Multimember Districts' Effect on Collaboration between U.S. State Legislators

• Date: 01 August 2012
• DOI: http://doi.org/10.1111/j.1939-9162.2012.00050.x
• Published date: 01 August 2012
• Author: JUSTIN H. KIRKLAND

JUSTIN H. KIRKLAND

University of Houston

Multimember Districts’ Effect on

Collaboration between

U.S. State Legislators

In this article, I demonstrate that multimember districts form a basis for collabo-

ration between two legislators. In order to maximize the limited incumbency advantages

they possess, legislators from multimember districts form coalitions in an effort to

generate greater credit-claiming opportunities and policy benef‌its for their district. In

order to test this conception, I utilize a natural experiment and an opportunity to observe

institutional change in North Carolina’s elimination of multimember districts during the

2000–2002 redistricting cycle. Coupled with cross-sectional analysis of severalstates that

use both single-member and multimember districts, empirical evidence strongly corrobo-

rates my conception of multimember districts as a basis for collaboration between

representatives.

American legislatures are more than a collection of individual

agents calculating the costs and benef‌its legislation once implemented.

They are social constructions inhabited by social beings that have com-

plicated inf‌luences on one another. In other words,a legislature is a social

network of rational actors making decisions in an interdependent system

of relationships. Legislators inf‌luence and are inf‌luenced by their rela-

tionships with one another. Party (Sarbaugh-Thompson et al. 2006), geo-

graphic distance (Caldeira and Patterson 1987, 1988; Clark, Caldeira,

and Patterson 1993), and gender (Bratton and Rouse 2011; Desmarais,

Cranmer, and Fowler 2009) all inf‌luence these relationships, and these

relationships in turn inf‌luence legislative outcomes (Arnold, Deen, and

Patterson 2000; Fowler 2006a; Kirkland 2011; Peoples 2008). Qualita-

tive interview evidence provides strong support for the notion that leg-

islators take cues and signals from one another about which legislation to

support (Kingdon 1973, 1989; Ray 1982; Songer et al. 1986; Sullivan

et al. 1996).

The evidence that relationships play an important role in legislating

is strong. What is missing from this literature is an understanding of how

the institutions of a legislature inf‌luence the formation and maintenance

LEGISLATIVE STUDIES QUARTERLY, XXXVII, 3, August 2012 329

DOI: 10.1111/j.1939-9162.2012.00050.x

© 2012 The Comparative Legislative Research Center of The University of Iowa

of legislative relationships. The structure-induced equilibrium school of

thought (Shepsle 1979; Shepsle and Weingast 1994) tells us that institu-

tional arrangements help a chamber avoid suboptimal outcomes by con-

straining behaviors. However, most of the work in this tradition has

focused on how institutions inf‌luence choices over bill outcomes and

committee behavior. If institutions in a legislature can inf‌luence choices

by legislators, then these institutions can also affect choices overcollabo-

rative relationships. For example, multimember districts force legislators

to share geographic constituencies, which should alter their incentives for

collaboration and provide natural allies for use in the securing of policy

benef‌its.

In this research, I take advantage of a unique opportunity to study

institutional change in a legislature. Using cosponsorship data as an

indicator of a collaborative relationship between two legislators, I study

the transition in the North Carolina legislature from a multimember

district system to a single-member system in 2002. I couple this with

cross-sectional analyses of cosponsorship networks in the four states

that use some combination of single-member and multimember

districts (Maryland, New Hampshire, Vermont, and West Virginia).

Results support my theory indicating that multimember systems

generate or strengthen relationships between actors with shared

constituencies.

Institutional Incentives for Collaborative Relationships

To understand how multimember districts might shape legislative

behaviors, I begin by assuming that legislative policy preferences are

multidimensional and are driven by a desire to satisfy constituents’ pref-

erences for government action (Crespin and Rohde 2010; Hixon and

Marshall 2007; Talbert and Potoski 2002). That is, legislators prefer to

sponsor, promote, and pass legislation that assists them in maintaining

their positions as elected representatives. I also assume that some issues

have clear partisan def‌initions, while other issues are less easy to def‌ine

along a partisan continuum.1For example, some legislative issues like

abortion fall cleanly along partisan lines. However, legislators from mul-

timember districts or similar geographic regions may have very similar

preferences over distributive legislation even if they are of different

parties.

Further, I also assume that a legislature has some f‌inite amount of

policy it can create targeted towards a specif‌ic district.That is, a legisla-

ture can only pass so many bills or distribute so much public funding

before a session concludes or a budget is emptied. This constraint is a

330 Justin H. Kirkland