The University of Michigan
555 South Forest Street
Third Floor
Ann Arbor, MI 48104-2531

T 734-936-9842
F 734-998-6341



The 1996 welfare reform legislation led to a dramatic reduction in the number of individuals receiving cash assistance. Although provisions were put in place to preserve Medicaid eligibility for individuals no longer receiving cash assistance, uncertainty remained about how welfare reform would affect the insurance coverage of poor and near poor individuals. A study by Harold A. Pollack, Mathew M. Davis, Sheldon Danziger, and Sean Orzol funded by the Economic Research Initiative on the Uninsured (ERIU), examines the health insurance coverage status of “welfare leavers” in the post-welfare reform period. They consider effects on coverage and access to care among women and children.

In Michigan and Nationally, Many Welfare Leavers Lack Health Insurance coverage

  • Among welfare leavers, 34% of mothers nationally and 27% of working mothers in Michigan lack coverage, compared to 8% nationally and 3% in Michigan of current welfare recipients (p<0.001).
  • Among children of welfare leavers, 22% (p<0.001) of the national sample and 9% (p<0.01) of the Michigan sample lack coverage while virtually all children of current welfare recipients have coverage. Maternal occupation was the strongest predictor of child coverage. Children of married, non-working mothers were more likely to lack coverage (odds ratio [OR]=2.44 of uninsurance compared to unmarried working mothers).
  • Maternal and child coverage are closely linked. In welfare leaver families, 33-38% of children with an uninsured mother lacked coverage, versus 2-3% of children with an insured mother.

Predictors of Coverage Among Welfare Leavers in Michigan

  • Women leaving welfare within the past year versus earlier are significantly less likely to lack coverage (OR=0.22, p<0.05).
  • Lacking coverage was less likely among welfare leavers in larger families versus smaller ones (OR=0.76, p<0.05), among African Americans versus non-Hispanic whites (OR=0.56, p<0.05), among those in managerial (OR=0.3, p<0.05) and operator jobs (OR=0.24, p<0.05) versus those in service or “other occupations,” and among those with a child with mental health concerns versus those without such concerns (OR=0.25, p<0.05).
  • Lacking coverage was more likely among welfare leavers over age 35 than among younger mothers (OR=1.49, p<0.05), among those with a child with physical health concerns versus those without (OR=2.5, p<0.05), and among women with their own physical health concerns versus those without (OR=2.02, p<0.05).

Access to Care Among Welfare Leavers

  • Adult welfare leavers who work are much more likely than current welfare recipients to have foregone (38% versus 14%) or delayed (21.8% versus 12.6%) needed medical care for financial reasons.
  • Reported delays in obtaining needed care were significantly less likely among African American mothers than among non-Hispanic white mothers (OR=0.52, p<0.05) and more likely among women with physical health barriers (OR=3.13, p<0.001), and among women with mental health barriers (OR=1.89, p<0.05) than women without such barriers.
  • Reported delays in obtaining needed care for children were significantly more likely among welfare leavers in states with more rapid caseload declines than in other states (OR=3.46, p<0.05).

One unintended consequence of welfare reform has been to reduce health insurance coverage among former welfare recipients. This decline has occurred despite policy initiatives intended to help welfare leavers retain coverage. Individuals leaving cash assistance are vulnerable to losing their health insurance coverage because the jobs they secure are less likely to come with offers of group health insurance. Post-welfare earnings are typically too low to allow these individuals to afford private coverage in the individual market. Former welfare recipients with physical or mental disabilities, who face greater health care needs, are more likely to lack coverage. Also, as others have found, children are more likely to lack coverage when a parent is uninsured.

Small sample size, particularly in the case of uninsured children in the Michigan sample, reduces the statistical power of this study. Most results with the statistical power necessary to reject the null hypothesis are bivariate comparisons across groups; multivariate models performed poorly. The analysis does not address the potentially endogenous relationship among health insurance status, welfare status, and health status. The data permit the analysis of factors influencing participation in a public program (e.g., Medicaid, SCHIP) but not of why eligible non-participants do not participate. For example, some families may have tried unsuccessfully to enroll in public programs. Others may have successfully enrolled initially did not complete administrative steps for continued enrollment.

The primary data source is the Women’s Employment Survey (WES), a 1997-2001 longitudinal study of single mothers who are current and former welfare recipients in one Michigan County. The 1999 National Survey of America’s Families (NSAF) household survey is used for national comparison data. Welfare leavers are defined as female head of households who left the welfare rolls following enactment of the 1996 welfare reform law. WES: 432 women in one Michigan county who received TANF in February 1997 but were not participating in fall 2001; all are U.S. citizens, ages 18-54, Caucasian or African American. NSAF: 992 women who had received their last benefits in August 1996 or later.

Bivariate comparisons and multiple logistic regressions are used to estimate health insurance coverage status and access/utilization among welfare leavers.

Health Insurance Coverage and Access to Care Among Former Welfare Recipients
Harold A. Pollack, Matthew M. Davis, Sheldon Danziger, Sean Orzol, University of Michigan

Conference paper presented at ERIU Research Conference, July 2002

ERIU Working Paper #13 (Adobe PDF)

Back to top

Funded by The Robert Wood Johnson Foundation, ERIU is a five-year program shedding new light on the causes and consequences of lack of coverage, and the crucial role that health insurance plays in shaping the U.S. labor market. The Foundation does not endorse the findings of this or other independent research projects.