This paper uses data on elected village councils in South India to examine the political economy of public resource allocation. We find that the pattern of policy-making reflects politicians' self-interest. Elected councillors benefit from improved personal access to public resources. In addition, the head councillor's group identity and residence influences public resource allocation. While electoral incentives do not eliminate politician opportunism, voters appear able to use their electoral clout to gain greater access to public resources.
One of the central issues in political economy concerns how to create a form of government that responds to voters needs. The preconditions for achieving this are widely debated. Recent interest has focused on various institutions such as the media or citizens? initiatives in calling the government to account.
The traditional Public Choice view is that government rarely serves the public interest and that a variety of checks and balances are necessary to have a well-functioning polity. This view reached its apotheosis in the Leviathan formulation of government motives in Brennan and Buchanan (1980). However, more modern incarnations of this view place agency problems at centre stage following on from the seminal contributions of Barro (1970) and Ferejohn (1986). The standard setting is a model where an incumbent can commit effort to produce better performance on a valence issue. Voters cannot observe the actions and must infer the amount of effort put in and/or the incumbent's type from the observable outcomes.
This paper focuses on the role of political competition as a device for enhancing accountability. The standard model of political agency pays little attention to some of the structural factors that shape the effort making decisions of incumbents. The typical set up is one of single voter whose action decisively can change the control of government between an incumbent and a challenger. For many theoretical purposes, this formulation is fine. Moreover, the model has a certain amount of plausibility when applied to Gubernatorial elections in the U.S. where the framework fits the institutional setting quite well.