This paper develops a generalized Roy model with human capital accumulation, moral hazard, and career concerns. We identify and estimate the model with a large panel that matches data on publicly listed firms to information on their executives. The structural estimates obtained are used to decompose the firm‐size pay gap. We find that although total compensation and incentive pay increase with firm size, certainty‐equivalent pay decreases with firm size. In larger firms, and for more highly ranked executives, weaker signal quality about effort results in higher risk premiums. This risk premium accounts for roughly 80 percent of the firm‐size gap in total compensation. Larger firms are also willing to pay more than smaller ones to attract executives. Finally, the estimated coefficients on human capital accumulation from formal education and experience gained from different firms are individually significant, but their collective effect on firm‐size pay differentials nets out.

An empirical approach to optimal income taxation design is developed within an equilibrium collective marriage market model with imperfectly transferable utility. Taxes distort time allocation decisions, as well as marriage market outcomes, and the within household decision process. Using data from the American Community Survey and American Time Use Survey, we structurally estimate our model and explore empirical design problems. We allow taxes to depend upon marital status, with the form of tax jointness for married couples unrestricted. We find that the optimal tax system for married couples is characterized by negative jointness, although the welfare gains from jointness are modest. These welfare gains are then shown to be increasing in the gender wage gap, with taxes here, as in the case of gender based taxation, providing an instrument to address within household inequality.