The policy of mandated representation (reservation) for disadvantaged social groups in Indian village councils or Gram Panchayats has been the subject of numerous studies. The implicit, and often unstated, assumption that underlies most of these studies is that the president of the council is the only one who wields effective power. However, the Gram Panchayat is comprised of several elected representatives, each of whom represents a village; and, in principle, a voting mechanism governs decision making within the Gram Panchayat. In this context, the focus on the president as the de-facto decision maker is equivalent to assuming a model of Sarpanch Raj, or a model of local government where the president (Sarpanch) dominates the council. This model is typically based on the premise that the president of the council possesses either on account of her informal powers or on account of her formal agenda setting powers- the de-facto power to dominate. However, whether these informal or formal powers of the president translate into such de facto power may well depend on other factors, such as local power structures. Indeed, extensive anecdotal evidence suggest that presidents elected on reserved seatsi.e. members of the disadvantaged Scheduled Castes (SCs) and Scheduled Tribes (STs), and women face considerable difficulties when they are situated amid representatives who come from powerful castes or belong to the local elite. Whether a Sarpanch Raj is indeed the de facto model of local governance in India is therefore an unanswered empirical question. This paper examines the question of Sarpanch Raj, using a unique data set from 80 Gram Panchayats and 225 villages in the Indian state of Karnataka. We exploit the design of the policy of mandated representation in order examine whether the Sarpanch Raj model is robust to the inclusion of elected representatives of the village council. The model of Sarpanch Raj is critically examined in the context of two key mandates of the Gram Panchayat: public good provision, and the targeting of household-level benefits under various anti poverty programs. The results suggest that the president is not the sole decision maker of the council, and that the council is in fact a more broad-based body where the voices of other elected village representatives matter. Decision making in the council is, however, not one among equals. In particular, the results suggest that the effectiveness of Scheduled Caste representatives depends on the caste of the president.
