Are rural Indian voters sophisticated enough to navigate the complexities of local elections in India? The 1992 passage of the 73rd amendment gave constitutional status to village councils—rural India’s lowest tier of government— and mandated regular elections for village council (gram panchayat or GP) members and the GP president, resulting in millions of elected positions in local government. This empowered village leaders, and particularly the GP president (sarpanch) with substantial discretion over the local implementation of government programs. At the same time, state governments retained discretion over the allocation of state funds for local public goods such as schools and roads as well as funds for projects included in the popular Mahatma Gandhi National Rural Employment Guarantee (MGNREGA) scheme.
Given that research shows that politicians at various levels of government in India tend to favor their supporters in the allocation of government resources, the multi-level system established by the 73rd amendment created a complex task for voters responsible for selecting a sarpanch. For example, in my research site of Rajasthan, where the BJP controls the state government and the Congress Party is the main opposition party, this means a Congress sarpanch may favor a Congress supporter over a BJP supporter; but his village may be neglected by the BJP-led state government when it comes to local public goods and pork barrel project. In choosing a sarpanch, Rajasthani voters must therefore consider not only the party of the sarpanch candidate on the ballot—an open secret despite the Election Commission ban on party symbols in GP elections— but which party is in power at the state level, which goods are controlled by state and local governments, and which goods they prioritize most. Are rural voters sophisticated enough to understand how this multi-level political context affects their access to state benefits, and does this shape their vote preferences in elections for sarpanch?
Past research suggests that Indian voters make blunt judgments based on simple heuristics (e.g., the caste of party leaders) to determine their party preferences and vote choices. This view has little room for vote preferences based on calculations that take local context into account. Contrary to this view, in “The Discerning Voter: Party-Voter Linkages and Local Distribution Under Multi-Level Governance,” a forthcoming article for Party Politics, I find that voters in rural India form expectations about the chances of receiving private and public goods in nuanced ways.
To determine that this is the case, I conducted a 2013 survey experiment that included 960 heads of household across 96 gram panchayats in poor, rural districts throughout Rajasthan. In what I call a “real candidate vignette experiment,” I first asked voters to identify the most popular BJP and Congress leaders in their GP. Then I asked them to report their expectations of receiving different types of benefits if either the BJP or Congress leader were to win the election for sarpanch scheduled for 2015; whether a respondent was asked about the BJP leader or Congress leader was determined randomly as is typical in experiments of this kind. I then analyzed the effect of the party affiliation of the BJP or Congress leader for sampled voters with different partisan preferences—BJP supporters (40 percent), Congress supporters (35 percent), and non-partisans who did not support any party (25 percent).
To measure voters’ expectations of private benefits, I described a local infrastructure project similar to those included in the MGNREGA right to work program, and asked respondents how likely it would be for the BJP or Congress leader (whichever was randomly selected in the experiment) to give them a job on this program if they were to win the next GP elections in 2015. I also asked respondents if they expected to receive a Below Poverty Line (BPL) card, a requirement for a number of welfare benefits for the poor, if the leader were to be elected sarpanch. To capture voters’ expectations on access to village-wide public goods, I asked if the leader selected for the experiment would bring state funds to their GP if they were elected in 2015. Crucially, since the BJP wave had reached Rajasthan by the time of my January 2013 survey—nine months before state assembly elections that saw the largest BJP victory in Rajasthan’s history—I could observe whether voters thought the next elected sarpanch would be more likely to be favored with state funds if he or she belonged to the BJP. My interpretation was supported by qualitative interviews conducted before and during the survey which revealed an overwhelming expectation of a BJP victory in the coming elections. It also fits Rajasthani voters’ pattern of rejecting the party in power in the state in every election since 1993.
My results confirm my view that rural voters understand political biases in distribution across different state benefits in quite sophisticated ways. First, I find strong evidence that voters perceive strong biases in targeted benefits such as a job on the infrastructure project. Voters who were Congress Party and BJP supporters exposed to a leader from their party reported expectations of receiving a job that were respectively .29 and .39 points higher (on a 4-point scale) than was the case when Congress and BJP partisans were exposed to a leader from a different party. Similarly, Congress and BJP partisans were 15 and 21 percentage points more likely to expect to receive a BPL card when the sarpanch in the experiment belonged to their preferred party. Non-partisans did not perceive the party of the sarpanch to make a difference in access to jobs or a BPL card. These results, hold when I consider the leader’s caste (including politically relevant castes in Rajasthan such as upper castes, Jats, Yadavs, Rajputs, and OBCs) and Muslim religion. In short, partisan voters perceive partisan biases in access to private benefits over which the sarpanch has discretion. This is plausibly not the case for non-partisans because voters who fall outside local party networks are unlikely to be favored by the sarpanch.
My second set of results shows that voters perceive partisan biases in access to funds from the state government, but understand that sarpanch from the state ruling party have a substantial edge. Overall, respondents, regardless of voters’ partisanship, were 5 percentage points more likely to expect a BJP sarpanch to bring state funding to their village. When I consider partisanship, BJP supporters and non-partisans were respectively 9 and 10 percentage points more likely to expect a BJP sarpanch to attract state funds (compared to a Congress sarpanch). Congress supporters do not perceive a difference between the BJP and Congress for this measure —contrary to their strong expectations for targeted goods from a co-partisan.
These results tell us that rural voters understand that access to state benefits is biased along party lines, and understand the role that different tiers of government (local and state) play in the allocation of different types of goods. BJP supporters and independents clearly expect a BJP sarpanch to be in a better position to bring home state funds. And even Congress supporters, who plausibly have some psychological biases toward a Congress leader, do not believe that a Congress leader is a good bet for delivering state funds—a striking difference compared to their strong expectation of co-partisan favoritism on jobs and BPL cards.
This has important consequences for local elections. If voters understand that a ruling party sarpanch is more likely to be favored with state government resources, those who value public goods most should vote for a candidate for sarpanch from the state ruling party. While hard-core supporters of the opposition party may remain loyal, the large share of BJP supporters in poor rural areas of Rajasthan where Congress support is typically high suggests a weakening in attachments to Congress, and non-partisans favored the BJP leader by 10 percentage points. With an ascendant BJP controlling 19 states across India, my results suggest that a swing toward the BJP in GP elections may be in the cards.
Mark Schneider is a Visiting Professor of Political Studies at Pitzer College and former CASI Non-Resident Visiting Scholar. He can be reached at mark_schneider@pitzer.edu.
India in Transition (IiT) is published by the Center for the Advanced Study of India (CASI) of the University of Pennsylvania. All viewpoints, positions, and conclusions expressed in IiT are solely those of the author(s) and not specifically those of CASI.
© 2018 Center for the Advanced Study of India and the Trustees of the University of Pennsylvania. All rights reserved.