Date of Award

Spring 2022

Document Type


Degree Name

Doctor of Philosophy (PhD)


Political Science

First Advisor

Rosenbluth, Frances


Corruption is commonly defined as the abuse of public office for private gain. However, private gain can be extracted from political processes or connections in ways that often do not constitute "abuse" or meet legal definitions of corruption. This volume explores some of the myriad ways private and public actors use political decisions to extract private gains—legal or extralegal—as well as explores potential mechanisms to reduce these actions. The first two chapters examine the degree to which individual citizens can reduce extraction of private gain from public office through participation in politics. The first chapter examines the average electoral cost of corruption to politicians across countries. A systematic meta-analysis of all randomized experiments examining voting for candidates revealed to be corrupt to date shows that corrupt candidates are punished by zero percentage points on average across field experiments, but 32 points in survey experiments. This suggests that the "true" or average effect of voter punishment of revealed corruption is likely to be small in magnitude in actual elections, and that the high levels of electoral punishment of corruption found in survey environments may stem from social desirability bias and the lower and hypothetical nature of costs. Next, I examine how to mobilize individuals to organize against interest group capture of local political institutions and regulations. Specifically, I examine the case of zoning regulations in large US cities. In these cities, homeowners are more likely than renters to participate in local politics across numerous dimensions, and zoning regulations reflect the preferences of this group. I conducted eight field experiments to investigate how to motivate renters (n=19,951 households) to comment at city council meetings in opposition to regulations that harm them. Opening a message highlighting high costs of abstention caused a large increase in both total public comments and pro-housing comments across all treated meetings. These results suggest that increasing the perception that abstention is costly is an effective motivator of collective action, and that outreach can make civic bodies greater reflect the broader public where increases in accessibility alone do not. The third chapter investigates the revolving door—a phenomenon long hypothesized to be a conduit to corruption and/or regulatory capture. However, data tracing flows of civil servants from the bureaucracy to the private sector remains rare. I create a dataset containing individual-level data of all Japanese bureaucrats who retired into positions outside of the bureaucracy over the past decade. I document how the data was constructed through a combination of digitization of thousands of PDF documents, web-scraping, and linkages with financial databases. Next, I describe what the data illuminates about the revolving door in Japan, such as a bifurcated job market in which high ranking officials are more likely to join the private sector while lower ranked bureaucrats commonly join non-profit organizations. I conclude by discussing how the data can be used to investigate empirical questions in subjects such as corruption, regulatory capture, procurement, pork, government waste, bureaucratic representation, and international political economy. The final chapter examines rent extraction by rational actors of the global financial system in a legal capacity, but with large consequences for capital access in developing economies. Extant theories posit that political instability discourages investment. I challenge this consensus by showing that instability does not systematically depress access to financial capital, but rather capital access can increase when instability is expected to lead to increased financial returns in the future. Using an event study approach, I examine daily returns of national financial indices in every country that experienced an irregular regime change subject to data availability. Returns following resignations are positive, while those following assassinations and coups are negative and smaller in magnitude. However, a pro-capitalist coup results in large positive returns (+10%), and authoritarian or anti-capitalist regime changes are more likely to lead to capital flight than democratic or pro-business changes. The immediate impact of political instability on capital access is therefore dependent on the type of regime change and its expected impact on future growth. In sum, I document four channels through which individuals or interest groups extract rents from political processes—political corruption, political capture of regulatory bodies, the revolving door, and the global financial system. I first examine the degree to which citizen action can be utilized to reduce corruption and capture. Next, I create the most detailed dataset of bureaucratic revolving door hires constructed to date and illuminate previously under-appreciated aspects of revolving door networks. Finally, I detail and measure how global investors extract financial returns from political instability.