Ethnic Diversity, Ethnic Inequality, and Public Goods

• In Event: The Politics of Public Goods in India: New Debates and Evidence

Abstract

Ethnic politics of subnational public goods provisioning is contentious, especially at sub-national scales. We contribute to the ongoing debate using a unique census-scale micro dataset from rural India including detailed caste, religion, language, and landholding data (n ≈ 13.25 million households) in combination with high resolution electoral data and detailed village-level incidence data for eight different public goods. The first census-scale enumeration and coding of elementary Indian caste categories (jatis) since 1931 enables us to test competing ethnic politics hypotheses at multiple ethnic scales. Our principal result is that the relationship between ethnic diversity and public goods provisioning is obfuscated by degree of ethnic inequality. While our results are consistent with recent evidence on ethnic inequality and public goods, we provide robust empirical evidence for specific political and administrative channels. We use self-reported ethnic identifiers and do not rely on subjective assessments of ethnicity used in extant cross-country studies. We introduce the concept of “conditional fractionalization” to better characterize the actual political saliency of ethnic boundaries when using high-resolution disaggregated ethnic information. We argue that the nested political-administrative structure and its interaction with placement of specific public goods is an important channel. We show that ethnic identities are more salient and politicized at the local level where the state-presence is mediated through informal societal relations. Greater autonomy to local actors through decentralization facilitates elite capture and clientelism. We provide robust evidence for how higher ethnic inequality (computed using household landholding data) exacerbates the effects of such clientelistic behavior. To the best of our knowledge, this is the first paper to combine high-resolution ethnic data, inequality data, and high-resolution electoral data to delineate the relationship between ethnic diversity and public goods provisioning at sub-national scales.

Authors

• Deepak Malghan, Indian Institute of Management Bangalore

• Andaleeb Rahman, Cornell University