I am an assistant professor at the Chair of Economic History under Davide Cantoni.
My research focuses on the political economy of less-developed countries using modern and historical data. I have worked on the assigning of property rights in the United States, ethnic partitioning in Africa, the political legacy of mass killings in Cambodia and the impact of school reforms in Cambodia.
I work with historical and high resolution spatial data, some digitized for the first time specifically for the projects you find below.
Property Rights, Resources and Wealth: Evidence from a land reform in the United States
The literature has suggested two alternative property rights regimes to overcome the Tragedy of the Commons. In this paper I compare the effectiveness of distributing access rights under public ownership, as proposed by Samuelson, to generating private ownership as proposed by Coase. However, as property rights are not randomly allocated, causal evidence on the relative effectiveness of public ownership is scarce. I exploit a spatial discontinuity generated by the 1934 Taylor Grazing Act, which created 20,000 miles of plausibly exogenous boundaries that separated publicly owned rangeland from open-access rangeland. Using satellite-based vegetation data, I find that both property rights regimes increased vegetation relative to the Tragedy of the Commons. Moreover, when compared to nearby private land of similar pre-1934 quality, public ownership can be as effective as privatization. Census-block-level income data reveals that public ownership raised private household income by 13% and decreased poverty rates by 18%. To study mechanisms, I exploit variation in pre-reform police presence and panel data on farm values, and show that legal enforcement through police presence is a necessary condition for the positive and long-lasting effects of public ownership to arise.
The Effects of Migration and Ethnicity on African Economic Development
International migration has a considerable impact on trade between nations. While supported by evidence from developed countries, the effects of migration on trade are less clear cut for developing countries. Given the evidence on ethnic identity in Africa, standard estimates based on the nationality of migrants are likely biased in developing countries. In this paper, I expand the standard approach to explicitly account for heterogeneous ethnicities migrating between countries. Using the precolonial distribution of ethnicities as an instrument for modern day migration in Africa, I estimate a considerably larger effect on bilateral exports and economic development than previously found. I provide evidence that ethnic identification shapes bilateral trade by facilitating the flow of information, especially for ethnicities who are not part of a government coalition. I discuss potential concerns of precolonial ethnic linkages and find no evidence of omitted variable biases caused by similar languages, preferences, or conflict. The results are consistent with a model of international trade where cross border connections decrease the fixed costs of exporting by shared information.
Work in Progress
State Repression, Exit, and Voice: Living in the shadow of Cambodia's Killing Fields
with Andreas Madestam
This paper asks whether state repression is an effective strategy for silencing dissent and changing political beliefs. We use evidence from history's most severe episode of state-led repression, the genocide in Cambodia under the Khmer Rouge, to estimate the effects of political violence on political behavior four decades later. To establish causality, we rely on the Khmer Rouge's desire to create an agrarian society, moving forced labor to areas experiencing higher agricultural productivity. Using historic rainfall to generate exogenous variation in productivity shows that more people died in productive communes. Higher productivity under the Khmer Rouge leads to more votes in favor of the opposition over the authoritarian incumbent and increased support for democratic principles. At the same time, citizens become more cautious in their interactions with the local community as captured by lower participation in community organizations and less trust. Our findings suggest that state repression makes people more convinced about the need for opposing views but more careful in expressing them, resulting in a less personal and more competitive political environment.
Who Benefits from Free Education? Long-Term Evidence from a Policy Experiment in Cambodia
Free primary education is considered an important public policy to promote poor children's schooling. We explore a nationwide policy experiment in Cambodia in 2000 that abolished primary school fees to assess this claim. The paper investigates the effects of the program by combining differences in fee exposure across province, time, and cohort. One additional year of free education had no impact on children living in households below the consumption poverty line, but increased the likelihood of completing primary school, led to more years of schooling completed, and raised literacy for children in households above the poverty line. To ensure a causal interpretation of the heterogeneous effects, we exploit weather-induced agricultural volatility to estimate the difference across the consumption poverty line. Though poor and non-poor children attended school to same extent after the reform, poor children were less likely to progress and complete the higher grades. The findings are consistent with the idea that poor children and their parents are affected by the local community's educational norm, where income segregation may explain why poor students fail to take advantage of the policy change.
Eroding the incumbency advantage: Evidence from a wealth shock in the western United States
I study the effect of a large wealth transfer to the rural population on their political preferences. The Taylor Grazing Act of 1934 placed 142 million acres under public ownership and documented the rights to access these lands for nearby farmers. These property rights on public land increased the wealth almost immediately since they were an accepted collateral and influenced the pricing of a farm. I focus on close elections and show that counties affected by this policy were initially more likely to elect a democratic congressman at a magnitude similar to the incumbency advantage. This preference for redistributive policy is completely eroded almost immediately after the policy took place. Focusing on congressional elections during the great depression, I show that the most likely explanation is a change in preferences away from redistribution.
|ArcGIS course for graduate students in economics|
|Teaching Assistant for 'Management and Analysis of Big Data' for David Strömberg|
|ArcGIS course for graduate students in economics, jointly lectured with Shuhei Kitamura|
|Teaching Assistant for 'Macroeconomics II, PhD course' for John Hassler|