Programs, Clientelism, and Ethnicity in a Modernizing India

• In Event: Ethnic Politics

Abstract

This paper studies the relative importance of clientelism, ethnicity, and the provision of collective goods in the decision-making of Indian voters in an era of economic growth and development. Our goal is to adjudicate between two competing positions. One contends that voters are primarily swayed by a party's record of "development" or its credibility in delivering broadly targeted goods and public services. This position is at odds with the alternative position that points to the fact that parties expend considerable effort in mobilizing citizens on the basis of ethnicity and narrowly targeted material benefits. We use three sets of data that combine observational and experimental methods to advance a set of theoretical and empirical claims that go some way towards resolving this debate.

First, an original two wave panel survey of 2500 voters in the state of Uttar Pradesh provides descriptive insights into the set of government services that are most preferred by voters, and how this varies across different groups of voters defined by their class, ethnicity, geographic location, and the availability to private alternatives. Random assignment to vignettes that prime the salience of different government policies or proposed programs allows us to study their effectiveness on voters’ support for the main political parties in the state. List experiments and other methods of eliciting sensitive survey responses enable us to examine the incidence of vote buying, abstention buying, and other clientelistic linkage strategies. Between the two survey waves, we communicate anti-vote buying messages (through recorded cellphone messages) to voters to understand the effectiveness of such messages in improving mechanisms of democratic accountability and the impact of these messages on the vote shares of the different parties. In addition, we study the impact of the "new clientelism" by advertising the incumbent party's promise to deliver free smartphones to a broad swathe of the population if it is re-elected. Second, we use district level data on income growth, access to health and education, crime, and infrastructure to examine if politicians' performance in delivering such goods pays electoral dividends, and under what circumstances. Finally, we take a broader view of the political economy factors that account for the different forms of politics across Indian states.

Author

• Gautam Nair, Yale University