Essays on Family and Labor Economics

Sammanfattning: Are Divorcing Parents Induced to Choose Joint Custody? An Evaluation Using Implementation Theory: What demands do divorcing parents make in court with respect to custody, under different legal regimes. This question is analyzed using the tools from implementation theory. Joint custody is viewed to be beneficial to the child in a majority of divorces. The legislation should therefore induce parents to agree on joint custody. My results indicate that the legislation works well in most cases. However, a parent who ranks joint custody as the worst alternative can and will veto joint custody. A proposed change in the legislation aims to further emphasize joint custody. However, my results indicate that the parents' tendency to accept joint custody may be reduced.The Evolution of the Marriage Premium in the Swedish Labor Market 1968- 1991: Married and cohabiting men in Sweden earn more than single men. Longitudinal data also indicates that these wage differentials may partly be explained by the selection of men into partnership based on unobserved productivity. This explanation to the marriage premium grows more in important over the investigated period. I also find, though only in the early part of the period and only for legally married, that wages increase with time married and that men with non-working spouses earn more than other married men. These results are in line with Becker's hypothesis of specialization between spouses. During this time period in Sweden, reforms have been undertaken to induce spouses to share labor market and housework more equally. If the wage differential between married and single men reflects specialization, we would expect it to decline.Estimation the Impact of Wage Structure on Wage Differentials: Juhn, Murphy and Pierce (1991), JMP, decompose changes in the unexplained wage differential between two groups of workers into two parts. One part is due to changes in the average amount of unobserved skills, and the other to changes in returns to unobserved skills. An alternative estimator to the JMP estimator is suggested which is approximately unbiased, consistent, and easy to calculate. The statistical properties of the JMP estimator depend on the distribution of the error term. Applied to the decline in the gender wage gap in Sweden, the two estimators produce qualitatively similar results.