Markets Versus Governments: Political Economy of Mechanisms

Joint with: Daron Acemoglu and Michael Golosov - MIT

This paper investigates the political economy of (centralized) mechanisms and the comparison of these mechanisms to markets. In contrast to the standard approach to mechanism design, we assume that the mechanism is operated by a self-interested agent (ruler/government), who can misuse the resources and the information he or she collects. The main contribution of the paper is the analysis of the form of mechanisms to insure idiosyncratic

(productivity) risks as in the classical Mirrlees setup, but in the presence of a self-interested government. We construct sustainable mechanisms where the government is given incentives not to misuse resources and information. An important result of our analysis is that there will be truthful revelation along the equilibrium path, which shows that truth-telling mechanisms can be used despite the commitment problems and the different interests of the government and the citizens. Using this tool, we provide

a characterization of the best sustainable mechanism. A number of features are interesting to note. First, under fairly general conditions, the best sustainable mechanism is a solution to a quasi-Mirrlees problem, defined as a problem in which the ex ante utility of an agent is maximized subject to incentive compatibility constraints, as well as two additional constraints on the total amount of consumption and labor supply in the economy. Second, we characterize the conditions under which the best sustainable

mechanism will lead to an asymptotic allocation where the highest type faces a zero marginal tax rate on his or her labor supply as in the classical Mirrlees setup and there is no aggregate capital taxes as in the standard dynamic taxation literature. In particular, if the government is sufficiently patient (typically as patient as the agents), the Lagrange multiplier on the sustainability constraint of the government tends to zero, and marginal distortions arising from political economy disappear asymptotically. In contrast, when the government has a small discount factor, we show that aggregate distortions remain, and there is both positive marginal labor tax on the highest type and positive aggregate capital taxes even asymptotically. We also investigate when markets are likely to be less desirable relative to centralized mechanisms.

For more information, contact Antonio Merlo.