The Evolution of Welfare Reform: Policy Changes and Current Knowledge

    Welfare reform has had dramatic repercussions for millions of adults and children in the country. So far, the results of the "natural experiment" of welfare reform have shown both benefits and problems. Although welfare reform became law in 1996, there have been legislative revisions and other policy changes aimed at making the law more effective and less punitive. The latest research indicates that welfare reform has had a major impact on the number of people who are receiving welfare benefits but not on the income of former welfare recipients. Research also suggests that there are families who are suffering as a result of welfare reform. In this article, I discuss how welfare reform has evolved and the implications for policy changes in the future.

    Welfare reform has been an enormous "natural experiment" that has affected the lives of millions of families. The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 changed welfare benefits from an entitlement program aimed at assisting the families of unemployed single mothers into a temporary assistance program geared toward getting those single mothers into the workforce as quickly as possible. Because the program was designed primarily for single mothers with dependent children under the age of 18, welfare reform has influenced the lives of millions of young women and children. Thus far, the dramatic changes in the law have not had the devastating impact that many researchers and policy experts predicted.

    Welfare reform policies have been enmeshed with election year politics, especially during the last three presidential elections (1992, 1996, and 2000). In 1992, candidate Bill Clinton promised to change "welfare as we know it." At that time, Aid to Families with Dependent Children (AFDC) was widely criticized because it fostered dependency and created a subculture that lacked the role models of working parents. There was considerable concern, however, that the 1996 congressional proposal for welfare reform was dangerous because women on welfare would not be able to find and keep jobs. Many progressive advocacy groups made dire predictions about what would happen to the families that would lose their welfare benefits if the 1996 bill became law. President Clinton and many of his policy experts shared these concerns, but in August 1996, Clinton reluctantly signed the welfare reform bill into law. His signing of a bill that was designed primarily by the Republican-controlled Congress was considered an election year necessity after the defeat of many Democrats in the 1994 elections. After the President was reelected, as the economy improved and his approval ratings reached new heights, the Clinton administration was able to make some changes to welfare policies. And in 2000, President Clinton and Vice President Gore have pointed to the drop in welfare recipients and their continuing legislative efforts to support families making the transition from welfare to work as evidence that welfare reform has been a success for which they should receive credit.

    The President and Congress made substantial legislative and policy efforts after 1996 that contributed to the success of welfare reform (White House, 2000). For example, the 1997 Balanced Budget Act included $3 billion in fiscal years 1998 and 1999 for welfare-to-work grants to help states and local communities move long-term welfare recipients and certain noncustodial parents into unsubsidized jobs. Funds could be used for job creation, job placement, and job retention efforts, including wage subsidies to private employers and other critical postemployment support services. Indian tribes could also receive up to $30 million of the welfare-to-work funds. For fiscal years 1998 and 1999, the federal government awarded 190 competitive grants to support local strategies to help noncustodial parents and individuals with limited English proficiency, disabilities, substance abuse problems, or who have been victims of domestic violence get and keep employment.

    In a different kind of effort to help welfare recipients find jobs, the Welfare-to-Work Tax Credit, enacted in 1997, offers employers incentives to hire individuals on welfare. The law provides a credit equal to 35% of the individual’s first $10,000 in wages in the first year of employment and 50% of the individual’s first $10,000 in wages in the second year, for a total credit of up to $8,500. This credit complements the Work Opportunity Tax Credit, which offers a credit of up to $2,400 for the first year of wages for eight groups of job seekers. From 1997 to 1999, employers were eligible to claim these tax credits for nearly 900,000 newly employed welfare recipients and other disadvantaged individuals. Both credits are available through December 2001.

    Welfare reform attempts to discourage teen pregnancy by requiring unmarried minor parents to stay in school and live at home or in a supervised setting and providing $50 million a year in new funding for state abstinence education activities. It is unclear whether these efforts have been successful; the most recent data indicate that welfare generosity under AFDC was associated with having more children for women 20 and over but did not affect teenagers’ childbearing (Hoffman & Foster, 2000). Nevertheless, teen birth rates declined nationwide by 20% from 1991 to 1999 and are now at the lowest level on record since tracking began 60 years ago. In September 1999, bonuses of $20 million to each of four states (Alabama, California, Massachusetts, and Michigan) and the District of Columbia were awarded for achieving the nation’s largest decreases in out-of-wedlock births between 1994 and 1997. These were the first bonuses awarded under the welfare reform law for up to five states with the largest reduction in the proportion of births out-of-wedlock that also show a decrease in their abortion rate.

    In 1999, President Clinton unveiled new incentives aimed at encouraging the formation and economic security of two-parent families through a new regulation that will add new categories for enrollment in Medicaid and the Children’s Health Insurance Program (CHIP) and the food stamp program. According to the White House, "These new measures will ensure that welfare reform will continue to move millions of families from dependence to independence, by encouraging work, supporting working families to help them succeed and stay off welfare, and increasing the number of low-income children living with two married parents" (White House, 2000). The final regulation, published in August 2000, provides $60 million in bonus funds for these new measures and $140 million for the continuing work measures.

    Child support enforcement has been part of welfare reform, under the assumption that families are less likely to need welfare if fathers help to support them. Child support collections doubled from $8 billion in 1992 to nearly $16 billion in 1999. Parents who owe child support currently have their wages garnished, their bank accounts seized, their federal loans denied, and their tax refunds withheld. Thus far, the number of families that receive child support increased from 2.8 million in 1992 to 4.5 million in 1998, and a new collection system, enacted as part of the 1996 welfare reform law, had located over 3.5 million delinquent parents by August 2000. In 1998, the Deadbeat Parents Punishment Act enacted tougher penalties for parents who repeatedly fail to support children living in another state or who flee across state lines. Meanwhile, new laws requiring the establishment of paternity resulted in a record 1.5 million in 1998, triple the 1992 figure of 516,000.

    In 1999, $283 million was provided for 50,000 new welfare-to-work housing vouchers for welfare recipients who need housing assistance to get or keep a job. Nearly all of these vouchers were awarded on a competitive basis to 35 states and two tribes, to communities that created cooperative efforts among their housing, welfare, and employment agencies.

    Despite the successful transition to work of many women who were welfare recipients, transportation to work is a barrier for many low-income families. Existing public transit often does not link to suburban job opportunities, cover late night and weekend hours, or serve many rural communities. The Transportation Equity Act for the 21st Century authorized $750 million over 5 years for the Job Access initiative and reverse commute grants in an effort to make transportation to jobs more affordable for low-income workers. In May 1999, $71 million of these funds were awarded to 179 communities in 42 states. In addition, the Clinton administration proposed a package of initiatives to help low-income families get to work by making it easier for them to purchase a car and improving public transit solutions. For example, current law limits food stamp eligibility to most families owning a car worth less than $4,650, a standard that dates to 1977. President Clinton proposed changes that would allow states to change their food stamp vehicle policy to make it conform with the more generous TANF vehicle policy. This proposal has not passed.

    The 1996 welfare reform law increased child care funding by $4 billion over 6 years to provide child care assistance to families moving from welfare to work and other low-income families. The Clinton-Gore fiscal year 2001 budget proposed additional assistance, such as expanding Head Start funding by $1 billion to provide slots to approximately 950,000 children and adding $3 billion over 5 years for the Early Learning Fund to help improve child care quality and early childhood education for children under 5 years old. The President’s budget proposal also would have increased funding for child care subsidies by $817 million in fiscal year 2001. However, these proposals have not passed. Meanwhile, in 1998, only about 10% of the 15 million low-income children eligible for assistance under federal law received subsidies.

    The Clinton administration efforts to pass health care reform in 1993–94 were partly aimed at ensuring that the transition from welfare to work did not mean the loss of health insurance. Since health care reform efforts collapsed, the Clinton administration has worked for more gradual solutions to address the same problem. The Children’s Health Insurance Program (CHIP) became law in 1997, allocating $24 billion dollars over the subsequent 5 years to extend health care coverage to uninsured children through state-designed programs. In August 1998, states were allowed to provide Medicaid coverage to more than 130,000 working, two-parent families who met state income eligibility requirements, eliminating disincentives to marriage and work. In April 2000, the U.S. Department of Health and Human Services (HHS) requested that states review their computer systems and eligibility processes, review their own records to be sure that no one who was entitled to keep Medicaid after leaving cash assistance lost out, and reinstate anyone who was improperly terminated from Medicaid.

    The 1996 welfare reform law authorized the use of welfare block grants to create individual development accounts (IDAs) to empower low-income families to save for a first home, for postsecondary education, or to start a new business. In 1998, a 5-year $125 million demonstration program was created. In fiscal year 1999, the federal government awarded nearly $10 million to 40 grantees that will establish over 10,000 savings accounts for low-income workers.

    One of the greatest controversies of welfare reform was the cut in benefits to legal immigrants. Upon signing the welfare reform law, the President made a commitment to reversing those cuts. The Balanced Budget Act of 1997 and the Noncitizen Benefit Clarification and Other Technical Amendments Act of 1998 provided $11.5 billion to restore disability and health benefits to 380,000 legal immigrants who were in this country before welfare reform became law (August 22, 1996). The Agricultural Research Act of 1998 provided food stamps for 225,000 legal immigrant children, senior citizens, and people with disabilities who entered the United States by August 22, 1996. The President’s fiscal year 2001 budget proposal would restore eligibility for Supplemental Security Income, Medicaid, and food stamps to legal immigrants who enter the country after that date under certain conditions; this proposal has not passed.

    In August 2000, in his farewell presidential address to the Democratic National Convention and in a separate statement marking the fourth anniversary of his signing of the welfare reform bill on August 22, President Clinton declared welfare reform a success, using statistics to support his claim (White House, 2000, p. 2):

Caseloads have fallen to historic new lows. In August 2000, HHS released data (from December 1999) showing that since January 1993, the welfare rolls have fallen by 7.8 million, or more than half (56 percent) from 14.1 million to 6.3 million, resulting in the fewest number of people on welfare since 1968 (32 years ago). The percent of Americans on welfare is now at 2.3 percent, the lowest level since 1965 (35 years ago) under the Clinton-Gore Administration. In August 1999, the Council of Economic Advisers reported that the single most important factor contributing to this historic decline is the implementation of welfare reform. Of the caseload reduction from 1996 to 1998, approximately one-third is due to federal and state policy changes resulting from welfare reform and about 10 percent is due to the strong economy.

At the President’s insistence, the 1996 welfare reform legislation included both re-wards and penalties to encourage states to place people in jobs. According to reports filed by the 46 states competing for the high performance bonus, more than 1.3 million welfare recipients nationwide went to work in just the one year period between October 1997 and September 1998. Retention rates were also promising: 80 percent of those who got jobs were still working three months later. States also reported an average earnings increase of 23 percent for former welfare recipients, from $2,088 in the first quarter of employment to $2,571 in the third quarter. The first three years of work data since welfare reform also show that all states subject to the work participation requirements met the law’s required overall work activity levels. Nationally, 38 percent of all welfare recipients were working or in work-related activities in 1999. The data also show that nationwide, the percentage of welfare recipients working has increased to nearly five times the level it was when the President took office, rising from 7 percent in 1992 to an all-time high of 33 percent in 1999. The vast majority of working recipients are in paid employment, with the remainder involved in community service or subsidized employment.

    Aware that the states had an incentive to inflate their success rates in order to qualify for federal bonuses, the White House statement also quoted independent studies that confirmed the success of welfare reform. For example, it pointed out that a national survey released by the Urban Institute found 69% of recipients had left welfare for work and that 18% had left because they had increased income, no longer needed welfare, or had a change in family situation. The White House also quoted a 1999 General Accounting Office (GAO) report that found that between 63% and 87% of adults have worked since leaving the welfare rolls, and a similarly high employment figure in a Census Bureau Current Population Survey. It cited a Minnesota study showing that "welfare reform can increase marriage rates and marital stability among low-income families."

    As the White House statement clearly shows, many studies evaluating welfare reform are now being used to confirm its success. These studies clearly show that by changing the incentives, welfare reform has encouraged many single mothers to work. Unfortunately, studies showing problems for some families are not touted by politicians as a way to determine how to revise these policies and strengthen the safety net.

    In 1999, three think tanks that had predicted that welfare reform would result in increased poverty released research reports supporting those predictions and claiming that the reality of welfare reform is not as rosy as politicians would have us believe. All the reports asked the same key question: Were single mothers who left welfare to go to work able to earn as much as they had received in welfare benefits?

    The Children’s Defense Fund (CDF) report compared the number of children in extreme poverty in 1996 and 1997, focusing on families headed by mothers (Sherman, 1999). CDF did its own analysis of a very large national Census Bureau survey, the Current Population Survey. It found that the number of children in these families who were living in extreme poverty increased from 1996 to 1997 as a result of welfare reform. Extreme poverty was defined as an income of less than half the poverty level (for example, less than $6,401/year or $123/week for a family of three). All kinds of income were included: wages and salaries, child support, government assistance checks, and noncash benefits such as food stamps.

    The CDF report pointed out that child poverty had been decreasing slowly in the years prior to the passage of federal welfare reform, from 1.7 million children in 1993 to 1.4 million in 1996. Then, in 1997, the first year after passage, the number of children increased 26%, to 1.8 million. One of the reasons is that 28% of former welfare families earned less than $125 per week. This would keep them below half the poverty level even if they worked all year, which many did not.

    The CDF report speculated that the loss of food stamps was a major reason for the rise in poverty. Increased employment would have made up for the loss of welfare benefits but not for the loss of food stamps too. Although families leaving welfare are often still eligible for food stamps, the statistics show that many stop getting them; the White House has made efforts to reverse that trend, as described above. Similarly, most legal immigrants lost food stamp eligibility under the 1996 revamping of eligibility for welfare-linked benefits, although most regained it in subsequent legislation, as described above.

    The CDF report explained that larger families that leave welfare struggle more, because welfare benefits were more generous for families with more children; in contrast, employers do not offer larger salaries to workers with more children. Although these "larger" families average only 3.5 children, many single mothers are unable to earn salaries large enough to make up for the loss of slightly larger welfare benefits.

    By focusing on the poorest families, this CDF analysis provided information about a vulnerable group that might not be noticeable in the aggregate statistics. The apparent drop in food stamps has received attention, but it is too soon to determine if this problem has been resolved.

    The Urban Institute has received more than $65 million in foundation grants for its New Federalism Project, which examines the impact of welfare reform and other laws that have returned major policymaking to the states. A primary focus is on the experiences of low-income families, rather than on those who have left welfare. In 1999, however, the Urban Institute released a report based on the government’s 1997 National Survey of America’s Families, comparing families of women who left welfare with "low income families" (with income under 200% of the poverty level) and the "near-poor" (with incomes below 150 % of the poverty level; Loprest, 1999). It found that women who have left welfare are working at low-paying jobs, and that one in four of the women who work at low-paying jobs, whether or not they were formerly on welfare, work mostly at night. This is obviously a potentially serious problem for women with young children.

    Loprest (1999) also found that 20% of former welfare recipients were not working, did not have a working spouse, and were not relying on government disability payments. This raises questions about how they are surviving and what is happening to their children. Presumably these families were receiving help from families, friends, or charitable groups, or were supporting themselves with "off the books" work that they did not report or with illegal activities.

    Loprest found that very few (6%) of the women who left welfare worked less than 20 hours a week, and most (69%) worked more than 35 hours a week. Their work hours were slightly longer than those of other near-poor or low-income workers who were not former welfare recipients. More than half the adults who left welfare were earning more than $6.61 an hour, considerably more than the minimum wage ($4.75 an hour at the time of the survey). This was somewhat higher than the earnings of near-poor and low income workers who were not formerly on welfare. The income differences apparently reflect other differences, such as educational attainment, race/ethnicity, marital status, and region where workers lived. The economic status of these families was similar to that of the other families in the study: on average, former welfare recipients earned approximately $100 less per month than low-income workers and only $100 more than near-poor workers.

    Many of these families were barely surviving economically: one in every three mothers who left welfare reported that they had to cut the size of or skipped meals because there was not enough food, compared with one in four of the other near-poor or low-income mothers. More than 38% reported that there was at least one time when they could not pay their rent, mortgage, or utility bills, which was almost twice as many as the other near-poor and low-income families.

    The report concluded that policymakers should focus on helping families who are poor, not just on former welfare recipients, since all these families face similar economic problems. Like the CDF, the Urban Institute concluded that there are probably families in all three groups that are not receiving food stamps or other government benefits to which they are entitled.

    The Center for Budget and Policy Priorities (Primus, 1999) compared two different time periods, from 1993 to 1995 (when states were just beginning to reform welfare) and from 1995 to 1997, which represents before and just after the beginning of federal welfare reforms. Their report concluded that the economic resources of the poor improved from 1993 to 1995, because their increased earnings and the earned income tax credit more than made up for the loss of welfare benefits that resulted from some state reforms. This progress was reversed, however, between 1995 and 1997 as a result of the 1996 federal welfare reform.

    The center analyzed the same national survey as did the CDF, although it separately evaluated five groups of single-mother families that each represented one fifth of the total (approximately 2 million families with 4 million children under 18 years of age). The poorest families (defined as the lowest 20% in income) became even poorer between 1995 and 1997, because their earnings and government benefits decreased. In fact, the incomes of the lowest 10%—approximately 1 million families—declined even more dramatically between 1995 and 1997. Most of the other single-parent families stayed at about the same economic level from 1995 to 1997; leaving welfare for work did not increase or decrease their family income.

    In a study released this year, 20 programs were evaluated by the Manpower Demonstration Research Corporation to determine the impact of mandatory welfare-to-work programs (Michalopoulos, Schwartz, & Adams-Ciardullo, 2000). All began operating prior to welfare reform; the earliest began in 1985 and the latest are still in operation. They found that participants in welfare-to-work programs earned approximately $500 more per year and had lower welfare payments than the control groups; however, the combined income was the same. The most disadvantaged groups, such as those at high risk for depression, also earned more money than they had before, but they still earned less than the other groups. The most disadvantaged groups tended to benefit more from the employment-focused programs, rather than those that were education-focused.

    The underlying message of all four of these reports is that welfare reform is not really a success to date, because so many families are still struggling and children are facing more poverty than before. For many policymakers, however, the major goal of welfare reform was simply to get families off welfare, not necessarily to increase their economic well-being. Therefore, the fact that so many families are doing equally well off welfare as they were on welfare is seen as success, rather than failure.

    Loprest and Zedlewski (1999) studied welfare recipients in 1997 compared to former recipients who had received benefits during the previous 2 years and found that current welfare recipients were less educated, more likely to be Hispanic, and less likely to be White or to live in the Northeast and the western United States. Those that were on welfare were more likely to have multiple obstacles to work, whereas those who were formerly on welfare tended to have no obstacles. There were no significant difference in the economic struggles they reported, again con-firming that moving from welfare to work does not solve the problems of poverty.

    The articles in this issue of the Journal of Social Issues are consistent with the findings from these five reports and elucidate the kinds of barriers that are most likely to interfere with self-sufficiency and success in the work force, the outcomes that should be used to measure the success of welfare reform. For example, the articles by Jayakody and Stauffer, Danziger et al., and Kalil and Danziger show how depression and other mental-health problems are likely to undermine a woman’s ability to succeed in the workforce, especially if she suffers from multiple problems. Tolman and Raphael describe the strong link between domestic violence and welfare and examine how the perpetrators of domestic violence often sabotage a woman’s ability to complete job training and succeed at work. Since human-capital problems, such as lack of education, training, and work experience, tend to coexist with mental-health problems and domestic violence, the ability of women who suffer from these problems to succeed under welfare reform will depend on programs and services that can assist with all of these kinds of barriers. The articles by Scott et al. and Coley et al. show that many of the women who were on welfare and the teenagers that seemed headed for welfare have apparently embraced the idea that work is better than welfare and that work requirements are an appropriate incentive to push them forward to a better life. The article by Hofferth et al. indicates that, after a difficult transition, there are often clear benefits for those children whose mothers are able to leave and remain off welfare. The article by Romero et al. indicates, however, that there are risks for children under welfare reform. Moreover, Henly and Lyons’ article and the mental health and domestic violence articles show that for some women and families, child care, personal characteristics, relationships, and other factors are likely to interfere with their ability to get jobs that offer a livable wage and to keep jobs that they now desperately need.

    Welfare reform is evolving every year, making it difficult to evaluate its impact: Findings from studies conducted in 1996 may not be relevant to welfare reform in 2001. As more families are discouraged from entering the welfare system, it will become even more difficult to determine the impact of welfare reform, because there will be no comparison group: It will be impossible to determine which families would have been on welfare if the system had not changed. And of course, if the economy weakens, it is expected that many more families will find it difficult to survive under welfare reform. The studies in this issue, however, provide important warnings about the types of women and families who are likely to be at risk under the evolving system and the services and waivers that will be needed to protect America’s most vulnerable families from harm.