Peralta, Abigail-Allison M. (2018-05). Three Essays on Public Economics. Doctoral Dissertation. Thesis uri icon

abstract

  • This dissertation introduces three essays on public economics with a focus on the developing country setting. The first two essays present evidence that the accountability generated by electoral institutions positively affects government performance. The last essay shows that exposure to inter-group conflict has lasting effects on the degree to which individuals favor members of their in-group. All three essays use data from the Philippines. In the first essay, "The impact of election fraud on government performance," I study the effects of an election reform that happened in the Philippines in 2010. Using data on election results for mayor, I show that it reduced election fraud regardless of the pre-existing level of election fraud. I find that the reduction in election fraud caused building permit approvals to increase by 15 percent. Since delays in granting approvals are associated with bribe requests, this increase implies a significant improvement in government performance. In the second essay, "Does electoral pressure lead to better government performance?," I study whether increased electoral pressure can increase the effort government officials put into their tasks. To measure effort, I use province-level data on the fraction of the affected population evacuated to prepare for tropical cyclones. This isolates effort because resources are pre-positioned and provided by national government agencies. I find that governors who face re-election incentives increase evacuations by twenty percentage points. In the third essay, "Does conflict exposure increase in-group bias? Evidence from experiments in the Philippines," we exploit community-level exposure to two types of conflict that exist in the Philippines: a Muslim separatist movement and a communist insurgency. We merge data on community-level conflict exposure with information about giving to partners of different religion or ethnicity measured using experimental games. Our results indicate that Muslims exposed to the inter-religious conflict become more biased against other Muslims, while being exposed to the indiscriminate communist conflict leads to opposite effects.
  • This dissertation introduces three essays on public economics with a focus on the developing country setting. The first two essays present evidence that the accountability generated by electoral institutions positively affects government performance. The last essay shows that exposure to inter-group conflict has lasting effects on the degree to which individuals favor members of their in-group. All three essays use data from the Philippines.
    In the first essay, "The impact of election fraud on government performance," I study the effects of an election reform that happened in the Philippines in 2010. Using data on election results for mayor, I show that it reduced election fraud regardless of the pre-existing level of election fraud. I find that the reduction in election fraud caused building permit approvals to increase by 15 percent. Since delays in granting approvals are associated with bribe requests, this increase implies a significant improvement in government performance.
    In the second essay, "Does electoral pressure lead to better government performance?," I study whether increased electoral pressure can increase the effort government officials put into their tasks. To measure effort, I use province-level data on the fraction of the affected population evacuated to prepare for tropical cyclones. This isolates effort because resources are pre-positioned and provided by national government agencies. I find that governors who face re-election incentives increase evacuations by twenty percentage points.
    In the third essay, "Does conflict exposure increase in-group bias? Evidence from experiments in the Philippines," we exploit community-level exposure to two types of conflict that exist in the Philippines: a Muslim separatist movement and a communist insurgency. We merge data on community-level conflict exposure with information about giving to partners of different religion
    or ethnicity measured using experimental games. Our results indicate that Muslims exposed to the inter-religious conflict become more biased against other Muslims, while being exposed to the indiscriminate communist conflict leads to opposite effects.

publication date

  • May 2018