The Personal Responsibility and Work Opportunity Reconciliation Act, enacted in 1996, signaled a profound shift in American welfare policy. "Welfare reform" eliminated the 60-year-old federal guarantee of ongoing cash assistance to all qualifying poor families and replaced it with time-limited benefits, for which recipients must work, provided through capped block grants to the states. It also transferred primary responsibility for the operation of welfare programs from the federal government to the states.

Because cash benefits under the Aid to Families with Dependent Children (AFDC) program historically had been reserved almost exclusively for single mothers, many welfare reform proponents argued that over time, the AFDC system itself had become a powerful disincentive to marriage. However inadvertently, they argued, AFDC undermined the "traditional" family by encouraging nonmarital childbearing among poor, disproportionately minority, women. Accordingly, one of the four stated purposes of the new block-grant program, Temporary Assistance to Needy Families (TANF), is to "prevent and reduce the incidence of out-of- wedlock pregnancies." And while previous policy interventions aimed at unmarried women's fertility had focused largely on services to help women prevent unintended pregnancies, TANF includes a range of policies to address a nationwide epidemic of what was presumed to be intended nonmarital childbearing. Nationally, the proportion of births that were to unmarried women rose eightfold between 1940 and 1999, from 4% to 33%—a ratio that has remained relatively stable since 1994 (see chart below).

Nonmarital Childbearing

The percentage of births to unmarried women has increased dramatically in the past six decades.

Notes: Data for 1999 are preliminary. Figures are based on weighted data rounded to the nearest individual. Source: Ventura SJ and Bachrach CA, Nonmarital childbearing in the United States, 1940-99, National Vital Statistics Reports, 2000, Vol. 48, No. 16.

At the same time, the 1996 law embodies broader social objectives aimed at reforming individuals' sexual behavior and restoring "traditional family norms" not only among those on welfare but among all Americans. For that reason, some of its provisions (such as the requirement that unmarried minor mothers live at home or in supervised settings and stay in school) apply specifically to welfare recipients, while others (such as those supporting abstinence-only education and instituting the so-called illegitimacy bonus) do not.

As Congress lays the groundwork to reauthorize the welfare law in 2002, one of the challenges it will face is the lack of a straightforward approach to assessing whether the law has had the desired effect on birthrates—and whether any identified effect stemmed from changes in sexual activity, contraceptive use or recourse to abortion. The pathways of causal effect are complex, and the data are difficult to come by.

Illegitimacy Bonus

No provision better demonstrates the framers' intent to use welfare reform as a vehicle to change behavior and influence individuals' childbearing decisions than the "illegitimacy bonus." For each of the four years beginning in 1999, the five states that achieve the greatest decline in their ratio of out-of-wedlock births to total births, while holding their abortion rate to below its 1995 level, are to receive $20 million each. If fewer than five states qualify in any given year, each is to receive $25 million. Significantly, the bonuses are for reductions in out-of-wedlock births among all women, not just welfare recipients or teenagers.

The first round of bonuses was awarded in 1999 to Alabama, California, the District of Columbia, Massachusetts and Michigan. The second year's bonuses were awarded to Alabama, Arizona, the District of Columbia, Illinois and Michigan. Among recipients in 2000, decreases in births to unmarried mothers ranged from moderate (4.13% in the District of Columbia) to infinitesimal (0.022% in Illinois).

In a survey conducted in 1998 by The Alan Guttmacher Institute, officials in health and social services agencies in 34 states and the District of Columbia reported that their state had taken steps to reduce out-of-wedlock births in light of the 1996 welfare law. Some states reported having acted in response to a formal decision to seek the illegitimacy bonus; others, however, had taken steps to lower nonmarital childbearing even in the absence of any official policy or, in some cases, after concluding that they had little chance of winning the bonus. In general, state efforts fall into three broad categories: programs and policies designed to increase contraceptive use among welfare recipients and other low-income women; activities to prevent teenage pregnancies; and grants to support local initiatives aimed at both adults and teenagers.

While the prospect of winning $20 million through the illegitimacy bonus apparently has helped to galvanize many states to address out-of-wedlock childbearing, states are not required to inform the federal government of steps they have taken to compete for the money. (Nor must they inform other states, which would allow them to learn from each others' experiences.) In fact, the law does not require states to do anything to win the bonus; the Department of Health and Human Services (DHHS) uses data that the states routinely submit to the federal government to determine the bonus winners.

Moreover, the initiatives launched since 1996 are too recent to have been an influence on changes on which the bonus is calculated, at least in the first two rounds. Round one was based on changes in out-of-wedlock births between 1994 and 1997; round two, on changes between 1995 and 1998. Realistically, any effects state initiatives have will not show up in the data for some time; even then, of course, it may not be possible to demonstrate cause and effect. When the winners of round two were announced, a DHHS spokesperson acknowledged that the bonus may be more a reflection of demographic changes (such as an increase in marital births or in the number of teenage women) than of program initiatives. A state's illegitimacy ratio could very well change without any policy intervention.

Family Cap

One of the basic premises of AFDC was that families on welfare would receive additional benefits whenever they had another child. In 1992, New Jersey became the first state to institute a "family cap"—a policy to deny these additional benefits when a child is born—after gaining DHHS approval for a research and demonstration "waiver." On the assumption that poor women were having children in order to obtain AFDC benefits and having additional children in order to increase those benefits, family cap proponents advocated mandating the cap in TANF. Opponents countered that the increase in benefits that normally accompanies a birth is too small to be a factor in a woman's decision to have another child. (In 1997, monthly increases for an additional child ranged from about $24 in Mississippi to $109 in California.) In addition, some conservative groups and antiabortion advocates argued that the family cap could result in more abortions among welfare recipients. In the end, Congress decided not to include a family cap provision in the law, leaving states free to adopt such policies if they wish. According to the Center for Law and Social Policy, 23 states have some type of family cap in place today (see table). Only a few states' family caps have been established since 1996; most were approved as waivers under the AFDC program.

Despite some concerns that the data could never prove a causal link between the family cap and individuals' childbearing decisions, a few studies have used experimental research designs in an attempt to assess the cap's impact. So far, the results are inconclusive. Investigators at the University of Arkansas School of Social Work reported no apparent effect on birthrates in Arkansas for the period 1994-1997, although they acknowledge that their sample was small. On the other hand, evaluators at Rutgers University found that the family cap in New Jersey did exert some influence over the family formation decisions of women on AFDC. They estimated that between 1993 and 1996, there were roughly 14,000 fewer births, almost 1,500 more abortions, and some 7,000 more family planning visits among welfare recipients than would have occurred in absence of the cap. In a third study, researchers at Cornell University, using state-level data for the period 1992-1996, concluded that although they were unable to control for some variables, the cap was associated with reductions in nonmarital childbearing among women of all races and ages—including a 9% decrease among teens and a 12% decrease among older women.

Minor Parent Provisions

Teens account for only about 5% of the welfare caseload; nevertheless, policymakers' interest in minor parents has burgeoned in recent years as evidence has become available to suggest that unmarried women who begin childbearing in their teenage years very often end up on welfare. According to a 1990 Congressional Budget Office report, almost three-quarters of unmarried teen mothers received AFDC within five years of giving birth. Marriage is no guarantee against welfare, however; the report notes that one-quarter of married teen mothers received AFDC during the same period.

In an attempt to interrupt this cycle, authors of the 1996 welfare law included a requirement in TANF that unmarried minor mothers, with few exceptions, live at home or in some other supervised setting and stay in school in order to receive welfare benefits. The living-arrangement requirement was premised on the notion that some teenagers who are unhappy living at home may have a child in order to get welfare benefits, which would enable them to set up an independent household. Opponents argued that requiring minor mothers to live with a parent or guardian could put them at risk of abuse or neglect. Most states (44) have dealt with this by allowing teen mothers to live independently with state approval, and all 50 states and the District of Columbia consider living with another adult relative as meeting the requirement.

Very little is known about whether this requirement has any effect on sexual behavior or reproductive decisions, and what is known may be surprising. Researchers at Cornell University found that the rule is associated with a 10% increase in nonmarital childbearing among teens. Speculating as to why a policy intended to lower out-of-wedlock childbearing may have actually resulted in an increase, the researchers suggest that for some teens, receiving their parents' support in caring for their child may add a measure of security.

The provision requiring teen mothers to attend school or training in order to collect benefits grew out of the idea that women with higher educational aspirations are less likely to become teenage mothers then their peers who have lower expectations. As the possibility of a successful future becomes more realistic, the argument goes, the cost of having a child increases. While several studies provide evidence that high levels of school engagement are associated with decreased pregnancy rates, the limited research on the school-attendance requirement has failed to demonstrate that these initiatives have an independent impact on initial or repeat teen childbearing.

Abstinence-Only Education

Intent on addressing the root cause of nonmarital childbearing among all women as a part of welfare reform, Congress enacted a major new abstinence education initiative, with the stated purpose of establishing educational programs across the country that censure all sex outside of marriage, at any age. Commenting on the intent of Congress in drafting the provision, congressional staff members Ron Haskins and Carol Statuto Bevan wrote in the Children and Youth Services Review in 1997 that "the explicit goal of the abstinence-only education programs is to change both behavior and community standards for the good of the country." Notwithstanding that the standard required by law may seem outdated to some, they went on, the "standard was intended to put Congress on the side of social tradition...that sex should be confined to married couples."

Under the auspices of the maternal and child health (MCH) block grant, the abstinence provision of the welfare bill guarantees $50 million annually for five years. Further, states are required to match every four dollars in federal funds with three dollars, bringing the annual total public expenditure to $87.5 million. Since the passage of the welfare law, Congress has also appropriated funding for FY 2001 for abstinence-only education under two other programs: $9 million under the Adolescent Family Life Act and $20 million under the MCH block grant's Special Projects of Regional and National Significance program.

To qualify for funding under all three of these programs, education programs must adhere to a strict eight-point definition of "abstinence education." (For example, the law requires that such a program teach that "sexual activity outside of the context of marriage is likely to have harmful psychological and physical effects" and offer instruction on "the importance of attaining self-sufficiency before engaging in sexual activity.") A state's program may not be inconsistent with any of the provisions of this definition; nevertheless, most states have taken advantage of flexible regulations that do not require them to emphasize all eight points equally. In 1998, the Association of Maternal and Child Health Programs (AMCHP) conducted a survey of state MCH directors and adolescent health coordinators to gather information about the implementation of the first year of abstinence-only education funding (FY 1998). AMCHP reports that many state initiatives include media and public education campaigns, mentoring and counseling activities, and curriculum development. Furthermore, although the federal law is written so as to apply to individuals of all ages, states are allowed to determine which age-groups to target. In a survey of state abstinence education coordinators, the Sexuality Information and Education Council of the United States found that the majority of programs target preteens and the youngest teenagers (see chart below).

A Focus on Youth

Although the abstinence-only program was intended for all age-groups, most states have targeted preteens and the youngest teens.

Age-group targeted

Note: Data are for 45 states and the District of Columbia. Source: Daley D and Wong V, Between the Lines: States' Implementation of the Federal Government's Section 510(b) Abstinence Education Program in Fiscal Year 1998, New York: Sexuality Information and Education Council of the United States, 1999.

Abstinence-only education is not a new concept; however, very little research exists that adequately evaluates its effectiveness, especially since the enactment of the welfare law's eight-point definition. Many studies rely on changes in attitudes to measure changes in behavior—a link that is not always valid. Few studies use control groups, and fewer still involve long-term follow-up. The scant research that is available suggests that such programs have little or no effect on initiation of sexual intercourse. In fact, conclusions drawn from international literature reviews conducted in 1993 (by the World Health Organization's Global Programme on AIDS) and in 1997 (by the National Centre in HIV Social Research at Macquarie University in Sydney, Australia) indicate that the programs that are most effective in changing young people's behavior—in terms of both delaying their initiation of sexual intercourse and promoting their contraceptive use—are those that address abstinence along with contraception and sexually transmitted disease prevention.

In response to concerns that the welfare law's massive expansion of funding for abstinence-only education lacked an evaluation component of any kind, Congress approved $6 million for national evaluation of abstinence education efforts with the passage of the Balanced Budget Act of 1997. The national evaluation is being conducted on a small scale, at six sites across the country. Interim findings are expected to be available in August 2001. Most states are also planning to undertake some type of evaluation. AMCHP reports that states have targeted slightly more than 5% of their abstinence education budgets for such activities.

Challenges Ahead

In the Children and Youth Services Review article on the welfare bill, proponents of the legislation acknowledged that the "attack on illegitimacy is based far more on the value position that sex outside of marriage is wrong" than on solid evidence that their actions would produce substantial results. Congress forged ahead with a range of new policies based on the judgment that the nation faced a serious social problem and that strong action was therefore justified. But four years later, with very limited empirical evidence, Congress faces a daunting challenge of considering how well these policies have worked to reduce nonmarital births, among American women in general and welfare recipients in particular. While actual debate on reauthorization of the welfare law will not take place until 2002, when authorized funding expires, Congress is expected to begin oversight hearings and general consideration in 2001.