Life-cycle models of labor supply predict the presence of an unobserved individual effect in the labor-supply equation that is correlated with observed explanatory variables, leading to an omitted variables bias in the cross section. I examine the sensitivity of parameter estimates to the presence of these effects, using fixed- and random-effect Tobit models. The estimated effects of children are too large in the cross section. The estimated intertemporal substitution elasticity ranges from 1.1 to 1.7. The results are similar for fixed- and random-effects models and for models using different specifications of the dependent variable.

We present estimates of the union wage effect controlling for unmeasured individual effects, and subject the conventional fixed-effects model to specification tests. For PSID men the union wage effect is 5-8% after controlling for person effects, as opposed to 20% in cross-section. Omnibus tests based on an unrestricted reduced form and instrumental variables tests based on differencing are consistent with conventional models. Tests based on comparing those who enter and leave union coverage provide evidence against the usual model. We find evidence for interactions between union status and other variables even after controlling for person effects.

This paper formulates and estimates a model of the determinants of female household headship. Headship responds to variations in the levels of well-being a woman can expect if she marries or if she heads her own household. We measure the opportunity cost of female headship and the effects of welfare benefits and women's work in the market on female headship and poverty. We find that if welfare benefits were reduced, there would be small reductions in the proportion of women heading households for whites and nonwhites, but a substantial increase in poverty for nonwhites. We also find that wives' work in the market reduces poverty and female headship for nonwhites, and reduces poverty, but increases headship for whites.