The standard economic model of occupational choice following a basic Roy model emphasizes individual selection and comparative advantage, but the sources of comparative advantage are not well understood. We employ a unique combination of Dutch survey and registry data that links math and...

We study whether student-advisor gender and race composition matters for publication productivity of Ph.D. students in South Africa. We consider all Ph.D. students in STEM graduating between 2000 and 2014, after the recent systematic introduction of doctoral programs in this country. We...

We use a large dataset of approximately 1500 physicists employed by the Centre National de la Recherche Scientifique (CNRS) in France to investigate the role of cumulative advantage in their publication career. Measuring output by time series of the number of publications and the number of...

Top income inequality in the United States has increased considerably within occupations. This phenomenon has led to a search for a common explanation. We instead develop a theory where increases in income inequality originating within a few occupations can "spill over" through consumption into...

Is government guiding the invisible hand at the top of the labor market? We use new administrative data to measure physicians' earnings and estimate the influence of healthcare policies on these earnings, physicians' labor supply, and allocation of talent. Combining the administrative registry...

Recent attention to the causal identification of spending impacts provides improved estimates of spending outcomes in a variety of circumstances, but the estimates are substantially different across studies. Half of the variation in estimated funding impact on test scores and over three-quarters...

Decisions to invest in human capital depend on people's time preferences. We show that differences in patience are closely related to substantial subnational differences in educational achievement, leading to new perspectives on longstanding within-country disparities. We use social-media data -...