Brian E. Roberts

The Determinants of Industry Political Activity, 1978–1986.

While the allocation of interest group monies to specific politicians has been extensively studied, little is known about the factors that determine of the overall level of political activity across groups. We study total contributions by corporate political action committees at the industry level. We create a large data set on industry political activity, covering 124 industries across five election cycles from 1978 to 1986 and sketch out a simple benefit-cost model to predict total corporate PAC contributions in each industry. The few previous studies of this phenomenon use relatively small samples and employ statistical techniques that are either biased or impose untested restrictions. The selectivity-corrected regression technique used here solves these problems. We find that industries with greater potential benefits from government assistance contribute systematically more but that the ability to realize these benefits is constrained by collective action problems facing firms in each industry.

The Industrial Organization of Corporate Political Participation

The study of industrial organization and regulation has been transformed over the past twenty years by the recognition that firms have incentives, and may possess the ability, to influence policies designed to regulate their activities. The basic paradigm of industrial organization, and the basis for regulatory and antitrust policy in the U.S., asserts market structure determines conduct, and that conduct determines performance. If public policy were exogenously formulated and implemented, optimal performance (productive and allocative efficiency, full employment, etc.) could be obtained by appropriate antitrust policy to affect industry structure and appropriate regulation to deter anticompetitive conduct. But policy is not exogenous. Neither the formulation nor implementation of antitrust or regulation is insulated from attempts by those affected to influence policy to their benefit. Further, there are important reasons to expect industry structure to determine the conduct of the firm not just in the market but also in the political process. Structure determines conduct and performance, but not only through the traditionally emphasized routes of pricing and output decisions. Structure may also affect performance by advantaging certain firms, or certain industries, in political competition. The resulting regulatory regime may well create and protect systematic differences in economic performance by restricting, rather than facilitating, normal market processes. Surprisingly little is known about the implications of industry structure for success in politi-

Trading Places: Choice and Committee Chairs in the U.S. Senate, 1950-1986

In this paper we ask why members of Congress chair the committees they do. While the answer is related to the norm of seniority, we argue that it additionally involves strategic considerations that have not been considered in prior research. Our analysis examines the choices of senators 1950-1986 who have had the opportunity to chair two or more committees. Because senators are allowed to chair only one committee in any given Congress, these legislators must decide which committee to chair. Empirical analysis indicates that the choice of committee chairs involves a strategy that extends the influence of senators in a position to make such decisions and effectively reduces the variance of policy preferences in the set of committee leadership positions.

Measuring Corporate Political Sensitivity

The sensitivity of U.S. firms to domestic political uncertainty is an under-conceptualized yet critical component of nonmarket strategy. Using financial and political prediction market data, we develop a panel measure of domestic political sensitivity at the industry level for the period of 2000-2012. We validate this measure in two ways. First, we model how covariates related to financial performance and industrial organization explain across-time and industry variation in sensitivity to the political status quo. Second, we demonstrate that our measure is associated with strategic political activity at the firm level: consistent with the literature on corporate political activity, higher levels of sensitivity correlate with a greater propensity to lobby and larger lobbying expenditures but do not significantly correlate with making campaign contributions.