Publications


There is a widespread policy view that a lack of job opportunities at home is a key reason for migration, accompanied by suggestions of the need to spend more on creating these opportunities to reduce migration. Self-employment is widespread in poor countries, and faced with a lack of existing jobs, providing more opportunities for people to start businesses is a key policy option. But empirical evidence to support this idea is slight, and economic theory offers several reasons why the self-employed may be more likely to migrate. This paper puts together panel surveys from eight countries to descriptively examine the relationship between migration and self-employment, finding that the self-employed are indeed less likely to migrate than wage workers or the unemployed. The paper then analyzes seven randomized experiments that increased self-employment, and finds that their causal impacts on migration are negative on average, but often small in magnitude. 

Working papers


We use new administrative data from Ecuador to study the welfare effects of the misallocation of procurement contracts caused by political connections. We show that firms that form links with the bureaucracy experience an increased probability of being awarded a government contract. The reallocation of contracts generates opportunities for misallocation, as politically connected firms charge higher prices and are less efficient than unconnected firms. We develop a methodology to quantify the welfare losses of political connections through these margins-price inflation and excess cost of provision-and estimate welfare losses of up to 8% of the procurement budget.

We study the evolution of social distancing behaviors across the U.S. on the number and density of individuals in public places and businesses, and show that the latter has decreased significantly less than overall foot traffic. Breaking this effect down by category of place, we find  a significantly large gap caused by sustained higher relative density in visits to religious places, and in particular due to differential drop-off in visits by destination size: physically larger religious places experienced a much larger drop-off in visits than smaller, leading to a comparatively higher density of people relative to the overall visits change. Inspecting this effect across regions, we find that this phenomenon is present to some degree in almost all states, but appears to be much smaller in Sun Belt states, such as Florida.