Legislative Cooperation among Impatient Legislators
| Published date | 01 January 2006 |
| Date | 01 January 2006 |
| DOI | 10.1177/0951629806059596 |
| Author | Justin Fox |
| Subject Matter | Articles |
Journal of Theoretical Politics 18(1): 68–97 Copyright &2006 Sage Publications
DOI: 10.1177/0951629806059596 London, Thousand Oaks, CA and New Delhi
http://jtp.sagepub.com
LEGISLATIVE COOPERATION AMONG
IMPATIENT LEGISLATORS
Justin Fox
ABSTRACT
Sufficient conditions for legislative cooperation are identified within the
context of a model of repeated legislative interaction. We show that in many
environments, cooperation is sustainable even among impatient legislators.
Special attention is given to the case of repeated spatial bargaining; we find
that when the dimensionality of the policy space is sufficiently large, para-
meterizations of the model which do not admit cooperation are rare and
atypical. Furthermore, contrary to conventional wisdom, we demonstrate
that legislative cooperation is possible in a one-dimensional policy space. The
developed theory is applied to address various claims in the substantive litera-
ture on legislative parties.
KEY WORDS .cooperation .legislative bargaining .legislative parties .
roll-call voting
The single, most important problem in the modern study of the role of parties in law-
making is the absence of a well-articulated (preferably formal) theory about the conse-
quences of intralegislative party activity on collective choices of legislatures. (Krehbiel,
1998: 165)
1. Introduction
A large literature exists exploring the various motivations for legislative
cooperation. For example, it has long been noted that legislators can profit
by participating in ‘log-rolls’ whereby a group of legislators give their sup-
port on issues of salience to other legislators in exchange for support on
issues of salience to themselves (Buchanan and Tullock, 1962; Weingast,
1979; Schwartz, 1989). While identifying incentives for legislative coopera-
tion, this literature leaves open the question as to how such cooperation is
I am grateful for discussions with Randy Calvert, Jamie Carson, Stacey Chen, John Duggan,
Mark Fey, Ko Maeda, Jianjun Miao, and Larry Rothenberg. The article has also benefited from
the comments of two reviewers, and the editor of the Journal of Theoretical Politics.
maintained. The various practical obstacles involved in sustaining legislative
cooperation are explicated in Weingast and Marshall’s (1988) seminal paper
on legislative organization.1
As legislators interact repeatedly, several scholars (Epple and Riordan,
1987; Calvert and Weingast, 1993; Carrubba and Volden, 2000; Calvert
and Fox, 2003) have appealed to the theory of repeated games to explain
how such obstacles are overcome. To date, however, these scholars have
restricted attention to distributive environments: those environments where
the benefits and costs of legislation can be targeted towards particular legis-
lators.2The present article builds upon this work by analyzing legislative
cooperation in a more general environment. This article’s model of repeated
legislative interaction can be parameterized to represent various settings;
such settings include repeated distributive bargaining and repeated spatial
bargaining.
We now offer two reasons why a theory of legislative cooperation must
address policy making on non-distributive issues. First, as the benefits and
costs of legislation may be targeted on distributive issues, harsh penal
codes which deter deviations from cooperative arrangements are easily
constructed. A typical punishment scheme employed in prior work involves
excluding deviant legislators from future benefits forever. Consequently, it is
not known whether the requirements on legislative patience necessary to sus-
tain legislative cooperation are empirically plausible when targeted punish-
ments are unavailable.3
Second, the recent theoretical and empirical literature on cooperation
within the US Congress, particularly as applied to legislative parties
(Rohde, 1991; Cox and McCubbins, 1993; Krehbiel, 1993, 1998; Aldrich,
1995; Snyder and Groseclose, 2000; McCarty et al., 2001), draws upon spatial
models of the political process. However, despite relying on the language of
game theory, this literature lacks explicit spatial game-theoretic foundation.4
Absent such foundations, the validity of various estimators employed to
measure the influence of parties on the policy process remains unclear.
FOX: LEGISLATIVE COOPERATION 69
1. Such obstacles include time-consistency problems that arise when vote trades are non-
contemporaneous. To see this, consider the case where legislators have a finite time horizon.
Once legislation deemed important to a given legislator has passed, that legislator then has an
incentive to renege on commitments made to others.
2. A canonical example of a distributive environment is one where a legislature decides by
majority rule upon the division of a unit resource.
3. Such environments include those in which the legislature produces a collective good such as
national security.
4. Smith (2000: 208) argues that the challenge for scholars of legislative parties is to ‘demon-
strate that party effects can be generated from a spatial model in which legislators are motivated
by policy preferences alone (i.e., minimizing the distance between their ideal points and the out-
come).’
To continue reading
Request your trial