Influence without Confidence: Upper Chambers and Government Formation

Published date01 November 2005
Date01 November 2005
529Influence without Confidence
Northwestern University
Rice University
University of California, Los Angeles
Influence without Confidence:
Upper Chambers and
Government Formation
In most parliamentary democracies, governments must maintain the confidence
of a single legislative chamber only. But in bicameral parliaments, upper chambers can
affect the fortunes of government policy proposals. Recent work shows that parlia-
mentary governments that lack control over the upper house also tend to collapse
sooner than those with upper-house majorities. In this article, we show that coalition
builders anticipate the importance of upper-chamber status (majority or minority) in
making their formation decisions. After controlling for a host of “usual suspect”
variables concerning the institutional, ideological, and partisan context of coalition
building, and examining 15,590 potential governments in 129 bargaining situations,
we found that potential coalitions that control upper-house majorities are significantly
more likely to form than are those with upper-house minorities. Our findings are
important for students of bicameralism, government formation, institutions, and,
perhaps most significantly, for those who study policymaking in parliamentary
Legislatures are, by and large, majoritarian institutions. A majority
is necessary to make law, and, in parliamentary systems, a majority of
members must at least tolerate the formation and survival of a particular
government (cabinet). But because most electoral systems are not
majoritarian but proportional, and typically do not produce single parties
that alone control a majority of legislative seats, both lawmaking and
government formation require coalitions that cross party lines.
530 Druckman, Martin, and Thies
Over the last 40 years, political scientists have produced an
impressive body of theoretical and empirical work designed to predict
which coalitions (or at least which sorts of coalitions) will form and,
relatedly, how long they will survive. The evolution of coalition-formation
research is well known (for example, Laver and Schofield 1990)—
starting with theories that portray parties as pure office-seekers (for
example, Dodd 1976, Riker 1962, and von Neumann and Morgenstern
1953), continuing with approaches that recognize that parties also pursue
policy (such as Axelrod 1970 and de Swaan 1973), and following up
most recently with work that shows how institutions affect coalition
formation (for instance, Bergman 1993; Laver and Schofield 1990, 195–
215; Mershon 1994, 2002; Müller and Strøm 2000; and Strøm, Budge,
and Laver 1994).
While this research on coalition formation has led to notable
predictive success, it has suffered from two serious limitations: one
methodological, the other conceptual. The methodological limitation,
brought to light by Martin and Stevenson (2001), is that most empirical
work on government formation has looked exclusively at governments
that have actually formed, excluding governments that might have formed
but did not. But any theory that purports to explain government-
formation decisions must be able to explain which potential govern-
ments would not form as well. Otherwise, one runs the risk of intro-
ducing bias into the analysis by selecting on (presumably high) values
of the dependent variable, that is, the probability that a particular coali-
tion will form.
The second, and more fundamental, limitation of prior research is
that it has abstracted away from the question of how coalition govern-
ments make policy decisions. Until very recently,1 scholars have focused
on government formation, portfolio distribution, and termination, paying
almost no attention to how coalitions actually govern once in office. Of
course, office-seeking models assume that parties are in the coalition
game only to enjoy the spoils of office, and so ignore policymaking
altogether (see Laver and Schofield 1990). But even supposedly “policy-
aware” models fudge questions of policymaking by use of strong
assumptions about how coalitions operate between birth and death,
namely, that policy bargains are internalized in negotiations over
government formation and then faithfully passed into law by the
members of those same parties. In other words, these models assume
that the same legislative majority responsible for the formation and
survival of the government can be relied upon to support the
government’s policy initiatives during its tenure.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT