The first chapter of this dissertation analyzes voting behavior across multiple elections. The voting literature has largely analyzed voter turnout and voter behavior separately, focusing on individual elections. I present a model of voter turnout and behavior in multiple elections. The assumptions are consistent with individual election preferences and decision is derived from utility maximization. Additionally, I provide necessary moment conditions for identification. The framework is applied to the 2008 California elections. The exit polls made national headlines by linking the historic turnout of African-Americans for Presidential candidate Obama in helping pass Proposition 8. The results show that the African-American turnout and voting share for Proposition 8 was lower than indicated by the exit polls. As a counterfactual, I look at the turnout and outcome of Proposition 8, without the presidential race on the ballot. As predicted, there is lower voter turnout: on par with midterm elections. I also find a lower share of Yes votes on Proposition 8 - enough that the referendum would not have passed. The second chapter looks at whether policies shift preferences, an important component in policy design. We isolate exogenous variation in abortion jurisprudence using the random assignment of Democratic appointee judges, which strongly increases the probability of a liberal abortion decision. We also document that newspapers report appellate abortion decisions and conduct a field experiment assigning workers to transcribe these news reports. Using both sources of variation, we find that exposure to liberal abortion precedent initially leads to more conservative public opinions, and more liberal public opinions over time. The third chapter studies payments to physicians by pharmaceutical companies, traditionally a topic of considerable debate. I examine which type of physicians are targeted through payments, and find that physicians with published research are paid by more companies and for larger amounts than non-published physicians. The effect increases in states with existing disclosure laws, consistent with reputation effects. The pharmaceutical company payments are also targeting networks of researchers versus individual specialists. Coauthoring or citing increases the likelihood of being paid by the same company and category. This result is consistent with higher payments under disclosure.
展开▼
