Association for Public Policy Analysis & Management

The allure of democracy for developing countries comes from the idea that it will lead to more responsive government. However, for democracy to work, elections must generate pressure to hold government accountable for their performance. This paper studies the case of the Philippines to determine whether increased electoral pressure leads to better government performance, as measured by evacuations performed in preparation for typhoons. This paper addresses a major gap in the existing literature as it is among the first studies to provide causal evidence that electoral pressure drives governments to exert greater effort in the performance of their duties.

I obtain data on evacuations, casualties, and damages due to storms from the National Disaster Risk Reduction and Management. To account for typhoon exposure, I merge this dataset with storm track data from the International Best Track Archive for Climate Stewardship. The combined dataset spans the period 2007-2014. To measure re-election incentives, I use election outcomes for all gubernatorial races during over the period 2001-2013 from the Commission on Elections.

In the Philippines, governors are responsible for minimizing casualties in their entire province whenever major catastrophes hit, which is especially true for frequently occurring typhoons. Their primary way of accomplishing this is through evacuation of their affected constituents when weather forecasts show that typhoons will severely affect their province. Typhoons affect huge swaths of the country several times each year, making disaster risk reduction an essential task of all elected officials. I exploit plausibly exogenous variation in electoral pressure arising from term limits in the Philippines. Governors can only serve three consecutive three-year terms. Thus, governors in their third term experience substantially weaker electoral pressure than governors who are still in their first or second terms. I examine whether governors that can still stand for re-election evacuate more of their affected citizens than governors who are on their third and final term using a difference-in-differences framework. In that framework, I also compare governors’ performance in the final year of their terms, when elections are most salient. In doing so, the only difference between a governor is eligible for re-election and a governor who is ineligible due to term limits is that voters can no longer hold the ineligible governor accountable for their performance at the next elections. The difference-in-differences research design allows me to account for time-invariant unobservable differences between both types of governors that might also affect how well they perform relative to each other.

Results from this difference-in-differences approach show that evacuations, measured relative to the affected population, increase by 20 percentage points when provinces are headed by governors that are eligible to seek re-election, and are thus under increased electoral pressure when elections are only one year away. This finding underscores the importance of having well-functioning electoral institutions, since better electoral pressure in this realm leads to stronger incentives for government to pursue life-saving actions.