for useful discussions, and to Steve Wilkinson for granting us access to a dataset on religious conflict. We thank Jay Dev Dubey for his able research assistance. We are grateful to five anonymous referees for their valuable comments. We particularly thank Co-Editor Jesse Shapiro, who went beyond the call of duty in his detailed reading of the manuscript, making many constructive suggestions that greatly improved both the content and the exposition. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peerreviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

for useful discussions, and to Steve Wilkinson for granting us access to a dataset on religious conflict. We thank Jay Dev Dubey for his able research assistance. We are grateful to five anonymous referees for their valuable comments. We particularly thank Co-Editor Jesse Shapiro, who went beyond the call of duty in his detailed reading of the manuscript, making many constructive suggestions that greatly improved both the content and the exposition. ABSTRACTWe model inter-group conflict driven by economic changes within groups. We show that if group incomes are low, increasing group incomes raises violence against that group, and lowers violence generated by it. We then apply the model to data on Hindu-Muslim violence in India. Our main result is that an increase in per-capita Muslim expenditures generates a large and significant increase in future religious conflict. An increase in Hindu expenditures has negative or no effect. These findings speak to the origins of Hindu-Muslim violence in post-Independence India. ABSTRACTWe model inter-group conflict driven by economic changes within groups. We show that if group incomes are low, increasing group incomes raises violence against that group, and lowers violence generated by it. We then apply the model to data on Hindu-Muslim violence in India. Our main result is that an increase in per-capita Muslim expenditures generates a large and significant increase in future religious conflict. An increase in Hindu expenditures has negative or no effect. These findings speak to the origins of Hindu-Muslim violence in post-Independence India. hini Pande for useful discussions, and to Steve Wilkinson for granting us access to a dataset on religious conflict. We thank Jay Dev Dubey for his able research assistance. We are grateful to five anonymous referees for their valuable comments. We particularly thank Co-Editor Jesse Shapiro, who went beyond the call of duty in his detailed reading of the manuscript, making many constructive suggestions that greatly improved both the content and the exposition. Observations 126 132 126 132 126 132

We investigate how increased electoral competition -by influencing the equilibrium policies of competing parties -affects the income distribution in society. Our model is embedded in a standard probabilistic voting setup where parties compete at two stages: (i) they allocate resources across various districts and (ii) then, for each district, they divide the resources among the different constituent groups. We show that an increase in electoral competition in a district results in a tendency towards equalization of incomes therein. We check for these relationships using data from the Indian national elections which are combined with household-level consumption expenditure data rounds from NSSO (1987-88 and 2004-05) to yield a panel of Indian districts. We find that districts which have experienced tight elections exhibit lower inequality and polarization which indicates a larger "middle class".JEL codes: D72, D78, O20

Mandated political representation for minorities involves earmarking certain electoral districts where only minority‐group candidates are permitted to contest. This paper builds a political‐economy model to analyse the effect of such affirmative action on redistribution in equilibrium. The model predicts that in situations where the minority is economically disadvantaged and where voters exhibit an in‐group bias, such a quota can reduce transfers to poorer groups. This suggests that the gains to the minority group from having such quotas are unevenly distributed. Redistribution in reserved districts leads to a rise in within‐group inequality for the minorities.

The version in the Kent Academic Repository may differ from the final published version. Users are advised to check http://kar.kent.ac.uk for the status of the paper. Users should always cite the published version of record.