On the Optimal Social Contract: Agency Costs of Self-Government

In a typical study of political economy, citizens are regarded as principals, and government as agent. This is a modern way of thinking in the sense that classical theorists of democracy such as Jean-Jacques Rousseau and James Madison were more interested in the dual nature of people; they are principals (citizens sharing the sovereign power) and, at the same time, agents (subjects under the laws). Government, in their framework, is an intermediate body which helps people solve their self-control problem. Equipped with tools of modern economics, this paper explores the classical problem to see how economic development and political institutionalization relate to the structure of government and the quality of public sector. In particular, I consider repeated games with a large population and incomplete information, in which players decide whether to sacrifice private consumption to provide public goods. Because both people and the executive of public projects are subject to moral hazard, the people spend resources to monitor the executive and the people themselves. The optimal self-enforcing contract, which can be interpreted as an efficiency upper bound of political systems, is characterized. The analysis of the contract shows that as a country gets more economically developed and politically institutionalized, the agency problem on the people's side becomes negligible, and the citizens' demand for accountable government becomes stronger, in which case the standard principal-agent framework is good enough to describe the political reality.