Social Monitoring and Corruption in Developing Countries

This dissertation studies the impact of social monitoring--monitoring of government officials by ordinary citizens--on institutional corruption. Specifically, I study a monitoring initiative in which citizens used cell phones to report instances of fraud during the 2009 Afghan presidential election. Since implementation of the program required cell phone coverage, I combine coverage maps with unique data on the geographic location and fraud levels of polling centers across Afghanistan to determine: (i) the effect of coverage on fraud, and (ii) whether social monitoring is the main corruption-deterring mechanism among several competing channels. Using a spatial regression discontinuity (RD) design along the cell phone coverage boundary, I find considerable evidence that cell phone-based participation deters corrupt behavior. Polling centers inside cell phone coverage areas report up to a 26 percent drop in the share of fraudulent votes relative to centers outside. Analyses of the effect of coverage on election-related violence and the tribal composition of villages suggest that the observed declines in fraud cannot be attributed to these alternative channels. From a policy perspective, these results illustrate how a widespread technology, namely cell phones, can exert a positive externality on institutional development via corruption deterrence.Doctor of Philosoph