Elite Parties and Poor Voters

American Political Science Review Vol. 108, No. 2 May 2014 doi:10.1017/S0003055414000069 c American Political Science Association 2014 Elite Parties and Poor Voters: Theory and Evidence from India TARIQ THACHIL Yale University hy do poor people often vote against their material interests? This article extends the study of this global paradox to the non-Western world by considering how it manifests within India, W the world’s biggest democracy. Arguments derived from studies of advanced democracies (such as values voting) or of poor polities (such as patronage and ethnic appeals) fail to explain this important phenomenon. Instead, I outline a novel strategy predicated on an electoral division of labor enabling elite parties to recruit the poor while retaining the rich. Recruitment is outsourced to nonparty affiliates that provide basic services to appeal to poor communities. Such outsourcing permits the party to maintain programmatic linkages to its elite core. Empirically, I test this argument with qualitative and quantitative evidence, including a survey of more than 9,000 voters. Theoretically, I argue that this approach is best suited to elite parties with thick organizations, typically those linked to religious social movements. hy do poor people in poor countries often In this article, I present a novel strategy through support parties that do not champion their which such a balance can be struck: Elite parties de- W programmatic interests? How do parties that ploy an organizational division of labor, in which they represent elite policy interests win mass support? Con- outsource the task of mobilizing poor voters to their ventional explanations to these questions have em- nonelectoral organizational affiliates. The latter recruit phasized one of three expansionary strategies: pro- the poor through the private provision of local public grammatic shifts, patronage, or “distracting” appeals goods—mostly basic health and educational services. to a voter’s moral values or social identity. Yet each Such outsourcing provides a material mechanism for of these arguments is limited in its ability to explain appealing to poor voters (contra identity-based ap- elite party success among poor voters, especially out- peals), circumvents the need for prior incumbency side the universe of wealthy Western democracies. (contra a patronage-based strategy), and permits the First, the powerful and privileged bases of these par- party to maintain economic and cultural policies fa- ties often constrain them from pursuing redistribu- vored by its elite supporters (contra programmatic tive programmatic shifts that would undermine the shifts). If successful, such a strategy implies that rich elite interests. Second, most elite actors in the global and poor voters will support elite parties for very dif- south have primarily operated as opposition parties. ferent reasons. Rich voters will support elite parties Consequently, unlike the longtime incumbent parties because of their programmatic affiliation with those common to these regions (notably catch-all parties of parties’ economic and ideological commitments. By independence), elite parties have never enjoyed the contrast, poor voters will support elite parties because sustained incumbency needed to develop extensive pa- of material benefits they receive from those parties’ tronage networks among the poor. Indeed, winning of- nonelectoral organizational affiliates. fice requires the prior support of these vote-rich elec- I test the observable implications of this strategy with torates, creating a chicken-and-egg dilemma for elite evidence from India, home to the world’s largest poor parties in opposition. Third, for reasons I describe later, electorate. I specifically examine how the Hindu na- poor voters in poor countries are especially unlikely tionalist Bharatiya Janata Party (BJP) has used such to be distracted by identity-based appeals. How then an approach to retain its traditional upper caste core, do elite parties recruit the poor while retaining the while winning unexpected support from poor Dalits rich? (the former “untouchable” castes) and Adivasis (in- digenous tribals) in several Indian states. I build on earlier research on the local welfare activities of Hindu nationalists (Froerer 2007; Thachil 2009; 2011), pro- Tariq Thachil is Assistant Professor, Department of Political Sci- viding a broader explanation of how these activities ence, Yale University, P.O. Box 208301, New Haven, CT 06520-8301 fit within the overall strategic efforts of an elite party. ([email protected]). I would like to especially thank Yogendra Yadav, Sanjay Kumar, I also draw on new quantitative and qualitative data Dhananjai Joshi, and their fantastic team at Lokniti-CSDS for shar- to test the implications of my division-of-labor the- ing their National Election Study Data with me, without which this ory. First, I use a national sample of more than 9,000 study would not have been possible. I would also like to thank the Indian voters to examine the determinants of BJP sup- APSR editors and three anonymous reviewers for their excellent and port among elite and non-elite voters. Next, I draw on thorough suggestions. For their constructive comments, I would like to thank Piyali Bhattacharya, Melani Cammett, Kanchan Chandra, 15 months of fieldwork, conducted between 2007–11, Samuel Decanio, Ana De La O, Ron Herring, Herbert Kitschelt, to substantiate the mechanism through which poor vot- Steve Levitksy, Ellen Lust, Irfan Nooruddin, Tom Pepinksy, Ken- ers were recruited. This unique data include elite and neth Roberts, Ken Scheve, Dan Slater, Susan Stokes, Christopher household interviews; the private, previously inacces- Way, Steven Wilkinson, and seminar participants at Columbia Uni- versity, Cornell University, University of Pennsylvania, and Yale sible records of Hindu nationalists; and a panel dataset University. of major Indian states. I conclude by considering the 454 American Political Science Review Vol. 108, No. 2 broader implications of the BJP’s experience beyond influential in providing them financial resources and South Asia, including an illustrative discussion of sup- shaping their policy profile—come from the upper port for Islamist politics in Yemen. strata of a national electorate (see Gibson 1996).1 Historically, elite parties have emerged in developing countries to defend the concerns of relatively wealthy ELITE PARTIES AND POOR VOTERS citizens, including large landowners resisting the redis- tribution or nationalization of landholdings (the PAN Disadvantaged voters routinely cast their ballots in in Mexico in the 1940s or ARENA in El Salvador in the favor of parties that represent the policy interests 1970s) or middle class and business communities advo- of wealthier citizens. They do so across a variety cating market reforms (the FIS in Algeria, UCEDE in of political contexts—in rich and poor countries, in Argentina, Partido Liberal in Brazil, or Movimiento plurality and proportional electoral systems, and in Libertad in Peru).2 parliamentary and presidential regimes. The preva- The specific poor voter paradox motivating this ar- lence of this paradox in advanced industrial democ- ticle is arguably even more dramatic in India than its racies has been extensively documented (De La O better known analogs within wealthy Western democ- and Rodden 2008;Gelmanetal.2008; Huber and racies, because the antipathy of Dalits and Adivasis Stanig 2009; Roemer 1998;Walsh2012) and perhaps toward the upper caste BJP is not simply a function of most compellingly framed by Frank’s (2004) well- class-based divides. The party emerged in the 1920s as known question—“What’s the matter with Kansas?”— the electoral arm of a Hindu nationalist social move- prompted by the robust support he observed for Re- ment founded by upper castes, which sought to en- publicans among poorer residents of his native state. trench a set of standardized Hindu traditions as the Instances of poor citizens voting against their pol- basis for Indian citizenship (Golwalkar 1966; Savarkar icy interests are seen as paradoxical because they cut 1923). Yet despite its pan-Hindu aspirations, the move- against the expectations of both sociological and in- ment’s elitist interpretation of Hindu praxis (including strumental theories of party politics. In sociological its defense of caste hierarchies) primarily appealed to accounts of European party formation, the organiza- the privileged co-ethnics of its founders.3 tion of politics around deep social cleavages was under- By contrast, non-elite communities were separated stood to produce enduring class-based partisan divides from the BJP’s upper caste4 core by economic in- (Bartolini and Mair 1990; Lipset and Rokkan 1967). equities and social hierarchies. Although cultural and By contrast, instrumental frameworks anticipate that economic cleavages often crosscut in Western poli- self-interested poor citizens will support progressive, ties, India’s “ranked” ethnic system of caste ensured redistributive parties, because the former’s individual that the two overlapped.5 Accordingly, Dalits and preferences closely align with the latter’s policy posi- Adivasis—a quarter of all Indians—were seen as espe- tions in any given election (Downs 1957; Meltzer and cially unlikely supporters of Hindu nationalism (Brass Richard 1981). The frequency with which this shared 1993; Rudolph and Rudolph 1987; Yadav 2004). expectation is contradicted has therefore understand- Indeed, the Indian paradox can be simultaneously

Load more