Universalism and the Electoral Connection: A Test and Some Doubts

Abstract

A recurring theme in the academic literature on distributive policy is the tendency for legislators to form oversized coalitions to bestow benefits on virtually every district represented in the legislature. In this paper we offer two tests of the universalism hypothesis. First, we examine the distributional expectation of the universalism thesis under the assumption that separate logrolls occur over the distribution of benefits for individual programs. Se cond, we test the thesis under the assumption that logrolls occur over bundles of programs organized by policy subsystems. Our findings show that the evidence on the extent to which benefits from distributive programs are univer salized is weak. We suggest a number of reasons why these weak results might be expected. We argue that the incentive to universalize benefits is only one goal of legislators and may not always be the most fruitful strategy for enhanc ing their reelection prospects.