Are Congressional Black Caucus members more reliable? Loyalty screening and committee assignments of newly elected legislators

American Journal of Economics and Sociology, The, April, 2007 by Franklin G. Mixon, Jr., Amanda C. Pagels

I

Introduction

STUDENTS OF MODERN ECONOMICS are often exposed to discussions about the organization and functioning of cartels in courses ranging from principles of microeconomics to game theory. Much of our modern understanding of cartel behavior dates back at least to the work of Stigler (1964) and includes an emphasis on how cartel profits are maximized by restricting industry output to monopoly levels. It is further understood, however, that attempts to avoid a rivalrous process are usually unsuccessful, as individual cartel members opt for cheating on cartel arrangements (by producing outputs beyond specified quotas) in an effort to earn higher individual profits.

Use of monopoly and cartel principles has entered into the public choice literature as well. In one example, Crain and Ekelund (1978) apply monopoly principles to the decision-making units in representative government (i.e., legislators). They suggest that legislative seats are comparable to private sector monopolies, In this framework, potential suppliers ("candidates") submit ex ante bids (i.e., "campaign platforms") for the production contract, and the production period is equivalent to the term of office for legislators. The Crain and Ekelund (1978) study uses this framework to point out how incentives and tradeoffs exist in implementing the production contract. These include borrowing money (deficit financing) or raising taxes.

The present study continues the general use of industrial organization tools to examine cartel aspects of legislative/legislator behavior. Specifically, we extend a research path developed by Mixon and Ressler (2001) that suggests that legislators form cartel-like organizations (often referred to as "memberships" or "caucuses") in an effort to maximize votes and benefits within the political process. The earlier Mixon-Ressler (2001) study examines voting records from the Congressional Black Caucus (CBC), a subgroup of Democratic legislators in the U.S. House. The records suggest that the CBC votes uniformly and consistently with the Democratic House leadership. As a reward, Democratic Party leaders in the House choose to place CBC members on the most important House committees. That this is the result of cartel behavior, as opposed to an example of leaders simply rewarding individual loyalty, is suggested by the fact that committee placement of white Democrats with CBC-like voting characteristics does not follow the same pattern (Mixon and Ressler 2001). A recent study by Chittom and Mixon (2003) both supports and extends the Mixon-Ressler work by examining efforts to detect and deter the expected "cartel cheating" by CBC legislators. Based on an analysis of committee placement of CBC members, provide econometric evidence suggesting that observed "cheating" by individual CBC members (i.e., voting records that differ significantly from the CBC's group record) leads to less desirable placement on House committees for those "cheating" CBC members.

The present study extends the work of Mixon and Ressler (2001) by examining the committee placements of CBC and non-CBC representatives as they enter the U.S. House as freshmen legislators. As freshmen, these legislators do not have past federal voting histories (such as Americans for Democratic Action (ADA) voting scores) that the Democratic leaders in the House can use to determine, and subsequently reward, loyalty through the committee selection process. The Mixon-Ressler (2001) study shows, however, that the CBC is a uniform voting body that acts as a reliable voting cartel for Democratic House leaders. As such, a freshman's membership in the CBC represents a low-cost screening device for House Democrats. Such a screening could prove useful in determining a representative's initial committee assignments. The use of this screening device by Democratic leaders in the House is explored in this study.

II

The Importance of Committee Assignments: Previous Literature

THE ECONOMICS LITERATURE INCLUDES several studies that emphasize the importance of committee assignments to vote- and support-maximizing legislators (see Grier and Munger 1991; Kroszner and Stratmann 1998; Regens, Elliott, and Gaddie 1991; Smith and Deering 1990; Alvarez and Saving 1997). Many of these studies support the argument that individual political action committees (PACs) make significantly larger contributions to legislators who hold jurisdiction (i.e., committee assignments) over the issues that are important to individual PACs. Thus, it is not surprising that legislators seek placement on committees that have jurisdiction over issues that will be important to PACs and their constituents.

A study by Coker and Crain (1994)--along with supporting work by Krehbiel (1990), Hall and Grofman (1990), and Shepsle (1978)-introduces a supplementary view of committees as loyalty-generating institutions. They point out that one would expect voting behavior on the more important committees to more closely conform to the preferences (i.e., voting behavior) of the leadership, reflecting the influence of those leaders on the committee appointment process. Thus, "[m]embers appointed to more important committees can be assumed to have passed a sort of loyalty screen" (Coker and Crain 1994: 196). Coker and Crain use Democratic legislator data for the 99th Congress to construct a loyalty index reflecting the degree to which party members vote in concert with party leaders. (1) As they point out: "Members who cannot or do not conform as closely to the leaders' desires as their cohorts will fare correspondingly worse in obtaining influential committee spots" (1994: 202). The authors are able to rank the loyalty mean for each committee in the House, and they find that the Armed Services Committee and the Agriculture Committee exhibit the lowest mean loyalty of all standing committees. Rules and Education and Labor exhibit the highest mean loyalty indexes (Coker and Crain 1994). As an empirical test of their committee "importance" proposition, they examine the degree to which the leadership feels compelled to "stack" committees with party members (vis-a-vis the opposition) since "[s]uch stacking provides a margin of safety when defections occur on votes, and would tend to be more heavily utilized where legislative stakes are greatest" (1994: 206-207). As Coker and Crain posit more loyal members will wind up on more influential and important committees. Upon regressing the mean loyalty index on the "stacking" measure of committee importance, they produce a positive and significant (at the 0.01 level) parameter estimate. This finding statistically links the Coker-Crain loyalty index to a traditional measure of committee importance.