Four Years of Welfare Reform

Document Sample
Four Years of Welfare Reform Powered By Docstoc
					No. 378                                          August 22, 2000

                        Four Years of Welfare Reform
                                            A Progress Report
                                               by Lisa E. Oliphant

                                         Executive Summary

        Fours year ago, President Clinton signed into          incentive to leave welfare, has produced few gains in
     law the Personal Responsibility and Work                  self-sufficiency and done little to discourage
     Opportunity Reconciliation Act, promising to put          women from making the choices that cause them
     an end to dependence on welfare by requiring              to enter the welfare rolls in the first place:
     work and responsibility and encouraging two-par-
     ent families. The conversion of the old entitle-               • Welfare reform has been largely unsuccessful
     ments-based Aid to Families with Dependent                       in moving long-term unemployed, difficult-
     Children program into the new work-oriented                      to-place individuals off cash assistance.
     Temporary Assistance for Needy Families pro-                   • Welfare reform is failing to make individu-
     gram has brought about some of the most talked-                  als self-sufficient..
     about public policy phenomena of the last decade,              • Welfare reform has failed to significantly
     including dramatically reduced caseloads and an                  curb out-of-wedlock births to young
     influx of single mothers into the labor market.                  women.
     Welfare reform is generally regarded as a great suc-
     cess, and policymakers today are talking about                Overall, the new welfare law is a second-best
     incremental add-ons to the 1996 legislation.              solution to welfare dependence. Until policy-
        A closer review of the research, however, reveals      makers take a preventive approach and make
     that the celebration of welfare reform has been pre-      welfare no longer available to, or an attractive
     mature and that the new law, in fact, merits funda-       option for, women considering making unten-
     mental restructuring. A survey of welfare reform’s        able life choices, the country will continue to
     progress to date suggests that the 1996 legislation,      spend endless taxpayer dollars trying to clean up
     although moderately successful in reducing the dis-       its dependence problem.


     Lisa E. Oliphant is an entitlements policy analyst at the Cato Institute.
    When the data                                                            result of out-of-wedlock childbearing—and
 are put together,                 Introduction                              continued reliance on noncash benefits are
                                                                             problems that work requirements, time lim   -
   the picture that       Nearly four years after the “end of welfare        its, and new training programs are failing to
     emerges four     as we know it,” weaning the nation’s poor              address. Keeping people out of the welfare
                      from the dole no longer rates as a top priori-         system would prove a far more effective way
years into welfare    ty among policymakers. Since enactment of              of reducing dependence than simply, as the
 reform is one of     the Personal Responsibility and Work                   current law does, altering the terms under
positive, but very    Opportunity Reconciliation Act (PRWORA)                which benefits are offered.
                      in 1996, caseloads throughout the country
  limited, change.    have plummeted and employment of single
                      mothers has surged beyond policymakers’                        Out with the Old,
                      expectations. Proclaiming victory last year,                   In with the New
                      President Clinton announced: “The welfare
                      rolls have been cut in half; they’re at their              As a result of bipartisan effort, PRWORA
                      lowest level in 32 years. And those who are on         was signed into law on August 22, 1996, end-
                      welfare today are four times as likely to work         ing the historic federal entitlement to assis-
                      as when I took office.”1                               tance. Under the new law, the Job
                          The evidence after nearly four years sug-          Opportunities and Basic Skills Training
                      gests that proponents of the new law have              (JOBS) and Aid to Families with Dependent
                      begun to uncork their champagne bottles                Children (AFDC) programs were replaced
                      prematurely. Although caseloads, for                   with a time-limited assistance and work-
                      instance, have fallen more rapidly than ever           requirement program, Temporary Assistance
                      before, most people who have left the rolls            for Needy Families (TANF).
                      have been short-term recipients who have                   The new welfare law aims to increase self-
                      benefited from the nation’s hot economy.               sufficiency and well-being through a system
                      Furthermore, the mass movement off cash                that emphasizes work and responsibility—“a
                      assistance has not actually resulted in                hand up, not a hand out.” The legislation
                      reduced dependence. A majority of individu-            outlines a four-pronged attack on depen-
                      als who have stopped receiving welfare since           dence: labor-market attachment, marriage,
                      enactment of the new law continue to                   two-parent families, and supplemental sup-
                      depend on noncash assistance from the gov-             ports to aid in the transition from welfare to
                      ernment to meet their health care, food,               work.
                      housing, and child care needs.                             Under PRWORA, states, territories, and
                          Although welfare reform is not succeed-            tribes receive federal block grant allocations
                      ing in making former recipients self-suffi-            totaling $16.8 billion each year until fiscal
                      cient, it would be inappropriate to declare            year 2002. Each state is permitted the flexi-
                      the new law a total failure. Critics’ fears that       bility to determine eligibility and benefit lev-
                      in their race to cut caseloads the states would        els, as well as services provided to needy fam -
                      spawn an outbreak of poverty, hunger,                  ilies. This so-called new federalism allows
                      homelessness, and domestic violence simply             states to use TANF funding in any manner
                      have not been justified. The lives of most             “reasonably calculated to accomplish the
                      individuals who have substituted paychecks             purposes of TANF,” so long as they maintain
                      for benefit checks have stayed the same or             historical levels of spending, known as
                      improved.                                              “maintenance of effort.” To continue receiv-
                          When the data are put together, the pic-           ing their full federal TANF allotments, states
                      ture that emerges four years into welfare              must also conform to specific requirements
                      reform is one of positive, but very limited,           regarding current recipients’ work participa-
                      change. Long-term dependence—largely the               tion rates and length of time on the rolls.

                                                     Despite the nation’s progress in cutting
  Welfare Reform’s Success:                      its welfare rolls under the new law, recent
        A Hasty Call?                            slowing of that trend in a number of states
                                                 suggests that welfare reform may have begun
    Over the past four years, welfare reform to exhaust its potential. The most rapid
has too often been evaluated on the basis of declines took place in the first two years after
caseload decline and work participation fig- enactment of the new welfare law, during
ures. That simplistic approach evades some which the nation’s rolls fell by 32 percent. By
of the most critical questions that need to be 1998, however, drops in the welfare rolls in
asked:                                           many states had started to slow or had begun
                                                 to increase. Figure 1 shows some of the more
     • Is welfare reform succeeding in moving notable examples. After nearly cutting its
       long-term recipients off the rolls?       rolls in half, for instance, New Mexico saw a
     • Are former recipients better off work- 40 percent increase in 1998. Hawaii, Indiana,
       ing than they were when on welfare?       and Wisconsin also experienced growth in
     • Are families that leave welfare for work their caseloads that year. Louisiana experi-
       becoming self-supporting and upward- enced no increase but did have a substantial
       ly mobile in the labor market?            slowing of its decrease. That reversal of for-     Despite the
     • Is the new law encouraging potential tune is largely linked to difficulties in moving        nation’s progress
       new recipients to make wiser choices or long-term welfare recipients, who now make           in cutting its wel-
       look for alternative means of support? up a larger fraction of the caseload, off the
                                                 rolls.                                             fare rolls, recent
    Misleading coverage and scarce follow-up                                                        slowing of that
data on former and potential recipients have Reasons for Caseload Decline
left many observers confused about just how          Cutting the nation’s rolls nearly in half is   trend suggests
much positive change has been wrought by impressive but not sufficient to constitute                that welfare
the 1996 law. To get a clearer sense of the leg- success. If, for instance, individuals are leav-   reform may have
islation’s impact, as well as insights into how ing welfare without jobs, or are merely trans-
welfare reform should proceed in the next ferring to other government cash-assistance               begun to exhaust
Congress, a comprehensive progress report programs, then the figures may mask pro-                  its potential.
that assesses the nature of the caseload longed dependence.
decline, gains in the well-being and self-suffi-     Rooting out from the welfare rolls the
ciency of former recipients (frequently called small number of fraudulent claimants with
“leavers”), and progress in stopping depen- alternative sources of income is an important
dence before it starts is in order.              first step in the “cleaning up” process. The
                                                 Washington Post reported in 1997 that crack-
                                                 ing down on fraud had accounted for as
      Shorter Welfare Rolls                      much as 20 percent of the decline in the case-
                                                 load.3 Most fraudulent claimants fell off the
Extent of Caseload Decline                       caseload during the early days of welfare
    The most talked-about public policy phe- reform when they were effectively “weeded
nomenon in recent years has been the out” by new regulations requiring recipients
nation’s plummeting welfare rolls. Between to spend 20 hours a week in a job or per-
1993, when states began experimenting with forming community service. Though most
welfare reform, and 1999, the country’s case- fraudulent claimants have by now left the
load fell by 53 percent—or, as Cato Institute rolls, the new work requirements are effective
scholars calculate, a decline of nearly 6,000 in deterring such individuals from signing
welfare families per day.2 The caseload now up in the first place.
stands at its lowest level in 30 years.              Although a sizable 63 to 87 percent of for-

                       Figure 1
                       State Caseload Trends

                             215,000                                                                                NM




                                  Jan. 1996            Jan. 1997                Jan. 1998             Jan. 1999             Sep. 1999

                       Source: U.S. Department of Health and Hunan Services, Administration for Children and Families, June 2000.

                       mer recipients succeed in finding work at                 benefits as well as refusing or being unable to
                       some point after leaving the rolls,4 only 22              comply with the new work participation rules.
                       percent of TANF leavers, according to HHS                 It should be noted that this low national figure
                       data, cited “employment” as their main rea-               for sanction closures masks widely differing
                       son for exit in 1998.5 That suggests that,                rates among some states (Figure 2). As the
                       although many former recipients are benefit-              Heritage Foundation has shown, the differ-
  Although many        ing from the healthy job market, it is not                ences are largely the result of variations in state
                       actually the booming economy that has been                sanctions policies. High percentages of exits
former recipients      drawing them off the rolls in recent years.               due to sanctions, it notes, correspond with
    are benefiting         A person may marry after leaving the wel-             stricter state sanction policies.9
                       fare rolls, but marriage is not frequently cited              At present there is little national data doc-
 from the healthy      as a primary reason for exiting welfare under             umenting case closures due to time limits.
  job market, it is    the new law. The Urban Institute notes that 6             The HHS’s most recent data show that 15.5
  not actually the     percent of early AFDC/TANF leavers                        percent of TANF families left the rolls in fis-
                       (1995–97) exited because of a change in fam    -          cal year 1998 because of “state policy,”
 booming econo-        ily situation, which could have included mar-             although the extent to which that figure
 my that has been      riage or ineligibility due to the youngest                reflects time-limit cuts is unclear.1 0 As do
                       child’s turning 19.6 In 1998 only 0.4 percent             policies on sanctions, state policies on time
drawing them off
                       of case closures, according to the HHS, were              limits vary widely, and that variation is
the rolls in recent    due to marriage.7                                         reflected in the differences in individual state
              years.       According to HHS data, “noncompliance”                figures for case closures due to time limits. As
                       was the reason 6 percent of leavers exited the            the New York Times notes, for instance,
                       rolls in 1998.8 Noncompliance may include not             although the expiration of time limits
                       following the proper procedure for claiming               explains only 6 percent of the caseload

Figure 2
Percentage of Case Closures due to Sanctions, 1996
        AR                                                                                              30.2
         FL                                                                                            30.1
        MS                                                                                             30
         SC                                                                                           29.6
         ID                                                                                   26.4
         IA                                                                                   26.3
        MO                                                                           22.8
        OK                                                       15
         AZ                                               13.5
        MD                                              12.9
        VA                                       11.5
         LA                                8.4
         SD                               8.3
        KS                               8
         NJ                             7.7
        OR                         6
        TN                        5.8
        OH                        5.8
         MI                      5.6
        WV                      5.2
         UT                    5
         DE                  4.5
        AK               3
        NE             2.5
        NH         2
        MT       1.4
         AL      1.3

Source: Jan Kaplan, “The Use of Sanctions under TANF,” Welfare Information Network, April 1999.

decline in Massachusetts in 1999, it accounts                    Several studies may provide clues to why peo-
for 33 percent of the decline in Connecti-                       ple are voluntarily exiting the system. The
cut.1 1 The impact of time limits will start to                  Urban Institute’s study of people who left
become more apparent as the federal five-                        AFDC or TANF between 1995 and 1997, for
year limit begins to take effect and recipients                  example, found that the second most com       -
are increasingly confronted with more strin-                     mon reason for leaving the rolls (next to         The impact of
gent state policies.1 2                                          work) was administrative problems or has-
   With only 22 percent of cases closed                          sles, reported by 10 percent of former recipi-    time limits will
because of employment; 6 percent because of                      ents. Another 7 percent in this study report-     become more
sanctions; 15 percent because of state policy,                   ed no longer needing or wanting to receive        apparent as the
which may include time limits; and a very                        benefits.1 5 A study by the state of Wisconsin
small proportion because of marriage or                          at the end of 1998 found that 34 percent of       federal five-year
fraud, a large group of TANF leavers, as seen                    those who had left the system said they sim   -   limit begins to
in Figure 3, remains to be accounted for. The                    ply “did not want to be on welfare,” and 16
HHS reports that, in 1998, 56 percent of all                     percent reported leaving because they “did
                                                                                                                   take effect.
cases that closed did so for reasons classified                  not want to, or could not, participate in, wel-
as “other.”13 Unfortunately, the HHS does                        fare program requirements.”1 6
not specify what this category consists of,
except to note that it encompasses all                           Welfare Reform vs. the Economy
unknown reasons, including that the indi-                           Since rapid caseload decline began to
vidual or family voluntarily closed the case.1 4                 accompany states’ experimentation with wel-

                       Figure 3
                       Reasons for Leaving TANF

                                                 6.0%                   State Policy


                       Source:U.S. Department of Health and Human Services, “National Emergency TANF Datafile,” May 1999.
                       Note: Entries in TANF Datafile do not add to 100 percent.

                       fare reform in the early 1990s, a debate has                    ticipation, a drop that is independent of the
                       raged over the primary causes of the declin-                    effects of the strong labor market.”1 7 Thus,
                       ing caseload. The evidence suggests that early                  both the hot economy and welfare reform
                       caseload declines were largely a response to                    have played important and shifting roles in
                       the booming job market, whereas the more                        generating the country’s caseload decline. As
                       recent and modest drops in the welfare rolls                    the Committee for Economic Development
                       have stemmed primarily from states’ tougher                     observes, “A fortuitous coincidence of wel-
                       welfare policies.                                               fare reform with a period of sustained eco-
                           The president’s Council of Economic                         nomic growth has produced increases in
  The recent mod-      Advisers has undertaken the most compre-                        employment that neither prosperity alone
                       hensive analysis of this question to date.                      nor changes in public assistance policy alone
   est drops in the    Tracking the rate of caseload decline along-                    could have achieved.”1 8
 welfare rolls have    side the unemployment rate and changes in                           The apparent weakening of economic
stemmed primari-       that rate, the study finds, as seen in Figures 4                influence on caseload declines in recent years
                       and 5, that the economy was responsible for                     can best be understood by looking at the
     ly from states’   26 to 36 percent of the falling rolls in the                    changing makeup of the welfare rolls. When
   tougher welfare     period 1993 to 1996 and a much smaller 8 to                     states began experimenting with work
                       10 percent between 1996 and 1998. For the                       requirements and other welfare reform mea-
           policies.   latter period, as Figure 5 shows, the council                   sures in 1993, not only was the economy
                       finds that the influence of changes in the                      booming, but a significant amount of under-
                       welfare law has exceeded that of the hot econ-                  used labor was sitting on the welfare rolls.
                       omy, producing roughly one-third of the                         When there were no work requirements and
                       reduction in the caseload between 1996 and                      benefits were generous, capable claimants
                       1998. The council notes that reform in recent                   had little incentive to rush off the welfare
                       years has “caused a large drop in welfare par-                  rolls to find jobs or to shorten the gap

Figure 4
Reasons for the Decline in Welfare Caseloads, 1993–96


                                  Other Factors

                                                          Welfare Waivers

Source: Council of Economic Advisers, “Explaining the Decline in Welfare Receipts, 1993–1996,” May 9, 1997.

between jobs.19 When states began experi-                ible or applied to only a small number of
menting with work requirements under                     recipients, individuals capable of working
waivers, even if those requirements were flex-           quickly began to find jobs—both because

Figure 5
Reasons for the Decline in Welfare Caseloads, 1996–98


                                                                    Welfare Reform

                               Other Factors

Source: Council of Economic Advisers, “The Effects of Welfare Policy and the Economic Expansion on Welfare
Caseloads: An Update,” August 3, 1999, executive summary.

                       Figure 6
                       Average Caseload Decline by Type of State Sanction










                                       Initial Full-Check Sanction   Delayed Full-Check Sanction        Weak Sanction

                       Source: Robert Rector and Sarah E. Youssef, The Determinants of Welfare Caseload Decline (Washington:
                       Heritage Foundation, 1999).

                       they had to (or perceived that they soon                   employed, many others exit because they
                       would have to) and because the jobs were                   have been sanctioned or choose to leave for
                       there. As a result, between 1993 and 1994, the             “other” reasons.
                       country began to see a sharp rise in the                      The strongest evidence of the increasing
                       employment of those most likely to have                    impact of welfare reform on the caseload in
                       been on welfare—never-married and previ-                   recent years is the fact that states with the
                       ously married women.20 Plentiful jobs and                  toughest work and sanctions policies and the
                       tougher new rules thus began to encourage                  least attractive benefits packages have been
   States with the     individuals to look for work, and the strong               the most successful in cutting their welfare
                       economy made their exit from the rolls into                rolls. A recent Heritage Foundation study,
   toughest work       employment possible.                                       which examines how states’ welfare policies
    and sanctions         By 1996 many states had been experi-                    and local unemployment rates correlate with
  policies and the     menting with work requirements and sanc-                   drops in the welfare rolls, concludes that
                       tions for several years and had already                    unemployment rates and the economy have
   least attractive    “creamed” from their rolls a substantial por-              had no significant or predictable effect on
benefits packages      tion of employable recipients. The people                  the size of the caseload in recent years. As
                       who remained on many states’ welfare rolls                 shown in Figure 6, the study notes that “dif-
    have been the      one to two years into welfare reform were                  ferences in state welfare policies—specifically
most successful in     individuals who faced numerous obstacles to                stringent sanctions and immediate work
 cutting their wel-    work. As a result, it has become increasingly              requirements—are highly associated with
                       challenging in recent years to move recipients             rapid rates of caseload decline. By contrast,
         fare rolls.   off the rolls and directly into the labor mar-             the relative vigor of state economies, as mea-
                       ket. As noted above, although many of the                  sured by unemployment rates, has no statis-
                       people who leave TANF eventually end up                    tically significant effect on caseload decline.

Indeed, states with higher caseload reduc-              through creaming the easiest cases from the          The majority of
tions, on average, had slightly higher unem     -       rolls in the midst of a hot economy, and such        case closures
ployment rates.”2 1                                     creaming is increasingly leaving behind a
    A Cato Institute study suggests that the            hard core of long-term unemployed and dif-           under the new
generosity of states’ welfare benefits has              ficult-to-place recipients. As a result, many        law have occurred
more influence on the size of the caseload              people contend that the remarkable caseload
than do sanctions. The study finds that for             decline of the past four to seven years cannot
                                                                                                             among the easi-
every thousand-dollar decrease in the overall           be sustained (and has, in fact, been slowing         est-to-place and
welfare package (which includes AFDC, food              down) and signifies little actual progress in        least-dependent
stamps, and Medicaid) states reduce their               chipping away at long-term dependence. As
welfare rolls by an average 3 percent.                  one caseworker puts it, “It is easy to take cred-    population of
Reductions in benefits, the author notes, can           it for placing people in jobs who would likely       recipients.
be significantly more effective at reducing             find them anyway, especially in a healthy
caseloads than the sanctioning provisions               economy; it is much harder to deal with
and work requirements highlighted by the                hard-core unemployed, who have all sorts of
Heritage Foundation’s study.2 2                         aptitude and attitude problems, in an econo-
    Despite the fact that in recent years “policy       my with few jobs for low-skill workers.”25
reform—not economics—is the principal                       Studies of people who leave welfare con-
engine driving the decline in dependence,”2 3           firm the view that the majority of case clo-
there remains cause for concern about what              sures under the new law have occurred
would happen during an economic slowdown                among the easiest-to-place and least-depen-
or recession. The Committee for Economic                dent population of recipients. Many who
Development predicts: “This reduction in wel-           have exited the system under welfare reform
fare rolls could quickly reverse itself if the          appear to be people who were not at risk of
economy weakens. With low seniority and lim     -       long-term dependence and frequently had
ited skills, many former recipients remain vul-         some form of employment even while receiv-
nerable to layoffs from their new jobs. In a            ing welfare.26 As Charles Murray notes, “The
slack labor market, many would experience               reductions in caseloads are occurring dispro-
considerable difficulty finding new positions,          portionately among women who wouldn’t
as would current welfare recipients seeking to          have spent much time in the system anyway
leave the rolls. A weaker economy would                 and are not part of the underclass.”2 7The ini-
sharply reduce the number of jobs available             tial wave of welfare leavers has tended to have
for former welfare recipients, and many might           better education, higher levels of basic skills,
attempt to return to public assistance.”24 And,         and more previous experience in the labor
in the event of such a slowdown, it is unlikely         market than those who remain on the rolls.2 8
that states would continue to enforce such                  The data indicate that a large proportion of
important guards against recidivism as strict           the 3 million American households still receiv-
work requirements, sanctions policies, and              ing welfare are hard-core dependence cases that
time limits.                                            will present a growing challenge for welfare
                                                        reformers in years to come. The HHS reported
Distribution of Caseload Decline:                       in 1998 that, as caseloads drop, the proportion
Current vs. Former Recipients                           of long-stay families on welfare rolls is increas-
   The economy’s shrinking role in the                  ing29 and noted that there had been a small but
nation’s caseload decline, as we have seen, is          steady decline since 1994 in the percentage of
largely attributable to the fact that fewer peo-        the caseload that had been on assistance for one
ple who are capable of taking advantage of              year or less (36 to 33 percent) and a corre-
the booming job market are left on the rolls.           sponding increase in the percentage of the case-
Critics suggest that the unprecedented rate             load on assistance for five years or more (19 to
of case closures has been achieved primarily            24 percent).30

                     Table 1
                     Barriers to Current Recipients’ Working

                     Barriers                                                                      Prevalence

                     None                                                                            15.5%
                     Less than high-school education                                                 30.1%
                     Fewer than four job skills                                                      21.1%
                     Perceived discrimination                                                        13.9%
                     Transportation problem                                                          47.3%
                     Major depressive disorder                                                       26.7%
                     Drug dependence                                                                  3.3%
                     Mother’s health problem                                                         19.4%

                     Source: Sandra Danziger et al., “Barriers to the Employment of Welfare Recipients,” Institute for Research on
                     Poverty, Discussion Paper no. 1193-99, June 1999.

                         Most of the families who remain on welfare             for pay diminishes rapidly, and those who
                     are headed by unmarried women under 30—a                   face two or more obstacles are highly unlike-
                     growing percentage of whom are African                     ly to engage in work activity of any sort.3 5
                     American or Hispanic.31 An increasing pro-
                     portion, as a recent Brookings Institution                 Summary
                     study shows, is concentrated in high-poverty                  It should be clear by now that the size of the
                     urban areas.3 2 HHS reports that 23 percent of             caseload decline alone is an inadequate gauge
                     adult recipients in 1998 were employed, with               of welfare reform’s success. More important
                     average monthly earnings of $553, and 7 per-               questions—why and whether individuals are
                     cent had unearned income averaging about                   actually exiting the system; whether the most
                     $229 per month. At the time of the survey, 45              needy have, in fact, benefited from the recent
                     percent of adult claimants were actively seek-             caseload declines; and what role the nation’s
                     ing work, and two of five were exempt from                 booming economy has played in enabling wel-
                     work participation requirements.3 3                        fare reform measures to take effect—often get
                         Studies of current welfare recipients suggest          lost when the talk turns to numbers.
                     that this population faces significantly more                 Most former recipients have voluntarily left
 Current welfare     impediments to work than did people who left               the welfare rolls under the new law because they
                     the rolls soon after passage of the new law.3 4            have found work; because of administrative
   recipients may    The Institute for Research on Poverty under-               hassles; because they no longer want or need
face significantly   took an extensive study of single mothers who              assistance; or, in a small number of cases,
    more impedi-     remained on welfare in 1997 and found that                 because of marriage. Few exits thus far have
                     around two-thirds of remaining recipients                  been initiated involuntarily by noncompliance
   ments to work     faced two or more obstacles to work. As Table 1            or exhaustion of time limits. If the pace of case-
 than did people     shows, of the barriers to working, low educa-              load decline continues to slow and the propor-
                     tion, lack of access to transportation, poor               tion of long-term cases remaining on the rolls
who left the rolls   health, drug dependence, a major depressive                continues to expand, however, it is likely that
  soon after pas-    disorder, and several experiences of workplace             involuntary exits will begin to increase signifi-
  sage of the new    discrimination were most likely to reduce                  cantly unless states choose to relax their sanc-
                     employment among nonleavers.                               tion and time limit policies.
              law.       As the Urban Institute observes in a simi-                Whether or not individuals are indepen-
                     lar study, as the number of obstacles increas-             dently choosing to leave the rolls, the
                     es, the percentage of TANF parents working                 nation’s caseload decline since enactment of

welfare reform remains impressive. As the                 sobering evidence is emerging that individu-       For most former
research shows, however, the economy                      als are making slow progress toward inde-          recipients, long-
deserves as much credit as do the new rules               pendence after exiting welfare. For most for-
for both bringing about and sustaining this               mer recipients, long-term employment and           term employ-
trend. Although the role of the economy in                upward mobility within the labor market            ment and upward
moving individuals off welfare appears to be              remain out of reach.
declining as fewer people remain who are
                                                                                                             mobility within
actually job ready, it is likely that sanctions           Participation in Work Activities                   the labor market
and work requirements would have a much                       Information on the types and extent of         remain out of
more limited impact on the size of the case-              work activities engaged in by former recipi-
load if jobs were not plentiful. That fact has            ents is more scarce than that on current           reach.
led critics to speculate about how the case-              TANF recipients, since states are not required
load trend might reverse itself in the face of            by the federal government to report on the
an economic turnaround. Because of the                    status of former recipients. This seems odd,
mutual interplay of welfare reform and the                given that a large portion of federal welfare
economy, there is good reason to believe that             dollars is intended for use by states to subsi-
a future period of dwindling prosperity                   dize strategies and programs that supposed-
could quickly lead to swelling caseloads.                 ly have long-term self-sufficiency goals. The
    Even if the economy remains healthy, critics          disparate figures that are available on former
contend that, unless benefits levels are cut and          recipients’ work activities often come from
tougher sanctions and time limits are enforced,           studies voluntarily initiated by the states.
the caseload declines of recent years will begin              The data that do exist indicate that a siz-
to slow because of the increasing proportion of           able proportion of welfare leavers who are
long-term recipients remaining on the rolls. In           supposedly engaged in “work activities” do
fact, as we have seen, such a leveling off—and, in        not actually have jobs.3 6 In the Urban
some cases, the beginning of an increase—is               Institute’s study of individuals who exited
already evident in the latest caseload figures.           the rolls between 1995 and 1997, for
Welfare reform appears to have been most suc-             instance, at the time of the interviews, 10 per-
cessful in creaming the short-term, easily                cent of the group were in school and 25 per-
employable cases from the rolls and has made              cent were actively seeking work.3 7 A more
comparatively little progress in moving long-             recent study on Wisconsin showed 21 per-
term recipients out of the system. The individu-          cent of former recipients not working and
als who remain on the rolls nearly four years             engaged in full-time training programs.3 8
after enactment of welfare reform tend to face                Nevertheless, 5 to 10 percent more former
multiple barriers to employment and have yet              recipients are managing to find unsubsidized
to reap the self-sufficiency gains at which the           jobs than under the old AFDC/JOBS pro-
1996 law aims. It appears that the new law has            gram.3 9Studies show, on average, two-thirds of
so far been more effective in forcing the depen-          former recipients having found unsubsidized
dence problem to the surface than in chipping             work at some point since exiting the rolls, and
away at it.                                               that number occasionally reaches 87 percent.4 0
                                                          The percentage of welfare leavers working tends
                                                          to vary widely by state, with Wisconsin showing
         Welfare to What?                                 especially high levels of work among TANF
                                                          leavers (between 66 and 83 percent),4 1 while
   In recent years, evaluations of the new wel-           states like New York and Tennessee report work
fare law have turned from “How many people                levels as low as 29 percent.42
are leaving the rolls?” to “Where are those                   Sanctioned former recipients appear to
people going?” As success begins to be mea-               have a consistently more difficult time secur-
sured more in terms of results than numbers,              ing unsubsidized work than do other former

                        recipients. Nationally, only 20 to 50 percent          low-skill minimum-wage workers.5 3 This
                        of sanctioned leavers tend to secure employ-           turns out, however, to be only mild cause for
                        ment after they exit welfare,4 3 and in                celebration. After several months on the job,
                        Wisconsin, where work rates tend to be the             a large proportion of welfare leavers begins
                        highest, only 42 to 53 percent of sanctioned           trickling off the payrolls either to return to
                        leavers find jobs.4 4                                  welfare or to search for new jobs. The Urban
                            Unsubsidized Employment. Considering the           Institute’s study of early welfare leavers
                        array of welfare-to-work programs and sup-             found that three-quarters of employed for-
                        port services available under the new law, we          mer recipients had worked at their current
                        might expect individuals to be finding more            jobs for less than one year, and one-third had
                        promising employment opportunities when                been at their jobs for fewer than six months.5 4
                        they leave the rolls than in the past. On the          Similar findings by the Committee for
                        contrary, the Washington Post reports, “There          Economic Development suggest that, on
                        is no reason to believe that the kinds of jobs         average, fewer than 25 percent of former
                        that . . . welfare recipients are getting today        recipients manage to stay in their jobs for an
                        are any different from the ones they have              entire year.5 5
                        already had.”45 Studies show the majority of               More important than former recipients’
     TANF leavers’      TANF leavers entering full-time jobs in food           tenure at their first jobs after welfare, howev-
   current jobs, as     service, retail trade, sales, manufacturing,           er, is their ability to use this experience to stay
          well as the   recreational services, hotels, and clerical            off welfare and to gradually move upward
                        duties. Those jobs are usually entry-level             within the labor market. Unfortunately,
   numerous sup-        positions with few educational experience or           TANF leavers’ current jobs, as well as the
  port services and     skill requirements.46                                  numerous support services and initiatives
                            Not surprisingly, many working former              that states offer, are failing to equip many
    initiatives that    recipients are not earning enough to lift              with the skills, experience, and opportunities
   states offer, are    themselves out of poverty through wages                needed to obtain more responsible and bet-
   failing to equip     alone,4 7 and many current recipients still            ter-paying positions and to become increas-
                        earn too little from their jobs to disqualify          ingly self-sufficient. On the one hand, cur-
    many with the       them for receipt of TANF benefits.48 An                rent research on upward mobility under the
 skills, experience,    Urban Institute study of early welfare leavers         new law does show modest signs of TANF
                        found the median wage in 1997 to be $6.61—             leavers’ earnings increasing over time. Eight
and opportunities
                        the 20th percentile of hourly wages for all            state studies funded by the HHS report
  needed to obtain      workers. Twenty-five percent of the early              steady increases in former recipients’ earn-
 more responsible       leavers were making less than $5.29 per                ings in every location over the course of the
                        hour.49 A 1999 General Accounting Office               year following exit.56 On the other hand,
and better-paying       survey found that annual earnings of former            however, there is little indication that these
  positions and to      recipients ranged from $9,512 to $15,144.5 0           earnings increases are helping former recipi-
       become self-     In addition to paying low wages, most former           ents to make significant advances toward
                        recipients’ jobs do not offer benefits pack-           self-reliance. Despite the evidence of some
          sufficient.   ages. A recent study by the Committee for              wage growth over time, earnings of former
                        Economic Development finds that only                   recipients remain low.57 In addition, access to
                        around 23 percent of employed former recip-            employer-provided benefits appears to
                        ients receive health care coverage through             increase only modestly over time.5 8 The
                        their employers.5 1In high-employment states           Committee for Economic Development
                        such as Wisconsin, only 16 percent of work-            speculates that the types of jobs in which
                        ing TANF leavers in 1998 had employer-                 TANF leavers currently work are responsible
                        sponsored health insurance.5 2                         for their slow progress in the job market and
                            On a positive note, former recipients actu-        continued difficulty in lifting themselves out
                        ally tend to stay employed longer than other           of dependence. The key problem, the com           -

mittee notes in a recent report, is that few for-        welfare-to-work appears to be generating few
mer recipients’ jobs offer opportunities for             or even negative results. Often, it is simply
training and advancement.5 9                             impossible to know the extent of programs’
   Subsidized or Supported Employment. As we             effectiveness because former participants are
have seen, many individuals engaged in “work             not tracked and studied.
activities” under the new law are still in the               In fairness, recent studies of welfare-to-
process of making their way into unsubsidized            work programs indicate that they can have
jobs. Various types of supported work schemes            positive impacts on maternal psychological
to help individuals advance into unsubsidized            well-being, parent-child interaction, and chil-
employment have emerged since states began               dren’s home environments when the family
experimenting with welfare reform, and the               environment is already relatively healthy.6 2
results have been mixed at best.                         Already volatile families and previously trou-
   Despite some disappointing outcomes                   bled mothers, on the other hand, notes the
before welfare reform, the new legislation               Urban Institute, tend to respond especially
channels a hefty amount of taxpayer funds                poorly to the new welfare-to-work programs.
into welfare-to-work programs. In the four               In troubled homes, such initiatives have pro-
years since enactment of the new law, nearly             duced negative impacts on maternal depres-
$9 billion has been expended on work-sup-                sion and stress levels and exacerbated already
port programs for current and former recipi-             unhealthy circumstances.6 3 The Manpower
ents.6 0 Welfare-to-work initiatives are sup-            Demonstration Research Corporation has
ported under the new law by money from                   noted, moreover, that most individuals who
states’ TANF block grants, state and local               do succeed in securing employment after
treasuries, a new welfare-to-work tax credit             participation in such programs end up in
for employers of long-term welfare recipients,           low-paying jobs with few benefits, have high
and the so-called welfare-to-work (WtW)                  turnover rates, and remain in poverty.
grants program. The sorts of initiatives cur-            Overall, welfare-to-work has not shown
rently operating with the help of WtW grant              much success in advancing self-sufficiency
money include wage subsidies in the public               and reducing welfare receipt.6 4
and private sectors, on-the-job training, job-               Workfare, employer subsidies, and supple-
readiness services, vouchers for job-readiness,          mental earnings programs are among the
placement and postemployment services,                   broadly implemented yet questionable strate-
community service and work experience, and               gies being used to encourage responsible
job-retention and supportive services. As of             behavior and unsubsidized employment
September 1999, WtW grants were flowing                  among current and former recipients. Fans of       Overall, welfare-
to about 700 local programs in 28 states and             workfare believe that unpaid work in return
territories at a cost of $2 billion annually. At         for a welfare check is an effective way of         to-work has not
this funding level, the WtW grants program               advancing self-sufficiency and making recipi-      shown much suc-
alone allocates nearly $1 billion more each              ents into responsible, work-oriented individu-     cess in advancing
year to welfare-to-work than was spent by                als. In truth, such programs have been largely
both the federal and state governments                   unsuccessful at realizing either of those objec-   self-sufficiency
together on welfare-to-work in 1996 under                tives. The Welfare Information Network notes       and reducing wel-
the JOBS program.6 1                                     that numerous studies have shown that
   Current calls for even greater investment             unpaid work “in almost every case did not
                                                                                                            fare receipt.
in welfare-to-work imply that such programs              result in increased earnings or steady employ-
have so far proven effective in alleviating wel-         ment for participants.”6 5 New York City’s
fare dependence. Republicans, in particular,             Work Experience Program, currently the
have advocated expanding welfare-to-work                 largest workfare program in the country, has
and finding methods to increase enrollment               achieved notoriously poor results in equip-
in those programs. Yet, on some measures,                ping participants with marketable skills and

Subsidized work     helping them to find full-time unsubsidized             to be only mildly successful with a limited
    as a route to   employment.6 6In return for cash benefits, par-         group of participants. Most grant diversion
                    ticipants are put to work cleaning parks,               programs are characterized by low enroll-
   unsubsidized     sweeping streets, answering phones, and rub-            ment due to both employers’ reluctance to
    employment      bing graffiti off walls. Not surprisingly, in           hire current and former recipients and to a
                    1998 fewer than one-tenth of the 125,000 peo-           meager client base. In its examination of
   appears to be    ple who had passed through the program had              Work First strategies in five states, for
     only mildly    reported finding permanent jobs, and nearly             instance, the Urban Institute found that
successful with a   two-thirds had left the program and given up            employers refrained from offering subsidized
                    benefits without any indication of employ-              work because of both the administrative bur-
limited group of    ment.6 7                                                den and the stigma attached to hiring welfare
    participants.       For those capable of more than unpaid               recipients.6 8
                    community service but lacking in work experi-               More important, only a small group of
                    ence, education, or certain basic skills, states        current and former recipients is likely to take
                    have encouraged subsidized employment as a              advantage of subsidized work. On the one
                    path to full-time, unsubsidized work. “Wage-            hand, the Urban Institute discovered that
                    based community service” or “supported                  “those who were good candidates for subsi-
                    work” jobs involve time-limited work at a pri-          dized employment generally could find (and
                    vate, public, or nonprofit organization in              preferred) unsubsidized jobs.”6 9On the other
                    return for taxable wages that are funded                hand, the long-term dependent individuals
                    through diversion of individuals’ welfare               who are the targets of many such initiatives
                    grants. Participants are generally paid mini-           are often unemployable. When the Brookings
                    mum wage, qualify for the Earned Income Tax             Institution interviewed employers offering
                    Credit, participate in training activities, and         subsidized work, many indicated that the
                    have access to comprehensive support ser-               individuals they were serving “had such poor
                    vices, including child care and transportation.         education and skill levels that the clients
                    Most programs also provide participants with            were often unable to hold even the simplest
                    a mentor or job counselor to facilitate transi-         entry-level jobs. Severe and often multiple
                    tion to unsubsidized employment. On-the-job             deficiencies in life skills, education, and job
                    training subsidies and employer tax credits are         skills combined with unstable living arrange-
                    also available to employers to encourage the            ments and children with health problems
                    hiring of current and former recipients,                threatened the ability of these recipients to
                    which will supposedly lead to full-time                 ever become economically self-sufficient . . .
                    unsubsidized work and upward mobility.                  many clients could work in a supported-
                    Some of states’ more noteworthy subsidized              work environment over the long term, but
                    employment initiatives under the new welfare            would have great difficulty meeting the
                    law include Wisconsin Works’ Trial Jobs,                demands of a normal private sector employ-
                    Florida’s WAGES Work Supplementation,                   er.”7 0 Particularly in an economy where jobs
                    Chicago’s Project Match, San Francisco’s                are plentiful, current and former recipients
                    Wage-Based Com     munity Service Pilot, and            have divided into two categories: those work-
                    Philadelphia@Work. Additional well-known                ing in low-paying, unskilled, unsubsidized
                    organizational approaches to subsidized work            jobs and those who face multiple barriers
                    include Goodwill Industries of Southern                 and either have been exempted from the
                    California and Marriott International’s Pathways        work rules or are engaged in dead-end work-
                    to Independence.                                        fare activities. For the few who fall in
                        Although anecdotal evidence of success-             between, subsidized work has proved moder-
                    ful job placements through those programs               ately successful at raising earnings and job
                    abounds, on the whole, subsidized work as a             retention, but its impact on the dependent
                    route to unsubsidized employment appears                population remains relatively limited.

    A final and increasingly popular strategy            costs like the $7,200 per participant for
for moving people from welfare to work is                Milwaukee’s New Hope, we need to ask
supplemental earnings programs. Those ini-               whether supplemental earnings programs are
tiatives use an array of state supports to               merely increasing the welfare burden and
encourage individuals to work. Such sup-                 extending the life of benefits or are, in fact,
ports include anything from child care to                generating long-term savings and indepen-
transportation assistance to mentoring to                dence among participants. As a skeptical
continuation of cash benefits for a specified            Robert Rector warned about Minnesota’s pro-
time period while the recipient is working               gram, “The financial incentives in Minne-
full-time. A number of supplemental earn-                sota’s plan ultimately could harm beneficia-
ings programs have been tested, and many                 ries by perpetuating welfare dependence.”7 2
continue to operate around the country.                  The evidence also suggests that the people
Milwaukee’s highly publicized New Hope                   who manage to reap benefits from programs
program, tested between 1994 and 1997,                   like New Hope tend to be participants who
guaranteed low-income, full-time working                 have few obstacles to overcome. Enrollees
participants an income above the poverty                 often participate voluntarily and express the
line and access to subsidized child care,                desire and have the ability to work full-time.7 3
health insurance, and, if necessary, a paid              Thus, even if we were to obtain evidence that         We need to ask
community service job. Minnesota’s Family                most participants moved into self-sufficiency         whether supple-
Investment Program, piloted in seven coun-               after expiration of their supplemental bene-          mental earnings
ties between 1994 and 1998, gave partici-                fits, we would have to wonder if those people,
pants who were working full-time the oppor-              the cream of the crop, would have achieved the        programs are
tunity to continue receiving some portion of             same results on their own without the added           merely increasing
their benefits checks until their incomes                supports.
reached 140 percent of the poverty level (in                 The lessons are still coming in about what
                                                                                                               the welfare bur-
contrast to the old AFDC program that with-              does and does not work in helping individuals         den and extend-
drew the entire check when earnings reached              to move from welfare to work. The fact that we        ing the life of
85 percent of the poverty level).7 1 Current             do not fully know the effects of current pro-
participants in Chicago’s Project Match who              grams, combined with the fact that many of            benefits or are, in
work at least 25 hours per week continue to              today’s efforts were notoriously ineffective          fact, generating
receive partial cash assistance, as well as men-         before welfare reform, should make policy-
                                                                                                               long-term savings
toring, individualized employment plans,                 makers wary of increased investment in and
and a range of follow-up services.                       expansion of welfare-to-work as a strategy for        and indepen-
    It is not at all surprising that participants        self-sufficiency under the new law.                   dence among
report gains in earnings and employment and
decreased receipt of benefits while on supple-           Nonemployment                                         participants.
mental earnings programs. Those results,                     When policymakers boast of welfare reform’s
however predictable and obvious, are what                successes, they rarely mention either the 30 to 40
evaluators are citing as evidence of “success.”          percent of former recipients who do not manage
Progress should instead be measured accord-              to find work after leaving the rolls or the 25 per-
ing to whether the programs are giving indi-             cent whom studies have found have no working
viduals a more solid foundation from which               member in their family.74
to achieve self-sufficiency when the program                 Most studies provide only limited infor-
ends. Unfortunately, there exists little, if any,        mation about the circumstances of and rea-
information about what happens to partici-               sons for failure to find work after welfare.
pants after their supplemental earnings                  The data that do exist indicate that the most
expire. Few studies examine whether individu-            common reasons for staying out of work
als manage to remain out of poverty and are              include lack of access to jobs or work sup-
able to stay employed and off the rolls. Given           ports, inability to find jobs that pay enough,

                      lack of child care, lack of transportation or             not employed but are instead engaged in
                      distance from available jobs, involvement in              education or training, or are merely seeking
                      nonwork activities, caring responsibilities at            work. Many of those individuals are in work
                      home, school, lack of need or desire to work              experience programs that offer few opportu-
                      (possibly because of receipt of other types of            nities to gain meaningful skills and lead few
                      benefits), and illness or injury.7 5                      participants into unsubsidized employment.
                          Without jobs, most former recipients are              Other recipients are engaged in job initiatives
                      forced to depend on less-reliable sources of              that offer artificial earnings supplements but
                      income such as the earnings of a working                  show little evidence of helping to increase
                      spouse or partner, Social Security, Supplemen-            self-sufficiency after the program ends.
                      tal Security Income (disability) benefits, unem  -            On a positive note, more former recipients
                      ployment compensation, child support, and                 are managing to secure unsubsidized employ-
                      other government supports such as food                    ment than did under AFDC, yet their jobs con-
                      stamps.7 6 Most are no closer to self-sufficiency         tinue to be the same low-skilled, low-paying,
                      than when they were receiving cash assistance.            no-benefits jobs secured by their predecessors.
                      A number have simply transferred to another               Moreover, few manage to keep their jobs for
                      cash benefit program but are formally consid-             an entire year, and few are given opportunities
                      ered “off welfare.” In the Urban Institute’s study        for advancement.
                      of individuals who departed the rolls between                 Worst off are the 30 to 40 percent of former
                      1995 and 1997, for instance, 23 percent report-           recipients who are not engaged in any form of
                      ed that they or a family member were currently            work activity. Most of those people continue
                      receiving Supplemental Security Income                    to rely heavily on other means-tested govern-
                      (SSI).77 A recent Wisconsin study showed an               ment programs for support.
                      astounding 47 percent of nonworking former
                      recipients receiving some amount of SSI.7 8
                      Though those figures do not generally reveal                     The Self-Sufficiency
                      the extent to which the other forms of govern-                       Challenge
                      ment assistance constitute a primary means of
                      support for TANF leavers, they do indicate that               Simply getting people off the rolls and
                      a significant proportion of nonemployed for-              into work, as we have started to see, does not
                      mer recipients remains at least partially trapped         necessarily imply real self-sufficiency of for-
                      in a cycle of dependence.                                 mer recipients. As one caseworker puts it: “If
                                                                                success is measured in a decreasing number
    Most former       Summary                                                   of public assistance cases, then welfare
                          It is dangerous to draw conclusions about             reform is certainly succeeding. But if success
recipients are no     welfare reform’s success from numbers alone.              is measured in client self-sufficiency, then the
    closer to self-   The same hazard exists when it comes to post-             experience of my caseload suggests we have a
 sufficiency than     welfare outcomes. Knowing, for instance, that             way to go.”7 9
                      a large proportion of former recipients even-                 A telling measure of whether or not welfare
  when they were      tually ends up working tells us nothing about             reform is causing more individuals to become
   receiving cash     the types of jobs they are getting and whether            self-reliant is the extent to which former recipi-
                      such employment is proving to be a path out               ents are able to meet their basic needs without
       assistance.    of dependence. Focusing attention exclusively             having to depend on additional governmental
                      on working TANF leavers also obscures the                 assistance. As the Urban Institute asks, “To
                      fact that one-third of former recipients do not           what extent do . . . leavers rely on work and
                      enter the labor market.                                   earnings to support themselves?”8 0
                          In examining the work activities of cur-                  Unprecedented numbers of recipients
                      rent and former recipients, we see that a sig-            have left welfare for work in the days since
                      nificant number of those who are “active” are             enactment of welfare reform, but it appears

that very few of those individuals are relying           part of welfare reformers’ multifaceted effort    Unprecedented
on their own wages and family earnings                   to “make work pay.” By “delinking” Medicaid       numbers of recip-
alone for subsistence. In 1999 the nation                from eligibility for TANF, the new law has
spent $80 billion on the Earned Income Tax               made it possible for individuals to continue      ients have left
Credit and other benefits like child care for            receiving government health insurance out-        welfare for work
low-income workers.8 1 Former recipients                 side the cash assistance system. In addition,
make up a significant proportion of the ben-             the new law permits families who were receiv-
                                                                                                           in the days since
eficiaries of such spending. The Urban                   ing Medicaid on welfare, but whose incomes        enactment of
Institute found that, in the first three                 disqualify them for Medicaid after they exit      welfare reform,
months off welfare, two-thirds of leavers                the rolls, to continue receiving transitional
report use of one or more of the following:              Medicaid assistance (TMA) for up to 12            but it appears
child care assistance, Medicaid, emergency               months. Families lose their eligibility for       that very few of
aid, assistance with transportation, and assis-          TMA when there is no longer a dependent
                                                                                                           those individuals
tance in meeting other work expenses.8 2                 child in the home, they fail to pay premiums
Similarly, a multistate study performed by               or to report their income quarterly, the care-    are relying on
researchers at the Institute for Research on             taker recipient has no earnings, or the fami-     their own wages
Poverty at the University of Wisconsin,                  ly’s average gross monthly earnings, minus
Madison, shows that around two-thirds of                 child care costs, exceed 185 percent of the       and family earn-
leavers receive some type of benefit in their            poverty level. The states, however, have tend-    ings alone for
first year after leaving welfare.8 3                     ed to take a more liberal approach and have       subsistence.
    To a growing number of critics, self-suffi-          suspended or relaxed many of the eligibility
ciency of former recipients is little more than          criteria for TMA.
an illusion. The array of supplemental bene-                 States vary widely with regard to the pro-
fits available to leavers of welfare, in particu-        portion of former welfare recipients continu-
lar, seems to suggest that the new law has               ing to claim Medicaid. Between 1995 and
succeeded merely in creating a sort of “work-            1997, for instance, from 4 to 94 percent of all
ing” welfare state. As one frustrated observer           former recipients made Medicare claims.8 5 A
puts it, “Replacing one open-ended entitle-              number of states on the lower end of this
ment with another does not serve self-suffi-             spectrum report that welfare leavers are fail-
ciency, welfare reform’s real goal.”8 4 A look at        ing to claim the Medicaid benefits for which
five categories of supplemental assistance               they are eligible. Since enactment of the new
provides important insights into how those               law, Medicaid claims have declined sharply.
forms of government “welfare” are perpetu-               As Figure 7 shows, between 1995 and 1997,
ating former recipients’ entanglement with               12 states experienced declines of 10 percent
the public safety net.                                   or more, and only 4 states saw Medicaid
                                                         enrollment increase by more than 5 per-
Medicaid                                                 cent.8 6 Among children, Medicaid receipt
   After exiting the rolls, a substantial por-           declined by almost 10 percent between 1996
tion of former recipients satisfy their health           and 1999, 87 although there remain more chil-
care needs through Medicaid and the                      dren of former TANF families receiving
Children’s Health Insurance Program; com       -         Medicaid coverage than heads of families.8 8
paratively few are self-insured or covered by                If the former TANF families who are fail-
their employers. Ideally, if self-sufficiency is         ing to claim Medicaid were doing so because
the goal, individuals will be moving away                they had secured alternative forms of health
from reliance on government-funded health                coverage, then the news of underclaims
care toward reliance on private insurance                would be welcome. A significant body of
after they exit the welfare rolls.                       research, however, suggests that increasing
   Medicaid has been made widely available               self-sufficiency is not the primary reason for
to former recipients under the new law as                former recipients’ failure to sign on to

Figure 7
Changes in Medicaid Enrollment, 1995–97
     Wisconsin          -19
  Pennsylvania            -17.4
           Ohio               -15.9
     Wyoming                      -14.5
        Hawaii                     -13.9
   Mississippi                      -13.4
        Florida                     -13.4
        Oregon                          -13
        Nevada                            -11.5
          Idaho                             -10.8
        Kansas                               -10.5
     Kentucky                                 -10.1
      Colorado                                  -9.6
     New York                                    -9.4
     Oklahoma                                     -9.3
        Indiana                                      -9
      Michigan                                     -8.9
         Texas                                      -8.6
      Montana                                       -8.6
        Virginia                                     -8.1
  North Dakota                                          -8
     California                                        -7.6
     Maryland                                           -7.2
      Louisana                                           -7.1
        Illinois                                          -6.8
           Iowa                                             -5.8
      Missouri                                                 -5.2
           Utah                                                   -5
    New Jersey                                                  -4.9
        Georgia                                                  -4.6
         Maine                                                    -4.2
       Alabama                                                    -4.2
 North Carolina                                                          -1.5
   Connecticut                                                               -1
     Minnesota                                                             -0.8
  South Dakota                                                              -0.5
      Arkansas                                                              -0.5
     Tennessee                                                               -0.3
        Alaska                                                                      0.2
New Hampshire                                                                         1.2
 Massachusetts                                                                        1.2
 South Carolina                                                                          2.4
       Vermont                                                                            2.6
       Arizona                                                                            2.8
      Nebraska                                                                                      5.1
    Washington                                                                                                    13.2
   New Mexico                                                                                                      13.7
      Delaware                                                                                                                             26.2

               -25                                  -10                                         5                             20                           35
                                                                                       Percentage Decline
Source: U.S. General Accounting Office, “Medicaid Enrollment: Amid Declines, States’ Efforts to Ensure Coverage after Welfare Reform Vary,”
GAO/HEHS-99-163, September 1999.
Note: GAO was unable to obtain comparable Medicaid enrollment data for the District of Columbia, Rhode Island, and West Virginia.

                                        Medicaid. More commonly, TANF leavers                                  dependence on the government with reliance
                                        have responded in surveys that they were                               on the private sector. “Over time,” the HHS
                                        unaware that they could work full-time, not                            notes, “as eligibility for Medicaid expires, more
                                        claim TANF benefits, and still receive                                 clients rely on private insurance. However, more
                                        Medicaid;8 9eligibility rules were too complex;                        also go uninsured.”92
                                        caseworkers did not inform them of their eli-
                                        gibility for Medicaid; there were too many                             Food Stamps and Nutrition
        Increasing                      administrative hassles and requirements; the                               Another important measure of self-suffi-
 self-sufficiency is                    separate application procedure was a burden;                           ciency is whether individuals are becoming
                                        or there was too much of a stigma attached                             self-reliant in terms of meeting their nutri-
  not the primary                       to Medicaid receipt.9 0                                                tional needs after leaving the rolls. Studies
reason for former                           The Medicaid claims of people who receive                          show that, on average, one-half of former
                                        Medicaid after leaving welfare tend to decrease                        recipients rely on food stamps.9 3 While siz-
recipients’ failure                     sharply over time. Studies show that, whereas                          able, that figure represents around a one-
      to sign on to                     the percentage of TANF leavers claiming                                third decline in the number of food stamp
         Medicaid.                      Medicaid increases throughout the first year                           claimants since January 1996 (Figure 9), sug-
                                        after exit, it drops threefold between individu-                       gesting that perhaps more individuals are
                                        als’ first and third year off welfare.91 Falling                       coming to rely on themselves to meet their
                                        claims, however, as Figure 8 shows, do not nec-                        basic nutritional needs.9 4
                                        essarily indicate that individuals are replacing                           In fact, many people who are no longer

Figure 8
Health Insurance Status of Former Welfare Recipients

                                               After 1 year off rolls

                                               After 3 years off rolls





                      Medicaid               Employer/Private Coverage             Uninsured

Source: Amy Johnson and Alicia Meckstroth, “Ancillary Services to Support Welfare to Work: Lack of Health
Insurance,” U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and
Evaluation, June 22, 1998,

claiming food stamps have become ineligible                 we are likely to witness increased dependence
because of increased earnings or a new job.9 5              on this part of the safety net in years to come.
According to a recently released report by the
U.S. Department of Agriculture, around one-                 Housing Assistance
third of the decline in the food stamp rolls                    Difficulties with housing after welfare are
that has occurred since states began experi-                common among TANF leavers. Given their
menting with welfare reform is attributable                 low earnings, it is not surprising that a num -    Policymakers
to changing economic conditions and indi-                   ber of families continue to rely on public         have been brain-
viduals’ escaping poverty.9 6 Similarly, the                housing assistance after they exit the rolls.
GAO reported last year that employment was                      Information on the percentage of TANF          storming ways to
the primary cause of falling food stamp                     leavers turning to the government for help         increase enroll-
claims.97                                                   with housing is scarce and comes mainly            ment in the food
   As the Urban Institute has pointed out,                  from individual state studies. We know, for
however, there also exists a sizable group of               instance, that around 25 percent of former         stamp program.
former recipients who, on the basis of their                recipients depend on housing assistance in         If they are suc-
earnings levels, have retained eligibility for              states like Wisconsin. 101 TANF leavers’
food stamps yet are failing to claim them.9 8               reliance on rent subsidies or public housing,
                                                                                                               cessful, we are
Some studies note that some potential recip-                however, can provide only an indication of         likely to witness
ients are unaware of their eligibility for food             the extent to which individuals are failing to     increased depen-
stamps,9 9 and others cite administrative                   become independent in this area because of
problems and hassles as a major deterrent to                the limited availability of housing assistance.    dence on this
potential food stamp claimants.1 0 0 In                     The fact that studies tend to show between         part of the safety
response, policymakers have been brain-                     one-third and one-half of former recipients
storming ways to increase enrollment in the                 experiencing problems with housing and
                                                                                                               net in years to
food stamp program. If they are successful,                 utilities indicates that individuals’ inability    come.

                       Figure 9
                       Decline in Food Stamp Participation






                            1996                                              1997                                              1998

                       Source:Laura Castner and Jacquelyn Anderson, “Characteristics of Food Stamp Households: Fiscal Year 1998,” U. S.
                       Department of Agriculture, Food and Nutrition Service, Office of Analysis, Nutrition and Evaluation, July 1999.

                       to meet such needs independently is more                     to differences in the local supply of care that
                       pervasive than the reliance on public subsi-                 is affordable and available.”1 0 3
                       dies would suggest. What surfaces, as a result,                  Greater demand for child care due to
                       appears to be another major deficiency on an                 increased employment is a development
                       important measure of self-sufficiency.                       worth celebrating. Dependence on the govern-
                                                                                    ment as chief supplier of child care, however,
                       Child Care Subsidies                                         represents stagnation in moving individuals
                          A fourth need that many former recipi-                    out of dependence. While the half of former
Former recipients      ents are failing to satisfy independent of the               recipients, on average, who are relying on gov-
                       public safety net is child care. Studies show,               ernment to finance their child care may be less
   who are relying     on average, 30 to 50 percent of welfare leavers              entangled in the welfare system than they were
on government to       relying on public child care subsidies.                      when they were receiving cash assistance, it is
      finance their    Dependence on child care benefits varies                     important to recognize that those individuals
                       widely across the country, however, with as                  have still not achieved independence of the
    child care have    few as 13 percent receiving subsidies in                     public safety net.
 still not achieved    Connecticut and as many as 70 percent
                       claiming assistance in states like Florida.1 0 2             Earned Income Tax Credit
 independence of       Much of the reason for the variation has to                     The recently expanded Earned Income
  the public safety    do with differences in the availability of sub-              Tax Credit, in addition to the supplemental
                net.   sidies, access to alternative informal care, and             benefits detailed above, plays a measurable
                       the cost of both formal and informal care. As                role in raising the incomes of former welfare
                       the HHS emphasizes, “There are differences                   recipients who are working. Whereas families
                       in how states have earmarked and allocated                   with two or more children were receiving tax
                       funding for child care, which will contribute                credits of 17 percent on earned income up to

$7,140 in 1991, by 1997 such families were                                                                        The new law has
eligible for tax credits of 40 percent on earn-                Worse for the Wear:                                made moderate
ings up to $9,140.1 0 4 That expansion of the                 Are Families Better Off
EITC thus allows many former recipient fam     -                                                                  progress in elimi-
                                                                  after Welfare?
ilies with two or more children to increase                                                                       nating the disin-
their incomes by almost $4,000.105 All in all,                Advocates for the poor warned before
notes the Urban Institute, the EITC is                    enactment of PRWORA, and cite growing
                                                                                                                  centive to leave
responsible for well over half the income gain            evidence since the law went into effect, that           welfare. Most for-
enjoyed by families who move from part-time               welfare reform actually makes low-income                mer recipients
to full-time minimum-wage work.106 In addi-               individuals worse off by removing the entitle-
tion to the federal EITC, many states also                ment to cash benefits. Critics have attempted           have maintained
offer their own supplementary EITCs. By the               to link changes in the welfare law with                 or increased their
end of 1999, the Center for Budget and                    increases in poverty, homelessness, hunger,
                                                                                                                  general well-
Policy Priorities reports, 11 states were offer-          loss of health insurance, and decreases in the
ing EITCs based on the federal credit.107                 general well-being of low-income individuals            being in moving
                                                          and their families.                                     from welfare to
Summary                                                       If such criticisms are valid, then policymakers
    When policymakers spoke of “ending wel-               will have not only failed to help former recipients     work.
fare as we know it” in 1996, they had far more            become self-reliant but will have achieved the very
in mind than simply shrinking welfare rolls.              opposite by making individuals worse off after
The long-term aim of the bill was to end the              welfare. Part of the motivation behind the 1996
country’s dependence problem by getting                   law was to remove the perverse disincentives built
recipients into jobs and on the path to self-             into the system that encouraged people to enter
sufficiency. What we see four years later,                and stay on welfare and to replace those disincen-
unfortunately, bears less resemblance to self-            tives with new system of incentives to seek work
sufficiency than to continued entanglement                rather than assistance. What critics are asserting is
with the system in the form of “supplemental              that, in fact, “work does not pay” for many who
welfare dependence.”                                      exit the system.
    Very few former recipients know true self-                The evidence reveals that the new law has
sufficiency under the new law. Most rely on               made moderate progress in eliminating the dis-
some form of means-tested government assis-               incentive to leave welfare. Most former recipi-
tance to help them meet their medical, food,              ents have maintained or increased their general
housing, or child care needs. The incomes of              well-being in moving from welfare to work.
most working welfare leavers, moreover, are sup-
plemented by generous federal and state EITCs.            Financial Well-Being
Overall, few can say they have surpassed the                 Advocates for the poor predicted before
need for state aid, and, down the road, predicts          welfare reform that tightening eligibility
the Upjohn Institute at the University of                 rules and cutting people from the rolls would
Michigan, “former welfare recipients will con-            cause dramatic increases in poverty among
tinue to need government income supplements               low-income families. The advocates reasoned
if they are to support their family at incomes            that people would not be able to compensate
above the poverty line.”108 Observers of welfare          for their loss of cash benefits because they
reform too often believe that TANF leavers are            had no—or low-paying—jobs. In addition, the
becoming self-reliant because they are earning            advocates predicted that leavers would expe-
more and having their needs met. In a more                rience increased financial hardship as the
thoughtful assessment of welfare reform, we               result of the loss of noncash benefits such as
have to ask, What have we achieved in the war             food stamps and Medicaid.
on dependence if those needs are continuing to               However, a majority of former recipients
be satisfied by reliance on the public safety net?        have managed to secure employment after

                      Figure 10
                      Child Poverty Rate (percentage)





                                   1979       1983       1989        1993         1994        1995     1996      1997       1998

                      Source: Unpublished Census Bureau data given in Katheryn Porter and Wendell Primus, “Recent Changes in the
                      Impact of the Safety Net on Child Poverty,” Center on Budget and Policy Priorities, December 1999.
                      Note: The child poverty rate is based on income that includes social insurance, cash assistance, food and housing
                      benefits, and EITC, less federal taxes.

                      leaving welfare, the earnings of most are subsi-            expanded EITC make work a more attractive
                      dized by a generous combination of state and                option than it was under AFDC. The Wall
                      federal EITCs, and few have lost their eligibili-           Street Journal points out:
                      ty for noncash benefits. Given that, it is hardly
                      surprising that poverty has continued to fall                      The new work-support system means
                      every year since welfare reform was enacted.109                    that mothers who take minimum-
                      With the help of a booming economy, more                           wage jobs are far better off than moth-
                      than 2 million individuals were able to rise out                   ers on welfare. Even under the official
                      of poverty between 1996 and 1998.1 1 0                             definition, poverty has been declining
                      Dependent single mothers and their children                        since 1994—the year the great exodus
Between 1993 and      account heavily for this decline. Between 1993                     from welfare began. But this definition
                      and 1998, the poverty rate for female-headed                       does not count most of the benefits
 1998, the poverty    families with children fell from 46 to 39 per-                     provided to workers through the new
   rate for female-   cent,111 and by 1998, as Figure 10 shows, the                      work-support system. If we use a
  headed families     child poverty rate was the lowest it had been in                   Census Bureau measure that counts
                      nearly 20 years.1 1 2 Perhaps most notable, the                    these benefits, we find striking evi-
 with children fell   number of black children in poverty has fallen                     dence that welfare reform is reducing
from 46 to 39 per-    since enactment of welfare reform at the great-                    poverty more than ever before.115
                      est rate ever recorded.113
cent, and by 1998,        In examining changes in poverty since                      The extent to which former recipients appear
 the child poverty    passage of the new law, the Committee for                   to profit in moving from welfare to work
 rate was the low-    Economic Development concludes, “There is                   depends on what benefits researchers choose to
                      little indication that welfare reform has made              include when calculating the welfare benefits
 est it had been in   the problem worse.”1 1 4 As numerous studies                package. Studies that take into account only
   nearly 20 years.   have shown, an array of supplemental bene-                  cash assistance and food stamps, such as a
                      fits, the increased minimum wage, and an                    recent GAO review of seven states, suggest that

former recipients’ earnings alone often exceed             by the new law, the experience of both financial
their TANF benefits packages.116 Other studies,            and psychological hardship should have become
such as one conducted by the Cato Institute,               less familiar. As the Washington Post puts it, the new
indicate that the gap between an individual’s              welfare rules “have magnified the strengths of
wages on and off welfare is not nearly as impres-          those who were strongly motivated and able to
sive when the entire welfare benefits package is           take advantage of new programs. For this group
taken into account. In fact, when the typical wel-         with the psychological stamina to push them          -
fare package, including cash benefits, food                selves, the new system has provided job training,
stamps, Medicaid, public housing, nutrition                child care, and other resources.”120
assistance, and utility assistance is considered,              Surveys of former recipients indicate that
recipients in most states would have to be mak-            quality of life for those who are working and
ing more than $8 an hour (and significantly                receiving supplemental benefits has, at the
more than that in a handful of states) to be               least, not deteriorated or has improved since
doing better in work than on welfare.117                   leaving welfare. Studies by Wisconsin and
    Nevertheless, while most working former                South Carolina show a majority of TANF
recipients are not earning the cash equivalent of          leavers disagreeing with the statement “life was
their full welfare packages as calculated by the           better when you were on welfare,”121 and nearly
Cato Institute, the majority are able to supple-           70 percent in Wisconsin agreeing that “getting           In all states a
ment their incomes from work with the EITC                 a job was easier than living on welfare.”122             woman increases
and an array of noncash benefits. As the Urban             Quality-of-life surveys also show more positive          her income if she
Institute has shown, in all states a woman increas-        responses from former recipients who are
es her income if she leaves welfare for full-time          engaged in full-time, year-round—as opposed              leaves welfare for
work at a minimum-wage job—so long as she                  to part-time or irregular—work. Such individu-           full-time work at
receives other supplemental benefits for which             als, as a University of Michigan study shows,
she is eligible, including food stamps, Medicaid,          experience fewer increases in financial hardship
                                                                                                                    a minimum-wage
child care subsidies, and the EITC.118                     after leaving the rolls and tend to feel more            job—so long as
    The people who fare the worst under the new            hopeful about their future well-being.123                she receives other
law tend to be nonworking families who have                    The families and children of most working
had their cash assistance reduced or removed;              leavers also appear to be no “worse for the              supplemental
both working and nonworking families who, for              wear” since exiting the rolls. Many critics pre-         benefits for
one reason or another, are not claiming the sup-           dicted that increased stress, lower incomes,
                                                                                                                    which she is
plemental assistance available to them; and fam   -        and more parental time away from home
ilies moving off welfare into unemployment or              would cause former welfare families to disin-            eligible.
low-wage jobs in high-benefit states. It is, with-         tegrate. The GAO reviewed data gathered by
out question, tragic that some of the worst-off            seven states and discovered that, in fact, such
families have seen their situations deteriorate in         fears had not been realized for the majority. It
recent years.119 That trend, however, does not             found little evidence of increased abuse or
necessarily imply the failure of welfare reform.           neglect that had led to involvement with child
Welfare reformers’ promise, after all, has been to         protective services or placement of children in
“make work pay,” as well as to continue to sup-            foster care, or both.124 As the Urban Institute
port those who cannot engage in work activities.           observes, “Children in certain subgroups will
For those who are complying with the new                   benefit from welfare reform to the extent that
requirements and taking advantage of the sup-              new policies succeed in moving parents into
ports offered by the new system, that promise is           jobs and increasing economic resources for
largely being realized.                                    families; bringing about greater and more pos-
                                                           itive father involvement (both economic and
Individual and Family Hardship                             social) in children’s lives; placing children in
  For the bulk of former recipients who are                care settings that are safe, stimulating, and
employed and using the work supports offered               supportive; and reducing family size.”125

                     Table 2
                     Odds of Child Mistreatment Allegations by Sanction Status

                     Type of Child                    Probability                    Probability           Ratio
                     Abuse and Neglect                of No Sanction                 of Sanction           (sanction/no
                     Allegation                       in 12/95                       in 12/95              sanction)

                     Lack of supervision                 .0085                        .0156                    1.84
                     Risk of harm                        .0065                        .0112                    1.72
                     Physical abuse                      .0042                        .0059                    1.41
                     Sexual abuse                        .0024                        .0032                    1.35
                     Environmental neglect               .0047                        .0059                    1.26
                     Parental substance abuse            .0049                        .0056                    1.14
                     Other neglect                       .0019                        .0021                    1.13

                     Source: Kristin Shook, “Assessing the Consequences of Welfare Reform for Child Welfare,” Joint Center for
                     Poverty Research Poverty Research News 2, no. 1 (Winter 1998): 5.

                         A certain subgroup of former recipient fami-           with the state welfare agency are reportedly 53
                     lies is, however, experiencing increased overall           percent higher than for other former TANF
                     hardship since exiting the rolls. As the Washington        families.130 The quantified effects of sanctions
                     Post puts it, “New welfare rules have exacerbated          on child well-being, as calculated by the Joint
                     the problems of families whose lives were already          Center for Poverty Research, are given in
                     chaotic. . . . The same regulations that are push-         Table 2. Other studies cite increasing domestic
                     ing some families toward self-sufficiency are              violence as a result of women being forced off
                     causing other, more troubled families to unrav-            the rolls and returning with their children to
                     el.”126 Families whose benefits have been termi-           battering husbands and boyfriends.131
                     nated and who are afflicted by substance abuse,
                     mental illness, domestic violence, or physical ill-        Satisfaction of Basic Needs
                     ness appear to be most at risk after leaving the               Most former TANF recipients remain
                     rolls.127 Income can also be a strong determinant          dependent on the public safety net to meet
                     of the extent to which former welfare families             their basic needs. It is precisely because of
    Most former      experience increased hardship under the new                continued dependence on both the govern-
                     law. As the Urban Institute finds, the quality of          ment and an alternative array of family and
TANF recipients      the home environment for young children of for-            community supports, however, that most are
  remain depen-      mer welfare families can decline when mothers              likely to be meeting their needs as well as or
dent on the pub-     begin jobs that are low wage and unstimulat-               better than they did as recipients of welfare
                     ing.128 Children in sanctioned families are usual-             Health Care. Former recipients’ Medicaid
 lic safety net to   ly the worst off because of the maternal stress            claims since enactment of welfare reform have
 meet their basic    and decreased economic resources that often fol-           been decreasing dramatically without a corre-
                     low termination of benefits.129                            sponding increase in enrollment in private or
           needs.        Some of the most notable hardships                     employer-provided health insurance.132 The
                     encountered by the families that are the worst             result is that most studies show the percentage of
                     off after leaving the rolls include increased              leavers who lack health coverage after welfare at
                     domestic violence and child abuse and neglect.             around 40 to 45 percent.133 Access to health
                     As usual, sanctioned TANF leavers appear to be             insurance also appears to deteriorate in the short
                     most at risk. As an Illinois study finds, sanc-            to medium term. Between the first and third year
                     tioned families’ odds of having a child placed             off the rolls, the number of former recipients

enrolled in Medicaid declines threefold, while the        examine whether there is increased reporting             Being uninsured
number of those who are uninsured doubles.134             of general difficulties in this area since the           does not neces-
    How has decreased access to health insur-             former welfare recipients left the rolls and
ance after leaving the rolls played out in terms          whether more former recipients are having to             sarily correlate
of the actual physical health of former recipi-           turn to food pantries and other community                with lack of
ents? Does increased lack of insurance really             resources for emergency food assistance.
mean that TANF leavers are worse off? In the              Both measures prove to be inadequate
                                                                                                                   access to health
absence of up-to-date figures on changes in the           gauges of whether former recipients have                 care, nor with
health of former recipients, it is tempting to            been made worse off by welfare reform.                   poor physical
assume that the large percentage of former                    Critics of the new law cite surveys, such as
recipients who lack health insurance implies              those conducted in South Carolina and                    health.
deteriorating health since enactment of welfare           Wisconsin, that suggest that TANF leavers may
reform. In fact, numerous economic analyses               be encountering more food problems, mainly as
have shown that, contrary to popular belief,              a result of failure to claim the food stamps for
being uninsured does not necessarily correlate            which they qualify. Those two studies paint a pic-
with lack of access to health care, nor with poor         ture of an emerging crisis by noting that, for
physical health. Poor physical health is largely          instance, when former recipients were asked “Was
the result of not being able to obtain needed             there ever a time when you could not buy food?”
medical services, and there is little reason to           the number responding “yes” increased 3 per-
believe that the uninsured are any worse off              centage points in South Carolina and 10 percent-
than the insured in terms of access to critical           age points in Wisconsin after respondents left the
care. Nearly every state requires state or local          welfare rolls.136 The danger in using the results
governments, or both, to provide care for the             from such studies to generalize about the well-
indigent and uninsured, and federal law                   being of the entire former recipients population is
requires hospitals treating Medicare patients to          that respondents to those surveys tend to be self-
accept all patients with emergency health prob-           selecting and the surveys therefore overrepresen-
lems. Generally higher levels of spending on              tative of a certain highly afflicted segment of wel-
health care and greater use of services by indi-          fare leavers. There exists little in the way of a com-
viduals with insurance does not, in fact, imply           prehensive, scientific examination of this issue as
better health. As John Goodman and Gerald                 it applies to leavers of welfare under the new law.
Musgrave point out, “The amount of medical                    Another indication cited by critics that for-
care that people consume varies with the out-             mer recipients have been made worse off is the
of-pocket price they have to pay, often with no           reported increase in requests for emergency food
effect on health.”135 Given such information, as          assistance since enactment of welfare reform.
well as the absence of data on changes in the             Between 1997 and 1998, the U.S. Conference of
health status of former recipients, there is no           Mayors noted an average 14 percent increase in
basis for concluding that TANF leavers are                requests for emergency food assistance in more
worse off in terms of physical health simply              than two-thirds of the cities it studied, and
because an increasing number lack insurance.              Catholic Charities found an average 38 percent
    Food. Since enactment of welfare reform,              increase in requests for such assistance among
critics have routinely blamed the new law for             nearly three-quarters of the parishes it surveyed
increasing hunger and food shortages among                in 1998.137 Overall, what this and data from
former recipients. Are people truly worse off             USDA indicate, notes the GAO, is that the
after they leave the rolls when it comes to their         demand for food assistance by low-income—pre-
food and nutritional needs, and is the new                sumably, former-recipient—families has
welfare law responsible for those difficulties?           increased over the last few years.138 Critics use
    Two ways researchers have attempted to                these figures to seduce us into believing that wel-
assess changes in former recipients’ ability to           fare reform, by making more people turn to food
satisfy their basic food needs have been to               pantries, has led to a hunger crisis among former

    Figure 11
    Percentage Increase in Requests for Emergency Shelter, 1998–99
           Trenton                                                        0
           St. Paul                           -11
          St. Louis                                                           0.97
            Seattle                                                       0
    San Francisco                                                                                 10
      San Antonio                                                                        5
    Salt Lake City               -19
       Providence                                                         0
          Portland                                                                                           14
           Phoenix                                                                                                            25
      Philadelphia                                                                   4
           Norfolk                                                                                                                      31
      Minneapolis                                                                                                        20
             Miami                                                                           8
     Los Angeles                                                                                              15
      Kansas City                                                -1
            Detroit                                                                                                17
            Denver                                                                                            15
         Cleveland                                                                     5
           Chicago                                                                   3.5
        Charleston                                                                                                       20
         Burlington                                       -4.5
            Boston                                                                       5
                      -30         -20               -10               0                          10                     20         30             40

    Source: United States Conference of Mayors, “A Status Report on Hunger and Homelessness in America’s Cities 1999: A 26-City Survey,”
    December 1999.
    Note: Study does not provide data for Charlotte, Nashville, and New Orleans.

                                 recipients. The misguided premise here is that                             however, may simply be that emergency food
                                 decreased dependence on the public safety net                              assistance has become more available in recent
                                 and increased reliance on private charity imply                            years, particularly now that the welfare law
                                 greater suffering. In truth, there is good reason to                       encourages government support for private-sec-
                                 believe, given private charities’ consistently docu-                       tor and faith-based delivery of such services.140
                                 mented effectiveness in delivering social services,                        Finally, numbers may appear higher if data col-
                                 that many individuals who have left food stamps                            lection has become more aggressive over the last
                                 for food pantries under the new law are, in fact,                          few years in response to interest in tracking the
There is reason to               having their needs more adequately met.                                    effects of the new welfare legislation.
                                     To the extent that leavers’ increased difficul-                            Housing. Advocates for the poor have sug-
believe that many                ties in meeting basic food and nutritional needs                           gested that, like hunger, homelessness has been
  individuals who                can legitimately be documented, there exists little                        exacerbated by welfare reform. They cite rising
    have left food               basis for faulting the new welfare law.139 Eligibility                     numbers at homeless shelters and increased
                                 for the food stamp program is in no way linked to                          demands for emergency assistance as evidence
  stamps for food                eligibility for cash benefits, indicating that efforts                     of the new welfare law’s detrimental impact on
   pantries are, in              to cut individuals from the rolls should have little                       the living conditions of poor families with chil-
                                 effect on their ability to claim food stamps and                           dren. Studies put the average increase in home-
 fact, having their              meet their basic food needs. In light of this, there                       lessness since enactment of welfare reform at
  needs more ade-                are several alternative explanations for increased                         between 10 and 15 percent (Figure 11).141 A
       quately met.              reports of hunger and need for emergency assis-                            majority of cities in at least one survey indicat-
                                 tance. One, already mentioned, is that individuals                         ed that they expected requests for emergency
                                 are failing to claim the food stamp assistance for                         housing assistance to continue to rise.142
                                 which they are eligible. Another provocative                                   Some observers, such as the National
                                 explanation for increased visits to food pantries,                         Coalition for the Homeless, have identified

Figure 12
Rents and House Prices Compared with Overall Inflation
                                               Consumer Price Index                            6.4

   6                                           Rent Inflation

                                               House Price Inflation


                     3.1                                                                3.1
   3                                                                              2.7

   2         1.7


                    1997                                1998                            1999
Source: U.S. Department of Housing and Urban Development, “The State of the Cities 2000: Megaforces
Shaping the Future of the Nation’s Cities,” June 2000.

poverty as a result of low wages and cuts in               the country is perhaps the strongest explana-
benefits as a primary cause of increased visits            tion of the reported increases in housing prob-
to homeless shelters.143 Scaled back cash ben-             lems in recent years. As a recently released HUD
efits and tough work and sanctions policies,               report shows, rents and housing prices have
certain studies have suggested, are causing                been consistently rising at higher rates than
the living situations of some former recipi-               overall inflation (Figure 12). According to the
ents to deteriorate. As the Christian Science              study, the number of low-income families pay-
Monitor recently pointed out, “The biggest                 ing more than half their income for housing or
hike is in families with children, many of                 living in dilapidated units has risen by 12 per-    Housing difficul-
whom recently had their welfare benefits                   cent since 1991.147 In Fairfax County, Virginia,
reduced or terminated.”1 4 4                               the Homeless Oversight Committee reports            ties after exiting
    Such studies, however, are misleading in               that a rise in the local homeless population        the rolls are con-
that they fail to reveal that homeless former              includes around 750 adults who have full-time
recipients are frequently sanctioned leavers               jobs but can’t find housing they can afford.
                                                                                                               nected, not with
who face more than the ordinary number of                  One of the key problems identified by the com   -   the new welfare
barriers to work. This group, as we have seen,             mittee is a shortage of rental housing.148          legislation, but
makes up a relatively small percentage of the                  Finally, former recipients’ housing diffi-
overall population of former recipients.145 In             culties may be linked to a decrease in the          with the decrease
general, surveys of homeless populations                   amount of housing assistance available, as          in affordable
tend to exaggerate the extent to which for-                well as the number of landlords willing to
                                                                                                               housing and
mer recipients are encountering increased                  accept Section 8 housing vouchers.149
poverty and homelessness.                                      The new welfare law itself appears to have      rental units and
    More important, when TANF leavers experi-              done nothing directly to worsen the housing         the supposed
ence housing problems, it is unlikely that wel-            situations of former recipients. Most families
fare reform is the culprit. A major decline in the         do not lose access to housing assistance as a       shortage of hous-
quantity of affordable housing146 throughout               consequence of leaving welfare. Housing dif-        ing assistance.

  The number of      ficulties after exiting the rolls are connected,             claim child care assistance. Studies show that
        child care   not with the new welfare legislation, but with               while the majority of TANF leavers qualify, only
                     the decrease in affordable housing and rental                between one-fifth and one-half are benefiting
    providers has    units and the supposed shortage of housing                   from subsidized child care.152 Advocates of
actually managed     assistance available to low-income individu-                 expanded child care assistance suggest that much
                     als. Perhaps even more important, lack of                    of the reason former recipients and other low-
to keep pace with    housing assistance or unaffordable housing,                  income working individuals are not taking
     the swelling    or both, does not necessarily mean that indi-                advantage of the subsidies has to do with short-
demand for child     viduals are becoming homeless and worse                      ages in the types of care such individuals need.
                     off. Many may be finding new ways to meet                    Nontraditional-hours care, infant care, special
             care.   their housing needs, such as moving in with                  needs care, and sick-child care are among the
                     relatives or, in some cases, getting married.                types of licensed care for which studies say
                         Child Care. For working families with chil-              supply is outweighed by potential demand.153
                     dren, affordable, quality, and accessible child                  For families who do benefit from govern-
                     care can be one of parents’ most pressing                    ment-subsidized child care, the quality of
                     needs. Opponents have criticized the new law                 care is said to be another crisis. A recent study
                     for failing to adequately provide current and                of formal care by the University of California
                     former recipients with access to child care so               at Berkeley and Yale University finds that
                     they can cope with the new work require-                     many toddlers are in low-quality care and are
                     ments. This is an area in which critics suggest              lagging behind in language and social devel-
                     families perhaps suffer most as they move                    opment. Numerous others, the study reports,
                     from welfare to work.                                        are spending hours in child care watching
                         Talk of a growing gap between need and                   television or wandering aimlessly, with little
                     provision, as well as increasing evidence of                 interaction with their caregivers. The study
                     poor-quality subsidized care, has grown loud-                concludes, ”This study did find early warn-
                     er since enactment of welfare reform. The                    ing signs of a child-care problem that’s going
                     Urban Institute recently found that only 15                  to get worse as the work requirements of the
                     percent of the population of children eligible               welfare laws ramp up.”154
                     for public child care assistance was served in                   On closer inspection, however, talk of a
                     1998, suggesting the existence of, as the HHS                national child care crisis for current and for-
                     puts it, “a large gap between child care need                mer welfare recipients is overblown.155 In the
                     and services.”150 A similar conclusion was                   first place, the gap between supply of and
                     arrived at by the GAO in a 1997 study of sup-                demand for licensed formal care is not nearly
                     ply of and estimated demand for child care in                as large as critics would have us believe.
                     four sites. In the report’s final remarks, the               Eligibility for government assistance with
                     GAO notes: “State and local officials in the                 child care does not necessarily imply need or
                     four cities and counties we reviewed regarded                desire for licensed care. Most of the studies
                     their current supply of known child care as                  that posit a gap do not take account of addi-
                     inadequate for meeting even the demand they                  tional sources of informal, or unlicensed, care,
                     currently face for children in certain age                   which low-income families tend overwhelm         -
                     groups, particularly for low-income popula-                  ingly to prefer and rely on.1 5 6 The heaviest use
                     tions in three of the areas reviewed. Unless the             of “alternative” care (usually by friends or fam -
                     supply of known child care for certain age                   ily) tends to be by parents needing infant,
                     groups at these sites is increased, the gap                  preschool, and nontraditional-hours care—the
                     between supply of known care and anticipated                 types of care that are supposedly in short sup-
                     demand is likely to become even greater as wel-              ply in the licensed market.157
                     fare reform is fully implemented.”1 5 1                          Even if demand for formal care in some
                         Former welfare recipients constitute a signifi-          areas were to approach the estimates in stud-
                     cant proportion of the eligible families that fail to        ies such as the GAO’s, there exists no real

threat of a national supply shortage. As                    from work, experience few new problems in
Darcy Olsen of the Cato Institute has shown,                accessing food, health care, housing, and
the number of child care providers has actu-                child care. When questioned about their new
ally managed to keep pace with the swelling                 lives off welfare, most individuals have rea-
demand for child care, and, down the road,                  son to be feeling more positive about the pre-
“there is no reason to believe that the child               sent, and hopeful about the future, than they
care market will be unable to adjust to                     did when receiving cash benefits.
accommodate more children.”1 5 8                                It is clear that any talk of the “crisis” awaiting
    Finally, if quality of care is an issue for for-        welfare recipients when they leave the rolls is
mer welfare families, it appears to be a problem            overblown. A more accurate way to put it would
that is largely confined to users of formal, sub-           be to say that difficulties—very occasionally
sidized care. Again, however, most low-income               severe—are in store for those who exit the rolls
families choose informal care by neighbors or               who do not secure employment or claim the
relatives, and, according to the Urban Institute,           additional benefits for which they qualify. It is
96 percent of parents are satisfied with their              here that we most profoundly witness the new
current child care arrangements.159                         law’s limited ability to rehabilitate individuals
    Given the extensive surveying of former                 who have fallen into dependence in the first place.
recipients that has taken place since enact-                                                                         The best way to
ment of welfare reform, there is little doubt                                                                        end dependence
that families who leave welfare for work                         Back to the Beginning:                              is to prevent its
encounter child care challenges along the                        Targeting the Roots of                              onset in the first
way. Nevertheless, in light of Olsen’s review
of findings regarding the quality and avail-                                                                         place.
ability of child care, there is little reason to               This review of welfare reform’s progress to
believe that the challenges families face have              date has shown the new law’s limited impact
become any worse since enactment of the                     on individuals already entrenched in the sys-
new law.                                                    tem. Most former recipients are no worse
                                                            off—in fact, they are often better off—because
Summary                                                     of the new legislation, but only a small frac-
    A large proportion of families who have                 tion no longer needs, or appears to be on the
left and managed to stay off the rolls since                path to not needing, means-tested assistance.
1996 possesses a distinct advantage over                    Welfare reform, as envisioned by the drafters
those who continue to collect TANF checks.                  of the 1996 law, appears incapable of fully
That is reflected, first, in the fact that as the           freeing recipients from the grips of depen-
nation’s caseloads have taken a steep dive                  dence. Nearly four years after enactment of
over the last four years, more people—partic-               PRWORA, we are left with a working welfare
ularly single-parent families and black chil-               state that engenders a new form of depen-
dren—have advanced out of poverty. Studies                  dence on supplemental assistance.
have shown that, combined, income from                         As Michael Tanner of the Cato Institute
work at a minimum-wage job and supple-                      has suggested, the best way to end depen-
mental supports such as food stamps,                        dence is, not through curative efforts, but to
Medicaid, and child care and housing assis-                 prevent its onset in the first place. He warned
tance allow families to enjoy a higher level of             in 1996 that
financial well-being than they did when on
welfare. Since the new law allows former                        there are serious problems with
recipients to continue to receive means-test-                   expecting hard-core, long-term wel-
ed noncash benefits for an unlimited period                     fare recipients to be able to find suf-
of time after they exit welfare, TANF leavers,                  ficient employment to support
in addition to having higher cash incomes                       themselves and their families. When

                           we established the incentives of the                  more than 600 percent and the rate of births
                           current system, we may have made a                    to unmarried teenagers has quadrupled over
                           Faustian bargain with those recipi-                   the past four decades. As Tanner explains,
                           ents. Now it may be too late to                       “By removing the economic consequences of
                           change the rules of the game. We                      out-of-wedlock births, welfare has removed a
                           should do whatever we can to move                     major incentive to avoid them.”165 “Responsi-
                           those people out of the system but                    ble behavior,” says Ben Wattenberg of the
                           recognize that success may be limit-                  American Enterprise Institute, “rises in reac-
                           ed. It is far more important to pre-                  tion to a lack of viable irresponsible econom   -
                           vent anyone new from becoming                         ic alternatives and to the cessation of mes-
                           trapped in the system.160                             sages by governments that out-of-wedlock
                                                                                 birth is a socially acceptable lifestyle.”1 6 6
                           Welfare reformers’ efforts to slow entry                  Welfare reformers claim to have taken ille-
                       into the system largely involve provisions                gitimacy seriously when drafting the new
                       aimed at tackling teenage and out-of-wed-                 law. In order to continue to receive federal
                       lock births. Those attempts, however, are                 TANF funding, states must (1) restrict bene-
                       being diluted by the states’ eagerness to                 fits to unwed teen parents under 18 who do
                       spend more on welfare than they did before                not live at home and attend school and (2)
                       the new legislation. There is little reason for           outline how they intend to establish goals
                       potential recipients to fear a rolling back of            and act to prevent and reduce the number of
                       the safety net when, as of 1998, almost all               nonmarital pregnancies—particularly those
                       states had increased spending per welfare                 of teenagers. To further discourage teen preg-
                       family and nearly half were spending more                 nancy, the new law also allows states to insti-
                       than required by the new law.1 6 1 Of the 35              tute a “family cap” that denies additional
                       states that had increased spending by 30 per-             benefits to families to which more children
                       cent or more per welfare family, 16 had                   are born while the family is receiving assis-
                       increased spending by 50 percent or more,                 tance. Finally, the new law offers bonuses to
                       and 3 had actually doubled the amount                     the five states that rank highest in decreasing
                       spent per family.162 As the HHS has observed,             out-of-wedlock births while also decreasing
                       “Overall, based both on the level of spending             abortion.
                       in FY 98 reported by States and on the cash                   Unfortunately, as the New York Times
                       assistance levels established by the States               reports, illegitimacy is “a front on which
                       under the TANF program, there is clearly no               progress has been slight if at all.”167 The new
Between 1993 and       ‘race to the bottom’ occurring.”163 The explo-            welfare law has been particularly ineffective
                       sion in welfare spending over the last four               in curbing out-of-wedlock births to
1997, the percent-     years is a likely reason for major setbacks in            teenagers—those whom lawmakers most
   age of all births   efforts to reduce out-of-wedlock childbear-               hoped to target. Between 1993 and 1997,
       that were to    ing by young women.                                       when states first began experimenting with
                                                                                 welfare reform, the percentage of all births
        unmarried      Reducing Out-of-Wedlock Births                            that were to unmarried teenagers (the
 teenagers rose by         Having a child out of wedlock is often a              teenage illegitimacy ratio) rose by as much as
                       precursor to a lifetime of poverty and depen-             28 percent in every state except Rhode
     as much as 28     dence. Out-of-wedlock pregnancy, in fact, is              Island.1 6 8By 1997, in five states more than 90
  percent in every     the reason that one-third of welfare recipi-              percent of teen mothers were unmarried.1 6 9
       state except    ents end up on the rolls and is the cause of              Over the last few years, the percentage of all
                       most long-term dependence.164 Studies sug-                births that is to unmarried teens has contin-
    Rhode Island.      gest that the availability of welfare, ironically,        ued to rise, though at a slightly slower
                       is largely to blame for the fact that the num    -        pace.1 7 0When the entire pool of single moth-
                       ber of out-of-wedlock births has increased by             ers is considered, welfare reform’s impact on

Figure 13
Out-of-Wedlock Births as Percentage of All Births, 1980–99








      1980   1985   1986   1987   1988   1989   1990   1991    1992   1993   1994   1995   1996   1997   1998   1999

Sources: Stephanie J. Ventura et al., “Births: Final Data for 1998,” Centers for Disease Control and Prevention,
National Center for Health Statistics, National Vital Statistics Reports 48, no. 3 (March 28, 2000): 8; and Sally
C. Curtin and Joyce A. Martin, “Births: Preliminary Data for 1999,” Centers for Disease Control and Prevention,
National Center for Health Statistics, National Vital Statistics Reports 48, no. 14 (August 8, 2000).

illegitimacy is even less impressive. PRWORA                  specifically at pregnancy prevention. PRWORA
has had no measurable impact on the overall                   allocates $50 million annually to abstinence
ratio of out-of-wedlock to married-couple                     education, requires establishment of nation-
births. As Figure 13 shows, after nearly dou-                 al teen pregnancy prevention goals, and man-             Illegitimate
bling between 1980 and 1994, the proportion                   dates that at least 25 percent of U.S. commu-            births to teens
of American babies being born to unmarried                    nities have teen pregnancy prevention pro-
women has remained relatively constant at                     grams in place. The HHS recently claimed,                remain an inex-
about one-third of all births in recent                       “The strategy sends the strongest possible               orable challenge.
years.1 7 1 The actual number of out-of-wed-                  message to all teens that postponing sexual
lock births to black women has been declin-                   activity, staying in school, and preparing for
ing, but that trend began in 1989, before the                 work are the right things to do.”174
states and the federal government began                           Has this message worked its magic?
enacting welfare reform measures, and the                     Welfare reformers can hardly contain their
percentage of illegitimate births to black                    excitement over recent dramatic declines in
women has remained steady.172                                 the teen birth rate (Figure 14), as well as in
                                                              the number of second births to teenage
Reducing Teen Pregnancy                                       mothers. Inspection of those figures, howev-
   Like unwed motherhood, teenage preg-                       er, reveals more modest progress in reducing
nancy is strongly linked with subsequent wel-                 births to teenagers, a dubious connection
fare dependence. More than three-quarters of                  between the downward trend and specific
unwed teen mothers end up on welfare                          provisions of the new welfare law, and sober-
before their children reach school age. 173 In                ing evidence that illegitimate births to teens
recognition of the relationship between teen                  remain an inexorable challenge.
pregnancy and dependence, the new welfare                         Progress in reducing the number of births
law, in addition to requiring unmarried teen                  to teens has been less impressive than national
mothers to stay at home and remain in                         figures tend to convey. Critics note, for
school, contains several provisions aimed                     instance, inconsistent (and often lagging)

                      Figure 14
                      Birth Rates for Teens Aged 15–19, 1985–99







                                1985   1986   1987   1988   1989   1990   1991    1992    1993   1994   1995   1996   1997   1998   1999

                      Sources: Stephanie J. Ventura et al., “Recent Declines in Teenage Birth Rates in the United States: Variations by
                      State, 1990–1994,” Centers for Disease Control and Prevention, National Center for Health Statistics, Monthly
                      Vital Statistics Report 45, no. 5 (December 19, 1996): 7; Stephanie J. Ventura et al., “Births: Final Data for
                      1998,” Center for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics
                      Reports 48, no. 3 (March 28, 2000): 4; and Sally C. Curtin and Joyce A. Martin, “Births: Prelimenary Data for
                      1999,” Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital
                      Statistics Reports 48, no. 14 (August 8, 2000).

                      progress among the states. Child Trends                         critics, the impressive downward trend in the
                      points out that whereas Vermont has reduced                     teen pregnancy rate began long before welfare
                      teen pregnancy to 11 births per 1,000 teens,                    reform measures were instituted. Second, while
                      states like Mississippi lag behind at 58 births                 the thrust of the new law’s attack on teen preg-
                      per 1,000 teens.175 Supporters of the new wel-                  nancy has been abstinence education, much of
                      fare law also cite the decline in the percentage                the decline in teen pregnancies, says Child
                      of teenage mothers who have second children                     Trends’ Jennifer Manlove, is attributable to
  It is not teenage   as evidence that the welfare law has succeeded                  increased use of injectible contraceptives. She
                      in dissuading those women from having more                      points out: “There has been little rigorous
pregnancy, per se,    children. The supporters note, for instance,                    research on the effectiveness of abstinence-only
that presages wel-    that the percentage of teenage mothers going                    programs. Consequently, the jury is still out on
 fare dependence      on to have a second child decreased between                     whether abstinence programs can significantly
                      1993 and 1997. What critics say they fail to                    reduce teen childbearing.”177 Third, it is diffi-
 but the choice by    point out, however, is that, two years into wel-                cult to discern the link between programs
   the majority of    fare reform, the number of teenage mothers                      intended to prevent teen pregnancy in general
                      who had second pregnancies began to level off                   and teen birth rates, says the GAO, because,
young mothers to      at around 22 percent and has shown no fur-                      while the states are tracking changes in teen
  give birth out of   ther sign of inching downward.176                               birth rates, few are actually evaluating the
          wedlock.        The declines in teen pregnancy that have                    effects of their prevention programs on teen
                      occurred, however, certainly represent progress                 pregnancy.178
                      over the last decade. Nevertheless, several facts                   More important than the scale and cause
                      cast doubt on the relationship between this                     of the declines, however, is the fact that fewer
                      hopeful trend and welfare reform. First, note                   teenage pregnancies over the last few years do

not appear to be associated with fewer young                • referring   individuals to private           As of August
women bearing children out of wedlock. It is                  charities, and                               1998, 35 states
not teenage pregnancy, per se, that presages                • requiring applicants to spend
welfare dependence but the choice by the                      weeks searching for work before              were employing
majority of young mothers to give birth out                   they receive their first welfare pay-        some form of
of wedlock. If, as has been the case since wel-               ment.181
fare reform, the number of teenagers having
                                                                                                           “diversion” tactic
children has inched downward, but the over-                 The HHS identifies three specific types of     aimed particular-
all percentage of births that occur to unmar-           diversion programs being tried throughout          ly at women con-
ried mothers has remained steady or contin-             the country. “Lump-sum payment” pro-
ued to climb, then the primary precursor to             grams make available a one-time payment to         sidered “job
welfare dependence has not been tackled.179             families with a short-term financial need;         ready” or who
                                                        “mandatory applicant job search” programs
                                                                                                           had another
Increasing Diversion                                    require job searches as a condition of TANF
   For those who have already become preg-              eligibility, with the aim of encouraging indi-     potential source
nant as teenagers or out-of-wedlock, or both,           viduals to secure employment before receiv-        of income.
how viable an alternative does welfare remain           ing the first welfare check; and “alternative
under the new law? As we have seen, once                resources exploration” encourages families
women become part of the system, it                     to consider other forms of support from
becomes very difficult to fully escape it,              family and community as a substitute for
despite the range of work-support measures              cash assistance from the state.182 As of 1997,
and incentives created by PRWORA. An                    the HHS reports 20 states were employing
important goal of welfare reform should                 the “lump-sum payment” tactic, 16 states
thus be preventing the onset of dependence              were requiring “mandatory applicant job
by “diverting” families from signing onto the           search,” and only 7 states were using an
rolls in the first place. As Professor June             aggressive “alternative resources” approach.
O’Neill of City University of New York told             Only three states had implemented all three
the Subcommittee on Human Resources of                  types of diversion strategies.183
the House Committee on Ways and Means in                    To what extent do decreasing TANF
1999, “Although attention is usually focused            claims reflect the success of diversion policies
on the effects of the policy changes on wel-            in causing fewer people to enter the system in
fare leaver rates, the ultimate efficacy of wel-        the first place? The Washington Post suggests
fare reform is going to turn on the extent to           that “the combination of these and similar
which it will have changed the incentive                [diversion tactics] explains a significant por-
structures in the program, and whether that             tion of the decline in welfare caseloads. . . .
change is enough to deter young women                   One noted researcher believes the policies are
from entering in the first place.”180                   responsible for one-third of the . . . drop in
   As of August 1998, 35 states were employ-            caseloads since the nation’s welfare system
ing some form of “diversion” tactic aimed               was overhauled in 1996. . . . The new tactic is
particularly at those considered “job ready”            critical to understanding the remarkable
or who had another potential source of                  decline in the number of Americans receiving
income, including                                       welfare over the past two years. While many
                                                        poor families have moved off public assis-
    • urging applicants to ask for help                 tance and into jobs, many others simply have
      from relatives instead of signing up              never gone on the rolls.”184 In 1999 the GAO
      for government assistance,                        agreed that “states’ diverting eligible families
    • writing one-time emergency checks                 from receiving cash assistance may have con-
      or offering low-interest loans in                 tributed to the large decline,”1 8 5 and the
      place of monthly benefits,                        Committee for Economic Development

                     remarked earlier this year that “the overall           lock births, teen pregnancy, and first-time
                     rate at which new welfare cases join the rolls         welfare claims constitute an important por-
                     has fallen sharply.”186 In states aggressively         tion of the 1996 welfare law. Unfortunately,
                     pursuing diversion strategies, such as Florida         the results associated with those measures do
                     and Kentucky, the percentage of applicants             not, on the whole, make up a significant por-
                     who end up on the rolls has decreased by as            tion of welfare reform’s success story.
                     much as 20 percent.187                                 Illegitimate births remain a significant prob-
                         In general, it appears that welfare reform,        lem, particularly among young mothers, and
                     through diversion, has made important                  have not shown any recent signs of inching
                     strides in preventing individuals from                 downward. In addition, it is difficult to estab-
                     becoming first-time recipients of cash assis-          lish a link between advances in reducing out-
                     tance. Nevertheless, the gates to TANF                 of-wedlock and teenage births and measures
                     remain open for individuals who are inca-              enacted under the new law.
                     pable of working, have no alternative source               Diversion strategies applied as a result of
                     of income or support, or require more than a           welfare reform have shown some success in
                     one-time check to bridge the time between              turning new applicants toward alternative
                     jobs. This is where prevention of out-of-wed-          means of support. Though that represents a
   Pregnancy pre-    lock births to young women becomes partic-             positive development in and of itself, there is
 vention, in prac-   ularly important. Once an unmarried young              no evidence that diversion strategies are
tice, has not been   woman decides to have children, she usually            effective as a disincentive for those, such as
                     becomes a valid candidate for TANF receipt—            unwed teens or high-school dropouts, think-
   the resounding    someone unlikely to be turned away during              ing of choosing life-long dependence.
  message of wel-    the diversion phase, or, as they call it in the        Women who have no other means of support
                     United Kingdom, the “Gateway” to welfare.              or who are unable to work remain, even
      fare reform.   Diversion is instrumental mainly in weeding            under diversion policies, key candidates for
                     out individuals who are in transition and              welfare receipt. Increased spending on recipi-
                     show potential for becoming self-supporting            ents under the new law, moreover, makes wel-
                     or have access to alternative means of sup-            fare seem that much more viable an alterna-
                     port. It can do little for those who have cho-         tive for at-risk women. Prevention, in prac-
                     sen poorly and consequently condemned                  tice, has not been the resounding message of
                     themselves and their families to a lifetime of         welfare reform.
                     dependence.                                                Consider the particular provisions of the
                                                                            1996 welfare law in light of Wattenberg’s com   -
                     Summary                                                ment that putting an end to the country’s
                         Pulling dependent families out of the              unwed teenage pregnancy problem requires the
                     “poverty trap” has proven a formidable task            dismantling of “viable irresponsible economic
                     for welfare reformers. Many recipients have            alternatives” and “the cessation of messages by
                     little education or work experience and are            governments” that such choices lead to a
                     trying to raise families without the help of a         “socially acceptable lifestyle.” As we have seen,
                     spouse. A fast-food job and food stamps, plus          lack of progress in tackling out-of-wedlock
                     short-term counseling and training classes,            pregnancy may indicate that welfare reform has
                     usually can’t do much in the long term for             not gone far enough in making welfare a less
                     families that, until now, have gotten by with          appealing and viable alternative. Far from elim -
                     the help of a welfare check. The real solution         inating welfare, the new law may be sending
                     involves stopping such dependence before it            contradictory messages by continuing to spend
                     is given a chance, with the help of cash assis-        generously on current recipients. Potential
                     tance, to flourish.                                    young single mothers, despite the changed
                         Welfare reformers know that prevention is          rules, may continue to get pregnant, believing
                     important. Measures to decrease out-of-wed-            that in no way has the safety net actually been

dismantled. And, in a very important sense,              by providing generously for young women
they are correct. As we have seen, the new feder-        who give birth out of wedlock.
al TANF block grant created by welfare reform                 What such results indicate, not surprising-
enables states that reduce their caseloads to            ly, is that work requirements, time limits, and
provide a more generous array of support to              efforts to make work pay are only moderately
each remaining welfare family, including con-            effective approaches to ending long-term
tinued cash benefits, intensive counseling, sub-         dependence. Few people would deny that both
sidized work, and transportation assistance.             recipients and taxpayers are better served if
The problem, aside from the fact that the extra          individuals are required to do something in
support does not appear to have been very effec-         return for help from the state instead of being
tive in assisting those who face multiple obsta-         given something for nothing. Although the
cles to becoming independent, is that such               new law is a “better way” in some respects, it
spending may convey to potential young recip-            does not appear to offer a remedy for depen-
ients the message that a fortified and ever-pre-         dence. Policymakers can cite little more than
sent safety net awaits them if they should               anecdotal evidence that welfare reform has
choose to become pregnant out of wedlock.                helped to dislodge deeply entrenched individ-
    The logical way to eliminate dependence              uals from the safety net and has made once-
and its various causes, which Tanner pro-                dependent families self-reliant.
posed before the new law was enacted, is to                   The solution to the country’s dependence
adopt a prohibition against new, single                  problem requires more than efforts to pull
mothers’ signing onto the rolls.1 8 8 If welfare         families off the welfare rolls. The ultimate
were eliminated for those women, Tanner                  solution lies in prevention. Most families end
predicts, out-of-wedlock births as a percent-            up on welfare because of out-of-wedlock
age of all births would likely decline by                births and, once in this precarious situation,
between 15 and 50 percent.1 8 9                          rarely are able to free themselves and future
                                                         generations from reliance on the state. Welfare
                                                         reformers need to turn their efforts to encour-
         Conclusion:                                     aging young women, as Tanner had suggested,
       Making the Grade?                                 to (1) finish school; (2) not get pregnant out-
                                                         side marriage; and (3) get a job, any job, and
    Four years after passage of the new welfare          stick with it.190 The most effective way to com-
law, the country’s cash assistance system has            municate this message would be to remove
undergone modest improvement. More indi-                 welfare as an alternative for young women
viduals are having to work for their benefits,           who fail to make smart choices.                    The logical way
fewer remain on the rolls, a greater number                   Until welfare is no longer available or an
are finding employment after they exit the               attractive option for women considering            to eliminate
system, and fewer are losing out in the tran-            making untenable life decisions, the country       dependence and
sition from welfare to work.                             will continue to spend endless amounts of          its various causes
    According to an overwhelming number of               energy and taxpayer money trying to clean
measures, however, welfare as we once knew               up its dependence problem. This nation can         is to adopt a
it remains firmly intact. Those who leave for            never know the extent to which self-reliance       prohibition
work continue to be short-term, transitional             and mutual aid are achievable until legisla-
cases; individuals who remain on the rolls are           tors decide, once and for all, to put the nanny
                                                                                                            against new,
still supported by a generous array of cash              state to rest.                                     single mothers’
and in-kind benefits; life after welfare contin-                                                            signing onto the
ues to be characterized by low-wage employ-
ment, heavy reliance on supplemental bene-                                 Notes                            rolls.
fits, and frequent returns to the rolls; and the         1. “President Clinton’s Remarks, Townhall Dis-
system continues to encourage dependence                 cussion Forum,” Welfare-to-Work Partnership,

August 2–4, 1999,                 3ices, “Characteristics and Financial Circumstances
                                                                of TANF Recipients: Fiscal Year 1998,” http://
2. U.S. Department of Health and Human Ser-           
vices, “Change in TANF Caseloads,” http://                      fy1998.sum.htm.; and
Stephen Moore and Michael New, “Next Step to                    9. Robert E. Rector and Sarah E. Youssef, The
Welfare Reform,” Washington Times, April 7, 2000.               Determinants of Welfare Caseload Decline (Washing-
                                                                ton: Heritage Foundation, 1999), p. 6. The study
3. Barbara Vobejda and Judith Haveman, “Wel-                    finds, specifically, that initial full-check sanctions
fare Clients Already Work, Off the Books,”                      and delayed full-check sanctions have a strong
Washington Post, November 3, 1997.                              and statistically significant effect on states’ case-
                                                                load decline. States that have the option of sanc-
4. U.S. General Accounting Office, “Welfare Re-                 tioning the entire TANF check at the first
form: States’ Implementation and Effects on the                 instance of nonperformance of or noncompli-
Workforce Development System,” Statement of                     ance with requirements for work or other activi-
Cynthia M. Fagnoni, director, education, workforce              ties are using “initial full-check sanctions”; states
and income security issues, Health, Education, and              that generally have a sequence of progressively
Human Services Division, GAO, before the                        more severe sanctions and will sanction the full
Subcommittee on Postsecondary Education,                        TANF check only after a number of months of
Training, and Life-Long Learning of the House                   noncompliance or repeated performance infrac-
Committee on Education and the Workforce,                       tions are using “delayed full-check sanctions”;
September 9, 1999, GAO/T-HEHS-99-190, p. 12.                    states with “moderate sanctions” or “weak sanc-
                                                                tions” either sanction a third of the TANF check
5. U.S. Department of Health and Human Ser-                     and rarely the full check or tend to sanction only
vices, “National Emergency TANF Datafile,” May                  the adult portion of the TANF check.
characteristics/fy1998/tab31_98.htm.                            10. U.S. Department of Health and Human
                                                                Services, “Characteristics.”
6. Pamela Loprest, “Families Who Left Welfare:
Who Are They and How Are They Doing?” Urban                     11. Jason DeParle, “As Welfare Benefits Expire,
Institute, Assessing the New Federalism program,                Second Thoughts,” New York Times, October 10,
Discussion Paper, February 1999, p. 8.                          1999.

7. U.S. Department of Health and Human Ser-                     12. Several facts suggest that time limits have
vices, “National Emergency TANF Datafile.” In                   had, and will continue to have, a relatively limited
striking contrast, before enactment of welfare                  impact on caseload decline. First, only a small per-
reform, marriage was one of the most common rea-                centage of the caseload consists of long-term
sons individuals gave for leaving the rolls. See                recipients who are likely to be affected by the time
Michael Tanner, The End of Welfare: Fighting Poverty in         limits. The added influences of a good economy
the Civil Society (Washington: Cato Institute, 1996), p.        and tough work rules makes it even less likely that
65; and U.S. House of Representatives, Committee                recipients will stay on the rolls long enough to see
on Ways and Means, 1996 Green Book: Background                  their deadlines expire. Regarding the more
Material and Data on Programs within the Jurisdiction of        prominent impact of work requirements than time
the Committee on Ways and Means (Washington:                    limits on caseload decline, the Manpower
Government Printing Office, 1996), p. 504.                      Demonstration Research Corporation points out
Presumably, the lower percentage of exits for this              that “while the presence of a time limit probably
reason under the new law is the result, not of fewer            spurs some recipients to work or leave welfare
marriages, but of the increase in exits due to work             more quickly in the pre–time limit period, such
requirements and tougher sanctions policies.                    impacts are probably fairly modest. . . . When
                                                                asked about welfare reform measures, recipients
8. U.S. Department of Health and Human Ser-                     discussed policies such as work requirements that
v+.0                                                            affect them immediately. Some expressed general
                                                                concern about time limits, but had not translated
                                                                this into a specific strategy for moving to self-suf-
                                                                ficiency.” Dan Bloom, “The Cross-State Study of
                                                                Time-Limited Welfare, Welfare Time Limits: An
                                                                Interim Report Card,” Manpower Demonstration
                                                                Research Corporation, New York, April 1999,
                                                                executive summary,
                                                                    A second reason time limits are likely to have

little impact on caseload decline is states’ willing-        of Welfare Policy and the Economic Expansion on
ness to bend the rules when push comes to shove.             Welfare Caseloads: An Update,” Washington,
As the GAO notes, although a number of states                August 3, 1999, executive summary, p. 1.
have adopted more stringent time limits than
those required by the 1996 law, many of those                18. Committee for Economic Development,
states have also adopted policies to extend assis-           “Welfare Reform and Beyond: Making Work
tance beyond the time limits in certain circum-              Work,” Washington, 2000, p. 21.
stances. In general, states “have not used their
flexibility under TANF to reduce cash assistance             19. See Michael Tanner, Stephen Moore, and
levels or deny eligibility to specific groups of peo-        David Hartman, “The Work versus Welfare Trade-
ple, except for convicted drug felons.” U.S.                 Off: An Analysis of the Total Level of Welfare
General Accounting Office, “Welfare Reform:                  Benefits by State,” Cato Institute Policy Analysis
States Are Restructuring Programs to Reduce                  no. 240, September 19, 1995. The study shows
Welfare Dependence,” GAO/HEHS-98-109,                        that the dollar value of states’ benefits packages
June 1998, p. 5. The Washington Post noted in 1998           often exceeds the after-tax income of low-wage
that, in the nearly 800,000 cases in the 12 states           workers and thus provides a strong disincentive
where time limits either had expired or were set to          to leave welfare for work.
expire, nearly half of the recipients had been
exempted from the rules because they had small               20. June O’Neill, professor of economics at City
children, disabilities, or faced other serious obsta-        University of New York, Testimony, in Effects of
cles to work. Judith Havemann and Barbara                    Welfare Reform: Hearing before Subcommittee on
Vobejda, “The Welfare Alarm That Didn’t Go                   Human Resources of the House Committee on Ways and
Off,” Washington Post, October 1, 1998.                      Means, 106th Cong., 1st sess., May 27, 1999, Serial
     A third reason time limits are not likely to            106-9 (Washington: Government Printing Office,
affect welfare recipients down the road is that a            1999), p. 23.
growing portion of the caseload will be formally
exempt from the deadlines. The new welfare law               21. Rector and Youssef, p. 1; Curtis J. Simon of the
exempts from time limits a significant percentage            Strom Thurmond Institute, a proponent of the “eco-
of recipients, including minor children who are              nomic” explanation, offers several reasons why
receiving assistance (around 17 percent of the               shrinking caseloads in areas of high unemployment
1996 caseload) and an additional 20 percent of               do not necessarily support the case for welfare
recipients who face unusual “hardships.” The exit            reform. He notes, first, that economic prosperity
of many of the “easy” cases from the rolls in the            operates with a lag. Part of the decline in the welfare
early days of welfare reform has left behind a large         caseloads in 1997 and 1998, he contends, was due to
portion of recipients with unusual “hardships” to            improvements in the economy as far back as 1995
whom it may be difficult to apply time limits.               and 1996. Simon also points out that even the small-
     A final reason we can expect time limits to             est—often unobserved—changes in an otherwise lag-
have little impact is the fact that they apply only          ging local economy can have tremendous effects on
to 4 of the nearly 80 federal welfare programs. As           the percentage change in the caseload, as the popula-
a consequence of numerous exclusions, in combi-              tion of welfare recipients is small, to begin with, in
nation with the large percentage of the caseload             most states. Finally, he questions the validity of cor-
that consists of short-term cases, time limits are           relations drawn from observations of the caseload
unlikely to prove to be a powerful weapon in the             over less than a two-year period (January 1997 to June
government’s arsenal against dependence.                     1998), noting that “welfare caseloads have strong sea-
                                                             sonal components in many states that are not neces-
13. U.S. Department of Health and Human                      sarily identical.” What Simon’s defense fails to
Services, “Characteristics.”                                 explain, however, is why a larger percentage of welfare
                                                             recipients, if the economy is really drawing them off
14. Reasons for exit under the category “other”              the caseload, is not leaving the rolls for the reason of
are not specified because of disparities in states’          “employment.” His explanation is a more convincing
reporting.                                                   answer to the question of what happens to individu-
                                                             als after they exit welfare. For more details, see Curtis
15. Loprest, “Families Who Left Welfare.”                    J. Simon, “Welfare Reform in South Carolina: A
                                                             Preliminary Analysis,” Strom Thurmond Institute,
16. State of Wisconsin, Department of Workforce              Clemson University, October 1999.
Development, “Survey of Those Leaving AFDC or
W-2, January to March 1998, Preliminary                      22. Michael New, Unpublished study, Cato Insti-
Report,” January 13, 1999, p. ii.                            tute, August 3, 1999.

17. Council of Economic Advisers, “The Effects               23. Rector and Youssef, p. 10.

24. Committee for Economic Development, p. 21;                Health and Human Services, June 22, 1998, p. 1.
Michael New contends that an economic turn-
around would not necessarily cause caseloads to               35. Sheila R. Zedlewski, “Work-Related Activities
soar if states maintained low benefits levels and             and Limitations of Current Welfare Recipients,”
strict sanctions policies. He notes that Colorado             Urban Institute, Assessing the New Federalism
managed a 40 percent reduction in its caseload                program, Discussion Paper, November 1999,
between 1997 and 1998, despite increasing unem-               pp. 11–12.
ployment, largely because of that state’s commit-
ment to strict sanctioning procedures. New.                   36. States’ distribution of current and former
                                                              recipients among various work activities, notes
25. Janet Schrader, “Lost on the Road to Reform,”             the Urban Institute, is largely a function of which
Washington Post, May 11, 1997.                                “Work First” model states have chosen to imple-
                                                              ment. Those with a “Work First, Work Mandate”
26. Vobejda and Havemann, “Welfare Clients Al-                model require participation in employment
ready Work, Off the Books.”                                   (unsubsidized, subsidized, or community work
                                                              experience) in return for cash assistance and have
27. Charles Murray, The Underclass Revisited (Wash-           generally low levels of participation in stand-
ington: American Enterprise Institute, 1999).                 alone education and training by current and for-
                                                              mer recipients. States with a “Work First,
28. Frank J. Bennici, “Final Report: Managing                 Participation Mandate” model, the more com-
Expectations for Welfare to Work: The Realities of            mon of the two models, allow job search, educa-
Serving the Hardest to Serve,” National Center on             tion, and training to count as work participation
Education and the Economy, Rochester, N.Y.,                   and consequently have a higher proportion of
August 1999, p. 1. See also Loprest, “Families                current and former recipients engaged in such
Who Left Welfare.”                                            activities. Pamela A. Holcomb et al., “Building an
                                                              Employment-Focused Welfare System: Work
29. U.S. Department of Health and Human                       First and Other Work-Oriented Strategies in Five
Services, “TANF Report to Congress,” 1998,                    States,” Urban Institute, June 1998, p. 2.
introduction, p. 2,
news/welfare/congress/tanfintr.htm.                           37. Zedlewski, p. 5.

30. Ibid., p. 5.                                              38. Rebecca Schwartz et al., W-2: Converting to
                                                              Wisconsin Works: Where Did Families Go When AFDC
31. A snapshot of the welfare caseload in the early           Ended in Milwaukee? (Indianapolis: Hudson
1990s showed that 64 percent of recipients were               Institute and Mathematica Policy Research,
white, 31 percent were African American, and 14               1999), p. 5
percent were Hispanic. Michael Tanner, “Ending
Welfare As We Know It,” Cato Institute Policy                 39. Sandra Danziger et al., “Work, Income, and
Analysis no. 212, July 7, 1994, p. 4. In 1998, by             Material Hardship after Welfare Reform,” Journal
contrast, black claimants outnumbered white                   of Consumer Affairs 34, no.1 (Summer 2000): 7.
claimants, and Hispanics were the fastest-grow-               Around half of former recipients under AFDC/
ing part of the caseload. Jason DeParle, “Welfare             JOBS secured unsubsidized employment.
Rolls Show Growing Racial and Urban
Imbalance,” New York Times, July 27, 1998.                    40. See U.S. General Accounting Office, “Welfare
                                                              Reform: States’ Implementation and Effects”;
32. Katherine Allen and Maria Kirby, “Unfinished              Maria Cancian et al., “Work, Earnings, and Well-
Business: Why Cities Matter to Welfare Reform,”               Being after Welfare: What Do We Know?”
Brookings Institution, July 2000.                             Institute for Research on Poverty, University of
                                                              Wisconsin, Madison, January 1999; Committee
33. U.S. Department of Health and Human                       for Economic Development; and U.S. Depart-
Services, “Characteristics.”                                  ment of Health and Human Services, Office of
                                                              the Assistant Secretary for Planning and
34. “Barriers to work,” as defined by the HHS, gen-           Evaluation, “’Leavers’ and Diversion Studies:
erally include problems with child care, disabilities,        Summary of Research on Welfare Outcomes
domestic violence, emergency financial needs,                 Funded by ASPE,” 1999,
housing instability, lack of health insurance, men-           hsp/leavers99/ombsum.htm.
tal health, substance abuse, and transportation.
Amy Johnson and Alicia Meckstroth, “Ancillary                 41. See Cancian et al.; and State of Wisconsin,
Services to Support Welfare to Work: Multiple                 Department of Workforce Development, “Survey
Barriers,” Office of the Assistant Secretary for              of Those Leaving AFDC or W-2, January to March
Planning and Evaluation, U.S. Department of                   1998, Preliminary Report,” January 13, 1999.

42. NETWORK, “Poverty Amid Plenty: The Unfin-                57. Steve Mayes, “Poverty Trails Welfare Clients to
ished Business of Welfare Reform,” National Welfare          Work,” Oregonian, January 31, 2000.
Reform Watch Project, Washington, 1999, p. 6.
                                                             58. The Committee for Economic Development
43. Danziger et al., p. 7                                    reports that after one year off welfare 25 percent
                                                             (up from an initial 23 percent) of TANF leavers
44. Cancian et al., p. 24.                                   receive employer-sponsored health care. After
                                                             three years, the proportion of leavers with
45. Mary D. Janney, “Off Welfare, On to College,”            employer coverage rises to only 38 percent.
Washington Post, July 15, 1998.                              Committee for Economic Development, p. 5; and
                                                             Johnson and Meckstroth, “Ancillary Services to
46. See Schwartz et al.; Loprest, “Families Who              Support Welfare to Work: Lack of Health
Left Welfare”; and Committee for Economic                    Insurance,” p. 2.
                                                             59. Committee for Economic Development, p. 5.
47. The 2000 federal poverty level for a family of
three in 48 states and the District of Columbia is           60. Ibid., p. 15.
                                                             61. Demetra Smith Nightingale and Kathleen
48. See National Governors’ Association, National            Brennan, “The Welfare-to-Work Grants Program:
Conference of State Legislatures, and American Public        A New Link in the Welfare Reform Chain,” Urban
Human Services Association, “Summaries of State              Institute, Assessing the New Federalism program,
Followup Studies,” April 1998,           Issues and Options for States no. A-26,
Welfare/StateFollowUp.htm; Judith Havemann,                  September 1998, p. 3.
“Welfare Reform Success Cited in L.A.,” Washington
Post,August 20, 1998; and Ellen Nakashima, “Welfare-         62. Martha Zaslow et al., “Welfare Reform and
to-Work May Fall Short for Most in Va.,” Washington          Children: Potential Implications,” Urban
Post, November 22, 1997.                                     Institute, Assessing the New Federalism program,
                                                             Issues and Options for States no. A-23, June 1998,
49. Loprest, “Families Who Left Welfare,” p. 12.             pp. 1–2.

50. U.S. General Accounting Office, “Welfare                 63. Ibid.
Reform: States’ Implementation Progress and
Information on Former Recipients,” Statement of              64. Judith M. Gueron and Amy Brown, “Work
Cynthia M. Fagnoni before the Subcommittee on                after Welfare,” Washington Post, August 13, 1998.
Human Resources of the House Committee on
Ways and Means, May 27, 1999, GAO/T-HEHS-                    65. Pamela Friedman, “Community Work Experi-
99-116, p. 7.                                                ence and Publicly-Funded Jobs—Helping the Hard-
                                                             to-Serve Meet Work Requirements,” Welfare Infor-
51. Committee for Economic Development, p. 5.                mation Network Issue Notes 3, no. 5 (July 1999).

52. State of Wisconsin, Department of Workforce              66. Alan Finder, “Training Programs Provide an
Development, p.18; Other reports place that                  Alternative to Workfare Jobs,” New York Times,
number as high as half of all former recipients in           June 16, 1998.
Wisconsin. See Schwarz et al. It appears that the
number might be much higher here because of                  67. Vadim Mahmoudov, “Are Workfare Partici-
the population studied—leavers in unsubsidized               pants ‘Employees’? Legal Issues Presented by a
jobs one year after W-2’s implementation—in                  Two-Tiered Labor Force,” Annual Survey of
other words, the cream of the crop.                          American Law, no. 3 (1998): 352.

53. Robert Pear, “Companies Find Former Recipi-              68. Holcomb et al., p. 7.
ents Stay on Job Longer Than Other Workers,”
New York Times, May 27, 1998.                                69. Ibid.

54. Loprest, “Families Who Left Welfare,” p. 10.             70. Paul Leonard, “Welfare-to-Work Block Grants:
                                                             Are They Working?” Brookings Institution,
55. Committee for Economic Development, p. 5.                September 1999, p. 22.

56. U.S. Department of Health and Human                      71. A modified version of this program has now
Services, Office of the Assistant Secretary for              been implemented statewide.
Planning and Evaluation, “’Leavers,’” p. 24.
                                                             72. DeParle, “Welfare Rolls Show Growing Racial

and Urban Imbalance.” In fact, in the Manpower              leavers claimed transitional benefits at a consis-
Demonstration Research Corporation’s recently               tently higher rate than those who had been off the
released evaluation of Minnesota’s Family Invest-           rolls for some time. Loprest, “Families Who Left
ment Program, it was discovered that increased              Welfare,” p. 19.
work activity came at the expense of increased wel-             The Urban Institute adds the cautionary note,
fare dependence. See Manpower Demonstration                 however, that the individuals whom such studies
Research Corporation, “Minnesota’s Welfare                  take into account are those who have succeeded
Reform Brings Dramatic Results for Long-Term                in staying off the rolls and, often, have managed
Recipients and Their Children,” June 1999.                  to stay employed. Those with the greatest depen-
                                                            dence problems, who would have continued to
73. Deborah Weinstein, director, Family Income              rely on supplemental help, are likely to have
Division, Children’s Defense Fund, and Robert C.            returned to the rolls. Thus, it notes, studies may
Granger, senior vice president, Manpower Demons-            overwhelmingly reflect the decreasing depen-
tration Research Corporation, Statements, in Effects        dence over time of the cream of the crop of welfare
of Welfare Reform, pp. 126, 106.                            leavers. What is needed is longitudinal data on the
                                                            same group of former recipients at several differ-
74. Loprest, “Families Who Left Welfare,” p. 14.            ent points after leaving welfare. Ibid., p. 19.
                                                            Consequently, since no such studies have been
75. See ibid.; and State of Wisconsin, Department           performed, it is fruitless to speculate about the
of Workforce Development.                                   exact extent to which supplemental support helps
                                                            dependent individuals work their way into self-
76. See Loprest, “Families Who Left Welfare”;               sufficiency and is not merely a different form of
State of Wisconsin, Department of Workforce                 “open-ended” welfare.
Development; U.S. Department of Agriculture,
Office of Analysis, Nutrition, and Evaluation,              84. “The New Welfare Trap,” Detroit News, January
Food and Nutrition Service; and David                       13, 2000.
Kocieniewski, “Study Finds Mixed Results in
Reducing Welfare Rolls,” New York Times, October            85. U.S. General Accounting Office, “Medicaid
22, 1999.                                                   Enrollment: Amid Declines, State Efforts to
                                                            Ensure Coverage after Welfare Reform Vary,”
77. Loprest, “Families Who Left Welfare,” p. 16.            GAO/HEHS-99-163, September 1999, p. 3.

78. State of Wisconsin, Department of Workforce             86. Ibid., p. 9.
Development, p. 10.
                                                            87. Families USA, “One Step Forward, One Step
79. Schrader.                                               Back: Children’s Health Coverage after CHIP and
                                                            Welfare Reform,” October 1999, p. 2.
80. Loprest, “Families Who Left Welfare,” p. 7.
                                                            88. U.S. Department of Health and Human Ser-
81. Nancy L. Johnson, “The Results Are In:                  vices, Office of the Assistant Secretary for
Welfare Reform Works,” Wall Street Journal, August          Planning and Evaluation, “’Leavers,’” p. 4.
24, 1999.                                                   Although the number of children claiming
                                                            Medicaid has fallen since enactment of welfare
82. Loprest, “Families Who Left Welfare,” p. 18.            reform, the number of children enrolled under
                                                            the Children’s Health Insurance Program (CHIP)
83. Cancian et al., p. 24; Proponents of self-suffi-        since its inception in 1997 has increased signifi-
ciency champion the provision of supplemental               cantly. Nevertheless, the increases have not been
benefits because they believe that those supports           enough to offset the declines in coverage by
enable former recipients to gradually work their            Medicaid. Families USA, p. 1.
way out of dependence. In fact, positive signs have
emerged that former recipients’ reliance on sup-            89. Amy Johnson and Alicia Meckstroth, “Ancil-
plemental assistance does tend to diminish over             lary Services to Support Welfare to Work: Lack of
time. Nine states’ studies reviewed by the                  Health Insurance,” U.S. Department of Health
University of Wisconsin indicate that, while two-           and Human Services, Office of the Assistant
thirds of TANF leavers reported receipt of supple-          Secretary for Planning and Evaluation, June 22,
mental benefits within the first year off welfare,          1998, isp/ancil-
only 35 to 45 percent of those who managed to               lary/hi.htm.
stay off the rolls were receiving supplements sev-
eral years later. Ibid., pp. 19-23. The Urban               90. Bowen Garrett and John Holahan, “Welfare
Institute reports similar results in its study of           Leavers, Medicaid Coverage, and Private Health
early AFDC/TANF leavers, noting that recent                 Insurance,” Urban Institute, Assessing the New

Federalism program, National Survey of                   Specialized Child Care,” U.S. Department of
America’s Families no. B-13, March 2000, p. 7,           Health and Human Services, Office of the            Assistant Secretary for Planning and Evaluation,
s_b/b13/b13.html.                                        June 22, 1998, p. 3, http://www.aspe.os.dhhs.
91. See ibid.; and Johnson and Mecksworth,
“Ancillary Services to Support Welfare to Work:          104. Norma B. Coe et al., “Does Work Pay? A
Lack of Health Insurance.”                               Summary of the Work Incentives under TANF,”
                                                         Urban Institute, Assessing the New Federalism
92. Ibid., p. 2.                                         program, Issues and Options for States no. A-28,
                                                         December 1998, p. 2.
93. See U.S. General Accounting Office, “Welfare
Reform: States’ Implementation Progress”;                105. U.S. General Accounting Office, “Welfare
Cancian et al.; Loprest, “Families Who Left              Reform: States’ Implementation Progress,” p. 15.
Welfare”; and State of Wisconsin, Department of
Workforce Development.                                   106. Coe et al., p. 4.

94. U.S. General Accounting Office, “Food Stamp          107. Nicholas Johnson, “A Hand Up: How State
Program: Various Factors Have Led to Declining           Earned Income Tax Credits Help Working Families
Participation,” GAO/RCED-99-185, July 1999,              Escape Poverty: 1999 Edition,” Center on Budget and
pp. 1–2.                                                 Policy Priorities, Washington, December 20, 1999, p. 1.

95. Sheila R. Zedlewski and Sarah Brauner, “De-          108. Sheldon Danziger, ed., Economic Conditions
clines in Food Stamp and Welfare Participation:          and Welfare Reform (Kalamazoo: Michigan Uni-
Is There a Connection?” Urban Institute, Assess-         versity, W. E. Upjohn Institute for Employment
ing the New Federalism program, Discussion               Research, 1999), p. 7.
Paper, October 1999, pp. 24–26.
                                                         109. Nancy L. Johnson, U.S. representative from
96. Cheryl Wetzstein, “Economy Credited for              Connecticut, Opening statement, in Effects of
Food-Stamp Dip; Glickman Says Too Many                   Welfare Reform, p. 5.
Going Hungry,” Washington Post, July 10, 2000.
                                                         110. “First Results of Welfare Overhaul Emerge in
97. U.S. General Accounting Office, “Food Stamp          1999,”,
Program,” p. 6.                                          historyreport.cfm?HistoryID=201.

98. Zedlewski and Brauner, abstract, p. 3.               111. Committee for Economic Development, p. 23.

99. See U.S. General Accounting Office, “Food            112. Rector and Youssef, p. 1.
Stamp Program.”
                                                         113. Nancy L. Johnson, Opening Statement, p. 5.
100. See Victoria Wegener, “Food Stamp Edu-
cation and Outreach: Working to Provide                  114. Committee for Economic Development, p. 23.
Nutrition Benefits to Eligible Households,”
Welfare Information Network Issue Note 3, no. 11         115. Nancy L. Johnson, Opening Statement, p. 2.
(December 1999); and Lissa Bell and Carson
Strege-Flora, “Access Denied: Federal Neglect            116. U.S. General Accounting Office, “Welfare
Gives Rise to State Lawlessness; Families Denied         Reform: States’ Implementation and Effects,”
Access to Medicaid, Food Stamps, CHIP, and               pp. 12–13.
Child Care,” Northwest Federation of Commun-
ity Organizations, Seattle, Washington, May 2000.        117. See Tanner, Moore, and Hartman.

101. State of Wisconsin, Department of Work-             118. Pamela Loprest, “Long Ride from Welfare to
force Development, pp. 1–2.                              Work,” Washington Post, August 30, 1999. It should
                                                         be noted that state policy can be pivotal in deter-
102. Tamar Lewin, “Study Finds Welfare Changes           mining exactly how much better off individuals
Lead a Million into Child Care,” New York Times,         are when they exit welfare. States vary in the gen-
February 4, 2000.                                        erosity of their TANF benefits packages, have dif-
                                                         ferent rules about earnings “disregards” for con-
103. Amy Johnson and Alicia Meckstroth,                  tinued receipt of benefits, and have different lev-
“Ancillary Services to Support Welfare to Work:          els of supplemental benefits for TANF leavers

under different eligibility criteria. As both the            128. Zaslow et al., p. 3.
Cato Institute and the Urban Institute have
shown, disparities in financial well-being (relative         129. Ibid., p. 5.
to welfare) after people exit the rolls reflect both
differences in states’ levels of TANF payments               130. Jan Kaplan, “The Use of Sanctions under TANF,”
and differences in how states’ TANF grants                   Welfare Information Network, April 1999, p. 7.
decline as families’ earned incomes rise. See Coe et
al.; and Tanner et al. A University of Michigan              131. Havemann and Vobejda, “Children of
study points out, “The payoff to earnings relative           Welfare Parents Feel Reform’s Help, Hurt.”
to welfare . . . would be lower in states with short-
er time limits, quicker and more aggressive sanc-            132. The Committee for Economic Development
tioning policies, or ones that make it more diffi-           reports that after one year off welfare 25 percent
cult to combine welfare and work.” Danziger et               (up from an initial 23 percent) of TANF leavers rely
al., p. 22.                                                  on employer-provided coverage. After three years,
                                                             that number purportedly rises to 38 percent.
119. Although overall poverty has been declining             Johnson and Meckstroth, “Ancillary Services to
since enactment of the new welfare law, the num-             Support Welfare to Work: Lack of Health
ber of people living in extreme poverty has grown,           Insurance,” p. 2. While a positive development in
as has the extent to which people in this category           and of itself, a 66 percent increase in employer-
fall below the poverty line. Between 1995 and                sponsored health insurance is hardly enough to
1997, for instance, the number of children living            offset former recipients’ loss of Medicaid coverage.
below one-half the poverty line grew by 400,000
Children’s Defense Fund and National Coalition               133. See Committee for Economic Development;
for the Homeless, “Welfare to What? Early                    Kaplan; and Loprest, “Families Who Left
Findings on Family Hardship and Well-Being,”                 Welfare.”
November 1998, p. 3. The average amount by
which children fall below the poverty line (the              134. Throughout the first year, as the Urban
“child poverty gap”) increased between 1995 and              Institute has found, the percentage of former
1998 to reach the highest level recorded in nearly           recipients without health insurance increases,
20 years. Center on Budget and Policy Priorities,            and presumably as many needlessly lose coverage
“Progress in Reducing Child Poverty Slows, Study             when their benefits expire. Garrett and Holahan,
Finds; Children Remaining Poor Have Become                   p. 3. As the HHS finds, the trend over time is
Somewhat Poorer,” December 23, 1999, p. 1.                   toward even more uninsured individuals, whose
                                                             numbers increase from 23 percent after the first
120. Judith Havemann and Barbara Vobejda,                    year off welfare to 45 percent after three years off
“Children of Welfare Parents Feel Reform’s Help,             the rolls. Johnson and Meckstroth, “Ancillary
Hurt,” Washington Post, December 27, 1998.                   Services to Support Welfare to Work: Lack of
                                                             Health Insurance,” p. 2. That is, between one and
121. Judith Havemann, “Most Find Jobs after                  three years after exiting welfare, the number of
Leaving Welfare,” Washington Post, May 27, 1999.             leavers with no health insurance nearly doubles.

122. State of Wisconsin, Department of                       135. John C. Goodman and Gerald L. Musgrave,
Workforce Development, p. 1.                                 Patient Power: Solving America’s Health Care Crisis
                                                             (Washington: Cato Institute, 1992), p. 231. A
123. Danziger et al., pp. 24–25.                             1984 report by the RAND Corporation, titled
                                                             “The Effect of Coinsurance on the Health of
124. U.S. General Accounting Office, “Welfare                Adults,” found that people with free care were
Reform: States’ Implementation Progress,” p. 11.             much more likely to use medical services than
                                                             were people who had to pay their bills out of
125. Zaslow et al., p. 5.                                    pocket, but there existed no apparent differences
                                                             between the two groups in terms of health out-
126. Havemann and Vobejda, “Children of                      comes as a result. See Goodman and Musgrave,
Welfare Parents Feel Reform’s Help, Hurt.”                   pp. 231–32, for more details.

127. Jane Knitzer et al., “Enhancing the Well-               136. U.S. General Accounting Office, “Welfare
Being of Young Children and Families in the                  Reform: States’ Implementation Progress,” p. 12.
Context of Welfare Reform: Lessons from Early
Childhood, TANF, and Family Support Pro-                     137. U.S. General Accounting Office, “Food
grams,” U.S. Department of Health and Human                  Stamp Program,” p. 12.
Services, June 1999, p. 38.
                                                             138. Ibid.

139. In the case of at least one group, there exists        March 28, 2000.
a supposed case for welfare reform’s detrimental
impact on satisfaction of food needs. Between               147. Ibid.
1994 and 1998, food stamp receipt by legal immi-
grants fell 72 percent, accounting for roughly 13           148. “The Housing Gap,” Editorial, Washington
percent of the total decline in claims for this time        Post, March 31, 2000.
period. Douglas J. Besharor, American Enterprise
Institute and University of Maryland School of              149. Johnson and Meckstroth, “Ancillary Services to
Public Affairs, Statement, in Effects of Welfare            Support Welfare to Work: Housing Instability,” p. 2.
Reform, p. 20. These individuals did not become
ineligible—they were made ineligible by a new pro-          150. U.S. Department of Health and Human
vision excluding the majority of legal immigrants           Services, “Access to Child Care for Low-Income
from receipt of food stamps. We have no evidence,           Working Families,” 1999, p. 4,
however, that this group of people who have left            news/press.1999.ccreport.htm.
food stamps are worse off in terms of hunger. It is
certainly plausible that the majority of those indi-        151. U.S. General Accounting Office, “Welfare
viduals are finding alternate ways to meet their            Reform: Implications of Increased Work Partici-
food needs. In any case, it would be premature to           pation for Child Care,” p. 3.
conclude that drops in food stamp claims and
cuts in eligibility have stimulated a hunger crisis.        152. See Schumacher and Greenberg; and
                                                            Committee for Economic Development.
140. Most notably, under the “charitable choice”
clause of the Welfare Reform Act, faith-based               153. See Waller; Johnson and Meckstroth, “Ancil-
organizations are permitted to compete for gov-             lary Services to Support Welfare to Work: Special-
ernment contracts to deliver various welfare ser-           ized Child Care”; U.S. General Accounting Office,
vices. In addition, under the new law, states have          “Welfare Reform: Implications”; and Committee
increasingly invited the private and voluntary sec-         for Economic Development.
tors to join the government in social service “part-
nerships” to more effectively meet the goals of             154. Quoted in “Study Finds Welfare Changes
welfare reform.                                             Lead a Million into Child Care,” New York Times,
                                                            February 4, 2000.
141. See Alexandra Marks, “Hot Economy, But
More Homeless,” Christian Science Monitor,                  155. See Darcy Olsen, “The Advancing Nanny
February 2, 2000; National Coalition for the                State: Why the Government Should Stay Out of
Homeless, “Homeless Families with Children,”                Child Care,” Cato Institute Policy Analysis no.
Fact Sheet no. 7, June 1999; and United States              285, October 23, 1997.
Conference of Mayors, “A Status Report on
Hunger and Homelessness in America’s Cities                 156.. See Tanner, “Ending Welfare As We Know
1999: A 26-City Survey,” December 1999.                     It”; Waller; Schumacher and Greenberg; Olsen;
                                                            and Zaslow et al.
142. National Coalition for the Homeless, p. 1.
                                                            157. U.S. General Accounting Office, “Welfare
143. Ibid.                                                  Reform: Implications,” p. 7.

144. Marks.                                                 158. Olsen, pp. 1–2.

145. It is highly likely, given that surveyed home-         159. Sandra L. Hofferth, National Child Care
less TANF leavers often indicate that their bene-           Survey, 1990 (Washington: Urban Institute, 1991),
fits have been “terminated,” that sanctioned for-           cited in Olsen, p. 2.
mer recipients constitute a large percentage of the
former recipients who visit shelters. Those indi-           160. Tanner, The End of Welfare, p. 131.
viduals tend to find jobs at much lower rates than
do other welfare leavers and, given their tendency          161. Barbara Vobejda, “Spending per Recipient
to be faced with more obstacles to work, do not             Has Risen since Enactment of Welfare Reform,”
generally obtain full-time, year-round employ-              Washington Post, February 7, 1998.
                                                            162. Jay Hein, “Intellectual Capital.Com: Pro and
146, Defined by the government as consuming                 Con: Early Successes, Important Questions and a
less than one-third of a household’s income. Irvin          Promising Future,” Hudson Institute, November
Molotsky, “Robust Economy Is Contributing to a              12, 1998, p. 2.
Loss of Affordable Housing,” New York Times,

163. U.S. Department of Health and Human                     175. Richard Wertheimer and Kristin Moore,
Services, Administration for Children and Families,          “Childbearing by Teens: Links to Welfare Reform,”
Office of Planning, Research and Evaluation,                 Urban Institute, Assessing the New Federalism
“Temporary Assistance for Needy Families Program:            program, Issues and Options for States no. A-24,
Second Annual Report to Congress,” August 1999.              1999, p. 1.

164. Tanner, “Ending Welfare As We Know It,” p. 7.           176. Wetzstein.

165. Tanner, The End of Welfare, p. 76.                      177. Wertheimer and Moore, p. 4.

166. Ben Wattenberg, “Linking Illegitimacy to                178. U.S. General Accounting Office, “Teen
Welfare,” American Enterprise Institute On the               Pregnancy: State and Federal Efforts to
Issues, August 1998, p. 1.                                   Implement Prevention Programs and Measure
                                                             Their Effectiveness,” Letter Report, GAO/HEHS-
167. Jason DeParle, “Bold Effort Leaves Much                 99-4, November 30, 1998, p. 1.
Unchanged for the Poor,” New York Times,
December 30, 1999.                                           179. See Murray, The Underclass Revisited.

168. Cheryl Wetzstein, “Teen Birthrates Drop in              180. O’Neill, p. 22.
U.S. As Unwed Mother Totals Rise,” Washington
Times, January 18, 2000.                                     181. Barbara Vobejda and Judith Havemann,
                                                             “States’ Welfare Shift: Stop It before It Starts,”
169. Ibid.                                                   Washington Post, August 12, 1998.

170. Ibid.                                                   182. U.S. Department of Health and Human
                                                             Services, Office of the Assistant Secretary for
171. “America’s Children 1999: Highlights,”                  Planning and Evaluation, “A Description and, p. 1,          Assessment of State Approaches to Diversion
highlight.asp; Charles Murray, “And Now for the              Programs and Activities,” executive summary, p.
Bad News,” American Enterprise Institute Articles,           2,
February 2, 1999, p. 2; and Sally C. Custin and              EXECSUM.htm.
Joyce A. Martin, “Births: Preliminary Data for
1999,” Centers for Disease Control and Prevention,           183. Ibid.
National Center for Health Statistics, National Vital
Statistics Reports 48, no. 14 (August 8, 2000).              184. Vobejda and Havemann, “States’ Welfare
                                                             Shift: Stop It before It Starts.”
172. Wattenberg, pp. 1–2; and Murray, The
Underclass Revisited, p. 14.                                 185. U.S. General Accounting Office, “Welfare
                                                             Reform: States’ Implementation Progress,” p. 15.
173. Michael Tanner, “Welfare Reform,”
Testimony before the Senate Finance Committee,               186. Committee for Economic Development, p. 19.
104th Cong., 1st sess., March 9, 1995, p. 1,                 187. “Deflecting Welfare Applicants,” Washington
                                                             Post, August 17, 1998.
174. U.S. Department of Health and Human
Services, “Clinton Administration Moving For-                188. Tanner, The End of Welfare, p. 131.
ward on the Promise of Welfare Reform,” HHS
Press Office, August 22, 1997,           189. Ibid., p. 132.
                                                             190. Ibid., p. 17.

Published by the Cato Institute, Policy Analysis is a regular series evaluating government policies and offering
proposals for reform. Nothing in Policy Analysis should be construed as necessarily reflecting the views of the
Cato Institute or as an attempt to aid or hinder the passage of any bill before congress. Contact the Cato
Institute for reprint permission. Additional copies of Policy Analysis are $6.00 each ($3.00 each for five or
more). To order, or for a complete listing of available studies, write the Cato Institute, 1000 Massachusetts
Ave., N.W., Washington, D.C. 20001, call toll free 1-800-767-1241 (noon - 9 p.m. eastern time), fax (202) 842-
3490, or visit our website at


Shared By: