Many developing countries have reorganized their subnational administrative boundaries as a part of administrative reforms and decentralization. Theoretically, administrative proliferation can lead to better developmental outcomes by better managing ethnic heterogeneity, bring public services closer to people, and better matching services to local preferences. However, empirical evidence on the antecedents and effectiveness of administrative proliferation is mixed. Using data on administrative change and nighttime lights from India in the period between 1991 and 2011, we show that ethnic and economic marginalization can lead to the creation of new administrative units, and that once created, new administrative units have a positive effect on economic outcomes of the region. However, the positive effects are not sustained in the long term.