A Clustering Approach to Legislative Styles

• Date: 01 August 2017
• Author: Tracy Sulkin,Daniel Sewell,William Bernhard
• DOI: http://doi.org/10.1111/lsq.12162
• Published date: 01 August 2017

WILLIAM BERNHARD

University of Illinois

DANIEL SEWELL

University of Iowa

TRACY SULKIN

University of Illinois

A Clustering Approach to

Legislative Styles

Once elected to off‌ice, members of Congress (MCs) face choices

about how to prioritize their activities. What is the right balance between

working in the district and on Capitol Hill? How active will they be in

introducing and cosponsoring legislation? Will they give speeches and

interviews? On which issues will they focus their attention? With whom

will they collaborate and form coalitions? How much time will they

spend raising money? To what extent will they toe the party line or chart

their own course? Collectively, these decisions make up a member of

Congress’s “legislative style.” Legislators’ styles are central to congres-

sional life, with the potential to shape their career trajectories, the

representation they provide to constituents, and the nature of the policies

passed by Congress.

In this article, we analyze legislative style, testing our theory about

its correlates and consequences. We focus on the 1,049 legislators who

served in the 101st–110th Congresses (1989–2008), gathering data on

the activities of members of Congress, categorizing these into indices

that ref‌lect components of legislative style, and using longitudinal

model-based clustering techniques to uncover how these components

group together. Our results reveal that MCs’ patterns of activity cluster

into f‌ive stable and predictable styles—district advocates, party builders,

ambitious entrepreneurs, party soldiers, and policy specialists.

Legislative Activity and Legislative Style

Our conception of style is rooted in the fact that MCs approach

their jobs in different ways, coming to Congress with varying career

goals, electoral and institutional constraints, past experiences, role

LEGISLATIVE STUDIES QUARTERLY, 42, 3, August 2017 477

DOI: 10.1111/lsq.12162

V

C2016 Washington University in St. Louis

orientations, and personal inclinations (Barber 1965; Clapp 1963;

Davidson 1969; Fenno 1978). Moreover, the complexity of the institu-

tion and the sheer volume of available opportunities require legislators to

make choices. They simply cannot be active on everything.

The idea that allocation of effort is central to understanding legisla-

tive careers is not new. A half-century ago, Bauer, Pool, and Dexter

noted that “[a] congressman must decide what to make of his job. The

decisions most constantly on his mind are not how to vote, but what to

do with his time, how to allocate his resources, and where to put his ener-

gy” (1963, 405). At about the same time, Matthews’s (1960) work on

norms and folkways in the Senate introduced the commonly invoked

distinction between “work horses” and “show horses”

1

—the idea that

some MCs did the heavy lifting legislatively, while others focused on

gaining acclaim and attention.

However, despite the recognition that MCs think about their

behavior in a gestalt way, with a set of fundamental goals and an under-

standing of the context in which they operate that precedes their

decisions about how to pursue their jobs, congressional scholars have

generally studied activities in isolation from one another—one behavior

at a time. The positive result is rich literatures on a variety of legislative

activities, but the downside is that we are often left without a sense of the

trade-offs MCs make, of how they strategize about their careers, and of

the effects of their choices on their prospects.

This is where legislative style comes in. We argue that legislators

have different ambitions—some come to Washington to invest in policy,

some want to be Speaker of the House, some view the House as a way

station en route to the Senate or presidency, some see themselves as par-

ty or district servants, and many, if not most, hold a combination of these

aims. Because MCs’ behavior is goal oriented, they engage in patterns

or “packages” of activity that correspond to a particular constellation of

goals. These patterns are characteristic of individual MCs, but not unique

to each. Instead, given similar aims, ambitions, and constraints, we

hypothesize that MCs’ activity will cohere into a small number of

common legislative styles.

We theorize that styles are a function of the incentives and con-

straints MCs face, which are affected by characteristics of both

legislators and their districts, as well as the context in which they f‌ind

themselves. Thus, while we expect styles to be relatively stable features

of individuals, they may shift with MCs’ circumstances (e.g., changes in

seniority, electoral vulnerability, majority status, institutional position)

or as their goals evolve. In what follows, we explore the dynamics of leg-

islative style by describing the styles that emerge from our clustering

478 William Bernhard, Daniel Sewell, and Tracy Sulkin