Comments on Aranson & Ordeshook's regulation, redistribution, and public choice

Abstract

This paper presents a good analysis of some of the weaknesses of earlier models of regulation. Its discussion of spacial models in Section II is particularly good. While median voter models are powerful analytic tools, they also have significant weaknesses. The assumptions underlying such models, such as the need for single peaked preferences, may often be violated. The paper identifies a number of major weaknesses in median voter models. One factor that the paper does not cover fully is that voters do not normally vote on regulatory issues or, for that matter, on any issues. Typically, voters consider candidates who may or may not have a position on regulatory problems. The elected officials, in turn, may appoint or approve officials who administer bureaucratic organizations which actually issue the regulations. The elected officials may also vote on general regulatory statutes. The problem, then, is that there are many steps between the voting and actual regulation. Moreover, candidates for office do not run on well-defined issues. This is a general problem with all, or almost all, median voter models. Typically, these models have candidates taking positions on a wide number of issues in an attempt to put together winning coafitions. But in actual practice, candidates avoid, if possible, taking positions and attempt to stress their competence, their honesty, their experience, their ability to get things done (without specifying what they will get done), and their general character. A typical advertisement for a candidate will not even mention the candidate's position on any issue, but will show the candidate with his or her _family and mention the candidate's experience. Voters do not generally know or care much about candidates' positions on various issues. It is not rational for them to do so, since the issues the candidate will face once elected will undoubtedly differ from those existing prior to election. Voters simply want to vote for those most competent to deal with the new issues that will arise after the election. Current position on issues is only relevant in providing information on the type of decisions a candidate might make. Moreover, since a candidate not in office will have less information about all the ramifications of an issue than will be available once elected, statements about his position on the issue may not even be a good indicator of how the