The rise in the number of female-headed families in the U.S. has generated interest in the effects of the Aid to Families with Dependent Children program (AFDC) and other transfer programs on women's choices concerning marriage, childbearing, and employment. This paper looks at marriage rates by never married women using a discrete time hazard model allowing for left and right censoring. It addresses the question of how the marriage market and welfare policy affect marriage rates using an improved measure of spouse availability.

Concern over the availability of desirable spouses stems from evidence on the low rate of family formation among blacks. Bane and Ellwood (1983), OINeill, et al. (1984), and Blank (1986) all report that black women have a much lower probability of leaving AFDC by marriage than whites. Wilson and Neckerman (1986) show that black women face a shrinking pool of “marriageable” employed black men. They suggest that the rise in black female headed families over recent decades is more closely linked with this diminishing pool of marriage partners than with expansion of transfer programs. Earlier work by Honig (1974) had suggested that unemployment and low male earnings significantly contributed to the formation of female-headed families. My study uses employment based sex ratios to test a Wilson and Neckeman-style hypothesis that low availability of “marriageable” men reduces marriage rate. Hypotheses are tested using data from the 1984 Panel of the Survey of Income and Program Participation (SIPP), although the small sample size of blacks who marry does not allow separate estimation by race.