This paper presents new identifiability conditions for the Cox proportional hazard model for duration data when unobserved person specific variables are present. We compare our conditions with those presented by Elbers and Ridder. We also present identifiability conditions for a rich class of parametric hazard models without regressor variables.

This paper tests the hypothesis that the classifications "unemployed" and "out of the labor force" are behaviorally meaningless distinctions. This hypothesis is rejected. Distinct behavioral equations govern transitions from out of the labor force to employment and from unemployment to employment. The evidence reported in this paper is broadly consistent with versions of search theory in which unemployment is a state that facilitates the job search process. In an Appendix, we demonstrate that log concavity of the wage-offer distribution implies that the exit rate from unemployment is an increasing function of the rate of arrival of job offers.

This paper examines the properties of instrumental variables (IV) applied to models with essential heterogeneity, that is, models where responses to interventions are heterogeneous and agents adopt treatments (participate in programs) with at least partial knowledge of their idiosyncratic response. We analyze two-outcome and multiple-outcome models, including ordered and unordered choice models. We allow for transition-specific and general instruments. We generalize previous analyses by developing weights for treatment effects for general instruments. We develop a simple test for the presence of essential heterogeneity. We note the asymmetry of the model of essential heterogeneity: outcomes of choices are heterogeneous in a general way; choices are not. When both choices and outcomes are permitted to be symmetrically heterogeneous, the method of IV breaks down for estimating treatment parameters.

This paper assesses the contribution of federal antidiscrimination policy to the dramatic improvement of black economic status in manufacturing that occurred in South Carolina in the mid 1960's. Using a unique data source on wages and employment by race and sex in South Carolina we evaluate competing explanations. Human capital stories, supply shift stories and tight labor market stories do not account for the black breakthrough. Our study documents a significant contribution of federal antidiscrimination programs.

This study explores relationships between parental resource trajectories and child development and their implications for intergenerational mobility. By modifying the child skill formation technology to incorporate new skills emerging during adolescence, we analyze the importance of the timing of family resources on life outcomes, educational attainment, and participation in crime. Parental financial resources partially offset deficiencies in nonpecuniary inputs to children’s human capital. Estimates of the intergenerational influence on child outcomes are strongly influenced by the choice of lifetime versus snapshot parental income measures. The most predictive ages of children when family resources are measured vary by the outcome analyzed.

This paper estimates a dynamic model of schooling attainment to investigate the sources of racial and ethnic disparity in college attendance. Parental income in the child's adolescent years is a strong predictor of this disparity. This is widely interpreted to mean that credit constraints facing families during the college-going years are important. Using NLSY data, we find that it is the long-run factors associated with parental background and family environment, and not credit constraints facing prospective students in the college-going years, that account for most of the racial-ethnic college-going differential. Policies aimed at improving these long-term family and environmental factors are more likely to be successful in eliminating college attendance differentials than short-term tuition reduction and family income supplement policies aimed at families with college age children.

This paper examines an empirical regularity found in many societies: that family influences on the probability of transiting from one grade level to the next diminish at higher levels of education. We examine the statistical model used to establish the empirical regularity and the intuitive behavioral interpretation often used to rationalize it. We show that the implicit economic model assumes myopia. The intuitive interpretive model is identified only by imposing arbitrary distributional assumptions onto the data. We produce an alternative choice‐theoretic model with fewer parameters that rationalizes the same data and is not based on arbitrary distributional assumptions.

In this paper, we discuss statistical problems that arise in studying sequences of quantal responses (e.g., labor force participation) in panel data on heterogeneous populations (i.e., populations in which there is unobserved variation in response probabilities). Assuming that response probabilities are governed by a beta distribution, we derive a generalization on of the cross-section logit model to enable it to deal with sequences of discrete events in panel data. This model is applied to panel data on labor force participation of married women. One of our findings is that the distribution of participation probabilities is U shaped, indicating that most women have participation probabilities near zero or near one.