The Median Voter Model

The Median Voter Model

Consider the median-voter model described ear­lier in this chapter. It has been shown that, assuming competition among political parties, the party that most appeals to the interest of the median voter will be elected. It is not likely that the strongest supporters of a political party will be most rewarded by favors from the party. In order to get elected, the party must sacrifice some of the benefits to its strongest supporters and real­locate them in a taxing-spending-program offer to the median voter. Holcombe has shown that when tax shares can be offered as part of a political platform, "democracy has a natural bias in favor of electing the political party that has the highest demand for public sector output" ("Public Choice and Public Spending," p. 382). He has also studied the empirical relevance of the Bowen median-voter model. Utilizing data from Michigan millage referenda on educational expenditures in 275 elections in 1973, Holcombe pro­vided empirical support for the assertion that the median-voter model is con­sistent with local governmental referenda on educational expenditures ("An Empirical Test of the Median Voter Model," pp. 272-273).

The Economics of Political Representation

Empirical models in public choice have extended to testing very practical questions. For instance, do methods of paying legislators (say, set in the state constitution or by state leg­islators themselves) determine "outside earnings?" A recent study by Robert McCormick and Robert Tollison suggests that in higher-paying states, with legislators setting their own salaries, individuals find it less in their own inter­est to seek outside payments or bribes ("Legislatures as Unions," p. 77). In another interesting empirical study, entitled "Legislators as Taxicabs: On the Value of a Seat in the U.S. House of Representatives," Mark Crain, Thomas Deaton, and Robert Tollison investigated the question of why the size of the U.S. House of Representatives has remained constant at 435 (with the minor exception of a temporary expansion after Alaska and Hawaii were admitted to the Union). The only two constitutional requirements respecting size are (1) that there be no more than one representative per 30,000 population and (2) that there be at least one representative from each state. The House, given these restrictions, could have supported 5,977 members in 1977. Why, then, were there only 435? The answer, according to Crain, Deaton, and Tollison, is that legislators, like the situation where taxicabs are controlled, are able to re­strict their own numbers. The result is that economic rents are earned by the existing units of supply—at least partially by the legislators themselves. Thus, some "economic" answers to "political" questions are provided by the ax­ioms of self-interest, the ability of U.S. representatives to control the num­ber of their own members, and the theory of rent seeking (see the following section).

The richness of the emerging literature on public choice is suggested in the brief discussion above. But beyond that, the public-choice paradigm has been a fertile source of advances in the theory of economic regulation. Indeed, an endogenous political process is central to most contemporary theories of eco­nomic regulation.