Caste-Based Reparations in the Real World? Evidence from a Natural Experiment

• In Event: Identity and Partisan Politics in Contemporary India

Abstract

One especially enduring and pernicious manifestation of group-level inequalities is the Hindu caste system. Some evidence suggests that historically disadvantaged caste groups such as Dalits (scheduled castes; a marginalized group comprising about 200 million people) have witnessed a limited degree of convergence in measures of education and, to a lesser extent, wealth and income, relative to “middle” and “upper” castes. Policies like affirmative action in education and government employment (“reservations”) have contributed to this outcome. Nevertheless, it remains the case that an individual’s caste is a strong predictor of their life chances; that historically disadvantaged castes are disproportionately poor; and that social stigmas and segregation persist and are widespread. In light of this persistence, we ask: can large, one-time wealth transfers mitigate historically grounded caste inequalities? A recent policy instituted by the state government of Telengana provides a unique opportunity to study this question. In 2021, the Telengana Government implemented a pilot program to give every Dalit family in a single electoral constituency a one-time in-kind grant valued at 1 million (10 lakh) Rupees (approximately $12,000), a sum that is several multiples of the median annual household income in the state. Our research draws on a completed survey of 3000 households in bordering villages within and just outside the selected constituency to understand the economic, political, and social consequences of the program, which we hope will shed light on the broader question of whether and how such wealth transfers can mitigate enduring disparities and whether they provoke backlash.

Authors

• Gautam Nair, Harvard University

• Akshay Dixit, Harvard University