Political elites and development

Abstract

This dissertation consists of three essays on the behavior of political elites and their effect on economic development. The first two chapters focus on political dynasties in the Philippines while the third chapter analyzes the long-run economic effects of the concentration of political power in the state of Cundinamarca, Colombia. In Chapter 1, I use a regression discontinuity design based on close elections to estimate the causal effect of entering the political system on dynastic persistence. I find that candidates who barely win their first election are four times (22 percentage points) more likely to have a future relative in office than those who barely lose and never serve. The magnitude of the effect is remarkable and substantially larger than the effect on the intensive margin reported by Dal Bo, Dal Bo and Snyder (2009) for the United States. These results suggest that the prevalence of dynastic politicians does not simply reflect the existence of a fixed set of historically powerful families, but rather that the political system itself creates persistence. In Chapter 2, I study whether the introduction of term limits in 1987 by the Philippine Constitution was effective at breaking the dynastic pattern in Philippine politics documented in chapter 1. In particular, I explore the potential countervailing effects created by dynasties in response to the introduction of term limits: (1) replacement of term-limited incumbents by relatives and (2) running for a different office. I find that term limits are not effective in reducing the probability that the same family remains in power both in the short and long-run. Moreover, term limits made incumbents safer in their early terms before term limits bind, by deterring high-quality challengers who prefer to wait for the incumbent to be termed-out and run in an open-seat race. These results suggest that political reforms that do not modify the underlying sources of power of dynasties will be ineffective in changing the political equilibrium. In Chapter 3, which was co-authored with Daron Acemoglu, Maria Angelica Bautista and James Robinson, we explore the relative importance of political and economic inequality in explaining long-run development outcomes in the state of Cundinamarca, Colombia. Using micro data on land ownership we find that municipalities that were more unequal in the 19th century (as measured by the land gini) are more developed today. However, we argue that political rather than economic inequality might be more important in understanding long-run development paths and we document that municipalities with greater political inequality, as measured by political concentration, are less developed today. We also show that during this critical period the politically powerful were able to amass greater wealth, which is consistent with one of the channels through which political inequality might affect economic allocations. Overall our findings shed doubt on the conventional wisdom and suggest that research on long-run comparative development should investigate the implications of political inequality as well as those of economic inequality.