No. B96-21



During the last thirty years, the composition of white and black families in the United States has changed dramatically. In 1960, less than ten percent of families with children were headed by a single mother, while in 1990 more than twenty percent of families with children were female headed households. A large body of research has focused on the role of the U.S. welfare system, and in particular, the Aid to Families with Dependent Children (AFDC) program, in contributing to these dramatic changes in family structure. Most studies use cross-sectional data and identify the effect of welfare on female headship through interstate variation in the AFDC program. Recent research finds that controlling for state effects has a large impact on the estimated welfare effect. This paper examines why state effects matter for estimating the role of welfare in female headship decisions by examining the importance of individual effects and the determinants of AFDC policy. A natural explanation for why state effects matter is that the composition of the population across the states differs, and the composition is related to the generosity of the state's welfare program. The results show that after controlling for individual effects, there is no evidence that welfare contributes to increasing propensities to form female headed households for either whites or blacks. Further, the results suggest that for blacks, previous estimates of the welfare effect may be biased because of changes in the composition of the population over time. These results illustrate the potential problem which may result in an empirical application where state factors are assumed to be fixed over a long period of time.

Hilary Williamson Hoynes
University of California, Berkeley
Department of Economics
549 Evans Hall #3880
Berkeley,CA 94720-3880