This dissertation seeks to understand the distributive strategies of local politicians in India, a context in which a robust secret ballot is in place and villagers have information on the allocation of selective state benefits due to the high information context of village politics. Specifically, it seeks answers to three questions. Can local politicians in India identity voters' partisan preferences, which is a critical assumption of theory on clientelism in comparative politics? Does a context in which citizens have a great deal of information on the beneficiaries of programs implemented by local governments and villagers and politicians are personally tied to one another pose constraints on targeting strategies relative to work on clientelism, elite capture, and citizen candidate models that predict co ethnic targeting preferences? And do voters perceive that co partisanship vis a vis sarpanch affects their access to selective state benefits and services? To address these questions, I draw on a unique original survey of village council presidents (sarpanch) and citizens across 96 village council areas (gram panchayats) in Rajasthan, India.
First, I argue that under a secret ballot, which voters overwhelmingly believe to protect the anonymity of their votes, the clientelistic logic that supports quid pro quo distributive politics does not hold. This has powerful implications for the role we should understand local leaders (who perform brokerage functions) to serve and whom we should expect sarpanch to target with antipoverty benefits implemented through the gram panchayat. I argue that if local politicians cannot identify the partisan preferences of uncertain voters, we should expect local politicians to target benefits in order to maintain their political constituencies, rather than pursuing a vote buying strategy to attract new supporters through a quid pro quo strategy. Second, I argue that the social and political context of the gram panchayat severely constrains sarpanch targeting behavior. In a context in which sarpanch and voters know each other and the latter can directly (or by rumor) observe who received visible and coveted selective benefits, and in which sarpanch and their kin are very likely to live in their village permanently, there is a powerful social cost to providing benefits to the non poor. At the same time, citizens accept that local elections have consequences, which means that some favoritism toward supporters, but not kinship or ethnic lines, is tolerated as long as the pro poor targeting norm is heeded. Finally, I check the validity of my argument on sarpanch distributive strategies by testing for the effect of co partisanship on voters' expectations of receiving selective benefits using a vignette experiment. I randomize partisan cues (Congress or BJP) based on prominent politicians identified by respondents themselves and find support for the claim that partisanship broadly affects access to state benefits.
Empirically, I draw on a unique survey of sarpanch and voters across rural Rajasthan. The survey includes two behavioral measures that cross-reference voters within sarpanch surveys. I ask sarpanch to guess sampled voters' partisan preferences and ask them to allocate tokens across these individuals to affect a lottery with a cash prize. I also embed a survey experiment within the voter survey.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/D8M32TH2 |
| Date | January 2015 |
| Creators | Schneider, Mark Allan |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0021 seconds