Welfare Reform, the Business Cycle, and the Decline in AFDC Caseloads

Recent research by the Council of Economic Advisers (1997) and Ziliak, Figlio, Davis, and Connolly (1997) provides substantively different estimates of the impact of the macroeconomy and welfare reform in accounting for the recent decline in AFDC caseloads. In this paper we conduct an extensive reconciliation between the findings in the CEA and Ziliak et al. In addition, we also address the issue of how welfare recipiency might respond in the event of a recession via a dynamic simulation model. Lastly, we examine the possibility of interactions between the macroeconomy and welfare reform. Using the data and sample period employed by the CEA, our reconciliation suggests that the differences in results between the CEA and Ziliak et al. emanate largely from the treatment of dynamics. The primary consequence of controlling for caseload dynamics is to reduce the role of welfare reform relative to the macroeconomy in accounting for the decline in AFDC caseloads. Once we control for dynamics, we attribute up to 75 percent of the 1993 to 1996 caseload decline to the macroeconomy and at most 1 percent to welfare reform. Our dynamic simulations underscore the cyclical sensitivity of welfare recipiency -- a 2 percentage point increase in the unemployment rate leads up to an 11.7 percent increase in welfare recipiency after four years. Finally, the results from interactions between the macroeconomy and welfare reform indicate that pre-TANF welfare reform required a robust economy (i.e. low unemployment rates) in order to have a negative impact on recipiency rates.