Mary Ann Bronson

Assistant Professor, Economics

Georgetown University 

Curriculum Vitae

Research Area

Labor Economics, Applied Microeconomics

Published and Working Papers

We analyze wage growth non-parametrically and show that the main gender difference in wage growth is that women are less likely to experience infrequent, exceptionally large (right-tail) pay increases, primarily in years that they do not switch firms. These increases are persistent and move workers substantially through firms’ wage distributions, indicative of large internal promotions. They account for a remarkable 70% of the gap in wage growth by age 45. We  then study the extensive margin of such large promotions, including lifecycle dynamics, the role of sorting across firms, and effects of childbirth. We interpret the large set of novel facts we document using a theoretical model of careers within firms. Finally, we discuss implications for existing models of wage dynamics. Observed wage growth and promotion patterns – including a large residual gender penalty in promotion that reverses after age 40 – are consistent with costs to firms associated with employee labor supply reductions, and employer uncertainty about women’s future childbearing. On the other hand, they are difficult to reconcile through the lens of several common explanations in the literature. Additionally, our findings indicate that models focused primarily on sorting and matching behaviors can account for only a minority of the gender divergence in wages with age. 


How do different income taxation systems — for instance individual vs. joint — affect household decisions and welfare? What system should a welfare-maximizing government adopt? We provide evidence on this question, in three steps. First, we document that taxing married households jointly, as in the U.S., generates outsize distortions to secondary earner labor supply. Using quasi-experimental variation from past tax reforms, we estimate quantitatively large responses by secondary earners to the disincentives in the existing joint system.  Second, we develop a lifecycle model in which single and married individuals make decisions about labor supply, household production, consumption, savings, marriage, and divorce. We estimate the model using observed labor supply responses to past tax reforms, as well as auxiliary time use and consumer expenditure data. Lastly, we use the model to derive the optimal system for taxing married households. We show that the individual system maximizes revenue and per capita labor income in the economy by creating the least distortion to secondary labor supply. However, the cost of this type of tax system is a large reduction in welfare for the most vulnerable part of the population: individuals with the lowest education and ability, who benefit significantly from transfers to single-earner households under the income splitting system. We show that the optimal system exhibits some positive jointness, and can be approximated by implementing an income splitting system (as in the U.S.), combined with a targeted set of secondary earner deductions. The optimal system represents a substantial improvement over the current system. It increases female labor force participation by more than 5 percentage points, which would raise the U.S. from 35th to 19th place in OECD rankings of women’s employment rates.  Additionally, it increases individual welfare on average by a consumption equivalent of approximately $480, with the largest gains in the bottom half of the income distribution.


Abstract: 

Women attend college today at much higher rates than men. They also select disproportionately into low-paying majors, with almost no gender convergence along this margin since the mid-1980s. In this paper, I explain the dynamics in the gender differences in college attendance and choice of major from 1960 to 2010. I document first that changes in returns to skill over time and gender differences in wage premiums across majors cannot explain the observed gender gaps in educational choices. I then provide reduced-form evidence that two factors help explain the observed gender gaps: first, degrees provide insurance against very low income for women, especially in case of divorce; second, majors differ substantially in the degree of "work-family flexibility" they offer, such as the size of wage penalties for temporary reductions in labor supply. Based on the reduced-form evidence, I construct and estimate a dynamic structural model of marriage, educational choices, and lifetime labor supply. I use the model to quantify the relative importance of changes in wages and changes in the marriage market over time for the observed educational investment patterns. Finally, I test the effects of two sets of policies on men's and women's choice of major: a differential tuition policy that charges less for science and technical majors, as has been proposed in some states; and interventions intended to improve work-family flexibility. My results show that the effects of family-friendly policies differ significantly depending on the program. Some policies, like part-time work entitlements, increase the share of women in science and business majors, while others, like paid maternity leave, further widen both college gender gaps. 


Abstract: 

We document that the U.S. marriage market is characterized by two systematic empirical patterns. First, there is a quantitatively large, strong and persistent negative relationship between changes in cohort size and marriage rates of women. Second, the same negative correlation holds for men. This relationship accounts for a large share of the variation in marriage rates over time and across states. We then establish the features a model should possess to generate the two patterns. We start with a standard matching model with search frictions. We show that it is rejected by the data because it produces a negative relationship for women, but a positive relationship for men. However, the standard model can rationalize both patterns if either of the following two features is added: (i) the marriage surplus deteriorates with cohort size; (ii) cohort size reduces the efficiency of the matching function. We show that the two models have different predictions for the relationship between changes in cohort size and out-of-wedlock births, with model (i) predicting a positive relationship and model (ii) a negative relationship. We find that only the model in which cohort size affects marriage surplus is not rejected.


Work In Progress