Contexts of Tyranny: Ethnic Majorities and Electoral Outcomes in Contemporary India

Does the size of an ethnic majority affect electoral outcomes? Can an increase in the size of an ethnic majority promote the success of parties championing a matching ethnic category? Previous research has consistently found that local ethnic context shapes electoral outcomes. However, scholars remain profoundly split as to the direction of this relationship. Proponents of the 'threat' hypothesis contend that the growth of ethnic minorities threatens the majority’s social position, resulting in interethnic prejudice (Blalock 1957; Fossett and Kiecolt 1989; Huckfeldt and Kohfeld 1989; Taylor 1998). In contrast, advocates of the 'contact' hypothesis argue that the growth of ethnic minorities increases the chances for intergroup contact and therefore promotes peace (Allport 1954; Pettigrew 1998; Tropp and Pettigrew 2005). This paper aims to contribute to this debate by making two important departures from the existing literature. First, while previous literature has emphasized minorities, this work examines the effect of majority size on the electoral performance of ethnic parties. This constitutes an important contribution since majorities have the potential tilt electoral outcomes to their advantage and exclude ethnic minorities. From this perspective, this study of contextual effect may provide us with a new insight into the notion of 'tyranny of the majority'. Second, this paper addresses this debate from the point of view of a developing democracy - India. Indeed, previous literature has been largely limited to studies of contextual effects in the advanced industrial democracies of Western Europe and North America. As the world's largest democracy, India is particularly well suited for shedding light on these questions. Beyond India, ethnic political parties flourish in established or consolidating democracies such as Canada, Spain, Finland, Israel, Sri Lanka, South Africa and Bolivia. Their success warrants our attention because the politicization of ethnic categories is often said to threaten the survival of democratic institutions. As Donald Horowitz puts it: "In a variety of ways, direct and indirect, ethnic conflict can be conducive to authoritarianism, and, in at least equally various ways, democracy can facilitate the majority rule and the exclusion of the minorities or minority rule and the exclusion of majorities" (Horowitz 1993: 20). This study of contextual effects in India may then yield important lessons for developing democracies elsewhere. In order to address these questions, this paper makes use of recent ‘open government initiatives’ by the Election Commission of India as well as advances in big data analytics to create an original cross-sectional dataset of electoral constituencies in India's largest cities: Mumbai, Delhi, Chennai, Bangalore, Hyderabad, Ahmadabad, Kolkata and Surat. A multilevel linear mixed effects model (LMM) was then employed to examine the relationship between the religious composition of constituencies and electoral outcomes. Preliminary results showed that not only are Hindu-majority constituencies more likely to vote for the BJP and the SS but also that this likelihood increases with the size of the Hindus majority in a constituency. This finding is robust to additional covariates, as well as control for unobserved variation at the municipal, state and national level.