"This article investigates the role of surprises in marital dissolution [in the United States]. Surprises consists of changes in the predicted earning capacity of either spouse. Data from the National Longitudinal Study of the High School Class of 1972 is used. We find that an unexpected increase in the husband's earning capacity reduces the divorce hazard, while an unexpected increase in the wife's earning capacity raises the divorce hazard. Couples sort into marriage according to characteristics that are likely to enhance the stability of the marriage. The divorce hazard is initially increasing with the duration of marriage, and the presence of children and high levels of property stabilizes the marriage."

An analysis of the economic impact of divorce settlements in the United States is presented using data for a white cohort taken from the National Longitudinal Study of the High School Class of 1972. "The effects of spouses' incomes on the divorce transfer are estimated and used to simulate the welfare effects of divorce on husbands, wives, and children under alternative assumptions about marriage contracts and the ability of a couple to continue coordinating resources in the aftermath of divorce. We find a positive (negative) relationship between divorce transfers and the growth of husband's (wife's) earnings during marriage. The estimated expenditure on children in the divorce state is only half the accustomed level during marriage."

The failure of many divorced fathers to comply with court-mandated child support awards has been identified as a major reason why a growing number of children live in poverty in female-headed households. This paper presents a model that seeks to explain why so many divorced fathers allow their children's welfare to suffer as a consequence of divorce. The point of departure is the recognition that children are collective consumption goods from the point of view of the father and mother. Within marriage, proximity and altruism serve to overcome the "free-rider" problem associated with the provision of public goods. However, on divorce the noncustodial parent suffers a loss of control over the allocative decisions of the custodial parent and it is not feasible for the couple to achieve a Pareto-optimal allocation of their joint resources. A model of optimal marriage contracts is constructed in which a couple decides on the allocation of resources within marriage and on the terms of a settlement in the event of divorce. The settlement consists of the determination of custody and transfer of income to the custodial parent. Divorce is endogenous in our model and its occurrence depends on a random variable introduced into the family budget constraint that measures the quality of the marriage match. The analysis yields several useful insights. In addition to explaining the apparently insufficient support by custodians we explain why custody rights and alimony transfers often go in the same direction, most commonly to the wife, and why uneven distribution of income between the spouses increases the probability of divorce.

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.

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.