In this dissertation I offer an explanation for why Indian states are undertaking economic liberalization at different rates, focusing on reforms to the electricity sector. In the period between 1991 and 2003, India's states restructured their electricity systems to vastly different degrees. The dissertation evaluates three variables that feature prominently in the literature on economic policy change: ideological predilections of governing elites, external pressures like those coming from international financial institutions, and state-society interactions. I argue that it is the last explanation, focusing on the degree to which the potential "losers" from reform dominate state politics--that most compellingly accounts for the unevenness in state-level reforms. In my work, I lay greater analytic weight on the role of rural actors than much of the existing literature on the political economy of market reforms. The primary independent variable that explains this variation in reform outcomes is the organization and political strength of societal actors in each state, particularly rural and industrial constituencies, and middle class interests. In some parts of India, the advent of Green Revolution technologies in the late 1960s meant that farmers--chiefly larger landowners--became the primary beneficiaries of extensive development subsidies, including those for electricity. During India's period of economic liberalization in the 1990s, these beneficiaries constituted the main opponents of privatization, which today threatens to change the rules of the game by allocating resources according to market logics. Given these dynamics, where farm sectors are large or well-organized, reform has not proceeded. In the absence of rural political clout, state elites elected to privatize in order to satisfy industrial and urban constituents and signal the state's openness to private capital inflows. By comparing outcomes across states within the single country of India, the research design can control for some variables that are proposed as determinative of government policy, like electoral institutions and macroeconomic shock. I have selected cases to both capture variation of the dependent variable and control for other plausible explanations, such as ideology, financial crisis, and external pressure. / text