The Whips: Building Party Coalitions in Congress by C. Lawrence Evans

Ann Arbor: University of Michigan Press, 2018. 384p. Available in cloth and paper (ISBN: 978-0-472-03730-8).

Michigan Press page
Amazon page

C. Lawrence Evans is Newton Family Professor of Government at the College of William & Mary. The Whips is the 2019 recipient of the Richard F. Fenno, Jr., Prize for the best book about legislatures published the previous year.

A comprehensive and historical treatment of party whips in the U.S. House and Senate, this book sheds new light on partisan coalition building on hundreds of the most significant measures considered on Capitol Hill from the 1950s to the 2000s. Its empirical heart is extensive archival research in the papers of dozens of former congressional leaders, which enabled me to create data sets charting the emergent positions of members on nearly 1,500 questions “whipped” by congressional leaders across five decades. With this new evidence, we can track with precision the magnitude of the challenges confronting party leaders, their relative effectiveness at convincing potential opponents to support the party position, and the consequences for party success. Importantly, the results do not jibe all that well with standard theoretical accounts of party influence, largely rooted in spatial models of legislative choice. Building on insights from Richard Fenno and John Kingdon, I argue instead for a behavioral perspective that highlights the processes through which members form policy positions as they strategically respond to the audiences important to their goals.

The structure of the book reflects its scope. Chapter 1 charts the organizational development of the House whip system, from the creation of the formal role in the 1890s to the extensive networks that characterize the contemporary Congress. Increases in the size or complexity of the whip networks are associated with burgeoning partisan polarization, but also with the presence of significant coalition-building challenges. Chapter 2 presents the behavioral perspective, and here I argue that leaders engage the whips only selectively, when an upcoming issue is important to the party agenda, there is some prospect for unifying the rank-and-file and the outcome is in doubt. When the whips are activated, their effectiveness depends on the degree of district-level homogeneity within the party, the size of the caucus/conference, whether the White House is controlled by the same party and the president’s position, among other factors.

The next two chapters provide summary quantitative tests of the behavioral logic, encompassing the entire time stretch covered by the book. Chapter 3 focuses on the decisions that party leaders make about when to whip, while Chapter 4 explores the distribution of member responses on whip calls and the ability of leaders to convert wavering members to party support. Figure 2 of Chapter 3, for example, shows the relationship between whip activity and majority party victory margins on roll calls, and is reproduced here. The horizontal axis denotes the magnitude of the victory margin on House votes, ranging from large majority party losses (-181 to -200) to massive wins (200 or more votes), with close calls located in the middle. The vertical axis maps the percent of all votes falling into the different margin levels for four subsets of the roll call record: nonunanimous votes that were whipped by the majority leadership; nonunanimous votes that were not whipped; party-line votes that were whipped; and party line matters that were not whipped. No matter the whip status or partisan content of a roll call, you can see, there is a tendency toward many close wins for the majority party and few close losses. But the magnitude of the pattern is stronger on partisan votes and on votes that are whipped by the majority, and is especially strong for partisan votes that were also singled out for whipping. Again, party leaders target matters important to the party program and where a victory is possible but not preordained – witness the disproportionate presence of close margins on whipped questions, particularly when the parties are at loggerheads. And the high proportion of narrow victories on whipped matters suggests that the majority leadership is generally able to eke out wins when the stakes are high.

MAJORITY PARTY VICTORY MARGINS BY WHIP STATUS, NONUNANIMOUS AND PARTY VOTES (1955-2002)
Source: Reproduction of Figure 3.2 provided by author

These and other results reported in The Whips should not be interpreted as support for the standard spatial theories of majority party power, such as the cartel model or conditional party government. For one, procedural matters are whipped about as often as amendments, bills and other primarily substantive questions, which does not resonate with claims that the majority leadership has monopoly control over the agenda. In addition, when questions are whipped simultaneously by both the majority and minority and in different directions, the movement of members between the two party positions roughly balances out, which is not consistent with even conditional dominance by the majority. Such apparent anomalies, in contrast, do mesh with the behavioral logic outlined in The Whips.

The chapters that follow take a closer look at four key time periods in recent House history, and feature case studies of major legislation. Chapter 5 explores partisan coalition building during the 1950s and 1960s, when House leaders were weak according to standard accounts. Interestingly, on issues that did not fully evoke sectional divisions within the Democratic majority, leadership success rates resembled those of the purportedly more effective majorities of recent years. Chapter 6 concerns the 1970s transition, where the majority Democratic Caucus was usually large, but the myriad internal cleavages and rampant member individualism meant that outcomes routinely were in play. Indeed, the independent impact of leaders was high during this era precisely because of the magnitude of the challenges they faced. Chapters 7 (about the Democratically controlled Houses of 1982-94) and 8 (covering the Republican majorities of 1995-2002) chart the changes in whip operations and effectiveness that occurred as party polarization grew. Importantly, the presence of initially undecided or opposing members continued to create significant challenges for party leaders, who only rarely had enough support before the whip process to prevail. Still, the likelihood of majority party success on whipped matters increased noticeably, from roughly 65% before the 1980s to nearly 80% in the decades that followed.

Chapter 9 reproduces much of this analysis for the Senate. Although majority party leaders in that chamber lack the formal powers granted to House Speakers, win rates on whipped questions for the Republican Senate majorities of 1995-2002 are fairly similar to the remarkable success levels of Tom DeLay, who was GOP whip on the House side at the time. A closer examination of Senate whip operations, however, reveals that the majority in that chamber uses its whips in a primarily defensive manner – to counter the obstructionist tactics of Senate minorities empowered by the filibuster.

My hope is that The Whips may convince scholars to rethink how they do research and teach about lawmaking. For one, the floor decision making process is far more fluid than suggested by the predictably partisan structure of the roll call record, with many members exhibiting significant uncertainty or initially leaning against the party position. Yet, impressive whip success rates and regular party-line votes are not indicative of majority party dominance, and the references to party government so common in contemporary congressional scholarship should be discarded. Leaders matter in the legislative process, to be sure, but they tend to matter the most when there are significant weaknesses in their supporting coalitions. Moreover, the default outcome in the legislative process should not be the “preference” of the median legislator or some other actor made pivotal by chamber rules. When the whips fail, the outcome tends to be inaction or gridlock. As a result, the measure of party effectiveness should not be some deviation from centrist preferences a la the standard spatial accounts. The real impact of the whips and of party leaders more generally is over what comes to constitute middle ground as the floor decision making process unfolds. Rather than estimating exogenously determined ideal points or the spatial location of alternatives, we need to focus more on the twists and turns of member positions as they take form and the consequences for legislation.

Comments are closed.