that had previously shown an inclination to support challengers and open-seat candidates did not dramatically alter their more recent pattern of contributions to incumbents. Among our case studies, only those PACs with issue agendas prominent in the 1992 elections--abortion and health care--changed their behavior to any great degree.
We conclude that despite the tremendous diversity in PAC histories, goals, organizational structures, informational resources, and decisionmaking processes, most PACs gave most of their money to incumbents during an election cycle in which there were many highly promising challenger and open-seat candidates running for Congress. The fact that most committees spent their funds in a way that was designed to maintain access to important decisionmakers, rather than replace them, suggests that most PACs are involved in electoral politics to advance lobbying goals instead of to influence the outcomes of elections.
There is tremendous diversity in the PAC community and among the committees in this study. This diversity is only partially reflected in contribution behavior, however. In an election in which the public professed to pollsters a desire to change the composition of Congress, and in which term-limit referenda passed in many states, the PAC community provided incumbents with additional financial advantages that limited the competitiveness of congressional elections. The proincumbent bias of most PACs reflects their access-oriented goals, the kinds of information available to most committees when they make their decisions, the nature of their decisionmaking processes, and the structure of congressional elections. These factors encourage most PACs to be supporters of the status quo, even in elections in which the majority of Americans want change.