The power of committee chairs

By Christopher R. Berry and Anthony Fowler

Congressional committees are thought to play a central role in policymaking and the distribution of federal spending, and seats on important committees such as Appropriations or Ways and Means are highly coveted. Political scientists have extensively studied and hypothesized about the importance of committee positions, but surprisingly, committee chairs have received notably less attention. Our results from two recent papers suggest that the focus of congressional scholars appears to be misplaced. Rank-and-file committee positions generally have little impact on the influence and power of members of Congress, while committee chairs are more important than previously thought.

In a recent study published in the American Journal of Political Science, we conduct a comprehensive study of the importance of committee positions for the distribution of federal spending. For decades, political scientists assumed that committee seats increase the power of members, allowing them to influence the content of policy and secure more goodies for their constituents. The primary evidence for this claim is that committee members seem to get more. For example, the figure below shows that senators on the Appropriations Committee procured twice as much federal spending from earmarks as senators not on Appropriations in the 110th and 111th Congresses (comparing the green and red lines). However, these differences are likely not attributable to the effects of the committee, because members who were about to join the committee procured just as much as committee members before joining the committee (blue line). In other words, the kinds of members who join Appropriations are different than those who do not, and we can’t learn about the effect of committee positions by simply comparing members to non-members. When we compare the same individuals before and after they joined the committee, we detect no evidence that the committee position influences the ability of members to procure pork for their constituents.

In our paper, we extend the logic of this simple analysis to test for the effects of congressional committee positions across virtually all committees and for virtually all federal spending. Specifically, we implement within-member differences-in-differences designs to test whether committee positions increase the flow of federal spending to a member’s constituents. Consistent with the simple result in the figure above, we find that committee positions are less important for distributive spending than previously thought. Although there is significant selection onto committees, we find little effect of committee positions for the ability of members to bring pork home to their states and districts. Strikingly, we even obtain null results when focusing on the specific policy domain of a committee. For example, a seat on the Agriculture Committee or the Appropriations Subcommittee on Agriculture doesn’t even allow a member to procure more agriculture spending for her constituents. In short, congressional committee positions appear to be less important for distributive politics than previously thought.

One notable exception is that when a member becomes the chair of an Appropriations subcommittee, she gets a lot more pork, at least within the domain of the subcommittee. In other words, it looks like Appropriations subcommittee chairs, the so-called “cardinals of Congress” exert disproportionate control over federal spending within their domain, and this comes at the expense of rank-and-file committee and subcommittee members who do no better than when they were off the committee.

In a follow-up study, published in the Journal of Public Economics, we go beyond federal spending to ask whether these general patterns are unique to federal spending or whether they hold broadly across many forms of legislative influence. Legislative influence is difficult to measure directly, but we employ three different measures. First, we examine Volden and Wiseman’s legislative effectiveness scores which capture the ability of members to move their proposed bills through the legislative process. Second, we examine campaign contributions with the thinking that access-seeking contributors will give more to members who have more legislative power. And third, following a similar logic, we examine the value of lobbying contracts garnered by professional lobbyists who used to work for a particular member. Across each of these measures, we test whether a member appears to become more powerful when they gain a particular committee position.

As we found in the context of federal spending, the effects of rank-and-file committee positions are typically small, even for the most important and coveted committees. However, becoming committee chair significantly increases the implied influence of a member. The graph below illustrates this phenomenon with an event study. Before becoming chair, a member’s legislative effectiveness score remains relatively flat, but it increases dramatically when the member becomes chair of any committee and it remains high so long as the member remains chair. In the case of campaign contributions, we detect modest effects for some rank-and-file committee positions. For example, when a member joins an authorizing committee, she gets more campaign contributions from PACs connected to that particular sector. However, in general, these effects pale in comparison to the effect of becoming chair of those committees. Interestingly, we also detect some anticipatory effects for campaign contributions. Because sophisticated contributors try to target members who will be influential in the future, members often receive a bump in contributions right before they become chair.

Synthesizing results from these two studies, we find that rank-and-file committee members are less influential than previously thought, but chairs are perhaps more important than previously thought. The chairs of Appropriations subcommittees appear to exert disproportionate influence over federal spending, while the chairs of authorizing committees appear to exert disproportionate influence over other forms of lawmaking. The lack of attention paid to chairs by congressional scholars is surprising because there are good theoretical reasons to expect chairs to exert the disproportionate influence that we detect. Among other things, committee chairs can hire, fire, and direct most committee staff, refer bills to subcommittees, schedule or block hearings and markup, and run committee meetings. In other words, the chair wields both positive and negative agenda control, and theoretical models show that the agenda setter on a committee can often wield disproportionate influence and achieve her preferred policy outcomes. Our results suggest that scholars, journalists, interest groups, and observers should probably pay less attention to the members of supposedly important committees and they should pay more attention to the chairs of committees and subcommittees.

Note: This post is an updated version of “Congressional Committee Membership is Less Important Than Previously Thought, but Chairs are Really Influential,” which we wrote for the American Politics and Policy Blog hosted by the London School of Economics and Political Science.

Christopher R. Berry is the William J. and Alicia Townsend Friedman Professor in the Harris School of Public Policy Studies at the University of Chicago. He studies distributive politics, municipal finance, and electoral accountability. His work has recently appeared in the American Journal of Political Science, Journal of Politics, and Yale Law Journal, and he is the author of Imperfect Union: Representation and Taxation in Multilevel Governments published in 2009 by Cambridge University Press.

Anthony Fowler is Associate Professor in the Harris School of Public Policy Studies at the University of Chicago. He studies political representation, with particular interests in elections and participation. His work has recently appeared in the American Journal of Political Science, Journal of Politics, Quarterly Journal of Political Science, and Proceedings of the National Academy of Sciences.