The proposed research will examine a woman's choice of marital status, with special attention to the way these choices are affected by economic incentives. Characteristics of the welfare system, income of a woman's current (or potential) spouse, and her wage rate will all be considered as they affect this decision. The underlying economic model utilizes indirect utility functions in order to facilitate direct modeling of income and price effects which are dependent on marital status. The virtue of this approach is that it explicitly and directly accounts for the actual alternatives facing each woman and does so through the standard economic analysis of consumer choice and utility theory. The empirical work will focus on marital status transitions rather than, as in cross-sectional work, a woman's current status. Special emphasis will be given to the analysis of remarriage by divorced women and first marriage by single women. We expect to derive estimates of the effects of economic incentives in general and of the specific effects of the income of a potential spouse, welfare income and benefit reduction rates (for women on AFDC), alimony, and a woman's own wage rate. Careful attention will be given to the analysis of race differences in marriage and remarriage rates. We will examine whether the large race differences persist even after specifying the relevant marriage and non-marriage alternatives, including welfare and potential spouse's income, available to white and black women. Particular attention will be given to estimating the economic position of marriage-eligible black males. The empirical work will address the econometric issues inherent in estimating a choice-based model in which some alternatives are observed for some women. Appropriate maximum likelihood procedures which correct for non-random sample selection will be used. No previous model of choice of marital status has both adopted a choice-based, utility-maximization framework and used appropriate econometric techniques. All of the empirical work will be done using longitudinal data from the Panel Study of Income Dynamics.
| Hoffman, S D; Duncan, G J (1988) Multinomial and conditional logit discrete-choice models in demography. Demography 25:415-27 |
