Some Limitations of Demand Revealing Processes

In: Public choice, Band 29, Heft 2, S. 107-124

Abstract

A number of authors have proposed that there is a decentralized mechanism for public choice yielding a Pareto-optimal allocation of resources in the presence of public goods. However, this approach, the demand-revealing process (DRP), has five limitations: (1) it typically wastes resources through the need for collection of a budgetary surplus, (2) it can drive some consumers into bankruptcy, (3) it may be dynamically unstable, (4) it can be manipulated strategically by some consumers, & (5) it can be manipulated by coalitions of consumers even when income effects are absent. A simple model of the process is examined to illustrate these problems. An alternative mechanism is outlined. In Some Limitations of Demand Revealing Processes: Comment, T. Nicolaus Tideman & Gordon Tullock (Virginia Polytechnic Instit & State U, Blacksburg) point out that the DRP has a sixth & more serious limitation: it undermotivates participants to reveal their preferences accurately. The warnings of Groves & Ledyard, however, do not pose serious enough problems to make the DRP unacceptable, & the problem of undermotivation of voters is present in all collective-choice situations. In Some Limitations of the Groves-Ledyard Optimal Mechanism, Joseph Greenberg, Robert Mackay, & Nicolaus Tideman (Virginia Polytechnic Instit & State U, Blacksburg & Tulane U, New Orleans, La) argue that the optimal mechanism proposed by Groves & Ledyard is limited both by lacking a dominant strategy equilibrium & by the problem of specifying an adjustment process which is stable & not subject to manipulation. The Groves-Ledyard approach could be expected to be even more inefficient than the DRP, due to these problems. An example of this problem is presented. In Reply to Comments by Tideman and Tullock and Greenberg, Mackay and Tideman on Some Limitations of Demand Revealing Processes, Theodore Groves & John O. Ledyard suggest that the major point to be grasped is that the DRP has not been shown to be conclusively superior to other processes. Further theoretical & empirical research are needed to determine which approach would be best. W. H. Stoddard.