Welfare to Work: Most AFDC Training Programs Not Emphasizing Job
Placement (Chapter Report, 05/19/95, GAO/HEHS-95-113).

Pursuant to a congressional request, GAO provided information on
employment-focused welfare-to-work programs, focusing on: (1) the extent
to which county and local Job Opportunities and Basic Skills Training
(JOBS) programs focus on employment; and (2) factors that hinder
administrators' efforts to move Aid to Families With Dependent Children
(AFDC) recipients into jobs.

GAO found that: (1) some welfare-to-work programs stress employment and
work closely with employers in promoting work among welfare recipients;
(2) although the programs reviewed keep participants focused on the
importance of work and help program participants find jobs or
work-experience positions, they vary in their approach; (3) many county
JOBS programs do not have a strong employment focus and many county
administrators do not work with employers to find jobs for participants
or use work-experience programs; (4) many local program administrators
believe that insufficient staffing and resources hinder their work with
employers and more flexibility in federal rules governing
work-experience programs would facilitate their use; (5) the low-wage
work available to many AFDC recipients discourages their movement into
the work force; and (6) AFDC programs may emphasize preparing
participants for employment without also making strong efforts to help
them get jobs, since states are not required to track the number of AFDC
recipients who get jobs or earn their way off AFDC.

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

 REPORTNUM:  HEHS-95-113
     TITLE:  Welfare to Work: Most AFDC Training Programs Not 
             Emphasizing Job Placement
      DATE:  05/19/95
   SUBJECT:  Education or training
             Aid to families with dependent children
             Disadvantaged persons
             State-administered programs
             Aid for training or employment
             Employment or training programs
             Welfare recipients
             Public assistance programs
             Single parents
             Welfare services
IDENTIFIER:  JOBS Program
             Job Opportunities and Basic Skills Training Program
             AFDC
             America Works Program (New York, NY)
             West Virginia
             Athens (OH)
             New York (NY)
             San Jose (CA)
             Riverside County (CA)
             Head Start Program
             Charleston (WV)
             Alameda County (CA)
             Butte County (CA)
             Los Angeles County (CA)
             San Diego County (CA)
             Tulare County (CA)
             Food Stamp Program
             Earned Income Tax Credit
             Milwaukee (WI)
             Silicon Valley (CA)
             Medicaid Program
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


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


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

May 1995

WELFARE TO WORK - MOST AFDC
TRAINING PROGRAMS NOT EMPHASIZING
JOB PLACEMENT

GAO/HEHS-95-113

Welfare to Work


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

  ACF - Administration for Children and Families
  AFDC - Aid to Families With Dependent Children program
  CET - Center for Employment Training
  EITC - Earned Income Tax Credit
  HHS - Department of Health and Human Services
  JOBS - Job Opportunities and Basic Skills Training program
  JTPA - Job Training Partnership Act

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


B-254614

May 19, 1995

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

Dear Senator Moynihan: 

This report, prepared at your request, provides information on
examples of employment-focused welfare-to-work programs, the extent
to which Job Opportunities and Basic Skills Training programs
nationwide focus on employment, and factors that hinder programs'
efforts to move recipients of Aid to Families With Dependent Children
into jobs. 

We are sending copies of this report to the Chairman, Senate
Committee on Finance; the Chairmen and Ranking Minority Members of
the House Committee on Ways and Means and its Subcommittee on Human
Resources; 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 upon request. 

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

Sincerely yours,

Jane L.  Ross
Director, Income Security Issues


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


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

To combat welfare dependence and improve the lives of poor children,
the 104th Congress proposes to fundamentally change the nation's
largest cash assistance program for poor families with children, the
Aid to Families With Dependent Children program (AFDC).  The Congress
is considering limiting the length of time families may receive cash
assistance, ending benefits for unwed mothers under 18 years old, and
converting AFDC from an entitlement to a block grant.  While there is
general agreement that reforms should promote work, the Congress is
also debating the type and extent of work requirements to be
established for those receiving cash assistance. 

In 1988, the Congress strengthened existing work requirements for
AFDC recipients by creating the Job Opportunities and Basic Skills
Training (JOBS) program, designed to encourage and require AFDC
parents to move from welfare to work.  Under JOBS, programs are to
provide participants with education and training if deemed necessary
to prepare them for work.  Program officials must also work with
employers to help place participants in jobs.  In addition, programs
can provide temporary subsidies to employers to encourage them to
hire and train AFDC recipients.  And when AFDC recipients are not
otherwise able to find work, JOBS may place them with public,
nonprofit, or other organizations to gain work experience while they
continue to receive their AFDC benefits. 

To help the Congress as it considers welfare reform, the Ranking
Minority Member of the Senate Committee on Finance asked GAO to
provide information on (1) examples of county or local JOBS or
JOBS-like programs that emphasize job placement, subsidized
employment, or work-experience positions for welfare recipients; (2)
the extent to which county JOBS programs nationwide use these
employment-focused activities; and (3) factors that hinder program
administrators' efforts to move welfare recipients into jobs. 


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

Under JOBS, states are to provide AFDC recipients with the education,
training, and employment-related and support services they need to
become employed and assume responsibility for the support of their
children.  The federal and state governments share the costs of JOBS,
which is overseen at the federal level by the Department of Health
and Human Services (HHS) and at the state level by state AFDC
agencies.  While states supervise the program, services are delivered
at the county or local level.  In fiscal year 1993, the federal and
state governments spent about $2 billion for JOBS, which served more
than 500,000 AFDC recipients each month. 

To encourage states to work towards the federal goal of reducing
welfare dependence, the Congress reduces the amount of federal funds
available to states unless they serve increasing portions of their
AFDC populations in JOBS programs and target resources to those most
at risk of long welfare stays.  While states have generally met these
participation and targeting requirements, little is known about the
extent to which JOBS has moved welfare recipients into employment or
reduced welfare dependency.\1 HHS has contracted for an evaluation of
JOBS' effectiveness in seven sites throughout the nation. 

To accomplish its objectives, GAO visited five welfare-to-work
programs in California, New York, Ohio, and West Virginia that
emphasized job-placement activities, subsidized employment, or
work-experience programs.  GAO identified these programs through
discussions with JOBS officials at the federal level, welfare-to-work
experts and practitioners, and a review of selected evaluation
literature.  GAO also surveyed a nationally representative random
sample of 453 county JOBS program administrators in mid-1994.  A more
detailed discussion of GAO's methodology appears in chapter 1. 


--------------------
\1 See Welfare to Work:  Measuring Outcomes for JOBS Participants
(GAO/HEHS-95-86, Apr.  17, 1995). 


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

Programs that stress the goal of employment for their participants
and forge close links with employers show promise in promoting work
among welfare recipients.  The five programs that GAO visited all
share that emphasis, although they differ in their approaches.  For
example, a Riverside County, California, program emphasizes moving
participants quickly into employment, while a San Jose, California,
program provides longer-term skills training.  In both cases and in
the other programs that GAO visited, program administrators keep
participants focused on the importance of work and forge links with
employers or other organizations to help participants find jobs or
work-experience positions. 

A majority of county JOBS programs across the nation, however, do not
have a strong employment focus.  About one-half of the county JOBS
administrators nationwide stated that they do not work enough with
employers to find jobs for participants.  In addition, although most
of the program officials reported that less than one-half of their
job-ready participants had become employed, the officials reported
little use of subsidized employment or work- experience programs,
options available under JOBS. 

Although program administrators expressed interest in working more
with employers and expanding their use of subsidies and
work-experience programs, they reported that many obstacles stood in
their way.  Most local administrators cited insufficient staffing and
resources as hindering their work with employers.  Many also stated
that more flexibility in federal rules governing employment subsidies
and work-experience programs, such as the option to place
participants with private-sector employers to gain work experience,
could facilitate their use.  In addition, most administrators noted
that the low-wage work available to many AFDC recipients discourages
their movement into the work force. 

GAO also notes that while the current federal participation rate and
targeting requirements hold states accountable for the number and
type of AFDC recipients participating in JOBS activities, states
currently need not track the number who get jobs or earn their way
off AFDC.  As a result, programs may emphasize preparing participants
for employment without also making strong efforts to help them get
jobs. 


   GAO'S ANALYSIS
---------------------------------------------------------- Chapter 0:4


      EXAMPLES OF
      EMPLOYMENT-FOCUSED PROGRAMS
-------------------------------------------------------- Chapter 0:4.1

Some welfare-to-work programs promote a strong employment message and
forge links with employers or other organizations that help to
promote work for welfare recipients.  GAO visited five such programs. 
They varied in their costs per participant and types of training
provided, but they all focused participants and staff on the
importance of obtaining employment or work experience.  In addition,
each program worked with local employers, creating workplace
connections that played an important role in making programs more
responsive to local employers. 

For example, in Riverside County's JOBS program in California,
program staff strongly encourage participants to take any job they
can find to get into the work force.  Researchers found that this
JOBS program saved federal and state governments $3 for every $1
invested, outperforming the five other California counties studied. 
Another program, the Center for Employment Training in San Jose,
works closely with employers to offer training for occupations that
are in high demand in the local economy, ranging from home health
aide to metal worker.  Research showed that this program increased
participants' earnings 47 percent over those not participating in the
program. 

In New York City, the state welfare agency pays America Works, a
for-profit firm, to find private-sector jobs for some AFDC
recipients.  These jobs are partly subsidized with AFDC grant dollars
for 4 months.  After this period of supported work, about 65 percent
of the AFDC recipients are ultimately hired by the employers. 
Although this program has not been evaluated, its approach shows
promise in helping AFDC recipients get jobs and remain employed. 
Other programs promote work for AFDC recipients even when regular
jobs are not available.  In Athens County, Ohio, and in several West
Virginia counties, some welfare recipients gain work experience while
contributing to their communities by working in government and
nonprofit agencies while receiving their welfare grants.  Research on
work-experience programs shows little evidence that such programs can
increase paid employment, but they can produce benefits for taxpayers
through the work performed and offer meaningful work for welfare
recipients. 


      MOST PROGRAMS DO NOT HAVE A
      STRONG EMPLOYMENT FOCUS
-------------------------------------------------------- Chapter 0:4.2

While some programs, such as the five that GAO visited, have created
employment-focused environments, most JOBS programs nationwide
seemingly have not, according to the responses of local program
administrators.  About one-half of the administrators reported that
they do not do enough to identify job openings or market participants
to employers.  In addition, although about 70 percent of the program
officials reported that one-half or less of their job-ready
participants had become employed, they indicated that their programs
made limited use of subsidized work and work- experience programs. 
Nationwide in mid-1994, about 10 percent of JOBS participants were
placed in work-experience positions and about 1 percent were in
subsidized employment. 


      ADMINISTRATORS CITE
      HINDRANCES TO PLACING
      PARTICIPANTS IN WORK
-------------------------------------------------------- Chapter 0:4.3

Most JOBS administrators cited insufficient staff as hindering their
work with employers to help participants find work.  Working with
employers to identify job openings or create subsidized or
work-experience positions takes considerable staff time and effort. 
However, administrators GAO spoke with and several studies have noted
that JOBS staff are busy enrolling participants, assigning them to
JOBS activities, and tracking and monitoring them, which leaves staff
little time to work with employers. 

GAO also notes that the current participation goals are
process-oriented--
focusing on the numbers of persons enrolled in JOBS activities--and
do not hold states accountable for such outcomes as the number of
JOBS graduates who leave welfare for work.  At one site that GAO
visited, a JOBS participant had successfully completed several
different training programs.  Under the current performance system,
this individual helps the program meet the federal requirement to
receive its full share of federal funding.  Yet the participant
remained unemployed and on AFDC.  While education and training may be
needed by some participants to help them find employment, such
activities should be an intermediate step on the way to the ultimate
goal of employment. 

Administrators cited other hindrances as well.  They reported that
subsidizing employers can be more costly than other JOBS activities. 
In addition, a majority expressed interest in more flexible use of
work-experience programs through revision of certain federal
restrictions.  For example, one official noted that she could expand
participants' work experience if federal restrictions were revised to
permit placing participants in all types of private-sector
employment. 

The low-wage work that many welfare recipients are likely to find
also serves to weaken programs' focus on employment.  About 70
percent of administrators believe that their job-ready participants
do not become employed because they cannot find jobs that pay
sufficient wages and benefits to support their families. 


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

GAO is not making recommendations in this report. 


   AGENCY COMMENTS
---------------------------------------------------------- Chapter 0:6

In commenting on a draft of this report (see app.  IV), HHS'
Administration for Children and Families (ACF) disagreed with GAO's
conclusion that JOBS programs do not have a strong employment focus. 
ACF stated that GAO did not sufficiently recognize programs' use of
job search or the extent of their job-development activities in
evaluating their employment focus.  In addition, it stated that GAO
did not acknowledge the many ways that programs could focus on
employment and, instead, relied too much on programs' low use of
subsidized employment and work experience to indicate a weak
employment focus. 

GAO continues to believe that the evidence gathered shows that JOBS
programs do not have a strong employment focus.  Many JOBS programs
nationwide appear to emphasize preparing participants for employment
without also making strong efforts to help place them in jobs.  In
addition, when unsubsidized work is not available, the programs have
made little use of subsidized employment or unpaid work-experience
positions. 


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

The 104th Congress is moving to make major changes in AFDC, the
nation's largest cash assistance program for needy families with
children.  Under consideration are limiting the number of years that
cash assistance may be received, capping benefit increases for
mothers on welfare who have additional children, denying cash
assistance to unwed mothers under 18 years old, and transforming AFDC
from entitlement status to a block grant administered by states.\2
Also under consideration are the type and extent of work requirements
to be established. 

In 1988, the Congress created the JOBS program to transform AFDC into
a transitional program geared toward helping parents become employed
and avoid long-term welfare dependence.  Under JOBS, states are to
assess the needs and skills of AFDC recipients, prepare them for
employment through education and training as needed, and place them
in jobs.  We reported earlier that while states have made progress in
implementing JOBS, only a small percentage of the almost 4.6 million
adults on AFDC participated in work-preparation activities in fiscal
year 1993.  Moreover, little is known about the JOBS program's
progress in moving parents into employment and reducing their
dependence on welfare.\3

To help the Congress as it considers welfare reform, the Ranking
Minority Member of the Senate Committee on Finance requested us to
provide information on (1) examples of county or local programs that
stressed job placement, subsidized employment, or work-experience
positions for welfare recipients; (2) the extent to which county JOBS
programs nationwide emphasized these employment-focused activities;
and (3) factors that hinder program administrators' efforts to move
welfare recipients into jobs. 


--------------------
\2 On March 24, 1995, the House of Representatives passed H.R.  4,
which included these and other reforms of AFDC. 

\3 Welfare to Work:  Current AFDC Program Not Sufficiently Focused on
Employment (GAO/HEHS-95-28, Dec.  19, 1994). 


   BACKGROUND
---------------------------------------------------------- Chapter 1:1

Through AFDC, the federal government and the states provide cash
assistance to needy families with children who lack support from one
or both parents because of death, absence, incapacity, or
unemployment.  As shown in figure 1.1, since 1970, the number of
female-headed families, including those headed by women who have
never married, has more than doubled, as has the number of families
receiving AFDC.  According to a Congressional Budget Office study,
the growth in female-headed families, especially those headed by
females who had never been married, accounted for about one-half of
the sharp increase of 1.2 million in the number of AFDC families
between 1989 and 1993.\4

   Figure 1.1:  Growth in Number
   of Families, by Type (1970-93)

   (See figure in printed
   edition.)

Source:  AFDC data from Department of Health and Human Services;
other family data from the Bureau of the Census. 

In 1993, the federal and state governments spent over $25 billion to
provide AFDC benefits to 14 million adults and children, most of them
in single-parent families.  About 56 percent of AFDC mothers live in
central cities, another 25 percent of them live in the suburbs, and
the remaining 19 percent reside in rural areas.\5 Since the 1970s,
the percentage of single mothers receiving AFDC who had never been
married has doubled, from 21 percent in 1976 to 52 percent in 1992. 
About 70 percent of families receiving benefits have 1 or 2 children. 

While most AFDC recipients are single mothers, these women are a
diverse group, making use of the program in different ways.  For
example, one study has estimated the total time that those receiving
AFDC at a point in time can be expected to receive benefits, as shown
in figure 1.2.\6

This analysis indicates that 9 percent of these recipients are using
AFDC for only a short time--2 years or less.  About 76 percent,
however, are receiving AFDC benefits for a total of 5 years or more,
when all moves on and off welfare are considered.\7 According to
these data, under a 5-year limit on receipt of cash assistance--a
measure included in the House welfare reform bill--about
three-fourths of those on AFDC may be expected to hit the time limit
and need to support themselves through employment or other means.\8

   Figure 1.2:  Estimates of the
   Total Time Those Receiving AFDC
   at a Point in Time Will Spend
   on AFDC

   (See figure in printed
   edition.)

Note:  Based on estimates of the total number of months that the AFDC
caseload at a point in time would receive benefits during the
subsequent 25 years, including both continuous and noncontinuous
periods of receipt. 

Source:  Beebout, Jacobson, and Pavetti, The Number and
Characteristics of AFDC Recipients Who Will Be Affected by Policies
to Time-Limit AFDC Benefits (1994), p.  12. 

This prospect poses a formidable challenge for many AFDC recipients
who have limited education, job skills, and work experience.  About
45 percent of all AFDC recipients, for example, have less than a high
school diploma.\9 Surveys of several thousand AFDC recipients
expected to participate in JOBS in selected sites showed that at
least one-third had extremely low literacy skills and between
one-fourth and more than one-half lacked prior work experience.  Over
one-fourth thought they could not prepare for work because they or
their family members had health or emotional problems.\10

Such recipients are at risk of long-term welfare dependence.  We
reported previously that states have made some progress in working
with some of these recipients but that many remain unserved.\11

The JOBS program, begun in 1989, was designed to improve upon the
performance of previous welfare-to-work programs and help combat
long-term welfare dependence.  Research studies conducted up to then
showed that employment training programs for welfare recipients could
have a positive but generally modest effect on increased earnings and
reduced welfare costs.\12 They also showed that programs that
emphasized low-cost services, such as job search, generally did not
help welfare recipients get higher paying jobs than they would have
without the programs or help the more disadvantaged.  It was hoped
that JOBS could improve upon previous programs' performance by
reaching further into the AFDC caseload and providing more
comprehensive services, including education and training, to help
parents find jobs that would end their dependence on welfare.\13

To this end, under JOBS, states are to (1) provide a broad range of
education, training, and employment-related activities; (2) increase
the number of AFDC recipients participating in these activities; (3)
target resources to the hard-to-serve; and (4) provide support
services, including child care, transportation, work-related, and
other support services, such as mental health counseling, if deemed
necessary. 


--------------------
\4 Forecasting AFDC Caseloads, With an Emphasis on Economic Factors,
Congressional Budget Office Staff Memorandum (Washington, D.C.: 
1993), pp.  1, 3. 

\5 Based on AFDC mothers 15 to 44 years old.  See Mothers Who Receive
AFDC Payments--Fertility and Socioeconomic Characteristics, Economics
and Statistics Administration, Bureau of the Census, U.S.  Department
of Commerce (Washington, D.C.:  1995). 

\6 The Number and Characteristics of AFDC Recipients Who Will Be
Affected by Policies to Time-Limit AFDC Benefits, Harold Beebout and
Jon Jacobson, Mathematic Policy Research, and LaDonna Pavetti, The
Urban Institute (Washington, D.C.:  1994), p.  12. 

\7 While this estimate indicates that about three-fourths of those on
AFDC at a point in time would receive benefits for a total of 5 years
or more, a much smaller percentage of those who ever receive
AFDC--about 35 percent--are estimated to receive benefits for that
duration.  Of those who ever receive AFDC, 42 percent are estimated
to receive benefits for a total of 2 years or less, and 23 percent
for more than 2 but less than 5 years. 

\8 This estimate assumes that there would be no exemptions from the
time limit set for welfare receipt.  The House bill (H.R.  4) allows
states to grant up to 10 percent of their welfare recipients a
hardship exemption from this lifetime limit. 

\9 Families on Welfare:  Sharp Rise in Never-Married Women Reflects
Societal Trend (GAO/HEHS-94-93, May 31, 1994). 

\10 The data on work experience came from a survey of 48,000
recipients in seven states; the other data were from about 20,000
recipients in four of these seven states.  See Gayle Hamilton and
Thomas Brock, The JOBS Evaluation:  Early Lessons from Seven Sites,
U.S.  Departments of Health and Human Services and Education (New
York:  1994). 

\11 Welfare to Work:  Current AFDC Program Not Sufficiently Focused
on Employment. 

\12 See Judith M.  Gueron and Edward Pauly, From Welfare to Work,
Russell Sage Foundation (New York:  1991). 

\13 HHS currently is sponsoring an experimental design seven-site
evaluation of JOBS to determine its effectiveness in increasing
employment and earnings for welfare recipients and reducing welfare
costs. 


      CURRENT JOBS PARTICIPATION
-------------------------------------------------------- Chapter 1:1.1

To encourage states to work towards the federal goal of reducing
welfare dependency, the Congress created minimum participation and
targeting requirements that states must meet to receive their full
share of federal funding.  The minimum participation requirements
rose from 7 percent of nonexempt AFDC recipients\14 in fiscal year
1991 to 20 percent in fiscal year 1995.\15 Under the targeting
requirements, states must spend 55 percent of their JOBS funds on
designated target groups.\16

The Congress also expected that performance standards based on
outcomes, such as increased employment and earnings and reduced
welfare dependency, would be established after the initial
implementation of the program.  Outcome-related performance standards
have not yet been established.  (For more information on the current
status of these performance standards, see p.  42.)

While most states have met the minimum participation requirements,
the number of AFDC recipients participating in JOBS remains limited. 
About one-half of the adults receiving AFDC have been exempted from
JOBS, most often because they are caring for a young child.\17 Of
those considered nonexempt, states decide how many to serve in JOBS
based on the availability of state resources.  As shown in figure
1.3, the number participating in JOBS each month, while increasing,
has remained limited for fiscal years 1991 through 1993.\18 In 1993,
about 11 percent of the 4.6 million adults receiving AFDC were active
in JOBS activities each month.  Although some individual programs
have succeeded in serving most of their nonexempt AFDC recipients,
JOBS programs overall served only about one-fourth of the nonexempt
population. 

   Figure 1.3:  Average Monthly
   Number of AFDC Recipients,
   Nonexempt Recipients, and Those
   Participating in JOBS (Fiscal
   Years 1991-93)

   (See figure in printed
   edition.)

\a Recipients with any level of involvement or participation in
JOBS-approved activity, including assessment. 


--------------------
\14 AFDC recipients 16 through 59 years old are considered nonexempt
unless they are ill or incapacitated; working 30 hours or more per
week; attending high school; or caring for children under 3 years old
(1 year old at state option).  However, teenage parents who have not
completed high school and have children under 3 years old are also
nonexempt. 

\15 The Congress also established separate minimum requirements for
participation for principal earners in two-parent families receiving
AFDC-Unemployed Parent benefits, beginning at 40 percent in fiscal
year 1994 and increasing to 75 percent in fiscal year 1997. 

\16 JOBS target group members include AFDC recipients or applicants
who have received AFDC for 36 months out of the past 5 years; are
under 24 years old and (a) have neither completed nor are enrolled in
high school or (b) had little or no work experience in the preceding
year; or are soon to become ineligible for AFDC because their
youngest child is almost 18 years old. 

\17 Those exempted may enroll in JOBS as volunteers. 

\18 During a year's time, a larger percentage of recipients may
participate in the program, but because HHS maintains its data on an
average monthly basis, it is not possible to calculate the percentage
that participated at any time during a year. 


      JOBS PROGRAMS ARE EXPECTED
      TO ASSESS, PREPARE, AND
      PLACE PARTICIPANTS
-------------------------------------------------------- Chapter 1:1.2

Three basic steps--assessment, employment preparation, and job
placement--are involved in attempting to move AFDC recipients into
employment through JOBS, and programs are provided with tools to help
them accomplish this, as shown in figure 1.4.  First, JOBS programs
must perform an assessment of each participant's needs and skills
and, with the participant, develop an employability development plan
setting forth an employment goal and a schedule of services and
activities necessary to accomplish that goal.\19 \, \20

   Figure 1.4:  Three Basic Steps
   in Moving JOBS Participants
   into Employment

   (See figure in printed
   edition.)

Then, as part of the preparation phase, all JOBS programs must
provide high school and equivalency programs, basic and remedial
education, job-readiness activities, job skills training, and support
services.  In addition, they may provide postsecondary education. 

Finally, JOBS programs must provide services designed to place
participants in jobs.  To accomplish this, programs are required to
conduct job-development activities, including identifying job
openings, marketing participants to employers, and arranging
interviews for participants.  Programs must also consult with local
private industry councils,\21 which include employers, or may choose
to work directly with employers to ensure that participants receive
education and training that prepares them for jobs that are available
in the local area. 

To help participants find work, JOBS programs must include at least
two of the following activities as part of their programs:  job
search, on-the-job training, work supplementation, or work
experience.  Participants enrolled in job search look for work on
their own or under program supervision.  The other options are called
work activities because they involve placing JOBS participants with
employers or at worksites with community sponsors. 

Two of these work activities, on-the-job training and work
supplementation, involve the use of short-term wage subsidies to
encourage employers to hire and train JOBS participants.  When JOBS
funds are used to reimburse employers, the JOBS activity is called
on-the-job training; when the participant's AFDC grant is diverted to
subsidize the employer, the activity is called work supplementation
or grant diversion.  In both cases, the participant receives a
paycheck instead of a welfare check.\22 These programs are designed
to encourage employers to hire welfare recipients whose productivity
may be lower than that of other potential employees. 

A third work activity, community work experience, and a similar
activity called alternative work experience--both referred to as
work-experience programs in this report--are designed to provide
welfare recipients with actual work-place experience to increase
their employability.  Under these programs, JOBS places participants
with public and nonprofit agencies to perform services for their
community.  These programs do not include payments to the employers,
and JOBS participants continue to receive their welfare checks.  For
a comparison of the federal rules governing the various work
activities for AFDC recipients, see appendix I. 


--------------------
\19 Federal law requires JOBS programs to make an assessment of
employability based on a participant's educational, child care, and
other support services needs; skills and work experiences; and family
circumstances.  The types of assessments used can range from 5-page
surveys filled out by participants to comprehensive career-oriented
assessments. 

\20 If participants are considered job-ready when they enter the
program, they may be required to look for work immediately without
further employment preparation.  Programs have varying criteria on
when a participant is considered job-ready.  While some local
programs encourage all of their participants to look for work before
being placed in education, training, or work-related activities, most
require some minimum level of education, skills, or work experience
before participants are expected to look for work. 

\21 Under the federal Job Training Partnership Act (JTPA), local
entities are charged with forming a private industry council, with
employer representation, that oversees training programs for
economically disadvantaged individuals. 

\22 In on-the-job training programs, JOBS programs may use JOBS funds
to reimburse the training and supervision costs of an employer who
hires a JOBS client.  Under a work-supplementation program, all or
part of the AFDC grant is diverted to an employer to cover part of
the cost of wages for a JOBS participant for up to 9 months. 


      STATE AND FEDERAL
      GOVERNMENTS SHARE JOBS
      RESPONSIBILITIES
-------------------------------------------------------- Chapter 1:1.3

Within the federal JOBS guidelines, states and localities assess the
needs of their JOBS participants, determine the type and intensity of
services provided, and set the criteria by which participants are
deemed job-ready.  They also have discretion to establish the wage
level and benefits associated with the employment goal established in
the employability development plan.  Some programs set wage goals as
high as $8 per hour, while others believe that a job at any wage
level is an appropriate goal. 

To help AFDC recipients move towards self-sufficiency, states rely on
two federal funding sources.  First, about $1 billion of federal JOBS
funds has been made available annually in recent years for allocation
to the states.  States must then commit their own funds to JOBS to
match these federal funds.\23 In fiscal year 1993, states used about
70 percent of the federal JOBS funds available to them.  Second, the
federal government has provided an uncapped source of funds to share
with states the costs of providing child care assistance to AFDC
recipients in education or training programs or who are employed.  In
fiscal year 1993, the federal government provided about $1.2 billion
of the almost $2 billion spent by states on JOBS and AFDC child care. 

HHS oversees the JOBS program at the federal level and state AFDC
agencies supervise it.  At the local level, JOBS is administered
either by the state AFDC office or by county officials.\24 Before
using JOBS funds to purchase services for participants, programs must
make full use of the services and resources available in their
communities without charge to AFDC recipients.  Programs may also
contract with other organizations for services.  As a result,
programs rely heavily on a variety of community resources, such as
Job Training Partnership Act (JTPA) agencies, adult basic education
programs, high schools, the state employment service, Head Start, and
community colleges.\25


--------------------
\23 The federal government shares in the costs of a state's JOBS
program at three different rates.  First, for each state's JOBS
spending up to the amount spent on certain fiscal year 1987
welfare-to-work activities, the federal share is 90 percent.  Second,
for the nonadministrative costs of providing services and full-time
staff, the federal share is 60 to 80 percent, depending on a state's
average per capita income.  Third, for administrative and support
services costs, other than child care, the federal share is 50
percent. 

\24 While the AFDC agency must maintain supervisory control over the
JOBS program, it may contract out certain activities and services. 
In some states or areas, JOBS is operated by the JTPA agency, the
state employment service, community-based organizations, or a
combination of agencies and providers. 

\25 For more information on the extent to which JOBS programs rely on
other providers for services, see JOBS:  Participants'
Characteristics and Services Provided (GAO/HEHS-95-93, May 2, 1995)
and Irene Lurie and Jan Hagen, Implementing JOBS:  Initial Design and
Structure, The Nelson A.  Rockefeller Institute of Government, State
University of New York (Albany, New York:  1993). 


   SCOPE AND METHODOLOGY
---------------------------------------------------------- Chapter 1:2

To identify welfare-to-work programs that strongly emphasize
employment or work for their welfare recipients, through
job-placement activities, subsidized employment, or work-experience
positions, we reviewed welfare-to-work evaluations and HHS program
data and contacted HHS officials and welfare experts.  We then
visited selected programs in Riverside County and San Jose,
California; Athens, Ohio; and New York, New York.  Also, in
Charleston, West Virginia, we spoke with six JOBS officials
representing 11 West Virginia counties.  Where results from impact
evaluations were available, they are included in the text; however,
only two of the five programs have been rigorously evaluated to
measure program effects.  We also note that the program cost data
cited may not be comparable among the different programs described. 

To determine the extent to which county JOBS programs nationwide used
these employment-focused elements and to identify factors that hinder
administrators' efforts to move AFDC recipients into employment, we
collected and analyzed data from a range of sources.  To obtain
nationally representative data, we randomly sampled 453 of the
nation's 3,141 counties and mailed questionnaires to their JOBS
administrators in May 1994.  The sample was stratified to ensure
representation of the nation's central-city, suburban, and rural
counties.  It included the nation's 10 largest central-city counties,
based on the number of female-headed families with children receiving
public assistance in 1990. 

Our analysis of the questionnaire data generally showed few material
differences among the responses of the counties comprising the 10
largest central cities, other central-city counties, suburban
counties, or rural counties.  Consequently, we present the results
using combined data from all the strata.  See appendix II for more
information about our sample.  The questionnaire and summaries of the
responses are in appendix III. 

For more information on JOBS program implementation, we spoke with
program administrators at HHS and the Department of Labor;
representatives of the National Alliance of Business and the American
Federation of State, County and Municipal Employees; and welfare
experts.  We also reviewed HHS and congressional welfare reform
proposals and analyzed economic data provided by the Bureau of the
Census and the Department of Commerce.  In addition, we visited JOBS
programs in Alameda, Napa, Santa Clara, and Sonoma Counties in
California; Franklin County in Ohio; and gathered additional
information at a meeting with JOBS administrators from 12 counties in
the San Francisco area. 

We conducted our work between September 1993 and April 1995 in
accordance with generally accepted government auditing standards. 


EXAMPLES OF EMPLOYMENT-FOCUSED
PROGRAMS
============================================================ Chapter 2

Some local welfare-to-work programs are well-focused on employment,
working closely with employers to help participants find jobs or
using subsidized employment or work experience to promote work for
welfare recipients.  We saw this in programs at five locations we
visited:  Riverside County and San Jose, California; New York, New
York; Athens, Ohio; and West Virginia.  The programs in these places
vary in their costs per participant and other features.  Yet they all
focus on work as the ultimate goal, with three of the programs--in
Riverside County, San Jose, and New York--working closely with
employers to move participants into paid employment, and the Athens
and West Virginia programs supporting work- experience positions when
regular employment was not available. 

Table 2.1 summarizes selected program features and highlights
important differences among the programs.  For example, the Riverside
County program is administered by a welfare agency and involves all
of the county's JOBS participants.  While welfare agencies also
operate the Athens and West Virginia programs, their work-experience
programs involve only a portion of their JOBS participants.  The
other programs are not JOBS programs and are not operated by welfare
agencies.  A nonprofit organization operates the San Jose program,
which serves welfare recipients among other individuals in the
community.  And a for-profit firm runs the New York City program
under contract to the state welfare agency; it serves but a small
fraction of the JOBS participants in the city.  We also note that the
Riverside County and San Jose programs have research-documented
success in getting more AFDC recipients employed than would have
occurred without the programs.  A more detailed discussion of these
programs follows. 



                                    Table 2.1
                     
                     Selected Features of Employment-Focused
                             Welfare-to-Work Programs

                Riverside                                       Athens, Ohio,
                County,         San Jose,       New York,       and
                California      California      New York        West Virginia
--------------  --------------  --------------  --------------  ----------------
Name            Riverside       Center for      America Works   Athens County,
                County GAIN\a   Employment                      Ohio, JOBS and
                                Training                        state of West
                                                                Virginia JOBS

Type and        JOBS program    Vocational      Job placement   JOBS programs
administrator   operated by     school          and support     operated by
                county AFDC     operated by     services        county and state
                agency          nonprofit       provided by     AFDC agencies
                                organization    for-profit
                                                company

Participants    Enrolls about   Enrolls about   Enrolls about   Athens: Enrolls
                2,000           470 AFDC and    120 AFDC        94 of 550 JOBS
                nonexempt AFDC  non-AFDC        volunteers in   participants in
                recipients in   individuals,    city with       a county with
                a county with   generally       309,000 AFDC    1,600 AFDC
                30,000 AFDC     volunteers, in  cases           cases.
                cases           county with                     West Virginia:
                                30,000 AFDC                     Enrolls 2,500 of
                                cases                           24,000 JOBS
                                                                participants in
                                                                state with
                                                                40,000 AFDC
                                                                cases

Highlighted     Finding         Training        Work            Community work
features        employment      closely linked  supplementatio  experience
                quickly; job    to employers'   n
                development     needs

Comparison or   Yes             Yes             No              Available on
control group                                                   similar programs
evaluation                                                      only
available
--------------------------------------------------------------------------------
\a The JOBS program in California is called Greater Avenues for
Independence (GAIN). 


   RIVERSIDE COUNTY PROGRAM
---------------------------------------------------------- Chapter 2:1

The Riverside County JOBS program stresses that its purpose is to
place participants in jobs quickly.  Researchers believe that this
strong employment message may have been one of the key factors in
producing results.  Using an experimental design to evaluate JOBS
programs in six California counties, researchers\26 found that the
Riverside County program increased the earnings of single AFDC
parents by 49 percent and decreased welfare costs by 15 percent over
3 years.  Results in the other five counties were about one-half that
level.  As shown in figure 2.1, the Riverside County program produced
greater net gains than the other counties for both welfare recipients
and government budgets, saving almost $3 for every $1 spent by the
federal, state, and local governments.  Moreover, long-term AFDC
recipients, those with little education, and those more job-ready
have benefited under Riverside's approach. 

   Figure 2.1:  Benefit-Costs
   Results of Welfare- to-Work
   Program in Selected California
   Counties (1988-93)

   (See figure in printed
   edition.)

\a For AFDC single-parent participants enrolled in the experimental
group, this includes per person effects on their earnings and fringe
benefits, offset by any reductions in AFDC and other transfer program
payments. 

\b The potential gains to government budgets per participant in the
experimental group include reduced AFDC and other transfer payments,
reductions in transfer program administrative costs, and increased
taxes paid by those in the experimental group.  The potential costs
include net expenditures for GAIN services and services provided by
other agencies. 

Source:  James Riccio, and others, GAIN:  Benefits, Costs, and
Three-Year Impacts of a Welfare-to-Work Program, Manpower
Demonstration Research Corporation (New York:  1994), p.  ES-9. 

Researchers who studied the six California counties believe that
Riverside County's greater positive impacts may be due to a
combination of program features.\27 For example, the program had
sufficient resources to make efforts to enroll all the AFDC
recipients deemed mandatory for JOBS.  In addition, it used the
threat of reduced AFDC benefits for uncooperative participants to
secure their participation in JOBS.\28

In contrast with the other counties evaluated, Riverside also
articulated a simple goal:  participants are there to get a job and
leave welfare as soon as possible.  They are, therefore, encouraged
to take any job offered, including low-wage jobs, part-time jobs, or
jobs without benefits.  To help participants get jobs, five full-time
job developers provide direct access to employers and support the
five JOBS offices that served about 2,000 active JOBS participants
each month in 1993.  The Riverside program also uses placement
standards for its JOBS workers; case managers are expected to place
at least 12 participants in employment each month.\29

Stressing the importance of job search along with education in the
basic skills of reading, writing, and math also appears to benefit
Riverside.  In Riverside and the other counties, new participants
whose test results indicate that they need basic education have the
option of entering the classroom immediately or attending 3 weeks of
job search.\30 \31 However, Riverside's orientation results in
proportionately more of its participants being in job search than is
the case in most other California counties studied.  Also, Riverside
encourages those participants in education and training to find jobs
quickly.  Staff closely monitor these participants and expect those
not making progress to look for work. 

Riverside's emphasis on short-term job search along with longer-term
education may also account for its relatively low average cost of
$3,000 per participant in 1993 dollars,\32 compared with other
California counties studied.  Consistent with its emphasis on moving
participants quickly into the work force, Riverside makes less use of
basic, vocational, and postsecondary education than some other
counties.  For example, in Alameda County, which used education
extensively, the 1993 average participant cost was $6,600. 

While the Riverside County results indicate that an emphasis on job
placement, among other factors, is important, questions remain about
what works best to help welfare recipients get jobs and earn enough
to support their families.  HHS has contracted with researchers to
conduct experimental design studies to provide additional information
on the cost effectiveness of higher-cost education and training
programs compared with lower-cost programs that emphasize quick entry
into jobs.\33

While the Riverside program produced greater earnings increases and
welfare savings than in the other counties, about 40 percent of its
participants were still on AFDC 3 years after the study began, and
many of those who did leave AFDC remained in poverty and possibly at
risk of returning to welfare.  Some of those that left AFDC may also
have continued to receive other forms of public assistance, including
Food Stamps and housing subsidies. 

The researchers also noted that it was not clear that Riverside's
program could be replicated or, if replicated, could produce similar
results in other localities nationwide--for example, in inner cities
where AFDC recipients may face greater barriers.  In addition, while
they concluded that the Riverside County results appear not to be
fully explained by its local labor market conditions, they cautioned
that similar results may not be possible in areas with very poor
economic conditions, such as rural areas with high unemployment
rates.\34


--------------------
\26 See James Riccio, and others, GAIN:  Benefits, Costs, and
Three-Year Impacts of a Welfare-to-Work Program, Manpower
Demonstration Research Corporation (New York:  1994). 

\27 The evaluation of the six California programs used experimental
and control groups to identify impacts on earnings and welfare
receipt.  In addition, detailed and standardized implementation data
were collected among the counties to shed light on the best approach
to moving participants off welfare. 

\28 Riverside initiated sanctions against 11 percent of its single-
parent participants because they failed to show up for scheduled JOBS
activities without good cause, and reduced the welfare grants of 6
percent of participants for specified periods of time. 

\29 Researchers noted that these placement standards do not lead to
creaming--working only with the more employable--because the program
covers most of those required to participate and no up-front
screening is performed to screen out those considered harder to
place.  They also noted that these placement standards apply to the
caseworkers that work with a mixed group, those in need of and not in
need of basic education.  The caseworkers who work solely with those
in need of basic education have lower placement standards. 

\30 The California JOBS program, called GAIN, uses a program model
that requires basic education for recipients that test below certain
levels.  This is not required by federal JOBS rules. 

\31 Riverside administrators estimate that participants attend basic
education classes an average of 8 months to reach eighth grade
reading and math levels or take 3 months to pass the high school
equivalency examination. 

\32 Researchers determined this amount was spent on those in its
experimental group over a 5-year period from JOBS and other funding
sources.  The impact analysis was based on 3 years of follow-up data. 

\33 For more information on the ongoing evaluation, see Gayle
Hamilton and Thomas Brock, The JOBS Evaluation. 

\34 For a discussion of the limited evidence available on the
influence of the local economy on a welfare-to-work program's impact,
see Judith Gueron and Edward Pauly, From Welfare to Work, p.  186. 


   SAN JOSE'S CENTER FOR
   EMPLOYMENT TRAINING
---------------------------------------------------------- Chapter 2:2

The Center for Employment Training (CET), a nonprofit organization
founded in 1967 and based in San Jose, California,\35 represents
another approach to promoting employment that has demonstrated
positive results.  CET contracts with job training and local welfare
programs to provide job skills training, combined with remedial basic
education, when needed.  Using an experimental design, researchers
found that this program increased employment and earnings for
minority female single parents on or at risk of becoming dependent on
AFDC who volunteered for training.\36 The research study noted that
at the end of 12 months, 46 percent of CET participants were working
compared with 36 percent of a control group and participants earned
47 percent more on average than the control group.  These results
were also greater than those for other sites in the study. 

To help its participants get higher-wage jobs with a potential for
upward mobility, CET offers job skills training in a range of
occupations for which employers have demonstrated consistent demand. 
About 28 courses are offered, including child care provider,
automated office skills, home health aide, commercial food service,
and electronic assembly.  Remedial education is integrated into the
job skills training curriculum for participants who have basic skill
deficiencies, rather than being offered separately.  Researchers who
have studied the CET program believe that its strong focus on
employment and integrated training design are important features. 

The employment focus is evident in several CET activities.  CET's
full-time job developers make contact with employers in the community
and meet with participants who are nearing completion of training to
help them find appropriate work.  The job developers are assisted in
their placement efforts by CET's vocational instructors, who maintain
close contacts with local employers.  CET also has an industrial
advisory board, composed of employers, that meets monthly to provide
advice on the types of training equipment to be used and other issues
to ensure that the training offered meets the needs of employers. 
Board members also conduct mock job interviews with participants. 

One employer we spoke with, manager of a local sheet metal
fabrication company, emphasized that his company relies heavily on
CET graduates.  He believes that this saves him advertising and other
hiring costs and guarantees him well-prepared workers.  At the time
of our visit, he was planning to open a company cafeteria to be
staffed with CET graduates. 

Another key feature, integrated training, provides basic education in
a practical context.  Participants lacking basic educational skills
are entered into job skills training immediately to help maintain
their motivation and focus on work.  Because basic education is
provided within the skills training class itself, participants appear
more likely to accept the remedial help and to succeed. 

Participants attend classes during the normal work week in a setting
designed to simulate the workplace, using the tools of their trade
under the guidance of instructors with recent industry experience. 
Individualized instruction allows new participants to enter class on
the first day of any week of the year, to proceed at their own pace,
and to leave as soon as they have demonstrated the necessary
competencies.  Training courses average 6 months in length and cost
about $6,000 to $7,000 per participant.\37


--------------------
\35 Thirty-six additional programs modeled on CET are operated or
planned in a total of 12 states. 

\36 See J.  Burghardt and A.  Gordon, The Minority Female Single
Parent Demonstration:  More Jobs and Higher Pay, Rockefeller
Foundation (New York:  1990).  Researchers found significant and
persistent gains in employment and earnings of the CET women studied,
although few were able to reduce their dependence on AFDC. 

\37 Welfare recipients may have their tuition paid by a variety of
sources, including JOBS, JTPA, or the federal Pell grant program,
which provides economically disadvantaged individuals with grants to
further their education. 


   NEW YORK CITY'S AMERICA WORKS
---------------------------------------------------------- Chapter 2:3

Another example of a work-focused program is seen in New York City. 
There, the welfare agency, as part of its work-supplementation
program, contracts with a private for-profit firm called America
Works.\38 America Works quickly prepares JOBS participants for
employment, places them in jobs, and provides counseling and support
to ease their transition to work.  Staff and resources are devoted to
working with employers and supporting clients after job placement to
help alleviate any personal problems that may arise and threaten
their ability to continue to work. 

America Works emphasizes the development of good work habits and
skills required for entry-level jobs during the short training period
it provides participants.  Specifically, participants are urged to
demonstrate punctuality, reliability, appropriate professional dress
and demeanor, a constructive and cooperative attitude, and an ability
to get along with others in a work environment.  Participants attend
a week-long pre-employment class and 6 weeks of business laboratory
where they use self-paced computer-assisted office skills programs. 
Tardiness and absences may result in suspension from the program. 

Participants who complete the business laboratory are placed by the
firm's job developers with private employers for 4 months of
supported work, during which time they are on the payroll of America
Works.  The America Works payroll in New York City is supported by
AFDC grant funds and funds from employers.  Upon placement,
participants are provided a support system, whereby America Works
staff help participants with personal problems, such as creditor or
landlord disputes, that interfere with their ability to work. 
America Works staff believe that their support system for
participants who have been newly placed in jobs is key to keeping
many of their participants employed. 

According to data compiled by the New York State welfare agency,
about 65 percent of participants in supported work are ultimately
hired by the private employers with whom they have been placed. 
America Works receives about $5,300 from the state's welfare agency
when an AFDC recipient enrolled in America Works remains employed and
off AFDC for at least 7 months.  Unlike the Riverside and CET
programs, the outcomes of the America Works program have not been
compared with a comparison or control group to determine whether the
effects were due to the program.  Some of the America Works
participants might have found jobs on their own, especially because
many of them were motivated volunteers. 

While the program's design screens out those not motivated, the
program does work with many long-term welfare recipients with low
levels of education.  The typical participant is an adult female head
of household on AFDC an average of 5 years.  Also, the typical
participant in America Works has volunteered for the program, has a
sporadic history of minimum-wage jobs, and can read and write well
enough to complete a brief application.  Applicants who need remedial
basic education or English language training are referred to other
community providers.  About one-half of the participants have not
completed high school. 

America Works officials believe that reaching out to employers and
responding to their needs is a prime program goal.  They noted that
employers who take on America Works participants save on placement
agency fees as well as costs of advertising for and screening job
applicants.  In addition, they obtain workers at reduced wage and
benefit costs initially, and with lower turnover and related costs. 
America Works guarantees that employers will be satisfied with
participants placed with them or replacements will be found.  About
60 percent of the jobs that America Works staff develop are the
result of repeat business with satisfied employers. 


--------------------
\38 Programs based on the America Works model operate at another site
in New York state, one site in Indiana, and another site in
Connecticut. 


   ATHENS, OHIO, AND WEST VIRGINIA
---------------------------------------------------------- Chapter 2:4

JOBS programs in Athens, Ohio, and West Virginia reveal a different
kind of work focus, typified by placing participants in community
work-
experience positions with public and nonprofit agencies.  Welfare
officials at the sites we visited indicated that having AFDC
recipients perform community service can benefit their communities,
in addition to developing participants' work habits and providing
work experience that may lead to paid employment. 

The JOBS program in Athens County, Ohio, uses work-experience
positions to increase the confidence and competency of participants,
and in some cases these positions lead to permanent employment.  The
county's welfare agency is the largest utilizer of work-experience
participants, many of whom are subsequently transferred to the
county's payroll and leave welfare.\39 One office unit within the
welfare agency is staffed primarily by work- experience participants,
and an estimated three-fourths of the welfare agency's personnel
consist of former welfare recipients. 

West Virginia, where unemployment rates are among the highest in the
nation, uses community work experience extensively to develop and
maintain work habits among its JOBS participants.  This involves work
for various public or nonprofit organizations.  Since the 1980s, West
Virginia's welfare-to-work program has promoted the idea that AFDC
recipients should contribute to their communities in exchange for
their benefits, and work for such organizations has been used to
promote work among AFDC recipients, especially men. 

The state has made greater use of community work experience than most
other states, with about 2,500 AFDC recipients enrolled in June 1994,
mostly at government agencies but also at nonprofit agencies. 
Participants often work for an average of 62 hours a month, putting
in full 40-hour weeks for some part of the month or part-time hours
throughout the month.  Single parents with young school-age children,
for example, may work during the 6 hours of a normal school day and
care for their children at home the remainder of the day, thus saving
on child care expenses. 

West Virginia administrators we spoke with noted that much time,
effort, and resources must be devoted to operate a work-experience
program.  Major work-experience program expenses involve intensive
use of JOBS staff to arrange for jobs with employers, screen and
match participants to available jobs, and provide follow-up support. 
JOBS case managers check monthly timesheets and ask to be called if
problems arise at the workplace.  They rarely visit worksites,
however, because they average caseloads of 300 to 400 participants. 

Based on experimental design studies of the use of work experience in
several sites in the 1980s, including some in West Virginia,
researchers have concluded that unpaid work experience alone does not
increase paid employment, earnings, or welfare savings.  However,
they also found that these programs could produce benefits for
taxpayers through the work performed by welfare recipients.  In
addition, program administrators and welfare recipients involved
generally thought that they had performed meaningful work, although
the participants said that they would have preferred to work in paid
positions.  Based on their review, the researchers estimate that the
annual cost of a work-experience position in 1993 dollars would range
from $2,000 to $4,000, excluding the AFDC benefit and child care
costs.\40


--------------------
\39 Under a federal waiver from AFDC and JOBS rules, participants may
extend the hours that they work and receive payment from the employer
for the added hours without a reduction in their grants in cases
where the employer demonstrates a commitment to hire. 

\40 See Thomas Brock, David Butler, and David Long, Unpaid Work
Experience for Welfare Recipients:  Findings and Lessons from MDRC
Research, Manpower Demonstration Research Corporation (New York: 
1993). 


MOST JOBS PROGRAMS HAVE A WEAK
EMPLOYMENT FOCUS
============================================================ Chapter 3

While some county and local organizations have forged links with
employers to promote work for welfare recipients, these programs are
more the exception than the rule across the nation.  A majority of
county JOBS programs do not work closely with employers to help their
participants find work.  Administrators and researchers cited many
factors that hinder efforts to find or create employment for welfare
recipients, including insufficient staff and resources and poor labor
market conditions.  In addition, we found that the federal JOBS
participation requirements emphasizing the enrollment of eligible
persons into JOBS programs without an emphasis on the graduation of
enrollees into employment provide programs little incentive to
redirect their resources to job-placement efforts. 


   EMPLOYMENT EMPHASIS LIMITED
   AMONG PROGRAMS
---------------------------------------------------------- Chapter 3:1

Most programs do not fully use the tools available to help move
participants quickly into work.  This is demonstrated by the limited
emphasis on job development, work incentives, and work activities,
including subsidized employment or work experience. 


      JOB DEVELOPMENT IS
      UNDERUTILIZED
-------------------------------------------------------- Chapter 3:1.1

Although job development is a potentially important tool for moving
JOBS participants into employment, about one-half of the nation's
county JOBS administrators believe that they are not doing enough job
development to help JOBS participants find work.  In addition to
preparing AFDC recipients for employment through education and
training, JOBS programs are required to engage in job development to
help participants secure jobs.  Program officials may also work with
employers to identify the types of education and training needed for
participants to meet employers' needs.  These job-development
activities can play an important role in making JOBS programs more
responsive to their local labor markets. 

While almost all county JOBS programs perform some job-development
activities, in most, their job-development resources are limited.  We
found that JOBS programs rely on a variety of local agencies and
organizations, such as JTPA, the Employment Service, and education
providers, to perform job-development activities for JOBS
participants.  While other organizations are involved in helping JOBS
participants find work, in most counties, the welfare agency itself
takes the lead in job development.  However, about one-third of the
nation's programs have no full- or part-time staff dedicated to
job-development activities.  And while caseworkers may also perform
job-development activities, we found that they devote little time to
working with employers.  More than three-fourths of all JOBS
administrators report that caseworkers devote 20 percent or less of
their time to job development. 

In many programs, the extent of job development performed on behalf
of JOBS participants is limited and may not meet the needs of the
job-ready looking for work.  For example, about 60 percent of the
nation's JOBS programs or their contractors arranged job interviews
for or marketed to employers only some or few of their job-ready
participants.  Moreover, about 46 percent or more cited that the
program or its contractors worked with each of the following only
sometimes or rarely:  public employers, private-sector employers, the
Chamber of Commerce, or other employer associations. 

Local administrators themselves also believe that job development is
underutilized in JOBS programs.  A majority of administrators believe
that they did not conduct enough job-development and job-placement
activities to meet the needs of their JOBS participants, as
illustrated in figure 3.1.  Furthermore, a 1994 study of JOBS
implementation in 30 localities in 10 states also noted that job-
development and job-placement activities are underutilized in JOBS
programs.\41

   Figure 3.1:  Administrators'
   Opinions on Extent of
   Job-Development Activities
   (Mid-1994)

   (See figure in printed
   edition.)


--------------------
\41 Irene Lurie and Jan L.  Hagen, Implementing JOBS:  Progress and
Promise, The Nelson A.  Rockefeller Institute of Government, State
University of New York (Albany, New York:  1994), p.  246. 


      WORK INCENTIVES NOT
      EMPHASIZED
-------------------------------------------------------- Chapter 3:1.2

Many JOBS programs nationwide do not make all participants aware of
some important incentives to seek employment.  To encourage work, the
AFDC program provides some assistance to recipients who become
employed by temporarily disregarding part of their earnings,
including some of those expended for child care, in calculating their
AFDC benefits.  These income and child care disregards allow AFDC
recipients who go to work to avoid the cutback in benefits that would
ordinarily result from an increase in earnings.  In addition, to
further ease the transition to employment, AFDC recipients who earn
enough to leave the welfare rolls are eligible for 1 year of child
care subsidies if needed and continued Medicaid coverage. 

Other assistance may be available after AFDC recipients leave the
welfare rolls.  When the 1 year of transitional Medicaid coverage is
exhausted, the children of AFDC recipients may still be covered due
to recent changes in Medicaid coverage for all children in families
below the poverty line.\42 And the recently expanded Earned Income
Tax Credit (EITC) will increase some low-wage workers incomes by up
to 40 percent.  These federal supports can increase the
attractiveness of low-wage work. 

However, many JOBS programs do not inform all their participants of
the work incentives that may be available to them.  Based on our
survey, from 67 to 84 percent of county JOBS programs inform all or
almost all of their participants about each of the following:  the
availability of transitional child care, transitional Medicaid, AFDC
income disregards, and child care disregards.  However, only about
one-half of the nation's JOBS programs inform all or almost all their
participants of the EITC.  While we identified about 18 percent of
the programs that worked with all or almost all their participants to
develop a sample budget demonstrating the benefits available to them
when working, about 60 percent of the nation's JOBS programs reported
that they do so for one-half or fewer of their participants. 

These findings are consistent with other studies showing that those
on welfare, as well as welfare and JOBS caseworkers, may not be aware
of or understand work incentives.  One study of a sample of 30 women
in Chicago concluded that the EITC may not provide an incentive to
work because few recipients have a clear understanding of how it
operates.\43 Another study of welfare administrators found that many
did not know that Medicaid coverage was available for certain
children in families with incomes up to or, in some cases, beyond the
federal poverty line.\44


--------------------
\42 States are required to provide Medicaid coverage to all children
under 19 years old who were born after September 30, 1993, and whose
family income is below 100 percent of the federal poverty line.  In
addition, states must cover pregnant women and children under 6 years
old with family incomes below 133 percent of the poverty line. 

\43 Lynn M.  Olson, "The Earned Income Tax Credit:  Policy
Implications of Street Level Experience," paper presented at Annual
Research Conference of the Association for Public Policy Analysis and
Management (Chicago:  1994). 

\44 See Vicki C.  Grant, Genny G.  McKenzie, and Sarah C.  Shuptrine,
A Study of the Relationship of Health Coverage to Welfare Dependency,
Southern Institute on Children and Families (Columbia, South
Carolina:  1994). 


      LIMITED USE OF WORK
      ACTIVITIES
-------------------------------------------------------- Chapter 3:1.3

Almost all JOBS programs encourage participants to engage in job
search activities at some point in their enrollment in JOBS,\45 but
many job-ready participants do not become employed for a variety of
reasons.  For JOBS participants who cannot find regular employment,
local JOBS programs have the option of using cash wage subsidies to
encourage employers to hire them into on-the-job training or work-
supplementation programs.  Another option is to place participants in
work-experience programs.  For example, as discussed in chapter 2,
West Virginia has used its community work-experience program to
promote work among its welfare recipients when jobs were not
available.  Yet the use of work activities is limited, even though
about 70 percent of the administrators reported that one-half or
fewer of their job-ready participants became employed during their
most recent program year.  The distribution of counties according to
their placement rates is shown in figure 3.2. 

   Figure 3.2:  Distribution of
   Counties According to Placement
   Rates for Job-Ready
   Participants (Program Year 1993
   or 1994)

   (See figure in printed
   edition.)

The limited extent of work activities is seen in the following
numbers:  nationwide in mid-1994, of about 586,600 JOBS participants
each month, about 59,000 were in work-experience programs, 3,000 were
in on-the-job training, and 1,000 were in work-supplementation
programs.  As shown in figure 3.3, these work activities were little
used compared with other JOBS activities.  Moreover, more than 80
percent of the nation's counties have no experience operating
work-supplementation programs and almost 50 percent have no
experience in on-the-job training. 

   Figure 3.3:  JOBS Participants
   by Activity (Mid-1994)

   (See figure in printed
   edition.)

Note:  Estimates are slightly understated due to missing data for 9
percent of the sampled counties.  For sampling errors, see appendix
II.
Participants may be enrolled in more than one activity at a time. 

This demonstrates that counties will face a major challenge in
supporting the work programs called for in some welfare reform
proposals.  For example, H.R.  4 requires states to provide work
activities for an increasing percentage of those receiving cash
assistance or face penalties of up to 5 percent of the state's block
grant.  In 1996, states would have to involve 10 percent of all
families in work activities, with the requirement rising to 50
percent by 2003.\46 And the administration's proposal before the
103rd Congress called for those young mothers who do not find
unsubsidized employment after 24 months of receiving AFDC to be
placed in subsidized minimum-wage jobs.  The House bill and the
administration's proposal place a much greater emphasis on work
activities than current law. 

Under both of these proposals, welfare agencies will need to work
with many welfare recipients who cannot find jobs on their own. 
Attention will have to be paid to preparing these recipients for the
workplace, because administrators we spoke with emphasized the
importance of screening and selecting able and motivated participants
to place with employers to maintain employer interest in
participating in the programs.  This is consistent with our survey
results showing that in most counties the typical JOBS participant
enrolled in on-the-job training or work supplementation has at least
1 year of previous work experience and high levels of motivation. 
Also, in most counties, participants in these work activities tended
to be more educated than JOBS participants in general. 

While work activities are little used in JOBS, most administrators
believe that they are effective tools that warrant expansion.  Of the
relatively small number of JOBS administrators currently using work
supplementation, 70 percent rated it moderately or highly effective
in moving AFDC recipients off welfare and 83 percent wanted to expand
their use of it.\47 Of those using on-the-job training, 72 percent\48
thought it at least moderately effective in moving individuals off
welfare and 88 percent expressed interest in expanding its use. 
Almost all counties used work experience, with 76 percent rating it
as effective and 84 percent wishing to expand its use. 

In sites we visited, JOBS participants had been placed with a range
of employers and other community organizations.  They performed
community service work with a county planning office, the Indian
Health Service, and a community food bank.  In addition, through the
work-supplementation program, participants had found jobs at a car
dealership, a large health care provider, and a small doctor's
office.  In one site, the work-supplementation program helped
refugees receiving AFDC gain employment at worksites where they could
improve their English-language skills.  According to the program
supervisor, some of the refugees had been in English as a Second
Language classes for several years but had not progressed to
employment. 

Of those program administrators not currently using on-the-job
training, about 32 percent believed it to be moderately or highly
effective in moving recipients off AFDC\49 and about three-fourths
supported expansion.  At least one-half of the administrators without
work-supplementation programs also wanted to develop or expand these
programs, although they were less sure about the effectiveness of
such programs. 

Evaluations of on-the-job training and work-supplementation programs
have shown positive results in terms of increased employment and
earnings for welfare recipients, but did not conclude that the
programs produced welfare savings.\50 As discussed in chapter 2,
evaluations of work-experience programs have shown that they offer
productive work for participants and benefits to taxpayers, but do
not generally produce increased earnings, employment rates, or
welfare savings.\51


--------------------
\45 About 37 percent of the nation's JOBS programs encourage JOBS
participants to look for work before placement in education,
training, or work-related activities, with the remainder reserving
job search activities until after education and training. 

\46 Cash assistance recipients must participate in one of the
following to count towards the participation rate:  unsubsidized
employment, subsidized private or public-sector employment (including
on-the-job training), work experience, job search, job-readiness,
education directly related to employment for those under 20 years old
and without a high school education or its equivalent, job skills
training directly related to employment, or high school for those
under 20 years old. 

\47 As a result of the limited percentage of JOBS programs using work
supplementation (8 percent), the sampling errors for these
percentages are large--plus or minus 21 and 18 percentage points,
respectively. 

\48 The sampling error for this percentage is plus or minus 10
percentage points. 

\49 Of those not currently using on-the-job training, 47 percent
believed they had no basis to judge its effectiveness.  However, of
those who did rate its effectiveness, about 61 percent believed it to
be moderately or highly effective. 

\50 See Howard S.  Bloom, and others, The National JTPA
Study--Overview:  Impacts, Benefits, and Costs of Title II-A, Abt
Associates Inc.  (Bethesda, Maryland:  1994); Judith Gueron and
Edward Pauly, From Welfare to Work; and What's Working (and What's
Not):  A Summary of Research on the Economic Impacts of Employment
and Training Programs, Office of the Chief Economist, U.S. 
Department of Labor (Washington, D.C.:  1995). 

\51 See Thomas Brock, David Butler, and David Long, Unpaid Work
Experience. 


   INTERNAL AND EXTERNAL FACTORS
   HINDER STRONG EMPLOYMENT FOCUS
---------------------------------------------------------- Chapter 3:2

While JOBS administrators acknowledged that they did not work enough
with employers to help participants find jobs, they identified
several administrative and programmatic factors that hindered their
efforts.  Further, administrators and researchers identified certain
labor market conditions that hinder efforts to place AFDC recipients
in jobs. 


      INSUFFICIENT STAFF HINDER
      LINKS WITH EMPLOYERS
-------------------------------------------------------- Chapter 3:2.1

Most administrators reported that insufficient staff hindered their
efforts to work with employers to place JOBS participants in
unsubsidized jobs or work activities.  Local program administrators,
researchers, and HHS officials have noted that working with employers
to find job openings or to create and maintain work-activity
positions requires a lot of time and effort on the part of JOBS
workers.  For example, to operate work- supplementation programs,
AFDC grant dollars must be diverted to employers to subsidize wages. 
Many administrators believe that it is difficult to develop and
administer a tracking system to operate such a program.  In addition,
staff must market their programs to employers and sometimes visit
worksites to maintain contact or monitor operations.  Economies may
be achieved if many participants are placed at a single worksite, but
we found that generally only one or two participants are placed with
each employer. 

Administrators believe that they need more staff to work with
employers because current JOBS staff and resources are mainly devoted
to participant intake and management of often heavy caseloads.\52
According to HHS, JOBS caseloads range from 30 to 400 participants
per worker.  Administrators we met explained that expansion of
job-development and work activities would necessitate shifting
current staff from intake and case management functions.  They also
noted that hiring additional staff is not an option where budgets are
constrained. 


--------------------
\52 Participant intake involves orientation, appraisal, and
assessment; case management involves assigning participants to
activities, arranging supportive services, and monitoring their
participation and progress. 


      RESOURCE CONSTRAINTS AFFECT
      PROGRAM CHOICES
-------------------------------------------------------- Chapter 3:2.2

While in some cases resource constraints may limit the number of JOBS
staff, they may also affect administrators' and caseworkers'
decisions about the activities in which they enroll participants. 
The study of JOBS programs in 10 states referred to earlier noted
that the availability of education, training, and employment-related
activities tends to drive the placement of participants.\53 For
example, as a result of resource constraints, programs would often
place participants in activities that were readily available or free
of charge rather than create or purchase services that were deemed
needed by participants. 

We also found that funding constraints limited the use of on-the-job
training.  About one-half of the JOBS administrators cited
insufficient funds and one-third cited the high costs of on-the-job
training compared with other JOBS activities as a major or moderate
hindrance to its expansion.\54 On-the-job training is sometimes more
costly to a JOBS program than other activities because many of the
educational or other activities in which participants are placed are
funded by other providers or programs and do not require expenditures
of JOBS funds.  For example, a JOBS program may not pay for adult
basic education or college courses funded by federal, state, or
county providers. 

Funding constraints also hinder the use of work supplementation, even
though this form of employer subsidy is funded by AFDC grants instead
of JOBS funds.  An official in Texas told us that in states with low
AFDC grants, the amount of money that can be diverted to the employer
is not sufficient for a wage subsidy.  For example, the average AFDC
grant in Texas equals $159 a month, providing few dollars to
subsidize wages.\55 In 1994, 31 state welfare agencies decided not to
include work supplementation in the state JOBS plans they must submit
to HHS for approval.\56 As a result, the local programs in these
states were barred from operating work- supplementation programs. 


--------------------
\53 Irene Lurie and Jan Hagen, Implementing JOBS:  Progress and
Promise, p.  179. 

\54 When programs did place participants in on-the-job training, it
was often done through the community JTPA provider, thereby requiring
no use of JOBS funds to subsidize the employers.  However, like JOBS,
JTPA on-the-job training slots and funding are limited. 

\55 Some states have obtained or applied for waivers from AFDC and
JOBS policy to allow them to include the cash value of an AFDC
recipient's Food Stamp benefit with the AFDC benefit to be diverted
to an employer as a wage subsidy.  This increases the dollar amount
available to subsidize an individual's wages. 

\56 States are required biennially to submit a JOBS plan to HHS. 


      FEDERAL PERFORMANCE
      MEASUREMENT SYSTEM DOES NOT
      PROMOTE EMPLOYMENT
-------------------------------------------------------- Chapter 3:2.3

The current federal JOBS participation and targeting requirements
provide little incentive for states to redirect scarce resources to
increase their focus on moving AFDC recipients into employment.  The
JOBS performance measurement system is process-oriented, based on the
numbers and types of participants enrolled in activities, and does
not include outcome measures, such as the portion of participants who
become employed and leave welfare.  While the participation
requirements have played an important role in encouraging states to
serve more participants, including the hard-to-serve, the ultimate
goal of JOBS is to increase employment and reduce welfare dependence. 
Yet states are not required by HHS to report the total number of JOBS
participants who find jobs each year and are not held accountable for
the number of JOBS participants who become employed.\57

Some program administrators and researchers have noted that programs
can meet federal participation requirements by placing participants
in readily available JOBS activities more easily and with less cost
to their programs than finding them unsubsidized jobs or creating
subsidized employment.  Because program administrators can meet
federal requirements without redirecting scarce resources to focus
more on employment, they have little incentive to do so.  JOBS
programs may, therefore, emphasize getting clients into program
activities without also focusing on establishing links with employers
to realize the ultimate goal of employment. 

For example, at one site we visited, a woman had successfully
completed several different training programs.  Under the current
performance system, this individual helps the program meet the
federal requirements to receive its full share of federal funding. 
Yet she remained unemployed and on AFDC. 


--------------------
\57 JOBS legislation directed HHS to recommend to the Congress by
October 1993 JOBS performance standards that included outcome
measures, such as increased earnings and reduced welfare dependence. 
While HHS has not made recommendations, it has reported to the
Congress on related issues and is pursuing changes to the JOBS
performance measurement system that will include outcomes.  For more
information, see Welfare to Work:  Measuring Outcomes for JOBS
Participants (GAO/HEHS-95-86, Apr.  17, 1995). 


      LABOR MARKET CONDITIONS
      LIMIT EFFORTS
-------------------------------------------------------- Chapter 3:2.4

Labor market realities also pose a range of problems for JOBS
administrators as they attempt to move AFDC recipients into the
workplace.  Several factors are important in this regard. 


         LACK OF JOBS
------------------------------------------------------ Chapter 3:2.4.1

Administrators and research studies cite high unemployment and low
job growth as hindering programs' efforts to get jobs for
participants.  Nearly three-fourths of local JOBS administrators
identify current labor market conditions, which are outside their
control, as a hindrance to their job-development efforts.  Many
counties operate JOBS in areas of high unemployment or negligible job
growth.  For example, in 1993, unemployment rates reached 8 percent
or more in 30 percent of the nation's counties; job growth was 1.5
percent or less in one-half the nation's counties and negative in
about one-third of the counties. 

While some research has shown that the outlook for job growth
nationwide over the next few years is encouraging,\58

in specific locations the number of job openings may not meet local
needs.  For example, a May 1993 survey of Milwaukee area employers
identified about 12,000 full-time job openings, which represented
only 20 percent of the jobs needed for the estimated 63,000 welfare
recipients and unemployed persons seeking or expected to work.  When
part-time jobs were included, the number of available jobs
represented 35 percent of the total jobs needed.\59

Likewise, JOBS officials in Silicon Valley in California, where many
once-booming high-tech computer companies are located, and other
areas in California believe that their JOBS participants and staff
acting on their behalf operate at a distinct disadvantage because of
the increase in competition for positions in general.  They noted
that they must operate their programs in areas where employers are
often faced with a surplus of job applicants, especially for
relatively unskilled, entry-level positions. 


--------------------
\58 Rebecca Blank, "Outlook For the U.S.  Labor Market and Prospects
For Low-Wage Entry Jobs," Northwestern University, paper prepared for
Urban Institute Conference (Washington, D.C.:  1994), p.  15. 

\59 See John Pawasarat, "Survey of Job Openings in the Milwaukee
Metropolitan Area:  Week of May 24, 1993," Employment and Training
Institute and Social Science Research Facility, University of
Wisconsin-Milwaukee (Milwaukee:  1993). 


         LACK OF EMPLOYER INTEREST
------------------------------------------------------ Chapter 3:2.4.2

Administrators we surveyed and spoke with emphasized that lack of
employer interest also hindered the expansion of work activities. 
Administrators cited as one contributing factor a federal
displacement restriction.  Under work-supplementation and
work-experience programs, participants may only be placed in
positions newly created by employers--not positions that become
vacant due to turnover.  This prohibition is intended to protect
workers from being displaced through layoffs and replaced by
federally subsidized JOBS participants.  About three-fourths of the
administrators operating work-supplementation programs reported that
this restriction hindered expansion of their programs\60 and about 46
percent of all administrators said that they probably or definitely
would like to use work supplementation for existing positions also. 

In addition, work-experience positions are restricted to sponsors who
serve a public purpose, another restriction that about 72 percent of
administrators would like to see changed, allowing them both more as
well as a greater variety of employers with which to place
participants to help them gain work experience.  Like work-
supplementation, work-experience positions are also subject to
displacement restrictions.  While most administrators did not believe
that the displacement restriction was currently a factor hindering
expansion, about one-half supported placing work-experience
participants in existing positions.  Administrators we spoke with
thought that other workers and individuals could be protected without
restricting work programs to new positions only. 

Local administrators also cited other reasons.  For example, for
on-the-job training, the JOBS program and employers must generally
enter into contracts covering the employment of participants,
maintain timekeeping and payroll records subject to audit, develop
individual training plans, establish qualitative measures of success,
and assess the progress of participants in acquiring jobs skills. 
Some employers may feel that the wage subsidy they receive--up to
one-half of participants' wages when training is completed--does not
adequately compensate them for any extra work they must do.  In a
work-experience program, while a participating employer gets an
unpaid worker, the employer is not compensated for any supervision
costs involved.  Administrators we met cautioned that the number of
available supervisors among employers places an upper limit on the
expansion of work experience. 

Administrators also cited employer concerns about welfare recipients
being unprepared for work.  Employers' perceptions may be skewed by
unfavorable stereotypes or unsuccessful prior experiences.  One study
of a welfare-to-work program in an inner-city neighborhood noted that
many of the participants who found jobs had problems keeping them for
various reasons, including chronic lateness and misunderstandings
with supervisors.\61 To overcome these perceptions and problems,
program administrators told us that they often select their most
capable participants for work activities. 


--------------------
\60 The sampling error for this percentage is plus or minus 21
percentage points. 

\61 See Toby Herr and Robert Halpern with Aimee Conrad, Changing What
Counts:  Re-Thinking the Journey Out of Welfare, Center for Urban
Affairs and Policy Research, Northwestern University (Chicago: 
1991). 


         LACK OF JOBS THAT SUPPORT
         FAMILIES
------------------------------------------------------ Chapter 3:2.4.3

While the lack of jobs is a problem in many areas, the low-wage work
that is available to many AFDC recipients discourages their movement
off AFDC.  Our work in 1991 demonstrated that many single mothers
will remain near or below the poverty line even if they work at
full-time jobs.\62 More recently, we found that in 1993 the typical
single mother with a low-wage job had more income than a comparable
mother and family on AFDC, but was nevertheless still in poverty.\63

Moreover, a low-wage worker may incur significant job-related costs,
such as child care, which could make her family worse off financially
than some AFDC families.\64 In addition, employment or increased
earnings may affect her receipt of other forms of assistance.  For
example, the previously cited survey of several thousand AFDC
recipients found that 60 percent of the respondents in Atlanta lived
in public housing projects or other subsidized housing.  As a result,
their incentive to find jobs may be affected because increased
earnings may cause them to incur significantly increased housing
costs.\65

The belief and often the reality that a poor single mother can better
provide for her family by being on welfare than by working at a
low-wage job plays a critical role in discouraging AFDC recipients
from looking for and accepting employment.  As figure 3.4 shows,
about three-fourths of the JOBS administrators cited the lack of jobs
with sufficient wages and benefits as a moderate or major reason that
their job-ready clients did not become employed.  About 70 percent of
administrators also noted that their participants did not become
employed because of concerns about losing their AFDC benefits,
Medicaid, or housing subsidies.  By comparison, about one-half of
administrators cited the lack of jobs as a major or moderate reason. 

   Figure 3.4:  Administrators'
   Opinions on Reasons Job-Ready
   Participants Do Not Become
   Employed (Mid-1994)

   (See figure in printed
   edition.)

One study found that 55 of 69 randomly selected current and former
AFDC recipients interviewed in Tennessee and North Carolina said that
they were not likely to accept a minimum-wage job that did not
provide health insurance for them and their children.  Most of the 55
thought that health insurance was a necessity and others said that
they could not support their families with a minimum-wage job.\66

Concerns about participants' abilities to support their families may
affect the attitudes of administrators and staff in promoting
employment as the ultimate program goal.  For example, we found that
while about 60 percent of local administrators said that they would
definitely encourage a 30-year-old JOBS participant with one child to
accept a minimum-wage job with health insurance benefits, only 26
percent would definitely encourage her to accept such a job without
health benefits. 

Recent studies of labor market conditions and the characteristics of
welfare recipients indicate that employment training strategies to
improve the earnings capacities of welfare recipients through
education and training may not lead to earnings increases great
enough to allow single parents to support themselves with their own
earnings.\67 These studies demonstrate that the supports available to
low-wage workers, for example, the EITC, expanded Medicaid coverage,
child support payments, and child care subsidies, play an important
role in helping families get jobs and remain employed.  Our recent
work on child care subsidies indicates that assistance with child
care has a large effect on the likelihood that poor women will work. 
Thus, subsidies may help welfare recipients become employed and
remain off the welfare rolls.\68


--------------------
\62 Mother-Only Families:  Low Earning Will Keep Many Children in
Poverty (GAO/HRD-94-177, Apr.  2, 1991). 

\63 Under the expanded EITC provisions scheduled to take effect in
1996, a family with more than one child will be eligible for as much
as a 40-percent credit on earnings up to $8,425 (in 1994 dollars). 

\64 Low-Income Families:  Comparisons of Incomes of AFDC and Working
Poor Families (GAO/T-HEHS-95-63, Jan.  25, 1995). 

\65 See Gayle Hamilton and Thomas Brock, The JOBS Evaluation:  Early
Lessons from Seven Sites. 

\66 Vicki Grant, and others, A Study of the Relationship of Health
Coverage to Welfare Dependency, p.  13. 

\67 See by Gary Burtless "The Employment Prospects of Welfare
Recipients," and by Rebecca Blank "Outlook for the U.S.  Labor Market
and Prospects for Low-Wage Entry Jobs" in The Work Alternative: 
Welfare Reform and the Realities of the Job Market, Demetra
Nightingale and Robert Haveman, eds., The Urban Institute
(Washington, D.C.:  1994).  Also see Linda Levine, Jobs for Welfare
Recipients, Congressional Research Service (Washington, D.C.:  May
1994) and Thomas Gabe and Gene Falk, Welfare:  Work (Dis)Incentives
in the Welfare System, Congressional Research Service (Washington,
D.C.:  1995). 

\68 Our analysis, based on an empirical model, predicts that
providing a full subsidy to mothers who pay for child care could
increase the percentage of poor mothers who work from 29 to 44
percent, and that of near-poor mothers who work from 43 to 57
percent.  Child Care:  Child Care Subsidies Increase Likelihood That
Low-Income Mothers Will Work (GAO/HEHS-95-20, Dec.  30, 1994). 


CONCLUSIONS, AGENCY COMMENTS, AND
OUR EVALUATION
============================================================ Chapter 4

The 104th Congress proposes to fundamentally change AFDC--the
nation's largest cash assistance program for poor families with
children.  While there is general agreement that reforms should
promote work, the Congress is considering the type and extent of work
requirements to be linked to the receipt of cash assistance. 

Whether AFDC continues as an entitlement program or is converted into
a block grant, program administrators at the county and local levels
will be concerned with moving large numbers of welfare recipients
into employment.  Our work highlights examples of programs that are
well-focused on the ultimate goal of employment--stressing the
importance of work for their participants and forging links with
employers to identify jobs or create work opportunities where none is
available. 

However, these programs appear more the exception than the rule. 
Most programs appear to emphasize preparing participants for
employment without also making strong efforts to help place their
participants in jobs.  While we acknowledge that some administrators
face factors beyond their control that may limit program choices,
including budget constraints and a lack of jobs, other programs
facing similar constraints have taken steps that promote work more
strongly for their participants.  These steps include focusing staff
and participants on the importance of employment, working more
closely with employers to identify job openings, determining
employers' needs, and helping match recipients' education and
training activities to labor market demands. 

Even programs that are well-focused on moving AFDC recipients into
employment have faced challenges, however.  For example, the
Riverside County program strongly emphasized moving recipients
quickly into jobs; yet after 3 years, about 40 percent of its
participants remained on AFDC.  Many who became employed remained on
AFDC or, if off AFDC, continued to receive other forms of public aid,
including Food Stamps or federal housing assistance.  And some of
those who left AFDC remained in poverty and at risk of returning to
AFDC. 

In those cases where unsubsidized employment is not available or the
characteristics of participants do not make them readily employable,
strategies like work supplementation or on-the-job training may help
welfare recipients become employed.  And where regular jobs or
subsidized employment are not feasible, work-experience programs may
serve as an alternative that promotes work for welfare recipients. 
Administrators generally supported the use of these work activities. 
However, they believe that they need more flexibility to design work
activities to meet the needs of their participants and local labor
markets. 


   AGENCY COMMENTS AND OUR
   EVALUATION
---------------------------------------------------------- Chapter 4:1

In commenting on a draft of this report (see app.  IV), HHS'
Administration for Children and Families (ACF) disagreed with our
conclusion that JOBS programs do not have a strong employment focus. 
ACF stated that we did not sufficiently recognize programs' use of
job search or the extent of their job-development activities in
evaluating their employment focus.  It also stated that we did not
acknowledge the many ways that programs could focus on employment
and, instead, relied too much on programs' low use of subsidized
employment and work experience to indicate a weak employment focus. 
We continue to believe, based on all the evidence we gathered, that
many JOBS programs nationwide do not have a strong employment focus. 

More specifically, ACF commented that the report does not recognize
job search as an employment-focused activity and its extensive use in
JOBS, thus, underrepresenting the employment efforts of JOBS
programs.  We acknowledge that programs can emphasize employment
through their use of job-search activities for participants.  As we
had shown in figure 3.3, the participants enrolled in job search
nationwide numbered 75,000 out of 586,600.  In addition, we note that
all programs use job search as an integral part of their programs and
have added this information to the report. 

We also found, however, that while job search plays a role in all
programs, its use varies considerably.  Only about one-third of
programs employ an early job search strategy that encourages
participants to look for work upon enrollment in JOBS, in effect
letting the local labor market decide who is job-ready and
employable.  Those who fail to find work initially are then placed in
job search again after participating in education and training.  On
the other hand, most programs do not expect all participants to look
for work upon enrollment, instead limiting job-search activities
until participants have received the education and training that the
program determines they need to become employed. 

We also note that the programs we highlighted for their strong
job-placement efforts took steps beyond enrolling participants in
job-search activities.  These programs facilitate job-search
activities by working closely with employers, through job-development
efforts, to help participants find work.  In addition, it is
important that programs encourage their participants to accept
employment by, for example, helping all participants understand the
work incentives available to them.  We found, however, that most
programs did not strongly emphasize job-development efforts or inform
all participants of important work incentives. 

In addition, ACF believes that we did not adequately recognize the
extent of programs' connections with employers through their job
development efforts.  Our data show and the report acknowledges that
almost all JOBS programs include some job-development activities,
performed either by a program's own staff or through other
organizations.  We also found, though, that the extent of job
development performed on behalf of JOBS participants, whether by the
welfare agency itself or other organizations, is limited.  For
example, about 60 percent of program administrators reported that
their program or its contractors arranged interviews for or marketed
to employers only some or few of their job-ready participants.  In
addition, over one-half of the nation's program administrators
believe that their program or its contractors did not do enough job
development to meet their participants' needs. 

ACF also noted that JOBS programs can take many approaches to help
their participants become employed.  In addition, ACF stated that the
relatively low use of subsidized employment and work experience does
not necessarily indicate a lack of employment focus.  We agree that
there are many ways that programs can focus on employment, as we
demonstrated with the examples of different approaches in chapter 2. 
We also agree that programs do not have to use subsidized employment
or work experience to be considered employment-focused.  The
Riverside County program, for example, does not emphasize these
options.  However, we found that most programs reported placement
rates for their job-ready participants of 50 percent or less.  Yet
programs were not widely using existing subsidized employment or
work-experience options to foster work among the many participants
unable or unwilling to find work. 

In addition to these issues, ACF expressed concern that our draft
report promoted holding states accountable for the employment
outcomes of their JOBS programs without noting the problems involved
in such an approach.  We acknowledge the challenges inherent in
holding JOBS programs accountable for results.  We maintain, however,
that strong congressional interest in AFDC becoming more focused on
helping recipients become employed, as well as requirements in the
Government Performance Results Act that performance monitoring become
more outcome-oriented governmentwide, indicate that more attention to
outcome measures and goals is appropriate. 

ACF also suggested certain technical revisions to the draft, which we
incorporated as appropriate. 


FEDERAL RULES GOVERNING WORK
ACTIVITIES FOR AFDC RECIPIENTS
=========================================================== Appendix I

Federal law and regulations govern the administration and funding of
JOBS work activities:  on-the-job training, work supplementation,
community work experience, and alternative work experience.  The
table below identifies and compares selected features of these
activities. 



                                    Table I.1
                     
                       Comparison of Federal Rules for JOBS
                                 Work Activities

                                Work
                On-the-job      supplementatio  Community work  Alternative work
Feature         training        n               experience      experience
--------------  --------------  --------------  --------------  ----------------
Payment to      Employee        Employee        AFDC benefit    AFDC benefit
participant     paycheck;       paycheck;       check; not      check; not hired
and employment  hired on day 1  hired by 14th   hired
status                          week

Payment to      JOBS funds for  AFDC benefits   None            None
employer        up to an        for up to 100
                average of 50   percent of
                percent of      wages
                wages

Hours and       Up to full-     Up to full-     Limited hours,  Up to full-time
months worked   time for        time for 9      such that AFDC
                length of time  months          grant divided
                appropriate                     by hours is
                for training                    not less than
                                                minimum wage;
                                                generally
                                                limited to 9
                                                months

Position and    Any position    New position    New position    New position
employer        with any        with any        with employer   with employer
                employer        employer        serving a       serving a public
                                                public purpose  purpose
--------------------------------------------------------------------------------

METHODOLOGY FOR SURVEY OF JOBS
ADMINISTRATORS
========================================================== Appendix II

To collect information on the extent to which JOBS programs were
working with employers to help their participants find work and what
factors hindered such efforts, we mailed a questionnaire to a random
sample of county JOBS program administrators in May 1994.  We did not
verify the data collected through the questionnaire. 


   SAMPLE SELECTION AND SURVEY
   RESPONSE
-------------------------------------------------------- Appendix II:1

Because most JOBS services are delivered and received at the county
level, we selected a random sample of counties for our survey.  We
derived a nationwide listing of counties from 1990 census data and
selected an overall sample of about 450 counties.  Before selecting
this sample, we stratified the counties into the following four
groups: 

1.  Large urban counties--Counties comprising the 10 cities with the
largest populations of female-headed families on public
assistance.\69

2.  Metropolitan counties with a central city--Counties containing
the central city for a metropolitan statistical area. 

3.  Metropolitan counties without a central city--Counties in
metropolitan statistical areas that do not contain a central city. 

4.  Nonmetropolitan (rural) counties--Counties that are not part of a
metropolitan statistical area. 

We selected all the large urban counties and random samples of
counties from each of the other three groups.  Table II.1 shows the
total number of counties and the number sampled in each stratum. 
After selecting the sample, we contacted states to determine the
name, address, and telephone number of the JOBS program
administrators responsible for programs in the sampled counties.  We
also identified 22 rural counties and one nonrural county in our
sample that did not offer JOBS programs.  Therefore, we adjusted our
initial sample to exclude these counties.\70

We obtained responses from 93 percent of the program administrators
for the counties in our adjusted sample (401 out of 430).  We used
these responses to produce national estimates for the JOBS program. 



                                              Table II.1
                               
                                Response Rates, Overall and by Strata

                                       Total
                                   number in        Total                 Number of
Stratu                             universe,     counties     Adjusted     counties   Response rate
m       Type of county                  1990      sampled       sample   responding     (percent)
------  -----------------------  -----------  -----------  -----------  -----------  ---------------
1       Counties comprising 10            18           18           18           17        94
         cities with most
         female-headed public
         assistance families
2       Counties in                      422          134          134          121        90
         metropolitan areas
         with central cities
3       Counties in                      311          120          119          106        89
         metropolitan areas
         without central cities
4       Counties in                    2,390          181          159          157        99
         nonmetropolitan areas
====================================================================================================
Total   All                            3,141          453          430          401        93
----------------------------------------------------------------------------------------------------

--------------------
\69 These cities were included in the sample:  Baltimore, Chicago
(Cook and DuPage counties), Cleveland, Detroit, Houston (Fort Bend,
Harris, and Montgomery counties), Los Angeles, Milwaukee (Milwaukee
and Washington counties), New York (the boroughs of Brooklyn, Kings,
New York, Queens, and Richmond), Philadelphia, and San Diego. 

\70 According to JOBS regulations, states do not have to operate JOBS
programs in sparsely populated areas as long as they meet other
requirements of statewideness, including providing full programs in
all metropolitan areas and at least minimal programs covering 95
percent of their AFDC population. 


      SAMPLING ERRORS FOR
      ESTIMATES
------------------------------------------------------ Appendix II:1.1

Because the estimates from this survey are based on a sample, each is
subject to sampling error.  Except where noted, the maximum sampling
errors for estimates in this report are plus or minus 7 percentage
points.  In addition, table II.2 shows the sampling errors for our
estimates of the numbers of participants in various JOBS activities
presented in figure 3.3.  We computed these sampling errors at the
95-percent confidence level.  Therefore, the chances are 95 out of
100 that the actual percentage or number being estimated falls within
the range defined by the estimate, plus or minus the sampling error. 



                          Table II.2
           
           Sampling Errors for Estimates Presented
                        in Figure 3.3

                    (Figures in thousands)

                                  Estimated number  Sampling
Activity                           of participants   error\a
--------------------------------  ----------------  --------
Postsecondary education                        121        24
High school or general                          85        15
 equivalency diploma
Job skills training                             80        13
Job search                                      75        13
Job readiness                                   70        36
Other approved activity                         65        21
Work experience                                 59        15
Adult basic or remedial                         56         9
 education
English as a second language                    19         6
On-the-job training                              3         1
Work supplementation                             1        <1
------------------------------------------------------------
\a These sampling errors were computed at the 95-percent confidence
level.  For example, the chances are 95 out of 100 that the actual
number in postsecondary education falls within the range of 121,000,
plus or minus 24,000. 


QUESTIONNAIRE WITH RESPONSES
========================================================= Appendix III

In this appendix, our questionnaire and summaries of the responses
are presented.  For each question, we show the unweighted actual
number of respondents that answered that question and the weighted
statistic for the nation.  The percentages may not add to 100 due to
rounding. 



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)




(See figure in printed edition.)Appendix IV
COMMENTS FROM THE ADMINISTRATION
FOR CHILDREN AND FAMILIES
========================================================= Appendix III



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


MAJOR CONTRIBUTORS
=========================================================== Appendix V

David P.  Bixler, Assistant Director, (202) 512-7201
Gale C.  Harris, Evaluator-in-Charge, (202) 512-7235
Bob DeRoy, Assistant Director, Computer Science
Patricia L.  Elston, Senior Evaluator
Steven R.  Machlin, Senior Statistician
Nora L.  Perry, Evaluator




RELATED GAO PRODUCTS
=========================================================== Appendix V

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

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

Child Care:  Child Care Subsidies Increase Likelihood That Low-Income
Mothers Will Work (GAO/HEHS-95-20, Dec.  30, 1994). 

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

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

Welfare to Work:  JOBS Automated Systems Do Not Focus on Program's
Employment Objective (GAO/AIMD-94-44, June 8, 1994). 

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

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

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

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

Multiple Employment and Training Programs:  Major Overhaul Is Needed
(GAO/T-HEHS-94-109, Mar.  3, 1994). 

Welfare to Work:  States Move Unevenly to Serve Teen Parents in JOBS
(GAO/HRD-93-74, Jul.  7, 1993). 

Welfare to Work:  JOBS Participation Rate Data Unreliable for
Assessing States' Performance (GAO/HRD-93-73, May 5, 1993). 

Welfare to Work:  States Begin JOBS, but Fiscal and Other Problems
May Impede Their Progress (GAO/HRD-91-106, Sept.  27, 1991). 

