A Theoretical and Emperical Study of Legislative Lobbying Strategies

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

Students of democratic institutions have long worried about the reach of private power into public affairs. Private inequalities inevitably get captured in the practice of interest group politics, giving rise to what Grant McConnell once called the "most serious and perplexing problems" of American democracy. In political discourse today, assertions about the corrupting influence of special interests are commonplace. Social scientific studies of interest group influence, however, are noteworthy for the non-cumulative, frequently inconsistent nature of their findings. What effect does lobbying have on the behavior of legislators? And to what extent does it magnify or otherwise contaminate the practice of representation. One of the means by which interest groups achieve influence is by lobbying elected officials. A second is donating to candidates' campaigns. A third is mobilizing constituents at the grass roots. This study examines the effects of all three on the behavior of legislators, but the cornerstone of the project is a new theory of lobbying. The large literature on lobbying characterizes it as a strategy intended to change legislators' preferences, but this approach is undermined by several empirical puzzles. The principal investigator models lobbying as a form of legislative subsidy - a grant of staff time, policy expertise, political intelligence, and other legislative assets - selectively given to congressional allies so that they can more effectively pursue policy objectives that they share with the group. PAC contributions and grass-roots campaigns are recast in light of this theory. The model is simple in form, realistic in its assumptions, and counterintuitive in its implications. One of the main implications is that lobbyists should lobby those who already agree with them. A second is that they should lobby most those allies who are least likely to change their mind. The model also helps to clarify the conditions under which preference-centered lobbying should work. The PI investigates the theory empirically, testing its main hypotheses and alternative hypotheses using data from the U.S. Congress. The main dependent variables are legislators' activity levels on specific bills and their interventions in specific agency decisions, which are modeled as a function of legislator expertise and capacity, constituency interests, and the interaction of group lobbying and member-group agreement. The nature of these data suggest that the models be estimated using zero-inflated poisson regression. Legislators' voting decisions, groups' lobbying choices, and group PAC contributions are analyzed using conventional maximum likelihood estimators for limited dependent variables. The data on lobbying and grass-roots campaign activity come from responses to a structured survey instrument, administered in face-to-face interviews with group lobbyists. Other data come from government documents, FEC databases, district-level census data, and various online sources. The study promises to improve our understanding of lobbying and the factors that affect its behavioral impact, such as PAC contributions and grass-roots mobilization campaigns. But it also will have practical implications for the design of lobbying reforms and the development of standards of legislative ethics. Subsidizing already committed legislators is, in effect, helping them to do their job, which looks very different than buying off uncommitted ones. Accepting lobbyists help thus poses no ethical dilemma for the representative. However, the inequality among groups in lobbying resources will nonetheless distort representation in favor of moneyed interests. Indeed, this may contribute more to the legislative advantages of such groups than their contributions to congressional campaigns.