Publication - Working Paper
Do political dynasties hinder development?
This study will examine the effect of political dynasties on local economic development. A political dynasty is defined as a network of family politicians who come to power in an election based regime. Dynasties are a prevalent form of transmission of political power in many new as well as established democracies. The results of this project are relevant in understanding the incentives of politicians to act in the interest of the voters in democracies where the de facto political power is concentrated in dynastic networks. The existing literature finds that political inequality can lead to worse policies and economic outcomes (Besley, Persson, Strum, 2010; Acemoglu, Robinson, 2010). However, more recent work suggests that the impact of dynasties should be heterogeneous. Besley and Reynal-Querol, (2015) find that growth is stronger in countries with hereditary leaders if constraints on the executive are weak. They argue that family reputation can be a mechanism to control moral hazard when formal institutions are weak. Bargan, Ferraz, and Rios (2015) find large increase in spending but no significant improvement in economic outcomes in Brazilian municipalities that elected dynastic politicians.
Given the mixed findings in the literature, it is important to understand how and under which conditions political dynasties may improve policies and development outcomes. Dynasties provide politicians with an opportunity to establish an electoral advantage over new entrants, by dispensing patronage which allows them to build a committed vote bank. This can weaken their incentive to implement policies that deliver benefits to voters outside their network. However, in election based regimes, dynastic politicians also need to cater to swing voters (outside their committed vote bank) to get re-elected. This increases their incentive to implement policies that benefit swing voters. The strength of this effect will depend on a number of factors such as relative size of the swing vote and information available to voters.
Using sub-national panel data from Pakistan from 2002 to 2012, the project will examine the impact of political dynasties on development spending and local economic growth. The empirical strategy will be to compare the outcomes in constituencies where the elected member of parliament belongs to a dynasty to where the elected member does not belong to a dynasty, using a difference in differences and regression discontinuity design. Data on election outcomes and politician characteristics will be compiled from the Election Commission of Pakistan. The dynastic affiliation of candidates will be coded using a press publication that provides comprehensive data on political families in Pakistan. The main outcome measure is development spending by politicians in their constituencies under a program that sanctioned the use of federal funds for local development schemes proposed by politicians. Since data on sub-national economic growth is not available, this project will utilise night time lights captured from satellite images as a measure of annual growth in a constituency. Night lights have been previously used by economists as a proxy for economic activity. The project will also collect additional data on development indicators at the district level available in national surveys. The study will explore the role of potential mechanisms such as competition and information that can explain the performance of dynastic politicians.