Breaking the Cycle of Welfare Dependence

  • The success of the 1996 welfare reform, also known as the Personal Responsibility and Work Opportunity Rec-onciliation Act of 1996 (PRWORA), is widely acknowledged and well documented. Significantly, Temporary As-sistance for Needy Families (TANF) ended the entitlement nature of the old welfare program, Aid to Families with Dependent Children (AFDC), which distributed cash to recipients without time limits and without requiring them to work or prepare for work in return.
  • In addition to fixing the funding structure and including work requirements and time limits, the landmark legislation also addressed the primary causes of welfare dependence and child poverty in the U.S.: family breakdown and unwed childbearing.
  • The immediate effects of welfare reform were striking.1 During the four decades before enactment of the 1996 reform, the welfare caseload never decreased significantly. By 1995, one in seven children was on AFDC, but within just a few years after TANF’s implementation, the caseload fell by half and employment rates and earnings among single mothers soared.2
  • Child poverty rates declined significantly. Roughly 3 million fewer children lived in poverty in 2003 than in 1995, including 1.2 million fewer African American children—the lowest level of black child poverty in the nation’s history.3

Research on Negative Effects of Welfare Dependence: Social science research published in peer-reviewed academic journals suggests that welfare participation is associated with negative effects for children and adults and with an intergenerational cycle of dependence.
  • Welfare participation and early childhood cognitive development. A 2011 study published in Children and Youth Services Review analyzed the effect of welfare participation in the TANF program on young children’s cognitive development.4 It found that, compared to children in non-welfare families, those whose families received welfare when they were between three and five years old had, on average, lower cognitive development (about 11 percent of a standard deviation difference on the Peabody Picture Vocabulary Test). The study considered a host of factors, including child and maternal background characteristics as well as family dynamics and the home environment, but the negative effect of participation in TANF persisted. Further analysis suggested that increased maternal stress and lower household income among TANF families explained about 7 percent and 19 percent, respectively, of the cognitive deficiency between five-year-olds whose families received welfare and those whose families did not.
  • Welfare receipt and children’s educational attainment. Welfare receipt, particularly during adolescence, also appeared to have a negative effect on children’s educational attainment. Using nationally representative data that tracked individuals born between 1967 and 1978 throughout their childhood, a 2003 Demography study found that one year of welfare receipt during ages 11 to 15 was associated with a reduction of more than a quarter of a year in total schooling.5 Moreover, the likelihood of high school completion diminished by 14 percent for each year of welfare receipt between age 11 and age 15 and by 6 percent for each year of welfare receipt between age 6 and age 10.
  • Intergenerational welfare receipt. Research suggests that welfare participation may adversely affect the next generation’s economic well-being. A 2003 study in the Journal of Marriage and Family found that, compared to women whose families, when they were between age 8 and age 13, did not receive welfare, those whose families participated in welfare were more likely to receive welfare themselves.6 The effect of intergenerational welfare receipt may be partially explained by the adult children recipients’ employment, education, and marital characteristics. That is, parental welfare receipt was linked to children’s employment, education, and marital status in adulthood. For example, compared to mothers who gave birth out of wedlock, married mothers who remained married averaged three fewer years of welfare participation.
  • Intergenerational welfare dependence. Using a nationally representative survey that followed the same group of respondents from childhood through adulthood, a 2004 study on the intergenerational transmission of welfare dependence also examined the likelihood of receiving welfare in adulthood if one’s parents ever received welfare.7 Welfare receipt included participation in AFDC (the pre–1996 reform cash assistance program); food stamps; or Supplemental Security Income (SSI). The study found that the average probability of welfare participation for the entire study sample was over a quarter (0.275); if their parents had ever received welfare while the respondents were growing up, their probability of welfare receipt as adults increased to nearly one-half (0.468)—nearly three times the probability (0.166) for respondents whose parents did not receive welfare.
Research on Positive Effects of Welfare Reform: In contrast to the negative effects associated with welfare participation, research has shown that the mid-1990s welfare reform and leaving the welfare system were linked to a number of positive outcomes for child and adult well-being.
  • Welfare reform and caseloads declines. A 2011 Social Science Journal study analyzed the effects of the mid-1990s welfare reform on welfare caseloads.8 Using state-level data from 1992 to 2005 that included all 50 states and the District of Columbia, it found that during the study period, on average, “welfare reform led to a 41 percent decline in welfare recipiency,” while unemployment rates accounted for only about 4 percent of the decline in caseloads. Four policies—family cap, work requirements, time limit, and cash sanction for rule infractions—together were associated with a 28 percent reduction in welfare recipiency over the study period. Considering the specific policies, over time, aid sanctions were associated with an 18 percent reduction in caseloads, and family caps were associated with an 11 percent reduction.
  • Welfare reform and abortion and pregnancy rates. Work requirements and time limits were two of the main components of the landmark 1996 welfare reform. A 2012 Marriage & Family Review study explored the link between these two components and fertility outcomes.9 It used state-level data from 1992 to 2005 and from all 50 states and the District of Columbia. It found that work requirements and time limits were each associated with reduced abortion rates. Work requirements were also linked to reduced pregnancy rates.
  • Welfare reform and life satisfaction. Using a nationally representative survey on Americans’ life satisfaction, a 2012 study published in Social Service Review examined the subjective well-being of single mothers after the implementation of the mid-1990s welfare reform.10 On average, low-skilled single mothers made significant absolute and relative gains in overall life satisfaction compared to low-skilled single women without children as well as low-skilled married mothers. The pre-reform deficit in life satisfaction experienced by single mothers compared to single women without children declined by 60 percent post-reform, and the gap between single mothers and married mothers declined by 22 percent. In addition to overall life satisfaction, low-skilled single mothers made gains and narrowed the gap between themselves and their single, non-mother peers and married mothers in the areas of expressing regrets about the past and self-reported financial satisfaction.Welfare participation and neighborhood quality. Analyzing a survey of children and mothers in low-income and moderate-income neighborhoods in Boston, Chicago, and San Antonio, a 2011 Social Science Quarterly study found that employment and leaving the welfare system was associated with better-quality neighborhoods for former recipients.11 Over a period of six years, compared to women who remained on welfare, those who left welfare lived, on average, in neighborhoods with less concentrated socioeconomic disadvantage. More specifically, the difference in neighborhood quality was significant only for women who left welfare with employment; those who left without employment remained, on average, in neighborhoods similar in quality to the neighborhoods of those who stayed on welfare. Further analysis suggested that greater residential mobility through employment enabled women who left welfare to experience better-quality neighborhoods. In particular, it appeared that the improvement came when the recipients left the welfare system first and then later moved to a new neighborhood, not vice versa.
  • Social capital and marriage and stable employment. Analyzing a representative survey of large cities in the U.S., a 2011 Journal of Sociology & Social Welfare study explored how social capital might have enhanced the likelihood of marriage and stable employment for low-income women in unstable employment who gave birth out of wedlock.12 Social capital was a compilation of several factors, including the ability to count on family for financial assistance, housing, and child care in times of need. The study found that, compared to similar low-income unwed mothers with lower levels of social capital, those who perceived greater social capital in their lives when their children were one year old had greater odds of being married and in stable employment two years later.
  • Welfare reform and illicit drug use. Using all available nationally representative surveys that were appropriate for the analysis of welfare reform and illicit drug use, a 2013 study in Economic Inquiry analyzed the link between these two factors.13 It considered several measures of welfare reform as well as drug use outcomes, ranging from overall illicit drug use to “hard core” illicit drug use, marijuana use, drug-related arrests, drug-related emergency room visits, and probable drug-related prison admissions. A consistent pattern emerged. As the authors of the study concluded, “Welfare reform reduced illicit drug use.” The magnitude of the decline ranged from 10 percent to 21 percent in illicit drug use, 6 percent to 7 percent in drug-related arrests, 7 percent to 11 percent in emergency room incidents, and 11 percent to 19 percent in prison admissions.
Conclusion: The research is compelling. Welfare dependence has lasting consequences for individuals, both children and adults, as well as for society generally. The decades before enactment of the 1996 welfare reform had demonstrated that the solution to helping our society’s most vulnerable and needy does not lie with a more expansive welfare state.
  • Since the beginning of the War on Poverty in the mid-1960s, the government has spent more than $19.8 trillion (in 2011 inflation-adjusted dollars) on nearly 80 welfare programs providing cash, food, housing, medical assistance, and targeted social services to poor and low-income Americans. The 1996 welfare reform was successful in reducing dependence on government and lifting individuals out of poverty,14 but it reformed only one welfare program.
  • Under the Obama Administration, the means-tested welfare state has continued to explode, with a total spending of $927 billion in FY 2011—about $9,040 for each American whose income falls in the bottom third of the population. Yet poverty rates remain stagnant because current government programs do not address the primary causes of poverty and dependence on government. Successful policies and solutions aim to restore self-sufficiency and economic opportunity to individuals and seek to strengthen the foundational institutions of marriage and civil society that can combat poverty and welfare dependence effectively.


  1.  Robert Rector, “An Overview of Obama’s End Run on Welfare Reform,” Heritage Foundation Issue Brief No. 3735. September 20, 2012, 
  2. Robert Rector and Patrick F. Fagan, “The Continuing Good News About Welfare Reform,” Heritage Foundation Backgrounder No. 1620, February 6, 2003,
  3. Robert Rector, “Obama’s End Run on Welfare Reform, Part One: Understanding Workfare,” Heritage Foundation Backgrounder No. 2730, September 19, 2012, note 6, 
  4. Colleen M. Heflin and Sharon Kukla Acevedo, “Non-Income Effects of Welfare Receipt on Early Childhood Cognitive Score,” Children and Youth Services Review, Vol. 33, Issue 5 (May 2011), pp. 634–643. 
  5.  Inhoe Ku and Robert Plotnik, “Do Children from Welfare Families Obtain Less Education?” Demography, Vol. 40, No. 1 (February 2003),pp. 151–170. 
  6.  Molly A. Martin, “The Role of Family Income in the Intergenerational Association of AFDC Receipt,” Journal of Marriage and Family, Vol. 65,No. 2 (May 2003), pp. 326–340.
  7. Marianne E. Page, “New Evidence on Intergenerational Correlations in Welfare Participation,” in Generational Income Mobility in North America and Europe, ed. Miles Corak (New York: Cambridge University Press, 2004). 
  8.  Radha Jagannathan, “Welfare Reform’s Impact on Caseload Decline in the United States: An Application of Latent Trajectory Model,” The Social Science Journal, Vol. 48, Issue 4 (December 2011), pp. 703–721. 
  9.  Radha Jagannathan, “A Latent Growth Curve Model of Nonmarital Births, Abortions, and Pregnancies: Did Welfare Reform Play a Role in Changing the Trajectories?” Marriage & Family Review, Vol. 48, Issue 4 (2012), pp. 363–385. 
  10.  Christ M. Herbst, “Footloose and Fancy Free? Two Decades of Single Mothers’ Subjective Well-Being,” Social Service Review, Vol. 86, No. 2 (June 2012), pp. 189–222. 
  11.  Pamela R. Bennett and Andrew J. Cherlin, “The Neighborhood Contexts in Which Low-Income Families Navigate Welfare Reform: Evidence from the Three-City Study,” Social Science Quarterly, Vol. 92, No. 3 (September 2011), pp. 735–760. 
  12. Jennifer A. Johnson, Julie Honnold, and Perry Threlfall, “Impact of Social Capital on Employment and Marriage Among Low Income Single Mothers,”Journal Sociology & Social Welfare, Vol. 38, No. 4 (December 2011), pp. 9–31. 
  13.  Hope Corman, Dhaval M. Dave, Dhima Das, and Nancy E. Reichman, “Effects of Welfare Reform on Illicit Drug Use of Adult Women,” Economic Inquiry, Vol. 51, No. 1 (January 2013), pp.653–674.