 reduced their grant amounts, and 2
combined TANF and food stamps into a single grant.  The remaining
states kept the same grant amounts that were in effect in July
1996.\56 Among our seven case study states, California reduced its
grant levels by 4.9 percent, while Wisconsin increased its grant
levels by over 30 percent for a family of three assigned to the
"Community Service Job" component (see table 2.5). 



                         Table 2.5
          
            Maximum Grant Levels for a Family of
           Three in Seven States Before and After
                       Federal Reform

                                   New grant
                      Previous         level
                   grant level      (October     Change in
State              (July 1996)         1997)   grant level
----------------  ------------  ------------  ------------
Calif.\a                  $596        $565\b          -$31
Conn.\a                    543           543     Unchanged
La.                        190           190     Unchanged
Md.                        373         377\c           +$4
Oreg.                      460           460     Unchanged
Tex.                       188           188     Unchanged
Wis.                       517     673/628\d     +$156/111
----------------------------------------------------------
\a Grant levels vary by county or city within the state.  The amount
shown is the grant level for the area containing the largest portion
of the state population. 

\b Amount shown is for nonexempt families. 

\c According to state officials, Maryland's grants are indexed to
inflation and adjusted on the basis of increases in the Consumer
Price Index.  In 1998, the grant level for a family of three was
increased again to $388. 

\d Families of three assigned to the "Community Service Job"
component receive $673; families of three assigned to the "W-2
Transition" component receive $628 (see fig.  2.6). 

Source:  L.  Jerome Gallagher and others, One Year After Federal
Welfare Reform, May 1998, pp.  VI-2 and VI-3, unless otherwise noted. 

Some states have opted to change their grant levels for families only
under certain circumstances.  For example, 21 states have adopted
policies not to increase cash assistance payments to families who
have additional children while on assistance, commonly referred to as
"family caps." In addition, under the new federal law, states are
allowed to pay cash assistance to families migrating from other
states on the basis of the grant levels of the prior state for the
first 12 months of assistance.  Nationwide, 14 states have adopted
such policies, including 2 of our 7 case study states:  California
and Maryland.  However, as of January 1998, California had not yet
implemented this policy because of a court challenge. 


--------------------
\56 The five states that increased their grant amounts were Maryland,
Montana, South Carolina, Vermont, and Wisconsin; the four states that
reduced their grant amounts were California, the District of
Columbia, Idaho, and Wyoming; the two states that combined their TANF
and food stamps into a single grant were Minnesota and North Dakota. 
See L.  Jerome Gallagher and others, One Year After Federal Welfare
Reform, May 1998, pp.  VI-1 through VI-3.  In April 1997, the
National Association of State Budget Officers reported that six
additional states were planning to recommend changes to grant levels
for 1998. 


   STATES HAVE PURSUED STRATEGIES
   TO REDUCE THE NEED FOR WELFARE
---------------------------------------------------------- Chapter 2:3

States' reform programs have also included various efforts to reduce
the need for cash assistance through heightened expectations of
self-reliance and personal responsibility.  In this spirit, states
have pursued strategies to divert individuals from applying for cash
assistance; enhance child support collections from noncustodial
parents; and prevent out-of-wedlock pregnancies, especially among
teens. 


      APPLICANTS ARE BEING
      DIVERTED WITH FOCUS ON
      "WHAT'S NEEDED"
-------------------------------------------------------- Chapter 2:3.1

A major new strategy states are using to reduce the need for welfare
is "diversion"--that is, families are diverted from receiving monthly
cash payments if they can be assisted through other means.  Before
reform, when families walked through the door of a typical welfare
office, the emphasis was on determining their eligibility for
assistance and completing their application.  Now, in states with
diversion programs, the emphasis is on determining what families need
to support themselves--perhaps without monthly cash assistance
payments.  In some cases, a one-time cash payment; support services
such as child care, transportation, or health benefits in lieu of
cash; or help in finding a job can enable a family to maintain its
self-sufficiency without ever going on the welfare rolls.  According
to TANF plans submitted as of November 1997, 30 states indicated they
have diversion programs, including 5 of our 7 case study states: 
California, Maryland, Oregon, Texas, and Wisconsin.\57


--------------------
\57 See National Governors' Association Center for Best Practices,
Summary of Selected Elements of State Plans for TANF, Nov.  20, 1997,
p.  5. 


         PROVIDING ONE-TIME CASH
         PAYMENTS
------------------------------------------------------ Chapter 2:3.1.1

One-time cash payments can help families support themselves in a
number of ways.  For example, such payments may enable families to
get their cars repaired so they can get to work, make overdue rent
payments so they can avoid eviction from their homes, or get through
a medical emergency that temporarily precludes work.  Unlike
emergency assistance provided to prevent destitution and homelessness
among families in the worst of situations, diversion assistance is
provided to families who just need a modest boost to remain off the
welfare rolls.  All 30 states with diversion programs had provisions
for one-time cash payments.  Payment amounts may be up to $1,000 or
more, depending on the state, and applicants receiving the payments
are typically prohibited from receiving TANF cash assistance for a
specified time period.  The most commonly reported period of time (or
equivalency value of assistance) for diversion assistance is a
maximum of 3 months, to be negotiated between the applicant and the
caseworker depending on immediate need. 

Among our case study states, California requires its counties to
provide diversion payments but allows counties to determine who shall
be eligible and how much assistance to provide.  In Wisconsin, the
payments are in the form of "job access loans" that must be repaid or
worked off.  In determining whether such payments are appropriate,
caseworkers in these states consider such factors as employment
history, employment prospects, housing stability, and whether the
expense is related to obtaining or maintaining employment or to a
discrete financial crisis (as opposed to an ongoing financial need). 
During 1997, Maryland provided one-time cash diversion payments to
643 families, with fewer than 4 percent of these families returning
to receive ongoing monthly cash assistance. 


         PROVIDING SUPPORT
         SERVICES IN LIEU OF CASH
------------------------------------------------------ Chapter 2:3.1.2

The availability of support services such as child care,
transportation assistance, and health care benefits--combined with
food stamps but not tied to the receipt of monthly cash
assistance--also may sometimes enable families to maintain their
self-sufficiency without going on the welfare rolls.\58 For example,
among our case study states, we found that Maryland would provide
child care and medical assistance in lieu of cash for families who
already have some prospects of employment.  In Oregon, a study of
applicants in the Medford branch office between January and June 1997
found that 83 percent of those diverted from monthly cash payments
received other assistance, such as food stamps, medical benefits, or
child care.  Some states have faced barriers in implementing this
strategy in the past, however.  In the Texas counties we visited, the
employment services program had periodically run out of funds for
child care assistance for TANF applicants.\59

In California, officials found that most families, when given a
choice, opt to receive both cash assistance and services, rather than
services without the cash.\60 In Wisconsin, families may opt to
receive such services as child care, medical benefits, and job
placement assistance as an alternative to enrolling in TANF. 
Furthermore, those families enrolling in TANF who are determined to
be "job-ready" may receive case management services in addition to
these other services, but they are not eligible to receive monthly
cash payments (see fig.  2.6).\61

   Figure 2.6:  Diverting
   Applicants From Cash Assistance
   in Wisconsin

   (See figure in printed
   edition.)

\a Families in any of the four components can be reassigned to
another component at any time on the basis of a caseworker's
evaluation of their progress. 


--------------------
\58 Unlike TANF cash assistance, medical assistance under the
Medicaid program remains an entitlement for all eligible people. 

\59 According to Texas state officials, appropriations for child care
to support welfare reform efforts have since increased significantly. 

\60 California implemented a waiver in 1994 separating the provision
of child care and health services from the provision of cash
assistance, but as families were eligible for both, nearly all opted
to receive both.  Under California's new reform law, families
continue to be eligible for both and will receive services in lieu of
cash only at the family's option.  It is too early to tell whether
time limits and the more rigorous work requirements will encourage
more families to take the "services-only" option. 

\61 TANF families determined to be job-ready, and therefore
ineligible for monthly cash assistance, may receive a "job access
loan" as described in the previous section. 


         PROVIDING JOB SEARCH
         ASSISTANCE
------------------------------------------------------ Chapter 2:3.1.3

Providing families assistance with job search and requiring
applicants to engage in job search activities before making them
eligible to receive monthly cash payments also may enable applicants
to find jobs and be diverted from the welfare rolls.  For example,
under Texas' welfare reform program, most applicants are required to
attend workforce orientations provided by the employment services
agency or local workforce development board providers before they are
approved to receive TANF cash payments.  Applicants may access
services available through the workforce centers to assist them in
obtaining employment.  Under Maryland's welfare reform program,
applicants are required to participate in up-front job search as a
condition of eligibility, and, on the basis of partial data, at least
9 percent of all job placements in fiscal year 1997 were for
applicants who never received cash assistance.  Under Oregon's
program, most applicants are required to be engaged in job search for
30 days before being fully screened for TANF eligibility.  In
Wisconsin, applicants judged to be "job-ready" are assigned to the
unsubsidized job component and are required to engage in job search
activities with no cash assistance (see fig.  2.6 above).  If an
applicant is unable to find work, caseworkers review the case for
barriers to employment and may assign the family to another program
component that would include cash assistance.  In California, we
found that in two of the three counties we visited, applicants were
required to participate in such activities, and, if they failed to do
so, their applications would not be processed or would be denied, and
no cash assistance would be provided. 


         STRONGER PROGRAM
         REQUIREMENTS ALSO RESULT
         IN DIVERSIONS
------------------------------------------------------ Chapter 2:3.1.4

Some potential recipients are also being diverted from pursuing their
applications simply by learning of the stronger work and child
support enforcement requirements at orientation.  Under prior law,
program staff were required to inform applicants of program rules,
including work requirements under the JOBS program and child support
enforcement requirements.  However, these requirements often were not
imposed immediately on applicants; and once applicants were on the
rolls, if they failed to comply, their grants could be reduced but
not terminated.  HHS data from fiscal year 1992 indicate that about
11 percent of welfare applicants failed to complete their
applications after the initial orientation process.  Under welfare
reform, most states are strengthening their work requirements and
child support requirements and imposing these requirements sooner. 
Preliminary data for a few sites indicate that, once these more
rigorous requirements are explained, more families may decide not to
pursue their applications than in the past.  For example, one
Maryland county official estimated that between July 1996 and January
1997, 19 percent of potential recipients withdrew their applications
once new program requirements were explained to them.  In Oregon, the
Medford study found that 22 percent of potential recipients did not
pursue applications after program requirements were discussed.  And
during January 1998 in one Wisconsin county, officials found that 84
percent of the families entering a welfare office decided not to
pursue an application for TANF assistance once they learned of the
new program rules. 


         PRELIMINARY RESULTS
------------------------------------------------------ Chapter 2:3.1.5

States were in the early stages of implementing their diversion
strategies at the time of our review, and comprehensive statewide
data on the effects of diversion were not yet available.  On the
basis of preliminary data, however, state officials in both Oregon
and Wisconsin estimated that about 40 percent of applicants likely to
have qualified for cash assistance were being diverted statewide.  In
a study of 1,169 applications for cash assistance in Medford, Oregon,
during the first 7 months of 1997, 74 percent of potential recipient
families were diverted from the rolls, including 52 percent who found
employment and 22 percent who did not pursue their applications after
learning of program requirements.\62 And as mentioned previously,
according to data compiled in Dane County, Wisconsin, during January
1998, 84 percent of potential recipients were diverted from pursuing
applications for TANF assistance, and of those enrolled as of
February 1998, over 25 percent were receiving TANF services but no
cash. 

While diversion strategies have contributed to declining caseloads,
state officials are cautious in declaring success.  Use of cash
assistance may be reduced with diversion, but the demand for other
forms of assistance and services may remain high, and resources for
these other services may be limited.  Furthermore, limited data exist
on the economic stability and well-being of families diverted from
receiving ongoing cash assistance. 


--------------------
\62 Of the total 1,169 applicants, 17 percent went on cash assistance
and 9 percent were found ineligible.  Those found ineligible included
those who had begun receiving unemployment compensation, had Social
Security benefits or child support, had no eligible children in the
household, or had moved from the area.  Outcome data on those not
going on cash assistance were based on a sample of 500 cases. 


      WELFARE PROGRAMS ADD TO
      EFFORTS TO COLLECT CHILD
      SUPPORT
-------------------------------------------------------- Chapter 2:3.2

In fiscal year 1996, only about 13 percent of the 7.4 million AFDC
child support cases nationwide received at least one child support
payment, according to preliminary data from HHS.  Recognizing that
child support is important to a family's self-reliance and that both
parents should be held financially responsible for their children,
states' welfare offices have taken steps to enhance child support
collections on behalf of current and potential recipient families. 
Among our case study states, we found the following:  (1) policies
that terminate the entire family's grant for failure to cooperate
with child support enforcement efforts without good cause, (2)
policies that allow welfare families to receive more of the support
collected on their behalf to encourage greater cooperation, and (3)
efforts to work with the courts to require noncustodial parents to
participate in work activities to bolster their ability to pay child
support. 

Prior law allowed states to sanction or deny assistance to adults who
failed to cooperate with child support enforcement requirements
without good cause, but not to deny assistance to the children.  The
new federal law requires states to reduce a family's grant amount by
at least 25 percent for such failure to cooperate and provides the
option for states to deny assistance to the entire family.  Before
federal reform, nine states had adopted waiver provisions to
terminate a family's entire grant for such failure to cooperate,
including two of our seven case study states:  Connecticut and
Maryland.  After reform, a survey conducted by APWA in the summer of
1997 found that the number of states with policies to terminate
grants on this basis has grown to 16; however, nationwide data on the
number of grants terminated under TANF for failure to cooperate with
child support enforcement requirements have not yet been compiled.\63

To encourage cooperation, prior law also required states to give a
recipient family the first $50 of current support collected per month
on the family's behalf, with the remainder used to offset the amount
of the welfare payment being provided.  The new law eliminated this
requirement, but states can opt to continue the practice with their
own funds.\64 According to TANF plans as of October 1997, 22 states
were continuing policies to pass through some portion of any child
support payment received.  Among our case study states, California,
Connecticut, Texas, and Wisconsin were continuing such policies, with
Connecticut increasing the amount of the pass-through to $100 and
Wisconsin adopting provisions to pass through the entire amount of
the child support payment to the custodial parent.\65

Under prior law, participation in the JOBS program was generally
limited to AFDC applicants and recipients and did not include
noncustodial parents.\66 Before passage of federal reform, however,
23 states had received waivers allowing noncustodial parents to
participate in the JOBS program, either on a voluntary basis or as
required by the courts, including 5 of our 7 case study states: 
California, Connecticut, Maryland, Oregon, and Wisconsin.  In
addition, under the new federal reform law, states are required to
report the number of noncustodial parents participating in work and
work-related activities under the states' TANF programs.  During our
site visits, we found that several local welfare offices had
implemented efforts to work more closely with the courts to require
noncustodial parents to participate in programs designed to place
them in jobs and increase their ability to pay child support.  For
example, in both Fresno and Santa Clara counties, California, the
welfare agencies had initiated such programs.  Officials in Santa
Clara told us that they not only expected to increase child support
collections by placing more noncustodial parents in jobs, but they
also believed that such an effort would identify a significant amount
of previously unreported income.  Similarly, in Wisconsin, one facet
of the "Children First" program is to help fathers meet child support
obligations by providing them services that are designed to enable
them to obtain and retain employment. 


--------------------
\63 See APWA, Survey Notes:  Issues Affecting Children, Vol.  1, No. 
3 (1998). 

\64 The federal share of amounts collected on behalf of recipient
families must be paid to the federal government, so any amount passed
through to the family must come from the state's share. 

\65 Under Wisconsin's program, the entire amount of the child support
payment is disregarded for computation of the grant amount, but not
for eligibility determination.  See L.  Jerome Gallagher and others,
One Year After Federal Welfare Reform, May 1998, pp.  VI-11 and
VI-12. 

\66 In a pilot effort, however, prior law did authorize HHS to permit
up to five states to provide JOBS services, on either a voluntary or
a mandatory basis, to noncustodial parents who were unemployed and
unable to meet their child support obligations. 


      PROGRAMS TO REDUCE
      OUT-OF-WEDLOCK PREGNANCIES
      ARE BEING EXPANDED
-------------------------------------------------------- Chapter 2:3.3

Because of the strong link between teenage childbearing and welfare
receipt, states are trying to reduce the need for welfare through
efforts to prevent out-of-wedlock pregnancies, especially among
teens.  Most states have had such programs under way for more than 25
years as part of their family planning services programs, funded
under title X of the Public Health Services Act of 1970.  And since
1978, adolescent family planning services have been specified in the
statute as one of its essential services.  More recently, however,
states have begun to focus attention on other strategies to reduce
teen pregnancies, such as abstinence education, the enforcement of
statutory rape laws, and male involvement programs. 

Under federal welfare reform, states are required to include in their
TANF plans a description of how they intend to establish goals and
take action to prevent and reduce the incidence of out-of-wedlock
pregnancies, especially among teens, and to conduct programs that
provide education and training on the problem of statutory rape.  HHS
will rank the states annually on the basis of their reductions in
out-of-wedlock births among families receiving TANF assistance and
will award bonuses to up to five states for reductions in
out-of-wedlock births among the general population. 

In addition, HHS is required to develop a national strategy to
prevent out-of-wedlock teen pregnancies and ensure that at least 25
percent of communities in this country have teen pregnancy programs
in place.  In response to this mandate, in January 1997, HHS issued a
national strategy emphasizing the need to prevent teen pregnancies by
encouraging teens to postpone sexual involvement, stay in school, and
prepare for work.  A report to the Congress is due this month. 


         ABSTINENCE EDUCATION
------------------------------------------------------ Chapter 2:3.3.1

The federal reform law included a provision to enhance states'
efforts to provide sexual abstinence education and authorized $50
million annually for this purpose.  States can use this money only
for the development of mentoring, counseling, and adult supervision
programs that encourage abstinence.  All 50 states have applied for
these funds with proposals to initiate new programs or expand upon
existing efforts that focus on abstinence.\67

As part of their efforts to reduce teen pregnancies, all our case
study states proposed increasing their efforts to provide abstinence
education aimed at both males and females.  For example, officials in
both California and Texas told us they plan to use the additional
abstinence funds to provide grants to various local communities to
administer abstinence and mentoring programs, and for a statewide
media campaign on abstinence aimed at males.  Both states plan to
perform evaluations to determine the impact of these activities.  In
Wisconsin, officials have also proposed using portions of the state's
abstinence funds for a media campaign, including an evaluation to
determine impact.  In addition, the Congress has authorized funds for
HHS to conduct an evaluation of states' abstinence programs. 


--------------------
\67 One state, New Hampshire, has decided not to spend its grant,
however, because of difficulties in administering an abstinence-only
program. 


         ENFORCEMENT OF STATUTORY
         RAPE LAWS
------------------------------------------------------ Chapter 2:3.3.2

All states have statutory rape laws but generally have not rigorously
enforced them.  In response to studies that show that a substantial
number of teen girls are impregnated by older men and that many of
these encounters, especially among younger teens, involve coercion,
states have begun to increase their enforcement efforts.\68

While some states had initiated programs before enactment of the new
federal law, most states are enhancing their efforts in response to
the new requirement to expand their teenage pregnancy prevention
programs to include men.  Under the new federal law, states are
required to indicate in their TANF plans how they intend to address
the problem of statutory rape, including education and training
programs designed to reach law enforcement officials, the education
system, and relevant counseling services.  The new federal law also
recommends that states aggressively enforce statutory rape laws.\69
According to state plans submitted as of August 1997, 29 states
provided descriptions of the statutory rape education programs they
planned to implement, while other states provided few details.  Among
our case study states, some--such as California and
Connecticut--outlined extensive efforts, while others--such as Oregon
and Wisconsin--indicated only that they planned to conduct a program
as required. 

California began strengthening its enforcement of statutory rape laws
as a pilot in 1995 and expanded the program statewide in 1996, before
federal reform was enacted.  To increase the prosecution and
conviction of adults who have unlawful intercourse with minors,
specialized units were established within the county district
attorneys' offices to investigate and prosecute cases, with the same
prosecutor following a specific case all the way through the judicial
process.  In addition, California's civil penalties for statutory
rape were increased effective January 1, 1997. 

In response to the new federal requirements, Connecticut's governor
declared that statutory rape laws are to be more aggressively
enforced throughout the state and appointed a special prosecutor to
enforce the statutory rape laws in Hartford, the city with the
highest teen birth rate in the state.  In addition, the state welfare
department is working with the state health department to develop a
media campaign to educate young women about their rights under the
statutory rape laws and to educate young men that statutory rape is a
crime and will be prosecuted. 


--------------------
\68 One study showed that men 20 years or older are responsible for
half the pregnancies among girls aged 15 to 17.  These pregnancies
have the potential to affect girls' ability to finish high school and
become independent adults, thus increasing the likelihood of welfare
dependency.  See Laura D.  Lindberg and others, "Age Differences
Between Minors Who Give Birth and Their Adult Partners," Family
Planning Perspectives, Vol.  29, No.  2 (Mar./Apr.  1997). 

\69 Under the new law, the U.S.  Attorney General is mandated to
establish and implement programs that study the links between teen
pregnancy and statutory rape, especially by predatory older men
committing repeat offenses, as well as to educate state and local
criminal law enforcement officials on statutory rape prevention and
prosecution. 


         MALE INVOLVEMENT
------------------------------------------------------ Chapter 2:3.3.3

Many years before federal reform, states had initiated programs
involving males, implemented in conjunction with their teen pregnancy
prevention efforts.  In response to the new law, however, some
states' TANF plans indicate that the states have added to or enhanced
their male involvement programs to emphasize abstinence and the
prevention of statutory rape. 

Among our case study states, for example, Connecticut's state plan
directly mandates its service providers who are administering teen
pregnancy prevention programs to incorporate a component that
comprehensively addresses young males.  Also, the state has developed
a public awareness program that includes a video and curriculum
materials targeted at males and females that will be distributed to
middle schools and high schools.  Louisiana is incorporating its male
involvement program into its effort to reduce the unwed teen
pregnancy rate in the state by at least 1 percent during 1998.  The
program will offer specialized services to preadolescent and
adolescent males with an emphasis on manhood development, including
counseling, mentoring, and tutoring.  The state's plan also
recommends that male leaders from business and academia be solicited
as guest speakers and mentors and to act as role models.  And in
Maryland, the state created a "Responsible Choices" task force, which
developed strategies targeting teen boys as well as teen girls, and
young men as well as young women, in an effort to reduce the number
of nonmarital births through various prevention and abstinence
programs.  An example of the type of material disseminated in
California's public awareness campaign targeting males is shown in
figure 2.7. 

   Figure 2.7:  Public Awareness
   Announcement Targeting Males in
   California

   (See figure in printed
   edition.)


         TEEN PARENT AND FAMILY
         CAP REQUIREMENTS
------------------------------------------------------ Chapter 2:3.3.4

Finally, some states maintain that more stringent requirements for
those already on welfare can serve as a disincentive for future
out-of-wedlock births.  Some states specifically cited teen parent
requirements and family caps as part of their efforts to reduce
out-of-wedlock pregnancies. 

The new federal welfare reform law enacted strict new provisions with
respect to teen parents, including providing "no assistance for
teenage parents who do not attend high school or other equivalent
training program" and "no assistance for teenage parents not living
in adult-supervised settings." According to some state officials with
whom we spoke, these provisions can not only deter teens from having
their first child, but--perhaps more effectively--can help to
encourage teens who are already parents not to have any more children
until they finish school and become self-sufficient.  As a result,
some states explicitly include these policies in the out-of-wedlock
initiatives outlined in their TANF plans.\70

For example, California's strategy to address teen pregnancy includes
various intervention programs directed toward pregnant and parenting
teens.  All these programs predate the passage of federal welfare
reform, and most have been in existence for more than 10 years.  The
Adolescent Family Life program, initiated in 1985 and administered by
the Department of Health Services, provides pregnant and parenting
teens with an individual case manager who helps link them to
education, health, social services, and other assistance.  In
addition, various teen parent programs have been initiated and are
administered by the Department of Education, including the Pregnant
Minor program, the School-Age Parent and Infant Development program,
the Pregnant and Lactating Students program, and the Gender Equity
Teen Parent program.  Finally, the Department of Social Services
administers the Cal-Learn program, authorized in 1993, for all teen
parents eligible for TANF (previously AFDC).  Cal-Learn encourages
teen parents to graduate from high school or its equivalent by meting
out cash bonuses and sanctions on the basis of teens' school
performance.  Cal-Learn also provides case management, child care,
and transportation services.  State officials told us these programs
were key to their effort to reduce the number of out-of-wedlock
births due to second and subsequent pregnancies among teen parents on
welfare. 

In addition, while the federal law is silent on whether to prohibit
grant amount increases for families on assistance when another child
is born (commonly referred to as "family cap" provisions), according
to TANF plans submitted nationwide as of October 1997, 21 states have
adopted some type of family cap provision as part of their effort to
discourage subsequent births, including 3 of our 7 case study states: 
California, Connecticut, and Wisconsin.\71 In Connecticut, a flat
payment of $50 per month is given for each additional child born to a
family receiving assistance.\72 In addition, while Maryland officials
do not consider their policy to be a family cap, the state's plan
calls for increasing the amount of the grant only for the specific
needs of an additional child born to a family on assistance and for
paying this assistance directly to the vendor of the goods or
services rather than to the caregiver. 


--------------------
\70 See Jodie Levin-Epstein, State TANF Plans:  Out-of-Wedlock and
Statutory Rape Provisions (Washington, D.C.:  Center for Law and
Social Policy, Aug.  1997), p.  3. 

\71 These figures include states with provisions to provide no
increase, a partial increase, or increased assistance only through a
voucher.  See L.  Jerome Gallagher and others, One Year After Federal
Welfare Reform, May 1998, pp.  VI-8 and VI-9. 

\72 See CRS, Transition to TANF:  State Benefit Levels and Selected
Program Rules, July 1, l996 - January 1, l997 (Washington, D.C.: 
CRS, June 25, 1997). 


         RANKING AND BONUSES FOR
         REDUCING RATES OF
         OUT-OF-WEDLOCK BIRTHS
------------------------------------------------------ Chapter 2:3.3.5

Under the new federal welfare reform law, HHS is required to rank the
states annually on the basis of their proportion of out-of-wedlock
births among families receiving TANF assistance and the net change
from the previous year, and to review the programs in the five states
ranking highest and lowest.  In addition, the law provides for
bonuses of $20 million for five states demonstrating a net decrease
in out-of-wedlock births among the general population for the most
recent 2-year period without increasing the state's abortion rate.\73

In March 1998, HHS issued proposed regulations outlining the specific
data to be used in awarding these bonuses.  At the time of our
review, all our case study states expressed an interest in these
bonuses; however, some states were uncertain whether the data they
collect on abortions would be adequate to allow them to qualify.\74


--------------------
\73 If fewer than five states are eligible for a bonus, the amount of
the bonus increases to $25 million per state. 

\74 According to the proposed regulations, HHS will identify the
potentially eligible states on the basis of reductions in
out-of-wedlock births and request data on the number of abortions
from those states to verify eligibility.  Some states, such as
California, do not collect abortion data for the general population
and had proposed using data for Medicaid recipients as a proxy rather
than be disqualified on this basis.  However, the proposed
regulations stipulate that the data must count all abortions and
cannot be based on subpopulations such as recipients of public
assistance or Medicaid. 


STATES ARE PROVIDING MORE SUPPORT
SERVICES AND RESTRUCTURING
PROGRAMS
============================================================ Chapter 3

At the same time that states are requiring more welfare recipients to
participate in work activities, they are modifying their programs to
better support recipients in their work-related efforts.  States have
expanded the roles of welfare workers and reorganized local offices
to better support employment.  Further, states are using the
flexibility provided by the new law, along with additional budgetary
resources, to help welfare applicants and recipients address problems
that interfere with work, including the lack of transportation and
child care, and mental and physical health problems.  Expanding
partnerships with other organizations--especially employers--is yet
another way that states are attempting to support recipients' work
efforts.  Finally, some states have restructured program
administration by giving local administrative entities greater
flexibility to tailor programs to the needs of their recipients. 


   STATES ARE EXPANDING WORKER
   ROLES AND REORGANIZING LOCAL
   OFFICES TO SUPPORT WORK FOCUS
   OF PROGRAMS
---------------------------------------------------------- Chapter 3:1

States are expanding the roles of welfare workers and reorganizing
local welfare offices in an attempt to change the "culture" of
welfare offices to reflect the greater emphasis on work.  Training
workers to perform their expanded roles has been among the most
challenging and widespread implementation issues reported by states. 


      WORKERS' ROLES HAVE EXPANDED
      FROM FOCUS ON ELIGIBILITY TO
      JOB COUNSELING
-------------------------------------------------------- Chapter 3:1.1

Recognizing that the success of their programs depends in large part
on how well welfare workers implement reform policies, the states we
visited have generally expanded workers' roles to meet new program
objectives.  This expansion of roles is an attempt to shift the
emphasis of staff from determining eligibility and issuing benefits
to helping applicants and recipients obtain work and become more
self-sufficient.  State and local officials cited a range of new
responsibilities for welfare workers, including communicating to
recipients their responsibility to become employed, motivating
recipients to seek work, exploring the potential for diverting
applicants from the need for monthly cash assistance, and collecting
more information about applicants and recipients to determine what
services would facilitate their becoming more self-sufficient.  As
expressed by a worker in Texas, staff interactions with recipients
have shifted away from a pattern of "Ms.  Jones, has anything
changed?  No?  Okay, see you in 6 months," to a much greater focus on
helping recipients obtain work. 

Training staff to perform their new roles has been one of the major
challenges of welfare reform.  Members of welfare advocacy groups in
several states we visited voiced concerns about what they viewed as
inadequate training of welfare staff in states' new program policies
and cited examples of staff providing inaccurate information about
these policies to applicants and recipients.  State and local
officials highlighted two key aspects of the challenge of training
staff to perform their new roles:  the need to change the perspective
of staff and to help them cope with workload pressures.  Many of the
workers whose roles have been expanded are current or former
eligibility workers, who were originally trained to apply complex
rules to make accurate determinations of welfare eligibility and
benefit levels.\75 In some instances, workers have encountered
difficulties in making the transition to broader responsibilities
that involve learning more about the lives of welfare recipients and
making decisions about the most appropriate course of action for
these recipients.  For example, the director of one Wisconsin office
told us that workers were uncomfortable with the considerable amount
of discretion they had in their new roles that required them to use
their judgment to determine the appropriate work component in which
to place recipients.\76 The director said that he has attempted to
help staff make the transition to their broader roles by working to
keep them focused on client empowerment and employment. 

Officials in Portland, Oregon, told us that the process of shifting
workers' perspectives to focus more on helping families become
self-sufficient had taken several years.  Many factors helped
facilitate this process, including making the expanded role a higher
job classification to provide salary increases and providing training
based on needs as identified by the workers themselves, according to
Portland officials. 

Welfare workers and managers cited a number of factors that
contributed, to varying degrees, to the workload pressures that
impinged upon workers' abilities to perform their new
responsibilities.  In addition to traditional problems, such as large
caseloads and substantial amounts of time required to determine
eligibility and benefit levels and process sanctions, workers cited
the extra time now required for initial meetings with applicants, the
additional time required to keep abreast of policy changes, and the
limited time for training.  For example, officials in the two
Maryland counties we visited reported that much of welfare workers'
time was consumed with processing paperwork to sanction noncompliant
clients, which limited the time available to work with other clients. 
Workers in New Orleans told us that the amount of time required to
interview applicants had tripled under the state's welfare reform
program, in large part because there is now much more information to
be communicated to applicants.  A worker in Wisconsin told us that
much more time is now needed for initial meetings with
applicants--from 2 to 5 hours--to learn more about them in order to
make appropriate decisions.  Some states have taken steps to reduce
such workload pressures.  For example, Oregon has streamlined the
paperwork for determining eligibility to allow staff more time for
case management. 


--------------------
\75 For example, Louisiana, Oregon, and Wisconsin combined
eligibility determination and case management functions in a single
role, which is now carried out by many former eligibility workers. 
Texas continued to keep the roles separate but expanded the
responsibilities of the eligibility worker. 

\76 In Wisconsin, local welfare agencies have some flexibility in
determining the criteria they use to assign individuals to one of the
four components of the "employment ladder" in the state's welfare
reform program.  This flexibility has resulted in some variations
among local agencies in overall caseload assignments.  For example,
at the time of our site visit, two of the local agencies in Milwaukee
County had assigned much higher percentages of individuals to the
unsubsidized employment component than had the other local agencies
in the county.  These individuals may receive case management and
support services but are not eligible for monthly cash assistance. 


      LOCAL OFFICES HAVE BEEN
      REORGANIZED TO CONSOLIDATE
      SERVICE DELIVERY
-------------------------------------------------------- Chapter 3:1.2

Another way in which some states and localities have attempted to
enhance their ability to serve welfare recipients is by reorganizing
local welfare offices to consolidate service delivery.  In some
instances, this has taken the form of moving various welfare-related
staff from various sites to one site.  For example, at the time of
our review, Fresno County, California, was in the process of bringing
together eligibility workers and employment specialists at various
sites to facilitate information sharing among staff and help
recipients obtain employment more quickly.  Fresno County officials
told us that having these staff at the same location would also
enable eligibility workers and employment specialists to hold joint
orientations for applicants that focus on diverting applicants from
monthly cash assistance. 

Service consolidation is also occurring on a larger scale as welfare
offices are being integrated into the workforce development system. 
As part of its effort to encourage states to consolidate workforce
development services, the U.S.  Department of Labor (DOL) has given
46 states grants to implement "One-Stop Career Centers," as of March
1998.\77 One-Stop Career Centers are intended to transform the
fragmented array of employment and training programs into an
integrated service delivery system for job seekers and employers. 
For example, Wisconsin and Texas have efforts under way to shift from
welfare offices to one-stop centers as the service delivery mechanism
for their welfare reform programs.  These centers will provide access
to a range of employment-related services, such as labor market
information, job listings and referrals to employers, job search
assistance, career counseling and job training, and case management. 
Serving welfare recipients through one-stop centers is a way of
reinforcing the message that recipients are responsible for seeking
employment and that staff at the centers are expected to facilitate
recipients' efforts toward this objective.  In addition, by making
services available to all job-seekers, one-stop centers may help
remove some of the "welfare stigma" previously encountered when
services were provided out of a welfare office, according to state
officials. 


--------------------
\77 DOL estimates that actual and projected grants to the states from
1994 through 2001 will total about $450 million.  This figure
includes planning and development grants and implementation grants. 


   STATES ARE USING ADDITIONAL
   BUDGETARY RESOURCES AND
   INCREASED FLEXIBILITY TO
   SUPPORT FAMILIES' WORK EFFORTS
---------------------------------------------------------- Chapter 3:2

Recent declines in welfare caseloads, combined with the TANF block
grant formula and the state maintenance-of-effort requirement in the
new welfare law, have given most states more budgetary resources for
their welfare programs than they would have had under prior program
funding rules.  Further, the new welfare law gave states flexibility
to design their welfare programs and to allocate financial resources
in substantially different ways than were allowed under prior law. 
The states we reviewed were using some of these additional resources
in combination with increased flexibility in spending rules to
enhance support services for applicants, recipients, and those
leaving welfare for work.  The law also provided federal funding for
child care subsidies, and all seven states we reviewed had increased
spending on child care for low-income families during fiscal year
1997. 


      FUNDING RULES FOR WELFARE
      PROGRAMS HAVE CHANGED
-------------------------------------------------------- Chapter 3:2.1

The welfare legislation dramatically changed the way the federal
government and states fund programs and services for needy families
with children.  Previously, the federal government and the states
shared the costs of AFDC, Emergency Assistance, the JOBS program,
AFDC/JOBS Child Care, Transitional Child Care, and At-Risk Child
Care.  Unlimited federal funds were available for all of these
programs except JOBS and At-Risk Child Care, for which federal funds
were capped.\78 As shown in figure 3.1, TANF block grants to the
states replaced the federal portion of the funding for the AFDC,
JOBS, and Emergency Assistance programs.  Each state's TANF block
grant allocation was determined by the federal expenditures for these
three programs in a base year from 1992 to 1995.\79 The law also
required that states continue to spend state funds on low-income
needy families at an amount at least equal to 80 percent of their
fiscal year 1994 spending for all six of the programs.  This
maintenance-of-effort requirement is reduced to 75 percent for states
meeting their required work participation rates. 

   Figure 3.1:  Changes in Funding
   for Assistance to Low-Income
   Families With Children

   (See figure in printed
   edition.)

The welfare reform law also provides states with federal funding to
help meet the child care needs of low-income families.  As shown in
figure 3.1, federal child care funding through the Child Care and
Development Fund (CCDF) was based on prior federal spending for the
three now repealed child care programs as well as the existing Child
Care and Development Block Grant.  Like TANF, the amount of federal
funds available for child care is now capped. 


--------------------
\78 For these programs generally, the federal government matched
state spending at rates ranging from 50 to 78 percent, in proportion
to a state's per capita income.  Administrative costs for these
programs and all costs of the Emergency Assistance program were
matched at a uniform rate of 50 percent. 

\79 Generally, states receive the highest of three amounts:  federal
spending on the programs replaced by TANF for federal fiscal year
1994 or federal fiscal year 1995 or the average of federal spending
on the programs replaced by TANF for federal fiscal years 1992
through 1994. 


      MOST STATES HAVE ADDITIONAL
      BUDGETARY RESOURCES
      AVAILABLE
-------------------------------------------------------- Chapter 3:2.2

Total program costs for AFDC each year were largely determined by the
number of individuals receiving assistance, and states received
federal funds for a portion of total program costs.  Under TANF, the
amount of federal funding is no longer determined by the number of
individuals receiving assistance.  Instead, states receive a fixed
amount of federal funds each year and are required to contribute at
least a minimum level of state funds, the amount of which is based on
past spending levels. 

Since TANF block grants to the states were based on expenditures for
historically large welfare caseloads, the recent decline in caseloads
has resulted in most states having more total budgetary resources
available for their welfare programs than they would have had under
prior funding rules.  In order to estimate the additional resources
available for states' programs under TANF, we constructed estimates
of the 50 states' baseline program budgets under prior funding
requirements and compared these with the states' estimated budgetary
resources under TANF.\80 For the United States as a whole, we
estimated that if all states had received a full year's TANF
allotment in 1997 and had maintained state funding at 80 percent of
historic levels, they would have had about $4.7 billion more than we
estimate they would have spent in 1997 under prior methods of
financing.\81 On average, given the actual caseload size in 1997, we
estimated that states would have had about 25 percent more budgetary
resources under TANF than they would have had under AFDC funding
rules.  Table 3.1 shows these additional federal and state budgetary
resources for the seven states reviewed, compared with estimates of
expenditures these states would have made in 1997 under prior program
funding.\82

Among these states, the estimated increase in resources ranges from 1
percent in Connecticut to 65 percent in Wisconsin. 



                         Table 3.1
          
            Estimated Total Additional Budgetary
               Resources for 1997 Under TANF

                                                Percentage
                  Estimated additional      increase these
                   budgetary resources          additional
          available in 1997, including           resources
                federal TANF and state      represent over
                 maintenance-of-effort      1997 estimated
            spending at 80 percent (in        expenditures
State                        millions)            baseline
--------  ----------------------------  ------------------
Californ                        $548.8                   8
 ia
Connecti                           5.5                   1
 cut
Louisian                          87.4                  48
 a
Maryland                         106.9                  25
Oregon                           109.6                  43
Texas                            239.2                  32
Wisconsi                         270.6                  65
 n
----------------------------------------------------------
Source:  GAO analysis comparing the total amount of 1997 federal TANF
block grants plus state maintenance-of-effort levels at 80 percent
with a baseline.  The baseline is an estimate of what states would
have spent for their 1997 caseloads based on average costs per case
in 1996 for AFDC, JOBS, and Emergency Assistance, adjusted for
inflation (except for cash benefits).  For more information, see our
forthcoming report on state fiscal planning under TANF. 

While current circumstances provide states with additional budgetary
resources, there is no guarantee that this situation will continue
into the future.  Even if, for example, economic conditions weaken
and more families need assistance, the TANF block grant amount will
not increase, and it will be up to each state to decide what level of
benefits to provide and to whom, and to provide additional resources
as it determines necessary.\83


--------------------
\80 Using states' actual 1996 expenditures for AFDC, JOBS, and
Emergency Assistance--the programs TANF replaced--we calculated an
average cost per case and used it, with some adjustments for
inflation, to estimate what total program costs would have been for
the actual 1997 caseloads for these programs.  This amount was
compared with states' federal TANF block grant funds and an amount
equal to an 80-percent state maintenance-of-effort.  For more
information, see our forthcoming report on state fiscal planning
under TANF. 

\81 Federal fiscal year 1997 was a transitional year:  states were
required to implement TANF by July 1, 1997, but many states began
earlier.  Sixteen states were eligible for their full annual TANF
grants for all of fiscal year 1997, and 28 states were eligible for
grants for at least 9 months of the fiscal year. 

\82 Differences between these estimates and those presented in state
budget documents are the result of a variety of factors, including
(1) differences between the state fiscal year and the federal fiscal
year, (2) the fact that most states did not receive TANF for all of
federal and state fiscal year 1997, and (3) assumptions made by state
budget analysts about the effects of program reforms in the state's
baseline that might not have been included in the expenditure data
and assumptions used in our estimates. 

\83 The welfare law did provide a contingency fund from which
qualifying states can receive matching funds up to 20 percent of
their TANF grants when their unemployment rate or Food Stamp program
caseload exceeds a certain level.  To be eligible for such funds, a
state must have maintained 100 percent of its historic state spending
during the fiscal year in which contingency funds are sought. 


      WELFARE REFORM LAW GIVES
      STATES MORE FLEXIBILITY OVER
      PROGRAMS FOR LOW-INCOME
      FAMILIES
-------------------------------------------------------- Chapter 3:2.3

In addition to changing the way the federal government funds
assistance for needy families with children, the new welfare law
gives states flexibility to spend program resources in ways that are
substantially different from those available under prior law.  The
law gives states authority to use TANF funds in ways that were
allowed under the earlier programs and, unless subject to a
prohibition, "in any manner that is reasonably calculated to
accomplish the purpose of the TANF program" for families with
children.  As discussed earlier, states do face many limits on how
they use federal TANF funds--notably time limits.  And federal law
restricts expenditures for medical services and assistance to certain
individuals and families, including certain aliens and teen parents. 
States are, however, liberated from many other detailed rules about
the types of poor families that can be served and the types and
amounts of assistance to be provided.  As a result, states have more
freedom to tailor services to families' unique situations, individual
needs, and local environments.  Further, states can expand
eligibility to provide TANF services to a broader range of families
than had been served under AFDC.\84 This means that supportive
services, which in the past had been linked to eligibility for cash
assistance, can now be provided in the absence of a welfare check,
further freeing states to tailor services to individual need.\85
States have even more flexibility in how they spend their
maintenance-of-effort funds.  For example, these funds are not
subject to time limits and thus can be used to provide services
indefinitely and to certain legal aliens for whom federal funds
cannot be used. 


--------------------
\84 While some services could be provided to low-income nonwelfare
recipients through Emergency Assistance, the focus of that program
was to protect children from destitution or homelessness as the
result of an emergency situation, rather than to promote employment. 

\85 The federal welfare law unlinked Medicaid and cash assistance,
leaving the entitlement to Medicaid for families that would have been
eligible for AFDC based on July 1996 standards. 


      STATES ARE ENHANCING
      SERVICES TO BETTER SUPPORT
      FAMILIES' WORK EFFORTS
-------------------------------------------------------- Chapter 3:2.4

The seven states we visited are using some of their additional
budgetary resources to provide services to help families address
barriers to employment, including lack of child care\86 and lack of
transportation, as well as more complex mental and physical health
problems.  States are also continuing to provide services to families
that have left the welfare rolls as a result of employment,
including, in some cases, providing case management services to help
ensure that families can deal with problems that might put parents'
jobs at risk.  In addition, some states are providing services to
low-income working families that are not receiving cash assistance
and that, under previous law, might not have been eligible for such
assistance. 


--------------------
\86 Child care is considered an essential support service if poor
families with children are to be able to work.  Changes in funding of
child care for low-income families are discussed in a later section. 


         TRANSPORTATION
------------------------------------------------------ Chapter 3:2.4.1

Helping welfare applicants and recipients obtain reliable
transportation to work or work-related activities continues to be a
challenge in many areas.  Many poor families do not own cars, and
even those with cars sometimes do not have reliable sources of
transportation.  Under AFDC eligibility rules, recipients were
limited to owning vehicles valued at no more than $1,500.\87 In
addition, public transportation is sometimes limited or unavailable,
especially for those in certain rural areas and in some urban areas
where many of the available jobs are located in outlying suburban
areas.  In a study of welfare recipients in Connecticut, focus group
participants throughout the state pointed to limited bus routes and
hours of availability, as well as to the expense of transportation,
as obstacles to obtaining employment.  Officials in a rural county in
Maryland told us that many of the better-paying jobs available are
for night-shift work, when public transportation is unavailable.  In
urban areas such as Baltimore and Milwaukee, most of the available
jobs are in the surrounding suburbs. 

As states seek to expand the percentages of their caseloads
participating in work activities, they are using a variety of
approaches to enhance transportation services for recipients,
including providing funding for rural transportation systems,
enlisting volunteers to provide transportation for recipients,
expanding bus routes, and providing funds for vehicle repairs. 
Oregon used some of its savings from caseload reductions to resume
operation of a rural transportation system and reimburse volunteer
drivers for transporting welfare recipients.  Hartford, Connecticut,
expanded bus routes on the basis of gaps in the public transportation
system identified by a study that compared the geographic locations
of welfare recipients with the job sites they were attempting to
reach.  Baltimore, Maryland, used funds from the Bridges-to-Work
program, which is administered by the U.S.  Department of Housing and
Urban Development (HUD), to provide transportation services for
recipients and thereby help them gain access to jobs in areas outside
the city.  A survey of states conducted by APWA found that some
states provided one-time, lump-sum funds for specific needs, such as
car repair, to applicants as a diversion strategy.  \88


--------------------
\87 As indicated in ch.  2, most states increased the value of the
vehicle allowance under their welfare waivers to address this
problem. 

\88 APWA, Survey Notes:  TANF Eligibility, Benefits, Work, Sanctions,
and Exemptions, Vol.  1, No.  2 (1997). 


         MENTAL AND PHYSICAL
         HEALTH SERVICES
------------------------------------------------------ Chapter 3:2.4.2

As states require larger percentages of their welfare caseloads to
participate in work-related activities--including some recipients who
were previously exempted because of a determination of physical or
mental disability--and as more recipients leave welfare for
employment, states are concerned that they will be left with a more
difficult-to-serve population.  Finding ways to involve these
recipients in work activities was one of the most challenging and
widespread implementation issues cited in the states we visited. 
Studies of these recipients have found that, in addition to being
less likely to have prior work experience and more likely to have
lower literacy levels, they tend to have multiple problems that make
participation in work-related activities more difficult.  These
problems include physical and mental health issues such as
depression, anxiety, personality disorders, substance abuse, and
domestic violence.  To move these recipients toward self-sufficiency,
states have sought to enhance their capacity to provide mental and
physical health services. 

Oregon officials estimate that about 50 percent of the state's
welfare caseload requires drug or alcohol treatment services.  Oregon
introduced mental health and drug and alcohol services by integrating
them into its life-skills training module for welfare recipients and
by placing counselors on-site at welfare offices.  The city of
Baltimore, Maryland, added a social service component to its welfare
program, relocating social service workers to welfare offices to deal
primarily with hard-to-serve recipients and refer them to appropriate
services to help remove their barriers to employment.  For example,
social service workers have arranged services for victims of domestic
violence. 

Under California's previous JOBS program, recipients who required
mental health or drug and alcohol services were deferred from
participation.  However, California's welfare reform program
eliminated these grounds for deferral.  The California Senate Welfare
Reform Conference Committee recently estimated that 15 percent of
welfare cases in the state require mental health treatment, 25
percent need treatment for substance abuse, and 20 percent need help
to deal with domestic violence.  California's welfare reform
legislation provided for the creation of separate funding streams and
programs for substance abuse treatment services and mental health
services for welfare recipients. 


         TRANSITIONAL AND
         POSTEMPLOYMENT SERVICES
------------------------------------------------------ Chapter 3:2.4.3

In recognition of the importance of child care subsidies and medical
assistance to low-income families, under prior law families leaving
welfare because of employment were eligible for a year of
transitional child care and medical assistance.  Under TANF, 29
states plan to extend transitional child care beyond 12 months, and
12 states plan to do the same with transitional Medicaid, according
to a study of states' TANF plans by the National Governors'
Association.  Some states do not set a specific limit on the length
of time a child care subsidy can be received but instead base
eligibility primarily on income criteria.\89

Previous experience with welfare-to-work programs has shown that
placing recipients in jobs is sometimes not as challenging as helping
them remain employed so that they do not cycle back onto welfare. 
Leaving welfare is often more of a process than a one-time event,
especially for families with multiple barriers to employment. 
Recognizing these challenges, some states have sought to maintain a
relationship with former welfare recipients through continued case
management in order to provide job retention, reemployment, and
advancement assistance services, which are referred to as
postemployment services. 

Thirty-five states are offering some form of case management services
to individuals who have left TANF because of employment, according to
the APWA survey of states, including four of our seven case study
states.  In many cases, the services are provided for up to 90 days
after the family exits TANF.  For example, in Oregon, case managers
help to identify the supports needed to increase workers'
employability, such as skills development and training.  Connecticut
developed a mentoring program in which experienced employees are
matched with newly hired welfare clients to help them make the
transition to the workplace environment.  The state managed to
overcome some early difficulties in recruiting mentors, according to
Connecticut officials.  Under California's welfare reform
legislation, counties have the option of continuing to provide case
management and supportive services for up to 12 months for former
participants who become employed. 


--------------------
\89 In addition to income eligibility criteria, however, many other
factors may determine whether child care subsidies are available to
low-income working families.  In our report, GAO/HEHS-98-27, Jan. 
13, 1998, we discussed several of these factors, including copayments
that families are required to make and states' having insufficient
resources to fund subsidies for all eligible families who apply. 


      INCREASED SPENDING FOR CHILD
      CARE SUPPORTS TANF GOALS
-------------------------------------------------------- Chapter 3:2.5

While states can use TANF resources for child care, one of the
largest sources of federal funds for child care subsidies for
low-income families is the Child Care and Development Fund.  The
welfare reform law combined four federal child care programs with
different target populations into one program with a single set of
eligibility criteria and requirements, now called the Child Care and
Development Fund (CCDF).\90 The new CCDF provides federal funds to
states for child care subsidies for families who are working or
preparing for work and who have incomes of up to 85 percent of a
state's median income.\91 The CCDF provided states with about $3
billion in federal funds in fiscal year 1997.\92

In a separate study of child care under the CCDF, we reported that
all of the seven states reviewed used federal and state funding under
the CCDF to increase overall expenditures in their fiscal year 1997
child care subsidy programs, with increases over fiscal year 1996
expenditures ranging from about 2 percent in Maryland to 62 percent
in Louisiana.\93 Six of the seven states also reported an increase in
the number of children served under their child care subsidy
programs.  All seven states expected to meet the fiscal year 1997
child care needs of families required to work under welfare reform as
well as those of families moving off welfare.  However, the states
did note difficulties with locating a sufficient supply of certain
types of care--infant care, nonstandard hour care, care for sick
children, and care for children with special needs.  States were also
uncertain that supply would be adequate in the longer term and had
varied initiatives under way to expand supply. 

Some states are taking advantage of the new consolidated program to
develop a more integrated child care system to support the work
efforts of welfare and nonwelfare families.  For example, Wisconsin
and Oregon's child care programs, which are primarily based on income
eligibility, are integrated programs that enable all potentially
eligible families to access program services under the same
procedures, criteria, and requirements.  Such programs are important
to ensure that families who have never been on welfare are treated
the same with regard to child care subsidies as those who have been
and that families can move from welfare to employment without fear of
losing eligibility for subsidized child care.\94


--------------------
\90 Three of the four child care programs--(1) AFDC/JOBS child care,
which provided child care assistance to welfare families involved in
work or approved education or job training activities; (2)
Transitional Child Care, which provided 1 year of child care
assistance to families leaving AFDC because of employment; and (3)
At-Risk Child Care, which assisted low-income working families who
were deemed to be at risk of becoming dependent on welfare without
child care assistance--were repealed.  The new law modified the
fourth existing child care program, the Child Care and Development
Block Grant, which previously had assisted families with incomes at
or below 75 percent of the state median income who were working or in
approved education and training. 

\91 Nationwide, for fiscal year 1997, 85 percent of state median
income for a family of four ranged from a low of $31,033 in Arkansas
to a high of $52,791 in Connecticut. 

\92 To receive its full federal allocation, a state has to maintain
its expenditure of state funds for child care programs at specified
previous levels and spend additional state funds above those levels. 
State spending on child care for low-income families may count toward
a state's TANF maintenance-of-effort requirement and, at the same
time, help it meet state spending requirements to get its full CCDF
allocation. 

\93 GAO/HEHS-98-27, Jan.  13, 1998. 

\94 Child Care:  Working Poor and Welfare Recipients Face Service
Gaps (GAO/HEHS-94-87, May 13, 1994). 


   STATES AND LOCALITIES STRIVE TO
   OBTAIN GREATER EMPLOYER AND
   COMMUNITY INVOLVEMENT IN
   HELPING RECIPIENTS OBTAIN WORK
---------------------------------------------------------- Chapter 3:3

Under the JOBS program, it was common practice for states and
localities to contract with other organizations to provide education,
training, and support services to welfare recipients.  As states and
localities now strive to move more recipients into work activities
and reduce welfare dependence, they are placing greater emphasis on
expanding their partnerships, especially with employers.  For
example, California requires that counties, in producing their local
welfare plans, describe their partnerships with the private
sector--including employers and employer associations--and indicate
how these partnerships will identify jobs for recipients.  Yet even
in the strong national economy that currently exists, persuading
employers to hire welfare recipients can sometimes be challenging. 
Moreover, some areas have weaker economies than others.  The states
and localities we visited are using a range of approaches to expand
their partnerships with employers and other organizations. 


      DIVERSE APPROACHES ARE BEING
      USED TO EXPAND INVOLVEMENT
      OF EMPLOYERS AND EMPLOYER
      ASSOCIATIONS IN WELFARE
      PROGRAMS
-------------------------------------------------------- Chapter 3:3.1

Approaches that states and localities are using to strengthen
linkages with employers include enlisting top state officials to
solicit employers, giving employers a central role in designing local
welfare-to-work programs, assessing the effectiveness of different
strategies for obtaining employer involvement, and offering
incentives to businesses to promote economic development. 

In some instances, the impetus for partnering with employers has come
from top state government officials.  For example, Connecticut kicked
off its welfare reform program with a large-scale direct-mail
campaign to more than 35,000 businesses in the state that included a
letter from the governor encouraging them to send in job orders for
welfare recipients.  In addition, the state commissioners of
insurance and banking initiated partnerships of insurance companies,
banks, the state Department of Social Services, and other
organizations to provide welfare recipients classroom training,
internships, and job opportunities in the banking and insurance
industries. 

Some states, such as Wisconsin and Texas, have sought to
institutionalize the involvement of employers by giving them a
central role in designing local welfare-to-work programs.  For
example, local welfare agencies in Wisconsin are required to
establish Community Steering Committees, which are responsible for
developing strong ties to local employers, creating and identifying
job opportunities, and performing other roles that will benefit
program participants.  These committees must include representatives
of local business interests.  At the time of our site visits, welfare
agencies in Milwaukee were just beginning to form their committees,
but Fond du Lac County had already established its committee two
years earlier under previous welfare reforms.  The director of the
welfare agency in Fond du Lac County told us that, while the
committee encountered some early difficulties in determining its role
and developing an action plan, the committee has been extremely
beneficial for welfare recipients in developing employment
opportunities because its members have much greater influence with
area employers than welfare officials ever could. 

Another approach has been to assess the effectiveness of different
strategies for obtaining employer involvement.  In Connecticut, the
Department of Social Services hired the Connecticut Business and
Industry Association to determine what incentives and services would
encourage small businesses to hire welfare recipients.  The
Association conducted pilot projects in three regions of the state
and tested employer incentives, such as training subsidies and tax
credits, and services, such as prescreening recipients' education and
skills; peer mentoring; and hiring through the use of temporary
agencies, chambers of commerce, and business consortiums.\95

The lack of job growth in some locations has stimulated efforts to
promote economic development.  For example, the welfare reform
program in the city of Baltimore faces a situation of low job growth. 
Maryland's Department of Human Resources, in conjunction with a
university research partner, has projected the number of jobs that
will be created in "target industries"--those in which recipients of
cash assistance are most likely to obtain employment--in each
jurisdiction statewide.  These projections were then compared with
projections of the number of adults that would have to engage in work
activities in order for the state to achieve mandated federal
participation rates.  This analysis showed strikingly different labor
market conditions in the two localities we visited.  Washington
County is projected to have three times as many new jobs in target
industries as work activity positions needed.  In contrast, Baltimore
is projected to have only about 4 percent of the number of newly
created jobs in target industries that are needed for work activity
positions (see table 3.2).  There have been various efforts to
promote economic development in Baltimore.  For example, the federal
government has provided funds to designate sections of the city as
"empowerment zones" and thereby attempt to attract and retain
businesses by providing tax incentives and regulatory relief.  As
part of his effort to generate new jobs in the city, the Mayor of
Baltimore lobbied successfully for the passage of legislation that
would provide businesses incentives to clean up and redevelop
commercial and industrial sites impaired through past activities. 



                         Table 3.2
          
           Projections of Job Growth and Required
          Levels of Caseload Participation in Work
           Activities in Maryland for Fiscal Year
                            1998

                                          Number of adults
                                               required to
                                      participate in work-
               Projected growth in   related activities to
Jurisdic      employment in target    meet 30-percent TANF
tion                  industries\a    participation rate\b
--------  ------------------------  ----------------------
Maryland                     5,932                   8,567
City of                        175                   4,129
 Baltimo
 re
Washingt                       408                     136
 on
 County
----------------------------------------------------------
Note:  These figures were forecast as of November 1996. 

\a "Target" industries are defined as those in which recipients of
cash assistance are most likely to obtain employment, namely,
industries that provide low-wage jobs, are cyclical, and in which
more than half the employees are women. 

\b These estimates take account of credits for caseload reductions. 

Source:  Regional Economic Studies Institute, Towson University,
Maryland. 


--------------------
\95 The Association's study found that employers were not very
interested in financial incentives to hire welfare recipients but
were instead looking for motivated, prescreened individuals who had
the skills they required.  The study also concluded that working with
business associations and chambers of commerce was effective because
they were able to reach large numbers of their member employers
easily and knew where the entry-level positions were and what skills
were required of entry-level employees. 


      STATES AND LOCALITIES ARE
      FINDING ROLES FOR OTHER
      ORGANIZATIONS TO PLAY IN
      WELFARE REFORM PROGRAMS
-------------------------------------------------------- Chapter 3:3.2

While partnering with employers has been a major focus, states and
localities have also sought to develop and expand partnerships with
other organizations to help address the needs of welfare recipients. 
For example, in August 1997, Maryland funded seven proposals for
demonstration projects to be administered by private, nonprofit
organizations.\96 These projects seek to help welfare recipients
obtain and retain employment but differ considerably in planned
approaches and populations targeted.  One of the largest of these
projects, located in Prince Georges County, enlists multiple program
partners and community resources in an effort to serve teenage
mothers, victims of domestic violence, long-term recipients,
substance abusers, and other hard-to-serve cases. 

In California, Santa Clara County is working with community-based
organizations to establish services that will help welfare recipients
retain jobs.  For example, the county has established a 24-hour hot
line that recipients can call for support when they encounter
job-related problems, such as obtaining child care.  In addition, the
Silicon Valley Council of Nonprofits received a grant to work on
strengthening the capacity of the nonprofit sector to participate in
the county's welfare reform efforts and other objectives. 

In some instances, states have encountered obstacles in their efforts
to expand community involvement.  For example, Maryland encountered
resistance from some religious organizations when it solicited their
involvement in administering noncash assistance to families whose
cash assistance had been terminated because of noncompliance with
program requirements.\97 These organizations cited various concerns,
such as that the state was attempting to pass off to religious
organizations its obligations to poor families, that administering
such assistance would constitute for some religious organizations a
conflict of interest by making them agents of the state, and that
religious organizations could be exposed to liability.  Although
state law regarding the provision of noncash assistance was
subsequently modified, religious organizations have played little
role in administering noncash assistance in Maryland, according to
members of advocacy groups in the state.\98

States and localities have also found opportunities to benefit
welfare recipients through partnerships with federal agencies.  For
example, the Weed and Seed program in Hartford, Connecticut, is a
joint effort of HUD, the U.S.  Department of Justice, and the state
Department of Social Services that places social service programs
directly in the community being served--in this case a public housing
development.  About a third of the program's caseload consists of
TANF recipients.  The program focuses on eliminating crime and drug
use in the housing development and changing the way people live.  The
program targets young people and seeks to address their barriers to
employment by providing services such as life-skills training,
substance abuse treatment, and English language instruction.  In
Baltimore, the city Department of Social Services is working with HUD
to address the issue of how welfare recipients in the city can use
their federal housing vouchers to move to a neighboring county and
pursue job opportunities.  While housing vouchers are transferable to
other locations, officials from the two agencies are working to
develop strategies to publicize and facilitate such transfers. 


--------------------
\96 The state's welfare reform law provides that 10 percent of any
savings from caseload reductions or other reductions in the total
amount of cash assistance payments is to be allocated to
demonstration projects. 

\97 The state's welfare reform law provides that in such cases, the
cash assistance that would have been paid to the family shall instead
be paid to an organization on behalf of the family for up to 3
months.  The organization is to provide transitional assistance in
one or more of the following forms:  counseling, housing, child care,
household supplies and equipment, direct assistance other than a cash
payment, or any other noncash assistance that may be necessary to
assist the family to make the transition from welfare. 

\98 For example, the law was modified to broaden the definition of
who can provide transitional assistance from nonprofit organizations
to "third-party payees," which includes nonprofit organizations,
for-profit organizations, individuals approved by the state
Department of Human Resources, and government entities.  In addition,
the law addressed the issue of liability by adding certain nonprofit
organizations serving as third-party payees to the list of state
personnel covered by the state Tort Claims Act. 


   DEVOLUTION OF RESPONSIBILITY TO
   LOCALITIES IS A KEY COMPONENT
   OF SOME STATES' WELFARE REFORMS
---------------------------------------------------------- Chapter 3:4

Just as the 1996 welfare reform law provided states greater
flexibility to design and administer their assistance programs for
needy families, some states have in turn given local administrative
entities greater flexibility to design programs tailored to the needs
of their recipients.  Devolution of responsibility is a
distinguishing feature of welfare reform in four of the states we
examined:  California, Maryland, Texas, and Wisconsin.\99 In
Wisconsin, this devolution was accompanied by a privatization
initiative that resulted in private organizations taking over
administration of the TANF program in some localities.  A central
issue these states faced in implementing these changes in program
administration was how to maintain local accountability.  They are
addressing this issue through such methods as providing financial
incentives for local administrative entities and establishing
performance measures that focus more on desired outcomes. 


--------------------
\99 California and Wisconsin have county-administered welfare
programs, whereas Maryland has a state-administered program.  In
Texas, eligibility services are administered by the state, and
employment and child care services are being transferred to local
workforce development boards. 


      STATES HAVE EXPANDED LOCAL
      FLEXIBILITY TO DESIGN
      WELFARE-TO-WORK PROGRAMS
-------------------------------------------------------- Chapter 3:4.1

The devolution of responsibility in California, Maryland, Texas, and
Wisconsin is reflected primarily in expanded local flexibility to
tailor welfare-to-work programs to local needs and available
resources.  However, eligibility standards and cash assistance levels
are still set at the state level in each of these states.\100


--------------------
\100 Certain states have transferred some responsibility to
localities in these policy-making areas.  For example, Colorado's
welfare reform program has statewide eligibility standards and a
statewide minimum benefit schedule, but counties can determine
whether to pay benefits higher than the statewide minimum or offer
additional work incentives to employed recipients. 


         CALIFORNIA
------------------------------------------------------ Chapter 3:4.1.1

Counties in California have broad discretion under the state's
welfare reform law in designing welfare-to-work programs.  For
example, while counties are required to provide an adequate range of
activities to ensure that participants have access to needed
activities and services to assist them in seeking unsubsidized
employment, counties can determine the activities they will
provide.\101 Moreover, counties can establish their own policies for
how they will work with families transitioning off assistance because
of employment or time limits.  Counties receive block grants from the
state to administer the program and can carry forward any unspent
funds for up to 2 years, whereas unspent funds previously were
reallocated among counties.  Despite such expansions in flexibility,
state officials told us that local discretion remains somewhat
constrained because the state welfare reform law did not reduce local
administrative burdens relating to eligibility determination and
verification requirements. 


--------------------
\101 However, state law prohibits counties from requiring job search
and work experience of participants to the exclusion of offering a
range of other activities. 


         MARYLAND
------------------------------------------------------ Chapter 3:4.1.2

One of the most important elements of Maryland's strategy to move
families from dependence to independence is a focus on local
solutions to local problems, according to state officials.  Maryland
designed its welfare reform program to provide local departments of
social services with considerable flexibility in designing their work
programs.\102 Local departments can determine what assessment tools
to use, require recipients to participate in work activities more
than the minimum required number of hours, establish their own
policies regarding immediate job search requirements for applicants,
and establish their own preferred strategies for moving recipients to
employment within the "work first" model.  The state gives local
departments funding allocations for child care, employment and
training services, Welfare Avoidance Grants, Emergency Assistance,
and program administration but not for cash assistance grants, which
are handled by the state.  Local departments have the flexibility to
use some or all of their Emergency Assistance funds for welfare
diversion programs. 


--------------------
\102 These local departments are units of the Maryland Department of
Human Resources. 


         TEXAS
------------------------------------------------------ Chapter 3:4.1.3

As part of its effort to strengthen the linkage between employers and
welfare recipients and achieve greater responsiveness to local needs
and resources, Texas transferred responsibility for its
welfare-to-work program to 28 local workforce development boards.\103
The broad mission of these boards is to design strategies to build an
efficient and effective local workforce development system.  With
regard to the state's welfare reform program, the boards are
responsible for designing the welfare-to-work program, contracting
for program services, and overseeing the provision of these services. 
Workers from the state Department of Human Services continue to
determine eligibility for cash assistance.  The state's philosophy
acknowledges that the needs of business must drive the design of a
workforce development system.  Local workforce development boards
must have a majority of representation from the private sector,
including business owners, chief executives or chief operating
officers, or other private sector executives.  In addition, the
presiding officers of the boards are to be selected from members
representing the private sector. 

Texas also had originally planned to solicit public and private
sector bids to design and implement its Texas Integrated Eligibility
Services (TIES) project.  Under TIES, Texas wanted to consolidate
eligibility determination for all health and human services and
workforce programs into one overall system that contractors could
manage.  When Texas officials queried federal officials about the
possibility of using private contractors to interview and determine
applicant eligibility for TANF, Medicaid, and Food Stamp program
benefits, however, they received letters from HHS that questioned the
advisability of proceeding with the state's plans.  One HHS letter
stated that Medicaid's authorizing legislation and the Food Stamp Act
preclude private contractors from evaluating applicant information
and certifying eligibility.\104 Texas officials told us that because
of HHS' interpretation of the law and other concerns that the state
legislature had with the TIES project, the legislature subsequently
limited the bid solicitation to developing new social service
eligibility determination processes and the information management
systems to support them. 


--------------------
\103 The Texas Workforce Commission, which is the state administrator
of workforce development funds, reviews the strategic and operational
plans submitted by local workforce development boards. 

\104 However, the 1996 welfare reform law specifically allows states
to contract with private firms for conducting TANF activities,
including determining applicant eligibility. 


         WISCONSIN
------------------------------------------------------ Chapter 3:4.1.4

In contrast to Texas, Wisconsin did privatize eligibility
determination for TANF in some parts of the state.  As part of
Wisconsin's plan to establish a competitive environment for the
administration of its welfare reform program, county and tribal
departments of social or human services were required to meet certain
performance standards--including reducing caseload size by 25 percent
in a year--to ensure that they would be selected to operate the
program.  The state issued a request for proposals for the
competitive selection of program operators in counties that did not
meet the standards or declined to run the program.  As a result,
private nonprofit or for-profit organizations were selected to
administer the TANF program in nine counties, including Milwaukee
County.  Milwaukee County was divided into six regions, because of
the state's view that smaller entities have historically performed
better under the state's welfare waiver programs.\105

Local welfare agencies receive fixed funding allocations from the
state and are responsible for determining program eligibility and
administering work and support services.  However, eligibility for
Medicaid and food stamps continues to be determined by county
government officials.\106

Wisconsin state officials told us that their strategy is to inform
local welfare agencies of key program objectives but tell them very
little about how to accomplish them.  State officials envision that
fully institutionalizing this state and local relationship may take a
couple of years.  While the welfare agencies in Milwaukee were in the
early stages of developing their welfare-to-work strategies at the
time of our site visit, they had already begun to reflect some
differences in emphasis.  For example, one agency was emphasizing
financial planning and helping participants become homeowners,
another stressed entrepreneurial opportunities, and a third was
focusing on providing community service jobs using its own
organizational resources. 


--------------------
\105 Five private nonprofit or for-profit organizations administer
the program in these six regions; one of these organizations
administers the program in two of the regions. 

\106 As a result, public assistance applicants in the localities
administered by private contractors are interviewed twice:  once by
the private contractor for TANF and once by a county or local
government employee for Medicaid and food stamps.  Wisconsin
officials believe that dividing program responsibility between
private contractors and public employees dilutes the state's ability
to monitor recipients' progress and compliance as well as its ability
to realize administrative efficiencies.  Wisconsin, like Texas, had
been unsuccessful in its appeal to federal officials for policy
changes that would allow states to unify and contract out their
eligibility determination processes for the Medicaid and Food Stamp
programs as well as for TANF. 


      STATES ARE USING FINANCIAL
      INCENTIVES AND ENHANCED
      FOCUS ON OUTCOME MEASURES TO
      PROMOTE LOCAL ACCOUNTABILITY
-------------------------------------------------------- Chapter 3:4.2

As states shift responsibilities to localities and privatize the
administration of welfare programs, they confront the challenge of
how to maintain local accountability.  This challenge is heightened
because states now operate their programs with fixed federal block
grants and are subject to various federal financial penalties and
rewards based on their performance.  Similar to the federal effort to
promote state accountability under TANF, California, Maryland, Texas,
and Wisconsin are working to promote local accountability in ways
such as creating financial incentives for local administrative
entities and establishing performance measures that focus more on
desired outcomes.\107

States have established financial rewards and penalties to encourage
local administrative entities to achieve desired program objectives. 
For example, Maryland's state law provides that 45 percent of any
savings from caseload reductions or other reductions in the total
amount of cash assistance payments is to be allocated to local
departments on the basis of their achieved levels of cost
savings.\108 California's law stipulates that each county will
receive 75 percent of the state share of its savings resulting from
the following outcomes:  (1) recipients exiting the welfare program
because of employment that lasted at least 6 months, (2) increased
earnings by recipients resulting from employment, and (3) diversion
of applicants from the welfare program for 6 months in addition to
the number of months equivalent to state diversion payments.\109 On
the other hand, any federal penalties for failure to meet federally
mandated participation rates are to be shared equally by the state of
California and those counties failing to meet the rates. 

In Wisconsin, local welfare agencies get to keep a certain percentage
of any "profit" they generate but also are at risk of having to put
up some of their own funds if their fixed block grants from the state
are insufficient.  If their surplus is less than or equal to 7
percent of their total expense allocation from the state, they
receive the entire surplus as profit.  Any surplus in excess of 7
percent is to be distributed as follows:  10 percent to the local
agency for unrestricted use, 45 percent to the state, and 45 percent
to the local agency for reinvestment in the community for services to
low-income people.  State contracts with local welfare agencies also
have a $5,000 "failure to serve" penalty for failure to provide
specified services.\110

With respect to performance measures, Wisconsin's written contracts
with local welfare agencies specify performance requirements for
various process measures.  For example, a Financial and Employment
Planner must meet the applicant within 5 business days of the date
the application is signed.  The contracts also note that the state
will develop outcome measures for future contracts, including
measures related to the delivery of all required services within
total expense allocations, percentages of applicants and recipients
moved to unsubsidized employment, recidivism, quality of job
placements as measured by wage rates and availability of
employer-provided benefits, and administrative cost per case.  Texas
established employment entrance rates and earnings gains as the key
performance measures for local workforce development boards. 


--------------------
\107 For a fuller discussion of some of the key issues and policy
implications associated with the privatization of social services,
see Social Services Privatization:  Expansion Poses Challenges in
Ensuring Accountability for Program Results (GAO/HEHS-98-6, Oct.  20,
1997). 

\108 This provision applies to savings generated for the 2-year
period following the effective date of the law.  The law also
stipulates that local departments can carry over into the next fiscal
year any of these allocations on the basis of cost savings that
remain unexpended after the current fiscal year. 

\109 The remaining 25 percent of savings is to be allocated to
counties that have not achieved savings resulting from these outcomes
but have performed in a manner worthy of recognition. 

\110 For example, welfare agencies can be penalized for failure to
serve if they fail to respond in a timely manner to a written contact
from an applicant or participant or knowingly deny or refuse services
that they knew or should have known were required by the contract. 


WELFARE DEPENDENCE HAS DECREASED,
BUT LITTLE IS KNOWN ABOUT IMPACTS
ON FAMILIES
============================================================ Chapter 4

A large decrease in the size of the national welfare caseload
indicates a significant reduction in families' dependence on cash
assistance.  In addition, states' progress in restructuring their
welfare programs is reflected by increased levels of participation in
work activities, including unsubsidized employment, among both
families receiving and leaving welfare.  Nonetheless, little is known
about the impacts states' programs are having on the well-being of
children and families.  Some states, however, have efforts under way
to track families that have left welfare, and the welfare reform law
includes various provisions that are likely to generate national
information about program impacts in the future. 


   STATES ARE EXPERIENCING
   DECLINES IN CASELOADS AND
   INCREASES IN JOB PLACEMENT AND
   PARTICIPATION RATES
---------------------------------------------------------- Chapter 4:1

As states have reformed their welfare programs first under waivers
and then under TANF, welfare caseloads nationwide have declined to an
extent unprecedented since World War II.  In addition, data from the
states we visited show that these states generally have increased
their job placement rates and the percentages of their caseloads
participating in work and work-related activities, as compared with
their earlier JOBS programs. 


      WELFARE CASELOADS HAVE
      DECLINED SHARPLY
-------------------------------------------------------- Chapter 4:1.1

After growing in the early 1990s and peaking in 1994, the national
welfare caseload decreased by 30 percent between January 1994 and
September 1997.  Moreover, the caseload has been decreasing by a
larger percentage each year, and most of the reduction has occurred
since January 1996 (see fig.  4.1).  Between January 1996 and
September 1997, caseloads declined in all states and territories
except Hawaii and Guam, with reductions ranging from 1 percent in
Nebraska to 77 percent in Idaho.  Twenty-five states had caseload
reductions of 25 percent or more (see app.  IV). 

   Figure 4.1:  Nationwide Decline
   in Number of Families Receiving
   Cash Assistance Has Been
   Accelerating

   (See figure in printed
   edition.)

\a These data have been annualized for comparability. 

Source:  GAO analysis of data from the Administration for Children
and Families, HHS, supplemented by revised data from state officials
in California and Texas. 

While economic growth and state welfare reforms have been cited as
key factors to explain nationwide caseload declines, there is no
consensus about the extent to which each factor has contributed to
these declines.\111 Among our case study states, Oregon and
Wisconsin, which were early implementers of welfare reform, have seen
the largest caseload declines from January 1996 to September 1997--42
percent and 52 percent, respectively.  Officials in these two states
said that, along with favorable economic conditions, implementing
their welfare reform programs--which included instituting "work
first," diversions, and full-family sanctions; emphasizing full-time
jobs and exiting welfare; and expanding child care assistance--has
contributed to their large caseload declines. 

National data are not yet available on the extent to which caseload
declines have occurred as a result of increased employment and income
for families, as opposed to other factors, such as full-family
sanctions or fewer families applying for assistance.  "Work first"
policies and more stringent work requirements may lead some families
to obtain jobs and leave welfare more quickly than otherwise; these
policies might also discourage other families from applying for
assistance.  Time limits and stronger sanctions may reduce the length
of time some families receive assistance, by either motivating them
to obtain jobs and leave welfare or leading to the involuntary
termination of their cases. 

Changes in eligibility rules and earnings limits can also affect the
size of caseloads.  For example, in September 1997, Wisconsin began
implementing its Wisconsin Works program, under which clients judged
to be job-ready are not eligible to receive cash assistance.  The
state's cash assistance caseload dropped by 55 percent from September
1997 to February 1998.  State policies on earnings limits can also
affect the extent to which caseloads decrease when recipients obtain
employment.  In California, where a three-person family can earn up
to $1,551 a month and still remain eligible for cash assistance, 23
percent of families that obtained employment left welfare in state
fiscal year 1997.  In contrast, a three-person family that earned
more than $616 a month in Oregon would no longer be eligible for cash
assistance, and 70 percent of families there that obtained employment
left welfare during the same period. 

TANF caseloads can also decline because of administrative changes. 
For example, in 1997, Wisconsin removed about 5,000 child-only cases
from its TANF program and transferred them to a kinship care program
administered by local child welfare agencies under the supervision of
the state Department of Health and Family Services.\112 These were
cases in which the custodial adults were not the legally responsible
parents and cash assistance was paid only for the children.  In
contrast, Oregon has retained the child-only cases in its TANF
program, and they constituted 28 percent of the state's TANF caseload
in 1997. 


--------------------
\111 Two recent national studies of the 1993 to 1996 period point to
economic expansion as a key factor.  However, both studies also
concluded that in states such as Oregon and Wisconsin, which
implemented stringent welfare reform provisions during this period,
these provisions also contributed significantly to caseload declines. 
See Council of Economic Advisers, Explaining the Decline in Welfare
Receipt, 1993-1996 (Washington, D.C.:  Council of Economic Advisers,
May 9, 1997) and James Ziliak and others, Accounting for the Decline
in AFDC Caseloads:  Welfare Reform or Economic Growth?  (Ann Arbor,
Mich.:  University of Michigan, Sept.  1997).  A critique of the
Council of Economic Advisers study disputes its claim to have
explained the decline in welfare caseloads.  See Alberto Martini and
Michael Wiseman, Explaining the Recent Decline in Welfare Caseloads: 
Is the Council of Economic Advisers Right?  (Washington, D.C.: 
Income and Benefits Policy Center, The Urban Institute, July 1997). 
Another national study found a closer correlation between economic
conditions and the size of two-parent caseloads than the size of
single-parent caseloads.  See Rebecca M.  Blank, What Causes Public
Assistance Caseloads to Grow?  Working Paper #6343 (Cambridge, Mass.: 
National Bureau of Economic Research, Dec.  1997). 

\112 Wisconsin transferred TANF funds to the Department to fund the
kinship care program. 


      STATES GENERALLY REPORT
      SIZABLE INCREASES IN JOB
      PLACEMENT RATES
-------------------------------------------------------- Chapter 4:1.2

Five of our seven case study states reported significant increases in
the rates at which people in their programs found jobs, as compared
with their rates under their JOBS programs, according to our
estimates.  As shown in the last column of table 4.1, California,
Louisiana, and Maryland more than doubled their job placement rates
from 1995 to 1997, and Oregon and Wisconsin increased their rates by
more than 70 percent.  Texas reported a slight decrease in its job
placement rate, which a state official attributed to a delay in state
appropriations for the program that left many staff positions vacant
for much of fiscal year 1997.  Although comparable data for
Connecticut were not available, data on case closures suggest
progress with regard to job placements:  the number of families in
Connecticut that exited the welfare rolls because of increased income
rose by 32 percent from September 1996 to September 1997. 



                                        Table 4.1
                         
                          States' Estimated Job Placement Rates
                          Generally Have Increased Substantially

                                                            Job placement   Change in job
                                                                     rate       placement
                               Total number                   (placements     rate, 1995-
                       Time          of job       Average       per 1,000            1997
State\a            period\b    placements\c      caseload        cases)\d    (percentage)
-------------  ------------  --------------  ------------  --------------  --------------
California             1995          84,458       906,262              93
                       1996         104,192       889,766             117
                       1997         159,796       829,344             193            +108
Louisiana              1995           4,699        79,825              59
                       1996           5,650        70,581              80
                       1997           9,855        56,577             174            +195
Maryland               1995           3,378        77,677              43
                       1996           8,849        74,106             119
                       1997          10,092        59,230             170            +295
Oregon                 1995          18,001        39,264             458
                       1996          23,201        33,444             694
                       1997          19,987        24,307             822             +79
Texas                  1995          37,288       273,759             136
                       1996          36,972       257,069             144
                       1997          28,387       219,579             129              -5
Wisconsin              1995          19,504        70,604             276
                       1996          23,280        54,954             424
                       1997          16,289        33,859             481             +74
-----------------------------------------------------------------------------------------
Note:  Under the AFDC and JOBS programs, the federal government did
not require states to report numbers of job placements or job
placement rates, so no federally defined reporting criteria exist for
these data.  As a result, we estimated job placement rates by using
data states collected on their own in combination with data states
routinely reported to HHS on AFDC caseloads.  These estimates should
be interpreted cautiously for several reasons.  The numerator on
number of job placements is a total for the year and may include
duplicate job placements--a person could be placed in more than one
job.  The denominator, on the other hand, is an estimate of the
average monthly number of cases in the same year.  The denominator is
smaller than the actual number of people who can potentially be
placed in a job during the course of a year.  That is, the number of
people who can potentially be placed in jobs increases over time as
new cases open and others close, even if the average number of cases
remains constant.  In addition, the numerator may include job
placements for individuals who received assistance with job placement
but who never received monthly cash assistance.  For example, on the
basis of partial data, at least 9 percent of the job placements in
Maryland in fiscal year 1997 were for applicants diverted from
receipt of monthly cash assistance. 

\a Comparable data were not available for Connecticut. 

\b Data for Louisiana, Maryland, and Oregon are for the federal
fiscal year.  California data are for the state fiscal year
(July-June); Texas data are for the state fiscal year (Sept.-Aug.);
and Wisconsin data are for the calendar year. 

\c Includes job placements both for people exiting welfare and for
those who continued to receive cash assistance. 

\d We used the number of job placements per 1,000 average cases as
the basis for examining each state's performance over time in order
to adjust for changes in caseload size. 

Sources:  HHS, Characteristics and Financial Circumstances of AFDC
Recipients, FY 1995 and FY 1996, and data obtained from states. 

Officials in Oregon attributed the state's success with job
placements in large part to the maturity of its program.  In fiscal
year 1997, the number of job placements in Oregon nearly equaled the
number of cases remaining on cash assistance at the end of the year. 
State officials said that most of the cases that remained on cash
assistance either had significant barriers to employment, such as
drug, alcohol, and mental health problems, or were child-only cases
not subject to work requirements. 

Economic conditions and unemployment rates varied among states and
counties within states, and in some cases state officials reported
progress in job placements even in areas of high unemployment.  For
example, despite having the highest statewide unemployment of our
case study states--7.2 percent as of June 1997--Louisiana reported
one of the biggest increases in job placement rates since
implementing the work component of its welfare reform program in May
1997.  The state achieved an annualized job placement rate of 300 per
thousand cases from May to September 1997--more than triple its job
placement rate for all of fiscal year 1996--and 77 percent of
recipients placed in jobs over this 5-month period exited welfare. 
However, Louisiana officials said they were having difficulty with
job placements in rural areas, where jobs are particularly scarce. 
In Oregon, the rural eastern counties, where unemployment rates
ranged from 7 to 22 percent, had a higher overall job placement rate
than the rest of the state in fiscal year 1997.  State officials
cited several factors that contributed to the performance of these
counties, including that they worked closely with local employers and
assisted clients with out-of-area job searches and relocation costs,
had a strong sense of community responsibility, and had service jobs
despite the high overall unemployment rate. 


      STATES ARE GENERALLY
      INCREASING LEVELS OF
      PARTICIPATION BUT FACING
      CHALLENGES WITH TWO-PARENT
      FAMILIES
-------------------------------------------------------- Chapter 4:1.3

Available data for our case study states indicate that the states
generally have increased the percentages of their caseloads
participating in work and work-related activities, compared with
their performance under their prior JOBS programs.  In addition,
while all seven states indicated that they would meet their required
TANF all-families participation rates for fiscal year 1997, two
reported that they would not meet their TANF participation rates for
two-parent families by wide margins. 


         STATES ARE INVOLVING
         GREATER PERCENTAGES OF
         ALL-FAMILY CASELOADS IN
         WELFARE-TO-WORK
         ACTIVITIES
------------------------------------------------------ Chapter 4:1.3.1

Using TANF definitions of participation, analysis of data on levels
of participation under TANF and JOBS indicates that the case study
states generally have increased the percentages of their caseloads
participating in work and work-related activities.  To facilitate
comparison with fiscal year 1997, we developed estimates of what
states' participation rates would have been for fiscal years 1994,
1995, and 1996 using TANF criteria.  These estimates contain some
degree of imprecision because we were unable to adjust for all
relevant factors.\113 A comparison of columns 2 through 5 in table
4.2 shows that estimated all-families participation rates generally
have risen over time and, in most instances, have at least doubled
since 1994. 



                                         Table 4.2
                          
                            Estimated Achieved and Required All-
                                Families Participation Rates

                                                                  Estimated required rate,
                                                                         fiscal year
            Estimated achieved rate, fiscal year (percentage)          (percentage)\a
          ------------------------------------------------------  -------------------------
State
(1)           1994 (2)      1995 (3)      1996 (4)    1997\b (5)      1997 (6)     1998 (7)
--------  ------------  ------------  ------------  ------------  ------------  -----------
Calif.              16            19            25            23            22           17
Conn.               15            18            21            46            20           18
La.                 11            15            17            22            13            1
Md.                  9            10            13            25            16            3
Oreg.               26            39            80            79            10            0
Tex.                 8             8            11            19            18            9
Wis.                28            31            59            55             8            0
-------------------------------------------------------------------------------------------
Note:  HHS will calculate the official required and actual TANF
participation rates for states.  The figures in this table are
estimates based on data available at the time of our review.  We used
the following method to approximate what the participation rates for
fiscal years 1994, 1995, and 1996 would be using TANF criteria:  For
the numerator, we added the number of recipients working 30 hours per
week or more (because they were exempted from JOBS participation but
counted under TANF) to the number of participants in approved JOBS
activities meeting the 20-hour per week requirement.  For the
denominator, we used the number of families with at least one adult
recipient minus the number with a child under 1 year of age.  We did
not subtract from the denominator cases in the first 3 months of a
sanction, as allowed under TANF. 

\a The estimated required rates for fiscal year 1997 were calculated
by subtracting the percentage of caseload decline between fiscal year
1995 and 1996 from 25 percent--the nominal TANF participation rate
for fiscal year 1997.  The estimated required rates for fiscal year
1998 were calculated by subtracting the percentage of caseload
decline between fiscal year 1995 and 1997 from 30 percent--the
nominal TANF participation rate for fiscal year 1998.  These
estimates understate the required participation rates to the extent
that HHS (1) determines that any portion of states' caseload
reductions was due to state eligibility policy changes and (2)
reduces the caseload reduction credit by this portion, as stipulated
by the welfare reform law. 

\b These estimates were calculated by the states, using definitions
of approved work or work-related activities based either on TANF or
state waivers.  The estimates for Connecticut, Maryland, Oregon,
Texas, and Wisconsin are for the fourth quarter of fiscal year 1997
(July-Sept.  1997), the first period in which TANF participation
rates were effective.  The time periods for the other two estimates
are as follows:  California (Apr.-June 1997) and Louisiana (July-Aug. 
1997). 

Source:  Data on average monthly caseloads for fiscal years 1995 and
1996 used for calculating required rates are from HHS,
Characteristics and Financial Circumstances of AFDC Recipients, Apr. 
1997, except for Maryland, which provided its own calculation of its
required rates.  Data for fiscal year 1997 were provided by the
states. 

The welfare reform law requires states to have adult recipients in 25
percent of their TANF families participating in work activities in
fiscal year 1997, rising to 30 percent in fiscal year 1998 and to 50
percent by fiscal year 2002.\114 The required participation rate for
each state is reduced by the percentage reduction in the state's
caseload from the base year of fiscal year 1995 unless the reduction
is the result of changes in eligibility criteria.  So the effective
required rate for most states in fiscal year 1998 is less than the
nominal rate of 30 percent.  While HHS has not issued data on states'
required or achieved TANF participation rates for federal fiscal year
1997, each of our case study states expected to exceed its required
all-families participation rate for this period, according to its own
calculations (see cols.  5 and 6 of table 4.2).  Even though the
estimated achieved participation rates in California, Louisiana, and
Texas for 1997 were less than 25 percent, these states expected to
meet their required rates because of credits for caseload reductions. 
We estimate that most of the seven states will have substantially
lower required participation rates for fiscal year 1998 because of
credits for large decreases in caseloads between fiscal years 1995
and 1997 (see cols.  6 and 7 of table 4.2). 


--------------------
\113 We did not adjust the estimated participation rates for prior
years to take account of TANF restrictions on education and training
activities, such as the provision that higher education is not a
countable activity.  Nor did we adjust the estimated participation
rates for prior years to remove families sanctioned for 3 months or
less in calculating participation rates, as allowed under TANF. 
While not making the first adjustment tends to lead to some level of
overstatement of participation rates for years before 1997, not
making the second adjustment tends to lead to some level of
understatement.



\114 Families with no adult recipients are excluded in calculations
of participation rates, and states also can exclude families with a
child under 1 year of age and cases sanctioned for nonparticipation
for up to 3 months. 


         SOME STATES ENCOUNTER
         DIFFICULTIES MEETING
         PARTICIPATION RATES FOR
         TWO-PARENT FAMILIES
------------------------------------------------------ Chapter 4:1.3.2

In contrast to meeting the TANF participation rate for all families,
meeting the required rate for two-parent families has presented some
states with greater difficulty.  States are required to have adult
recipients in 75 percent of their two-parent families participating
in work activities in fiscal year 1997, rising to 90 percent in
fiscal year 1999.\115 Of the six states that provided data on their
estimated two-parent participation rates for fiscal year 1997,
Connecticut, Louisiana, Oregon, and Wisconsin reported they would
meet their estimated required rates, though Louisiana would only meet
the rate because of a sharply reduced two-parent caseload (see cols. 
5 and 6 of table 4.3).\116 However, California and Texas, which
accounted for 96 percent of the two-parent cases in our study states,
indicated that they had fallen short of meeting their required rates
by wide margins.\117 Caseworkers noted that two-parent cases often
have mental health and substance abuse problems that make
participation in job placement programs difficult. 



                                        Table 4.3
                         
                            Two-Parent Caseloads and Estimated
                             Achieved and Required Two-Parent
                                   Participation Rates

                                                          Estimated participation rate
                                                                  (percentage)
                                                       ----------------------------------
                                           Percentage
                                            change in
                             Percentage    two-parent
                Two-parent     of total     caseload,
                 caseload,    caseload,  1995 to 1997   Achieved,   Required,   Required,
State (1)       1997\a (2)     1997 (3)           (4)  1997\b (5)  1997\c (6)  1998\c (7)
------------  ------------  -----------  ------------  ----------  ----------  ----------
Calif.             143,946         17.9           -12          25          74          63
Conn.                3,239          6.0            +9          75          75          75
La.                    136          0.2           -81          14           6           0
Md.                    378          0.6           -44          \d          42          31
Oreg.                  989          4.1           -67          75          47           8
Tex.                10,558          4.9           +56          31          62          75
Wis.                 1,476          3.6           -74          40          39           1
-----------------------------------------------------------------------------------------
Note:  HHS will calculate the official required and actual TANF
participation rates for states.  The figures in this table are
estimates based on available data at the time of our review.  Data
are for the federal fiscal year. 

\a Average monthly two-parent caseload for the federal fiscal year. 

\b These estimates were calculated by the states, using definitions
of approved work or work-related activities based on either TANF or
state waivers.  The estimates for Connecticut, Oregon, Texas, and
Wisconsin are for the period July-September 1997, the first period in
which TANF participation rates were effective.  Louisiana's estimate
is for July-August 1997, and California's estimate is for April-June
1997. 

\c The estimated required participation rates were calculated by
subtracting the caseload declines between fiscal years 1995 and 1996
(for the 1997 rate) and between fiscal years 1995 and 1997 (for the
1998 rate) from 75 percent--the nominal TANF participation rate for
fiscal years 1997 and 1998.  These estimates understate the required
participation rates to the extent that HHS (1) determines that any
portion of states' caseload reductions was due to state eligibility
policy changes and (2) reduces the caseload reduction credit by this
portion, as stipulated by the welfare reform law. 

\d Maryland did not estimate its achieved two-parent participation
rate because the state was funding its two-parent cases with state
funds only and thus concluded that the participation rate
requirements did not apply.  The state's two-parent program has work
requirements that are identical to those of its TANF program,
according to state officials.  However, because many of Maryland's
two-parent families have multiple barriers to employment, the state
is providing additional services to help them overcome these
barriers. 

Source:  Data obtained from the states. 

For fiscal year 1998, three of the seven states have dramatically
lower required two-parent participation rates because of large
declines in their two-parent caseloads.  Since the two-parent
caseloads in Louisiana, Oregon, and Wisconsin have decreased sharply
since fiscal year 1995--the decreases range from 67 to 81
percent--these states' estimated required two-parent participation
rates for fiscal year 1998 range from 0 to 8 percent (see cols.  4
and 7 of table 4.3).  State officials cited various reasons for the
decline in two-parent caseloads, including an improving economy
resulting in more two-parent families finding work; stronger work
requirements; and the redefinition of two-parent families to exclude
those with a disabled parent, as allowed under the Balanced Budget
Act of 1997.  On the other hand, Connecticut and Texas, which have
expanded their eligibility rules since fiscal year 1995, have
experienced an increase in their two-parent caseloads and therefore
may not receive any reduction in the 75-percent required
participation rate for fiscal year 1998.\118


--------------------
\115 Just as in the case of the all-families rate, states' required
participation rates for two-parent families are reduced by the
percentage of any reduction in the size of their two-parent
caseloads. 

\116 Maryland did not estimate its two-parent participation rate for
the last quarter of fiscal year 1997 because the state was funding
its two-parent cases with state funds only and thus concluded that
the participation rate requirements did not apply. 

\117 In its proposed regulations, HHS has suggested that the penalty
for failure to meet the two-parent participation rate not be assessed
on the entire TANF block grant but rather on the proportion equal to
the two-parent family proportion of a state's caseload.  This could
result in substantial variations among states in the severity of any
financial penalty, since states vary greatly in the percentages of
their caseloads that are two-parent families.  For example, the
percentages in our case study states range from 0.2 percent in
Louisiana to 17.9 percent in California (see col.  3 of table 4.3). 

\118 The issue is unresolved at this time because HHS has not
published the final rules for the methodology for calculating
caseload reductions. 


         SOME STATES VOICED
         CONCERNS OVER FOCUS ON
         PARTICIPATION RATES
------------------------------------------------------ Chapter 4:1.3.3

TANF participation rates and their associated financial penalties
were designed to serve as an incentive for states to increase the
percentages of their caseloads participating in work activities. 
Nevertheless, some state officials we interviewed expressed concerns
about the law's focus on participation rates and said that
emphasizing outcome measures instead would better stimulate program
effectiveness.  These officials maintained that the law's provisions
regarding countable activities and required number of hours were too
restrictive and that tracking and reporting these elements to enable
participation rates to be calculated was burdensome and distracted
staff from helping recipients find jobs and become self-sufficient. 
The welfare reform law tasks HHS with working cooperatively with the
states to determine whether outcome measures can be developed as an
alternative to TANF participation rates to evaluate states' success
in moving individuals out of the welfare system through employment. 


   LITTLE IS KNOWN ABOUT PROGRAM
   IMPACTS ON WELL-BEING OF
   CHILDREN AND FAMILIES
---------------------------------------------------------- Chapter 4:2

The recent sharp decreases in welfare caseloads raise questions about
the broader impacts of states' welfare reform programs on children
and families.  For example, to what extent do families who have left
welfare return to welfare, how economically stable are they, and
where do they stand on various measures of well-being?  Moreover, how
do children fare after losing assistance because their parents did
not comply with new program requirements?  Little information is
available to begin answering such important questions.  However, some
states have efforts under way to collect data on such topics, and the
welfare law also includes various provisions to help ensure that
national information on program impacts is collected. 


      SOME STATES ARE TRACKING
      FAMILIES THAT HAVE LEFT
      WELFARE
-------------------------------------------------------- Chapter 4:2.1

Some of our case study states have efforts under way to collect data
on families that leave welfare.  For example, Maryland is tracking a
random sample of over 2,000 families that exited welfare during the
first 12 months of the state's welfare reform program (Oct.  1996 to
Sept.  1997).\119 Maryland has issued two reports with interim
findings.\120 Oregon has been tracking welfare recidivism to provide
data for one of the state's performance measures:  the percentage of
families that remain off welfare for 18 months after case closure
because of employment.  California law mandates an evaluation of the
direct and indirect effects of the state's welfare reform program,
including effects related to employment, earnings, and
self-sufficiency as well as those pertaining to child well-being,
such as entries into foster care, at-risk births, school achievement,
child abuse reports, and rates of child poverty. 

Data on welfare recidivism indicate that some families are returning
to welfare, but it is too early to determine whether recidivism
patterns differ from those under AFDC.\121 In tracking families that
left welfare because of job placements, Oregon found that 35 percent
of families had returned to cash assistance sometime within 18 months
but that only 10 percent were receiving cash assistance 18 months
after case closure.\122 According to Oregon officials, it sometimes
takes two or three job placements before recipients are able to hold
a job and remain off welfare.  This is reflected by findings from an
evaluation of Oregon's Post-Employment Services Demonstration, which
found that after the first year of the demonstration (1994-95), 61
percent of participants were no longer at their first job because
they had left for a better job, quit without other job prospects, or
been fired or laid off.\123

The 1998 Maryland study found that 19 percent of all families who
left welfare returned to the rolls within 3 months and 23 percent
returned within 12 months, which suggests that families who return
tend to do so fairly quickly.  The study also found that families
whose cases were recorded as closed because they started work or
because their incomes exceeded eligibility limits had significantly
lower 3-month recidivism rates:  3 percent and 11 percent,
respectively.\124

Little is known about the well-being of children and families after
they leave states' welfare reform programs.  The 1998 Maryland study
found that of the cases that closed during the first 9 months of the
program, 51 percent of adults reported earnings in the quarter after
they left welfare.  The average earnings of these adults were about
twice as high as Maryland's maximum cash assistance payment for a
family of three.  Among our case study states that tracked the
average wage for job placements in 1997, wages ranged from $5.60 to
$6.60 an hour.  A low-wage job combined with the earned income tax
credit, food stamps, and subsidized child care may lift a family
above the federal poverty line, but families may continue to be
hard-pressed, especially in areas with high housing costs and in
states that limit transitional child care and medical assistance to 1
year.  Prior research on the AFDC program has indicated that
recipients on average experienced little rise in wages over time.\125

However, Oregon officials have found some evidence in the state's
welfare reform program that those placed in jobs who have remained
employed have experienced a growth in wages.  Oregon matched job
placements with data on employer-reported wages between 1993 and 1996
and found that those who remained employed 3 years later had a wage
growth averaging 14 percent per year.\126

Early results from the 1998 Maryland study indicate that case
closures have not been associated with significant increases in the
size of foster care caseloads.  Of the 3,467 children in families
that left welfare in the first 11 months of the program, only 15
children (0.4 percent) were placed in foster care after their
families left welfare.  Approximately 3 percent of the 3,467 children
had been in foster care at some point before their families left
welfare. 


--------------------
\119 The state also plans to track a random sample of families that
leave welfare in subsequent months. 

\120 University of Maryland School of Social Work, Life After
Welfare:  Second Interim Report (Baltimore, Md.:  University of
Maryland School of Social Work, Mar.  1998) and Maryland Department
of Human Resources, Family Investment Administration, and University
of Maryland School of Social Work, Life After Welfare:  An Interim
Report (Baltimore, Md.:  University of Maryland School of Social
Work, Sept.  1997). 

\121 For example, a study by Pavetti found that 45 percent of
families returned to AFDC within 1 year of exiting and 72 percent,
within 5 years.  A study by Blank and Ruggles found lower recidivism
rates:  21 percent of families returned to AFDC within 25 months of
exiting, with the rate of return peaking at 5 months after exiting. 
See LaDonna Pavetti, "Questions and Answers on Welfare Dynamics,"
paper presented at a research meeting on welfare dynamics, The Urban
Institute, Washington, D.C., Sept.  11, 1995, and Rebecca M.  Blank
and Patricia Ruggles, "Short-Term Recidivism Among Public-Assistance
Recipients," The American Economic Review, Vol.  84, No.  2 (May
1994). 

\122 These data are as of June 30, 1997. 

\123 The goals of this federally initiated demonstration were to help
individuals who became employed keep their jobs, help those who lost
their jobs regain employment quickly, and reduce the amount of time
on AFDC.  See Toby Herr and others, Something Old, Something New:  A
Case Study of the Post-Employment Services Demonstration in Oregon
(Chicago:  Erikson Institute, Nov.  1995). 

\124 The study does not cite the 12-month recidivism rates for these
families. 

\125 For example, using data from the National Longitudinal Survey of
Youth, one study compared the change in wages among two groups of
women between 18 and 22 years of age in 1979:  one group that
collected AFDC during all or part of at least 1 year between 1979 and
1981 and one that collected no AFDC during this period.  From 1979 to
1990, the average real wage of the AFDC-dependent women, in 1991
dollars, rose from $6.18 to $6.85 an hour, compared with an increase
from $6.07 to over $10.00 an hour for the other group.  See Gary
Burtless, "Employment Prospects of Welfare Recipients," The Work
Alternative:  Welfare Reform and the Realities of the Job Market,
eds.  Demetra Smith Nightingale and Robert H.  Haveman (Washington,
D.C.:  The Urban Institute Press, 1995). 

\126 In the study, 1,915 clients reported wages in 1993; the wage
progression data were based on the 1,351 clients also found to have
reported wages 12 quarters later. 


      WELFARE LAW INCLUDES
      PROVISIONS TO ENSURE
      COLLECTION OF NATIONAL
      INFORMATION ON PROGRAM
      IMPACTS
-------------------------------------------------------- Chapter 4:2.2

The welfare law tasked HHS with collecting various information that
will help policymakers and researchers evaluate the impacts of TANF. 
For example, HHS is required to develop a formula to reward "high
performing" states--those that achieve the goals of the law.  While
the formula for fiscal years 1999 to 2003 has not been finalized, HHS
issued guidance to states in March 1998 on the performance measures
that will be used for fiscal year 1998.  In addition to job
placements, these measures include job retention and wage
progression, which would provide information on the economic
stability of families after they are placed in jobs.\127 The welfare
law also directs HHS to conduct research on the benefits, effects,
and costs of the state programs funded under TANF to determine
effects on welfare dependency, illegitimacy, and other areas.  HHS is
also required to evaluate innovative programs designed to decrease
welfare dependency and increase child well-being.  A key element in
HHS' pursuit of this research mandate is its continued funding of
evaluations of state waiver programs, which assess the impacts of
provisions such as time-limited assistance and work requirements.\128
Furthermore, beginning in 1999, HHS is required to submit annual
reports to the Congress on the circumstances of families reaching
their time limits and families headed by teen parents. 

The welfare law also appropriated $10 million per year beginning in
1996 for 7 years for the Bureau of the Census to continue to collect
data on a national sample of families surveyed in the 1992 and 1993
panels of the Survey of Income and Program Participation to obtain
information on the impacts of TANF.  The law directs Census to pay
particular attention to the issues of out-of-wedlock births, welfare
dependence, how long people remain on welfare, causes of welfare
recidivism, and status of children.  Census has subsequently
developed an expanded survey, called the Survey of Program Dynamics,
to collect information on these topics. 


--------------------
\127 Welfare Reform:  HHS' Progress in Implementing Its
Responsibilities (GAO/HEHS-98-44, Feb.  2, 1998) provides additional
information on HHS' efforts to develop regulations for the
high-performance bonus and some of the challenges HHS has faced. 

\128 See GAO/HEHS-98-44 for information about how HHS allocated its
research funding under the welfare law. 


CONCLUDING OBSERVATIONS, AGENCY
AND STATE COMMENTS, AND OUR
EVALUATION
============================================================ Chapter 5


   CONCLUDING OBSERVATIONS
---------------------------------------------------------- Chapter 5:1

States are transforming the nation's welfare system into a
work-focused, temporary assistance program for needy families.  They
are modifying their policies to require or encourage welfare
recipients and potential recipients to work; rely on alternatives to
cash assistance when appropriate; use welfare as temporary, rather
than long-term, assistance; cooperate with child support enforcement
requirements; and not become pregnant outside of marriage.  At the
same time, states are modifying their programs to better support
welfare recipients in meeting these expectations, such as by
expanding the role of welfare workers to include job counseling,
transforming local welfare offices into job centers, enhancing
support services, expanding efforts to establish partnerships with
employers and other organizations, and giving local administrative
entities more flexibility to tailor programs to local needs. 

The confluence of a strong national economy and the availability to
most states of greater budgetary resources under TANF than they would
have received under prior law has created an optimal time for states
to reform their welfare programs.  Increases in job placement rates
and levels of participation in work activities are early encouraging
signs of progress toward some of the objectives of the federal
welfare reform law.  However, any comprehensive assessment of states'
welfare reform programs must also take into account what happens over
the longer term with respect to some critical issues, such as the
following. 


      HOW DO FAMILIES FARE AFTER
      LEAVING WELFARE FOR WORK OR
      BEING DIVERTED FROM WELFARE? 
-------------------------------------------------------- Chapter 5:1.1

States are helping many families leave or avoid welfare by providing
a range of services to support their work efforts, including Medicaid
and child care assistance.  However, these families' prospects for
achieving some measure of economic stability remain an important
issue, in light of prior research showing that welfare recipients,
who often find jobs with low wages, generally have experienced little
rise in wages over time after they began working.\129

To the extent that these families' earnings do not increase over time
and their employment-based fringe benefits are limited, their ability
to maintain employment and support themselves may depend to a great
extent on the availability of income supports, such as subsidized
medical and child care and the earned income tax credit.  Federal and
state policies and programs for assisting low-income working families
are likely to play a critical role in the future success of welfare
reform. 


--------------------
\129 Gary Burtless, "Employment Prospects of Welfare Recipients,"
1995. 


      HOW EFFECTIVE ARE STATES IN
      WORKING WITH HARD-TO-SERVE
      CASES? 
-------------------------------------------------------- Chapter 5:1.2

As more recipients leave welfare for work, the characteristics of
states' caseloads can be expected to change.  Data from states that
have implemented early reforms and experienced large caseload
reductions indicate that many of the remaining recipients face
multiple barriers to participation in work activities, such as mental
health and substance abuse problems and domestic violence.  As a
result, even under continued favorable economic conditions, states'
initial successes with moving applicants and recipients into
employment will probably slow over time.  In response, states will
need to adjust their approaches to better enable families with a
range of problems to take steps toward becoming more self-supporting. 
Although states can exempt up to 20 percent of TANF families from the
federal time limit on assistance on the basis of hardship, even with
more intensive supports, there may be some recipients who cannot be
placed in the workforce because of the severity of their problems. 
In addition, there may be some adults whose lack of compliance with
program requirements, for whatever reasons, puts their children at
risk of a range of negative outcomes.  Given time limits on
assistance and pressure on states to meet rising federally mandated
work participation rates, monitoring how these recipients and their
children fare in states' welfare reform programs will be especially
important. 


      HOW WELL DO STATES MANAGE
      THE DEVOLUTION OF AUTHORITY
      TO LOCALITIES OR THE
      PRIVATIZATION OF WELFARE
      PROGRAMS? 
-------------------------------------------------------- Chapter 5:1.3

As we have seen, some states have given localities greater
flexibility to design programs tailored to the needs of their
recipients or privatized some functions of their welfare programs,
such as eligibility determination.  By encouraging experimentation,
or in some cases, competition, these initiatives may hold some
promise of developing more effective and efficient ways of serving
needy families.  However, these initiatives also heighten the
challenges states face in ensuring local accountability and avoiding
unintended negative results.  Therefore, it is essential to pay
attention to how well states are performing such tasks as specifying
clear program goals and developing effective systems to monitor local
performance. 


      HOW WOULD AN ECONOMIC
      DOWNTURN AFFECT STATES'
      WELFARE REFORM PROGRAMS? 
-------------------------------------------------------- Chapter 5:1.4

A comprehensive perspective on states' TANF programs cannot be
obtained until it is known how they perform in both a strong economy
and a weak one.  Yet little is known about how a weak economy would
affect states' programs.  For example, some adults who had previously
left welfare for work could become unemployed.  While they could be
eligible for unemployment insurance, some could once again apply for
cash assistance after their unemployment insurance ran out. 
Furthermore, if caseloads did increase significantly in a worsening
economy, it is unclear what budgetary responses states would take in
an environment of fixed federal TANF funding.  They would have a
range of options, such as using any reserve funds that they might
have set aside; appropriating additional state funds; accessing TANF
loan or contingency funds; or reducing program funding by cutting
cash assistance levels, restricting eligibility, or reducing
expenditures on services.  Since these options are subject to varying
constraints and could have very different effects on poor families,
it will be important to pay close attention to how states respond in
the event of an economic downturn. 


   COMMENTS FROM HHS AND THE
   STATES AND OUR EVALUATION
---------------------------------------------------------- Chapter 5:2

We obtained comments on a draft of this report from HHS and the seven
case study states.  HHS stated that the report provides a useful and
comprehensive description of how states are implementing various
reforms under TANF and provided two general comments.  First, HHS
noted that there were some inconsistencies between the information
that it had regarding policies and activities in the seven states and
the information included in the draft report but acknowledged that
the states were reviewing the draft report.  In their comments, the
states generally agreed with the report's findings and provided
additional technical information about their welfare reform programs,
which we incorporated in the report as appropriate. 

Second, HHS suggested that the report should indicate that, while
states were permitted to make their own reasonable interpretations of
the 1996 welfare reform law prior to the publication of final rules
for the TANF program, some of the state policy interpretations
reflected in the report may not be sustained when HHS issues final
rules.  In response, we added this information to the report.  HHS
also provided additional technical information, which we incorporated
in the report as appropriate. 


GAO'S WELFARE REFORM ADVISORY
COMMITTEE MEMBERS
=========================================================== Appendix I

Harold Beebout, Mathematica Policy Research, Inc. 

Douglas Besharov, American Enterprise Institute

Barbara Blum, National Center for Children in Poverty

Sandra Danziger, University of Michigan

Judith Gueron, Manpower Demonstration Research Corporation

Anna Kondratas, The Urban Institute

Lawrence Mead, New York University

Gerald Miller, Lockheed Martin IMS

Richard Nathan, Rockefeller Institute of Government, State University
of New York

Barry Van Lare, Welfare Information Network

Michael Wiseman, University of Wisconsin


SUMMARY OF STATES' PREREFORM
WAIVER PROVISIONS
========================================================== Appendix II

                                                                                                        Established full-
                                                                                          Liberalized   family sanction                  Imposed teen
                          Established     Increased asset                 Imposed time    100-hour or   for                Allowed or    living and/
          Lowered age-    full-family     limits over       Changed       limits on the   labor force   noncooperation     required      or teen
          of-youngest-    sanction for    $1,000 and/or     earned        receipt of      attachment    with child         noncustodial  school
          child           noncooperation  vehicle           income        benefits for    rules for     support            parents to    attendance
          exemption to    with work       allowances over   disregard     entire          two-parent    enforcement        participate   requirements
State     under 1 year\a  requirements\b  $1,500\a          policies\a    family\b        families\a    requirements\b     in JOBS\a     \b
--------  --------------  --------------  ----------------  ------------  --------------  ------------  -----------------  ------------  ------------
AL        X                               X                                               X                                X

AK

AZ                                                                                        X                                              X

AR

CA                                        X                 X                             X                                X

CO                                        X                 X

CT                        X               X                 X             X               X             X                  X             X

DE        X               X               X                 X             X               X             X                  X             X

DC                                                                                                                                       X

FL        X                               X                 X             X               X                                X

GA                                        X                                                                                X

HI        X                               X                 X             X               X

ID        X               X                                                                                                              X

IL                        X               X                 X             X               X                                X

IN        X               X               X                 X                             X                                              X

IA        X               X               X                 X                             X                                              X

KS                        X               X                 X                             X             X

KY

LA                        X                                               X

ME                                        X                                                                                              X

MD        X               X               X                 X                             X             X                  X             X

MA                        X               X                 X                             X                                X             X

MI        X               X               X                 X                             X                                X             X

MN                        X               X                 X                             X                                X

MS                        X                                 X                             X                                X

MO                                        X                 X                             X                                X             X

MT                                        X                 X                             X

NE        X               X               X                 X             X               X

NV

NH                        X               X                 X                                                              X

NJ                                                                                                                         X

NM

NY                                                                                                                         X

NC                        \c              X                 X             X               X

ND                        X               X                 X                             X             X

OH                        X               X                 X             X               X             X                  X

OK                                        X                 X                             X

OR        X               X               X                 X             X               X                                X             X

PA                                        X                 X                             X

RI

SC                        X               X                               X               X                                X

SD                        X               X

TN        X               X               X                 X             X               X             X                  X

TX                                        X                                               X

UT                        X               X                 X                             X             X                  X

VT        X               X               X                 X                             X                                              X

VA                        X               X                 X             X                             X                                X

WA                                                          X                             X

WV                        X

WI                        X               X                 X             X               X                                X

WY                                        X                                                                                X

=====================================================================================================================================================
Total     13              26              36                31            14              32            9                  23            15
-----------------------------------------------------------------------------------------------------------------------------------------------------
\a Data are based on waivers approved before enactment of the new
federal welfare reform law on Aug.  22, 1996 (HHS, Setting the
Baseline, June 1997). 

\b Data are based on waivers approved before the enactment of the new
federal welfare reform law on Aug.  22, 1996 (GAO/HEHS-97-74, May 15,
1997, apps.  III and IV). 

\c North Carolina's waiver provided for terminating benefits to the
entire family for failure to enroll in the welfare-to-work program. 


PENALTIES SPECIFIED IN THE FEDERAL
WELFARE REFORM LAW OF 1996, AS
AMENDED
========================================================= Appendix III

      Grounds for penalty               Amount of penalty
----  --------------------------------  ------------------
(1)   Use of grant in violation of the  Amount so used
      law                               (5 percent if
                                        intentional
                                        violation)

(2)   Failure to submit the data        4 percent
      report required quarterly

(3)   Failure to satisfy minimum        5 to 21 percent
      participation rates

(4)   Failure to participate in the     Not more than 2
      income and eligibility            percent
      verification system

(5)   Failure to comply with paternity  Not more than 5
      establishment and child support   percent
      enforcement requirements under
      the law

(6)   Failure to timely repay a         The outstanding
      federal loan fund for state       loan amount, plus
      welfare programs                  interest

(7)   Failure to maintain certain       Amount of
      level of historic effort          shortfall
      (commonly referred to as the
      "maintenance-of-effort"
      requirement)

(8)   Substantial noncompliance of      1 to 5 percent
      state child support enforcement
      program with requirements of the
      law

(9)   Failure to comply with 5-year     5 percent
      limit on assistance

(10)  Failure of state receiving        Total amount paid
      amounts from contingency fund to  from the
      maintain 100 percent of historic  contingency fund
      effort                            and not yet
                                        remitted

(11)  Failure to maintain assistance    Not more than 5
      to adult single custodial parent  percent
      who cannot obtain child care for
      child under age 6

(12)  Failure to expend additional      Not more than 2
      state funds to replace grant      percent plus the
      reductions                        amount the state
                                        failed to expend

(13)  Failure to maintain historic      Amount of the
      effort during a year in which a   Welfare-to-Work
      Welfare-to-Work Grant is          Grant
      received

(14)  Failure to reduce assistance for  1 to 5 percent
      recipients refusing without good
      cause to work
----------------------------------------------------------
Source:  42 U.S.C.  secs.  603 and 609. 


CHANGES IN THE NUMBER OF FAMILIES
RECEIVING CASH ASSISTANCE, BY
STATE, JANUARY 1996 TO SEPTEMBER
1997
========================================================== Appendix IV

                                   September    Percentage
State             January 1996          1997        change
----------------  ------------  ------------  ------------
Alabama                 43,396        21,171           -51
Alaska                  11,979        10,957            -9
Arizona                 64,442        50,025           -22
Arkansas                23,140        16,719           -28
California             904,940       756,950           -16
Colorado                35,661        24,579           -31
Connecticut             58,124        55,159            -5
Delaware                10,266         9,236           -10
District of             25,717        22,410           -13
 Columbia
Florida                215,512       140,937           -35
Georgia                135,274        90,746           -33
Hawaii                  22,075        23,532            +1
Idaho                    9,211         2,083           -77
Illinois               225,796       189,196           -16
Indiana                 52,254        41,201           -21
Iowa                    33,559        27,317           -19
Kansas                  25,811        16,949           -34
Kentucky                72,131        59,505           -18
Louisiana               72,104        49,951           -31
Maine                   20,472        16,636           -19
Maryland                75,573        53,359           -29
Massachusetts           90,107        72,430           -20
Michigan               180,790       140,854           -22
Minnesota               58,510        48,793           -17
Mississippi             49,185        31,061           -37
Missouri                85,534        65,527           -22
Montana                 11,276         7,835           -31
Nebraska                14,136        13,979            -1
Nevada                  15,824        11,202           -29
New Hampshire            9,648         6,679           -31
New Jersey             113,399        93,800           -17
New Mexico              34,368        17,800           -48
New York               437,694       359,707           -18
North Carolina         114,449        88,969           -22
North Dakota             4,976         3,686           -26
Ohio                   209,830       160,162           -24
Oklahoma                40,692        26,894           -34
Oregon                  35,421        20,608           -42
Pennsylvania           192,952       147,211           -24
Rhode Island            21,775        19,557           -10
South Carolina          46,772        28,401           -39
South Dakota             6,189         4,377           -29
Tennessee              100,884        60,385           -40
Texas                  231,154       166,919           -28
Utah                    15,072        11,264           -25
Vermont                  9,210         7,845           -15
Virginia                66,244        47,529           -28
Washington              99,395        86,792           -13
West Virginia           36,674        28,148           -23
Wisconsin               65,386        31,336           -52
Wyoming                  4,975         1,564           -69
Guam                     2,097         2,603           +24
Puerto Rico             51,370        45,932           -11
Virgin Islands           1,437         1,222           -15
==========================================================
United States        4,593,862     3,539,689           -23
----------------------------------------------------------
Source:  GAO analysis of data from the Administration for Children
and Families, HHS, supplemented by revised data for California and
Texas from state officials. 




(See figure in printed edition.)Appendix V
COMMENTS FROM THE DEPARTMENT OF
HEALTH AND HUMAN SERVICES
========================================================== Appendix IV



(See figure in printed edition.)


GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
========================================================== Appendix VI

GAO CONTACTS

Gale C.  Harris, Assistant Director, (202) 512-7235
Andrew Sherrill, Evaluator-in-Charge, (202) 512-7252

STAFF ACKNOWLEDGMENTS

The following individuals also made important contributions to this
report:  David P.  Bixler, Margaret E.  Boeckmann, Donald J. 
Porteous, Cynthia J.  Scott, Margie K.  Shields, and Cheri Y. 
Truett. 


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

American Public Welfare Association.  "State-by-State Analysis of
Selected State Plan Provisions Allowing Flexibility." W-Memo, Vol. 
9, No.  3 (May-June 1997). 

________.  Survey Notes:  Issues Affecting Children, Vol.  1, No.  3
(1998). 

________.  Survey Notes:  TANF Eligibility, Benefits, Work,
Sanctions, and Exemptions, Vol.  1, No.  2 (1997). 

________.  "Welfare Reform:  Reducing the Rates of Out-of-Wedlock
Births and Teen Pregnancies." W-Memo, Vol.  9, No.  1 (Jan.-Feb. 
1997). 

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

Blank, Rebecca M.  What Causes Public Assistance Caseloads to Grow? 
(working paper #6343).  Cambridge, Mass.:  National Bureau of
Economic Research, Dec.  1997. 

Blank, Rebecca M., and Patricia Ruggles.  "Short-Term Recidivism
Among Public-Assistance Recipients." The American Economic Review,
Vol.  84, No.  2 (May 1994), pp.  49-53. 

Bloom, Dan.  After AFDC:  Welfare-to-Work Choices and Challenges for
States.  New York:  Manpower Demonstration Research Corporation,
1997. 

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

Burtless, Gary.  "Employment Prospects of Welfare Recipients" in The
Work Alternative:  Welfare Reform and the Realities of the Job
Market, Demetra Smith Nightingale and Robert H.  Haveman, eds. 
Washington, D.C.:  The Urban Institute Press, 1995. 

Committee on Ways and Means, U.S.  House of Representatives.  l996
Green Book.  Washington, D.C.:  U.S.  Government Printing Office,
Nov.  4, 1996. 

Council of Economic Advisers.  Explaining the Decline in Welfare
Receipt, 1993-1996.  Washington, D.C.:  Council of Economic Advisers,
May 9, 1997. 

Gallagher, L.  Jerome, and others.  One Year After Federal Welfare
Reform:  A Description of State Temporary Assistance for Needy
Families (TANF) Decisions as of October 1997.  Washington, D.C.:  The
Urban Institute, May 1998. 
http://newfederalism.urban.org/html/1year.html (cited June 3, 1998). 

Greenberg, Mark.  Welfare-to-Work Grants and Other TANF-Related
Provisions in the Balanced Budget Act of 1997.  Washington, D.C.: 
Center for Law and Social Policy, Aug.  1997. 

Herr, Toby, and others.  Something Old, Something New:  A Case Study
of the Post-Employment Services Demonstration in Oregon.  Chicago: 
Erikson Institute, Nov.  1995. 

Hershey, Alan M., and LaDonna Pavetti.  "Turning Job Finders into Job
Keepers." The Future of Children:  Welfare to Work, Vol.  7, No.  1
(spring 1997). 

Institute for Women's Policy Research.  IPWR Welfare Reform Network
News:  Reproductive Issues and Welfare Reform.  Issue No.  11/12
(Oct./Nov.  1997). 

Levin-Epstein, Jodie.  State TANF Plans:  Out-of-Wedlock and
Statutory Rape Provisions.  Washington, D.C.:  Center for Law and
Social Policy, Aug.  l997. 

Lindberg, Laura D., and others.  "Age Differences Between Minors Who
Give Birth and Their Adult Partners." Family Planning Perspectives,
Vol.  29, No.  2 (Mar./Apr.  1997). 

Martini, Alberto, and Michael Wiseman.  Explaining the Recent Decline
in Welfare Caseloads:  Is the Council of Economic Advisors Right? 
Washington, D.C.:  Income and Benefits Policy Center, The Urban
Institute, 1997.  http://www.urban.org/welfare/cea.htm (cited May 14,
1998). 

Maryland Department of Human Resources and University of Maryland
School of Social Work.  Life After Welfare:  An Interim Report. 
Baltimore, Md.:  University of Maryland School of Social Work, Sept. 
1997. 

Moffitt, Robert A., and Eric P.  Slade.  "Health Care Coverage for
Children Who Are On and Off Welfare." The Future of Children: 
Welfare to Work, Vol.  7, No.  1 (spring 1997). 

National Conference of State Legislatures.  Welfare Reform &
Immigrants:  State Trends.  Washington, D.C.:  National Conference of
State Legislatures, July 15, 1997. 
http://www.igc.apc.org/handsnet2/welfare (cited Aug.  7, 1997). 

National Governors' Association and the National Association of State
Budget Officers.  The Fiscal Survey of the States.  Washington, D.C.: 
National Governors' Association and National Association of State
Budget Officers, Apr.  1997. 
http://www.nasbo.org/pubs/fiscsurv/fs0497.htm (cited Nov.  19, 1997). 

National Governors' Association Center for Best Practices.  Summary
of Selected Elements of State Plans for Temporary Assistance for
Needy Families, as of November 20, 1997.  Washington, D.C.:  National
Governors' Association, Nov.  20, 1997. 

Pavetti, LaDonna A.  "Questions and Answers on Welfare Dynamics"
(paper presented at research meeting on welfare dynamics). 
Washington, D.C.:  The Urban Institute, Sept.  11, 1995. 

Pavetti, LaDonna A., and Amy-Ellen Duke.  State Welfare Reform
Efforts:  Executive Summary.  Washington, D.C.:  The Urban Institute. 
http://www.urban.org/pubs/statwel/stat_wel.htm (cited Nov.  12,
1996). 

Sonenstein, Freya L., and others.  Involving Males in Preventing Teen
Pregnancy:  A Guide for Program Planners.  Washington, D.C.:  The
Urban Institute, Jan.  1998. 
http://www.urban.org/family/invmales.html (cited Jan.  12, 1998). 

Turetsky, Vicki, and Andrea Watson.  Some States Continue $50 Child
Support Pass-Through Despite Federal Disincentive.  Washington, D.C.: 
Center for Law and Social Policy, Jan.  1998. 

University of Maryland School of Social Work.  Life After Welfare: 
Second Interim Report (prepared for Maryland Department of Human
Resources).  Baltimore, Md.:  University of Maryland School of Social
Work, Mar.  1998. 

U.S.  Congressional Research Service.  Transition to TANF:  State
Benefit Levels and Selected Program Rules, July 1, l996 - January 1,
l997.  Washington, D.C.:  CRS, June 25, l997. 

________.  Welfare Reform:  Adolescent Pregnancy Issues (94-591 EPW). 
Washington, D.C.:  CRS, updated Dec.  7, 1995. 

________.  Welfare Reform:  State Programs Under the Block Grant for
Temporary Assistance for Needy Families.  Washington, D.C.:  CRS,
updated Sept.  4, 1997. 

U.S.  Department of Health and Human Services, Administration for
Children and Families.  Characteristics and Financial Circumstances
of AFDC Recipients, Fiscal Year 1994.  Washington, D.C.:  HHS. 

________.  Characteristics and Financial Circumstances of AFDC
Recipients, Fiscal Year 1995.  Washington, D.C.:  HHS. 

________.  Characteristics and Financial Circumstances of AFDC
Recipients, Fiscal Year 1996.  Washington, D.C.:  HHS. 

________.  Report to Congress on Data Processing and Case Tracking in
the TANF Program.  Washington, D.C.:  HHS, Dec.  1997. 

U.S.  Department of Health and Human Services, Office of the
Assistant Secretary for Planning and Evaluation.  Setting the
Baseline:  A Report on State Welfare Reform Waivers.  Washington,
D.C.:  HHS, June l997. 

U.S.  Department of Health and Human Services, Administration for
Children and Families and Office of the Assistant Secretary for
Planning and Evaluation, and U.S.  Department of Education, Office of
the Under Secretary and Office of Vocational and Adult Education. 
Evaluating Two Welfare-to-Work Program Approaches:  Two-Year Findings
on the Labor Force Attachment and Human Capital Development Programs
in Three Sites (national evaluation of welfare-to-work strategies). 
Washington, D.C.:  HHS, Dec.  1997. 

Ziliak, James P., and others.  Accounting for the Decline in AFDC
Caseloads:  Welfare Reform or Economic Growth?  Ann Arbor, Mich.: 
University of Michigan, revised Sept.  1997. 




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

Medicaid:  Early Implications of Welfare Reform for Beneficiaries and
States (GAO/HEHS-98-62, Feb.  24, 1998). 

Welfare Reform:  HHS' Progress in Implementing Its Responsibilities
(GAO/HEHS-98-44, Feb.  2, 1998). 

Welfare Reform:  States' Efforts to Expand Child Care Programs
(GAO/HEHS-98-27, Jan.  13, 1998). 

Social Services Privatization:  Expansion Poses Challenges in
Ensuring Accountability for Program Results (GAO/HEHS-98-6, Oct.  20,
1997). 

Welfare Reform:  Three States' Approaches Show Promise of Increasing
Work Participation (GAO/HEHS-97-80, May 30, l997). 

Welfare Reform:  Implications of Increased Work Participation for
Child Care (GAO/HEHS-97-95, May 29, 1997). 

Welfare Reform:  States' Early Experiences With Benefit Termination
(GAO/HEHS-97-74, May 15, l997). 

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:  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 Dependency:  Coordinated Community Efforts Can Better Serve
Young At-Risk Teen Girls (GAO/HEHS/RCED-95-108, May 10, 1995). 

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


*** End of document. ***