Bargaining, Transaction Costs, and Coalition Governance - Lupia
Representative democracies face two fundamental challenges, delegation and coalescence.
Delegation is necessary because most citizens have neither the capacity nor the time to
make many important political decisions on their own. To facilitate large-scale governance,
multiple acts of delegation must occur. Citizens must find and select representatives whom they
can trust to make public policy in accordance with particular principles. Elected representatives,
in turn, delegate to leaders of political parties or bureaucratic agencies to further the pursuit of
policy goals. While delegation makes large-scale representative democracy possible, it is not
without risk. Problems of delegation need to be overcome. For with the power the elective or
appointive office brings also comes opportunities to act against the public interest.
Coalescence is necessary because representative democracies typically produce a
multitude of political actors and because democratic rules often require that decisions be
supported by a simple or qualified majority of the representatives in a national legislature. In a
democracy, no one person can legislate or make power without the support of others. To
facilitate large-scale governance, coalitions must be built. To pass laws or implement public
policy, at least some of the agents that are empowered to act on citizens’ behalf must find a way
to work with others who are also empowered.
All representative democracies of necessity face both of these challenges, and the ways
in which they solve or confront them in the long run has a decisive influence on the feasibility
and quality of democratic governance. If the problem of delegation is not solved, then we may
end up with a “democratic deficit” in which political representatives do not enjoy the trust of
their constituents. If coalescence does not occur, then the same representatives may not be able
to make authoritative decisions. In either case, the prospects for sustained democratic rule are
The challenges of delegation and coalition are not wholly distinct. In fact, they intersect
in the coalition life cycle described in Chapter One. While most democratically elected
politicians work in coalitions with others on a daily basis, many such actions occur in the
shadow of elections. Many politicians act knowing that the electorate who once delegated
power to them will have an opportunity to take it away in the future. When coalescing,
politicians must keep their delegation relationship with voters in mind. When delegating,
citizens and elected representatives are generally aware the needs and desires of relevant
political coalitions will affect the consequences of delegating to particular people.
In a previous study (Strøm, Müller and Bergman 2003) we have explored at length the
challenges of delegation in contemporary Western European democracies. Although in this
book we shall frequently refer to and build on the lessons of that previous effort, our purpose
here is different. Here, we assume that conditions that allow for at least minimally successful
delegation from citizens to politicians are satisfied and turn our attention to questions of
In the previous chapter, we have accounted for the ways in coalition formation,
maintenance, and termination in parliamentary democracies have been understood. In this
chapter, we shall delve more deeply into the questions of what such political coalitions are, why
they exist, and how they operate. The answers that we shall provide to these questions will
inform the analysis that will be presented in the various empirical chapters in this volume.
To begin at the beginning, a coalition is a team of individuals or groups that unites for a
common purpose. In many countries, teams of political parties coalesce for the purpose of
running a government. Together, these coalition members convert a wide range of social
demands into a manageable set of state-sanctioned activities.
While working for a common cause, coalition members may disagree about important
matters. Some disagreements come from members’ attempts to please distinct constituencies
(e.g., members who represent urbanites want different policies than members who represent
rural regions). Conflicts can also arise when multiple members crave a particular seat of power
(e.g., that of the prime minister). How coalition members cultivate their common interests and
manage internal conflicts affects the fate and effectiveness of the governments they run.
Throughout this book, we study these decisions in order to better understand coalition
governance around the world.
In this chapter, we offer a theoretical framework. It yields clarifying generalizations
about coalition behavior in political contexts. While this framework can be applied to
governance dynamics in many nations we focus primary attention on Western Europe’s
parliamentary democracies. 2
Generalizing about coalition politics is no trivial matter. Every country challenges those
who govern it with a unique mix of historical precedents, democratic principles, political
institutions, social conventions, and popular demands. It is natural, therefore, to expect that
clarifying accounts of one country’s coalition politics will seem bizarre when applied
Consider, for example, the coalition politics of Italy, which since 1945 has averaged
approximately one new cabinet every 13 months. This frequency of cabinet turnover is the
highest in Western Europe. It is indeed more than twice the average for the region (Müller and
Strøm 2000: 561). At the same time, Italian politics was uniquely stable in that for more than a
third of a century (1945-81) a single political party (the Christian Democrats) dominated every
government and accounted for every single person that passed through the revolving door that
the Italian prime ministership often resembled. If a theoretical framework can explain
important aspects of coalition governance in this unusual circumstance, can we use it to better
explain coalition governance in other countries? Throughout this book, we answer “Yes.”
Our approach builds on the premise that parliaments are deliberately organized to
recognize and reinforce a central role for political parties (Müller 2000). Hence, we examine
coalition governance by focusing on how political parties juggle the interests of constituents
and coalition partners when making critical decisions (e.g., coalition formation, policy making,
ministerial replacements, and the timing of elections). This is not to say that what happens
inside parties is not important for understanding cabinet politics. Several chapters will in fact
show that intra-party politics can systematically affect inter-party bargaining. Our approach is
rather built on a pragmatic recognition that a full account of intra-party politics would be far
beyond our capabilities in this project, and that as analysts we benefit from the fact that, for
better or worse, parties have a strong incentive to maintain cohesion within their ranks.
Our framework has six components. Five of these come from established approaches to
the study of coalition governance. While these components are often used one-at-a-time, we
1 The behavior of political parties can, in principle, be understood as an interaction between party members,
officers, and activists (see, for example, Strøm 1994). Yet, any systematic and tractable version of such an account
is beyond the grasp of current social science. Thankfully, most political parties are sufficiently institutionalized
and hierarchical that many of their behaviors can be understood as if they were those of a person. So, throughout
this book, we assume that we can treat parties, or more specifically their parliamentary leaderships, as if they were
unitary actors (see Laver and Schofield 1990; for a more elaborate defense see Müller and Strøm 1999, ch. 1).
produce additional insights by integrating them into a unified framework. The sixth component,
our framework’s backbone, provides the means for integration. Components 1-5 are as follows:
1. Exogenous contextual factors, while important, are not the whole story.
A venerable tradition in the social sciences emphasizes the uniqueness of the contexts in which
collective decisions are made. These accounts tend to highlight factors associated with the
particular community in which events take place. Such accounts may seek causality in aspects
of a country’s culture and history. It is often argued, and no doubt true, that political decisions
reflect peculiar characteristics of a national or other systemic context that are not easily
captured in the form of discrete and parsimoniously defined variables. Applied to the Italian
propensity for cabinet turnover cited above, such accounts might privilege particularly Italian
conditions such as its culture of distrust (Banfield 1958).
A second type of account in this broad tradition stresses the historical precedents of
social events. The argument is that each community is profoundly and pervasively influenced
by its own past, in ways that can neither be modeled parsimoniously nor ignored. Moreover,
each new historical event adds to a society’s uniqueness and further constrains all future
decisions. This is the argument of path dependency. A strong version of this argument would
imply that political decisions are so heavily conditioned by past choices that there can be no
meaningful comparisons across political settings with different histories (see Thelen and
Steinmo 1992 for a review). If path dependence is indeed a fruitful perspective, then one must
seek the explanation of what happened in Italy in 1950 in the events of previous years in the
same country, such as its late unification or the legacy of fascism.
Contextual arguments may also stress the importance of shared, rather than separate,
experiences. According to this logic, political events are determined by contemporaneous
events, even if these occur in mutually remote locations. This view presumes that politicians in
different countries live interdependent lives, in which they may be subject to a “mood” of the
times or simultaneously influenced by shared experience of traumatic periods such as the Great
Depression or World War II. The mood of the times may be diffused through the common
icons or symbols of a period. If such common experiences are indeed decisive, then coalition
decisions of the 1950s, for example, can only be understood in that historical context, and not
those of the 1980s or 1990s. Whether one stresses cultural uniqueness, path dependency, or a
shared history, however, the implication for scientific work is that once an analyst controls for
context, there is little left to explain.
Yet, such approaches seem inconsistent with the observed trajectories of political
representation in contemporary democracies. Despite profoundly different historical paths,
many European societies, including such unlikely cases as the Netherlands and Switzerland,
have witnessed the growth of populist right-wing parties. On the other hand, the pervasive
“mood” of European secularization has been reflected in party systems in surprisingly varied
ways. Why, for example, have Christian Democrats since 1980 collapsed in Italy, weakened in
the Netherlands, but experienced exhilarating new highs in highly secular Sweden?
Invocations of time and place clearly cannot fully substitute for more fine-grained
scholarly analysis. Our own conviction is to be somewhat more ambitious on behalf of political
theory. In this and subsequent chapters, we show that while temporal and country-specific
effects can be important, they do not provide the whole story.
2. Resource distribution among political actors, while important, is not the whole story.
Some observers see political life as a contest between large and powerful political forces (e.g.,
Miliband 1990) in which resources ultimately decide the winners. As Stein Rokkan (1966: xxx)
succinctly put it, “Votes count, but resources decide.” Such perspectives are often coupled,
though they need not be, with structural perspectives on political contestations. In the context
of parliamentary politics, they may contend that bigger is always better. Hence, the most
important asset that a political party can bring to the political bargaining table is the size of its
parliamentary delegation, complemented, perhaps, by currencies such as money. In the Italian
case cited above, this approach could portray cabinet turnover as the consequence of a standoff
between two powerful forces in Italian politics: the Christian Democrats and the Communists.
While politics is often dominated by the strong and the resourceful, such domination is
neither inevitable nor is it sufficient to explain many past events. In the early 1980s, Italian
politics did indeed seem dominated by the respective parties of Christian Democracy and
Communism, respectively. Yet, in 1982 the Christian Democrats had to give over the prime
ministership to the much smaller and less imposing Republican party (PRI), and within about
ten years both of the once so dominant parties were effectively dead and gone.
3. Politicians’ preferences, while important, are not the whole story.
The two perspectives we have identified above account for many of the classic studies of
coalition politics. If the emphasis on country-specific and historical factors dominated the
discipline up to the 1960s, the behavioral revolution brought resource questions to the fore.
Partly, such a perspective was also reinforced by the first generation of more analytically
rigorous reasoning, such as Riker’s (1962) seminal coalition theory. Yet, the next generation of
coalition analysts typically contended that spatio-temporal parameters and resources were
insufficient explanations of coalition politics. In their view, there is a direct link between
politicians’ preferences and coalition decisions -- a link that cultural factors, resources, and
institutions do not affect. One manifestation of this idea is that if two parties share common
policy agendas (or have policy ideal points that are close to one another), then they will have a
stronger, more effective, and longer-lasting coalition than will coalitions with less common
agendas (De Swaan 1973, Warwick 1994). Applied to the Italian case cited above, the argument
might be that the large differences in policy preferences between Christian Democrats,
Communists, and Neo-fascists make coalition bargaining difficult – and that such interests
would clash under any institution, political resource distribution, or flag.
But politics sometimes makes for strange bedfellows. Studies of subnational coalitions, in
particular, have shown a surprising incidence of partnerships that fly in the face of those that
exist at the national level and that seem to defy explanation in terms of common policy
preferences (Downs 1999?, Bäck 2003?). In this and the chapters that follow, we argue that
while preference-based hypotheses sometimes have great explanatory power, there are also
contexts in which they explain very little or yield mistaken conclusions.
4. Institutions, while important, are not the whole story.
Another view is that institutions drive coalition politics and, indeed, our knowledge of the
effects of the institutions of coalition politics has grown greatly since the 1980s. A strong
version of the institutionalist position holds that particular institutions (such as a proportional
electoral system, a constructive vote of no confidence, a powerful legislative committee
structure, or an investiture requirement) have consistent effects on coalition politics, regardless
of national setting, resource distribution, or the personalities involved. In the Italian example,
this view not only implies that electoral and legislative rules generate the high frequency of
coalitional change, but also implies that if you changed the nationality of the participants, while
leaving the institutions constant, the outcome would be the same. Institutional ideas are implicit
in many writings on coalition government, including the “structural attributes” literature on
cabinet duration and the literature on bargaining constraints (see, e.g., Powell 1982; Lijphart
1999; Schofield and Laver 1990; Strøm, Budge and Laver 1994).
Institutions surely affect coalition politics, and it is important to understand their effects.
We concur with scholars such as Luebbert (1986, especially pp. 29-44) that institutions
privilege certain behaviors. We also agree that institutions can shape the expectations and
preferences of political actors. But if institutions are the sole factor in explaining coalition
decision making, then the implication for empirical work is that once an analyst has controlled
for institutional differences, there is little left to explain. Throughout this book, we clarify how
institutional effects depend on, and are affected by, other factors.
5. Critical events, while important, are not the whole story.
Finally, there are also those that stress the importance of critical events in shaping modern elite-
level politics. The Great Depression brought to power many political parties with radical policy
agendas and indeed brought down democracy in a number of countries. The oil crisis of the
1970s resulted in severe losses for governing parties, whoever they happened to be, all over the
Western world. The collapse of the Soviet Union and its European empire in the late 1980s and
early 1990s embarrassed and severely weakened communist parties all over Western Europe.
Critical events may also take more modest forms and be confined to specific countries. Thus,
coalition politics may be substantially affected by national events such as political scandals,
bankruptcies, crimes, the sexual escapades of politicians, or even the fortunes of cherished
national athletes (for example, the national football team).
Since the early 1980s, students of coalitions have tried to take the importance of political
events seriously, and some have gone so far as to argue that the importance of such
unpredictable factors obliterates any effort to account for coalition politics in deterministic
ways (see Browne et al., 1986). But while critical events can certainly play a major role in
coalition politics, they do not seem to preclude any attempt to account for coalition bargaining
in systematic terms. Indeed, we shall argue that we can best capture the significance of critical
events within a larger theoretical framework.
6. Bargaining Requirements Provide an Organizing Framework
Each the five perspectives we have briefly identified names a determinant of coalition behavior.
A primary theme in this book is that when factors such as these are examined together, under a
unified framework, they can teach us even more. Our framework’s final component provides
our means of unification. This component begins with the fact that all members of potential
governing coalitions, regardless of country, resources, institutions, or preferences, face a
common prize and a common problem. The prize is the opportunities for gain, honor, and
policy advantage that control of the national government affords. The problem of the same
contestants is that to reap the benefits of governing, they must satisfy two requirements
1. Form and maintain agreements with other parties.
2. Please voters.
A coalition that wants to survive must therefore satisfy both requirements simultaneously.
Parties that fail to cultivate voter support are replaced in parliament by parties that succeed,
while parties that fail to maintain coalition agreements are replaced in cabinets by parties that
can. Satisfying the two requirements of coalition participation is seldom easy. It is made
difficult when a party’s electoral constituents and coalition partners want different things. It is
even more difficult when a party’s political goals – such as enacting a wide-ranging policy
2 Under any
agenda – necessitate that both requirements are met for an extended period of time.
circumstances, coalition members have an incentive to engage in coalition bargaining thinking
not just about their present desires, but also about the shadow of the future (i.e., the likely
reactions of voters and political opponents).
This representation of coalition politics is the backbone of our framework. At every
stage of a coalition’s life cycle, coalition decisions are the result of bargaining, where every
bargaining outcome is the result not only of past bargains that affect history, institutions, and
2 This is why credibility can be such a difficult problem in democratic politics. The problem is not just that it takes
time to forge and implement political agreements. For such agreements to be effective, it is often also necessary
for other political elites and ordinary citizens to act in compliance, which they may not do unless they believe that
the agreements, or the coalitions that sustain them, will endure. Consider, for example, tax schemes intended to
promote investment or any policy that requires costs paid now in exchange for future benefits. If potential
investors perceive a coalition government as unwilling or unable to repay such debts in the future, they may be less
likely to contribute or make sacrifices in the present. Indeed, economic downturns may make it extremely
members’ resources but also of the fact that bargaining occurs in the shadow of citizen
opinions and under the constant threat posed by political rivals who want to replace them.
Our bargaining-based framework will clarify how each of the components listed above
influences coalition decisions. It can be applied at every stage of a coalition’s life cycle, it
answers important questions about when one component (i.e., country-specific attributes)
dominates another, (i.e., institutional) and it clarifies how uncertainty affects all components.
We caution against attempts to simplify the analysis of coalition governance by turning a blind
eye to the necessity of bargaining, the anticipated response of the electorate, or the shadow cast
by threats to the incumbent coalition. While removing such factors can simplify analyses, they
increase the likelihood of erroneous conclusions.
This chapter continues as follows. In Section 2, we introduce our theoretical framework.
In Section 3, we use the framework to correct errors in widely held beliefs about the timing and
nature of a coalition’s decision to terminate. Section 4 summarizes our findings.
2. Bargaining: The Engine of Coalition Governance
Coalition decisions are the product of agreements between coalition members – each of whom,
if they want to remain on the governing team, must satisfy the two requirements of coalition
participation. Bargaining is the means by which such agreements are usually reached. With
these facts in mind, we present a framework whose purpose is to clarify fundamental aspects of
coalition bargaining. We introduce it first in the context of answering simple questions about
coalition governance. As this chapter proceeds, we use it to engage increasingly complex
situations. Throughout, we denote key concepts in bold.
Our main premises are as follows:
• Bargaining is the means by which parties attempt to satisfy the first requirement of
coalition participation – to form and maintain agreements with other coalition partners.
expensive to live up to commitments made in rosier times. Coalitions seen as unlikely to keep promises in such
times will have a harder time making such policies work.
• Who gets what in coalition bargaining depends on what political parties can offer to one
another. Which offers parties will make and accept depends – in particular ways -- on
country-specific factors, institutions, party preferences, voter support, the complexity of
the agreements being sought, and party leaders’ beliefs about an uncertain future.
• Voter support – the second requirement for coalition participation – in turn depends on
the qualities of the agreements that parties strike.
• At any stage in a coalition’s life cycle (e.g., formation, policy making, conflict
management, termination) the outcome of the bargaining process depends on the past as
well as the present. The past determines the resources that different players have
available to them. Past bargains also affect the history and institutions under which
bargaining occurs. The future affects the participants’ expectations about the kinds of
bargains they should accept. While most scholars have treated the different stages of
bargaining as distinct, our approach implies that bargaining dynamics and outcomes at
every stage of the process are inextricably linked to one another.
Let us now define the key terms of our approach. Bargaining is a process by which
actors engage in communication for the purpose of finding a mutually beneficial agreement.
Bargaining is required to reach such agreements, if
a. there exist individual benefits that can only be achieved through collective action,
b. there are multiple ways of distributing the benefits associated with such actions, and
c. no actor can simply impose a collective arrangement upon everyone else.
By individual benefits we mean that potential partners can accomplish more as a
coalition than they can otherwise (i.e., party 1 acting alone can produce social output X, party 2
acting alone can produce social output Y, the parties acting together can produce social output
Z and there is a way to distribute the benefits of Z between the parties such that each party is
better off than if it had produced X and Y alone). Policy decisions, legislation, military
operations, enforcement of laws and property rights help coalition members satisfy their goals,
but often require coalescence. Consider, for example, cases where lawmaking requires a
legislative majority and no single party has such support. Working alone, no single party can
pass a law. Working together, however, they can. As we shall see, this is indeed the typical
situation in European parliamentary democracies.
A challenge to every coalition is that there are usually multiple ways to distribute the
individual benefits associated with coalescence. If public problems allowed only one solution,
then coalition politics would be simple (and not very interesting). The only question would be
whether or not to employ the solution – a relatively simple bargaining problem. But such
simplicity is rare. More often, there are many ways to distribute social resources, many different
policy platforms that a majority of MP’s can support, and more politicians desirous of high
office than offices available. If the different means of achieving a collective goal vary in the
costs and benefits they imply for individual coalition members, then these members can have
different preferences about what the coalition should do. Bargaining is the means by which such
conflicts are managed.
Contracts are the usual currency of bargaining. Contracts are agreements, oral or
written, in which participants commit to certain actions in return for specified benefits. A
contract’s purpose is to clarify terms of agreement, to delineate punishments for noncompliance
and rewards for compliance, and to reduce risk.
The costs of reaching such agreements are known as transaction costs, which Kreps
(1990: 743) defines as follows:
“When undertaking a transaction, parties to the transaction must incur several sorts of
costs… [Some] costs are incurred before the transaction takes place. If the transaction is
to be governed by a written contract, the contract must be drafted. Whether governed by
a contract or simply by verbal commitments, the terms of the transaction must be
negotiated. [Other] costs are incurred in consummating and safeguarding the deal that
was originally struck."
Indeed, those who participate in coalition bargaining must spend time and effort obtaining an
agreement that they and their constituents find acceptable. The amount of these expenditures
depends on the complexity of the agreement being sought. As potential partners encounter an
increasing number of issues to resolve – such as which policy stances the coalition will take and
which parties will obtain particular leadership positions – more time and effort is needed to
weave each resolution into a general agreement. Similarly, when potential partners are seeking
an agreement that they want to last for several months or years, they may find it worthwhile to
spend time and effort setting up ways to enforce the contract’s terms. Because such
expenditures are an essential part of contracting, transaction costs are an important factor in
understanding the role of bargaining in coalition politics.
What can we learn about coalition governance from a framework based on the premises
listed above? Our initial answer to this question comes in the context of explaining why parties
coalesce at all. Once we establish that parties have strong incentives to form and maintain
lasting coalitions, we then turn our attention to the key decisions that enduring coalitions must
A single political party may gain sufficient support among the voters that it can govern
alone, and in most (though not all) such cases it will. Much more commonly, however,
elections in parliamentary democracies yield a result in which no single party is so privileged
(i.e., a minority situation, see Strøm 1990). This is, of course, especially common under
Proportional Representation. In most such circumstances, the resulting government is a self-
recognized and enduring coalition of political parties. Why does this happen? It is not obvious
that parties should prefer such formal and enduring patterns of cooperation. Commitment to a
coalition agreement can constrain individual parties in important ways (e.g., through the loss of
voter support associated with legislative compromises that the party would not have sought on
its own). Given that the world is unpredictable, why shouldn’t political parties prefer to just
drop in and out of free-floating majorities? Indeed, the advantage of not coalescing is freedom
– parties need not be burdened by their coalition partners’ demands and could indeed change
partners from issue to issue. Thus, they might avoid the most costly demands of any coalition
partner and would be free to remain more responsive to their own constituents at any time.
These are indeed important considerations and they help explain why not all parties seek stable
and comprehensive coalitions all the time.
Prevailing against such freedom, however, is a host of problems and inefficiencies.
Among the most serious are the following:
• Increased transaction costs. Free-floating majorities force participants to negotiate every
decision anew. The time and energy needed to proceed in this manner can exhaust a party’s
resources and reduce its abilities to accomplish broad or multiple goals. An important part
of the rationale for stable coalitions is that they economize on transaction costs.
• Less policy continuity and impact. Without a formal coalition, why should the majority that
forms on a Tuesday morning enforce the laws made by Monday evening’s legislature? Such
instability would greatly reduce the value of any decision that a governing coalition would
be able to make. After all, what is the value of governing if you have no idea whether your
decisions would stand long enough even to be implemented? In a situation of free-floating
majorities, the impact of any government decision could be radically reduced, so that the
public or private gain from controlling the levers of government might simply vanish.
• Less policy credibility and outside support for government programs. As just mentioned, a
polity governed strictly by free-floating majorities might be one in which public policy
would have little long-term stability or credibility. This might adversely affect citizens in
many ways. Imagine, for example, what buying a house or investing in stocks would be like
if a nation’s basic notion of property rights were subject to frequent and unpredictable
change. The costs of such arrangements to anyone with a long-term interest in a society
would be tremendous. Indeed, for people attempting to base social, political, or economic
plans on government policies, free-floating majorities spell disaster.
Policy credibility is not a concern simply for those who “consume” government decisions.
For most government policies to be effective, people outside of government must cooperate.
Citizens must abide by the laws, businesses must adhere to the terms of their contracts, and
countries must act within the terms of certain treaties. If a government cannot credibly
uphold agreements tomorrow that its leaders sign today, anyone who deals with that
government has less incentive to trust or cooperate with it. Unstable governments might
encounter such problems of cooperation very close to their own ranks. Even if a free-
floating majority were in power long enough to name its cabinet ministers, there is little
reason to expect others in government to abide by their ministerial directives. Indeed, we
should expect bureaucrats who dislike today’s legislative directives to disregard them if
they believe that tomorrow’s governing coalition will change the law in ways that they
prefer (see, e.g., Huber 1998, Huber and Lupia 2001). In other words, government on land is
like the government on high seas – the prospect of leadership instability makes mutiny more
• Less reliable voter support. In democratic countries, parties are in a position to bargain for a
place in government only if citizens delegate policy-making authority to them through
elections. The electoral connection – the threat that eventually voters will judge coalition
members – governs member behavior. This connection means that parties cannot simply
bargain as they please. Voters gave them their power and they can take it back. Therefore, if
voters prefer a government whose actions are at least somewhat predictable, then politicians
who can credibly commit themselves to something other than a transitory coalition stand to
gain. Stable coalitions can tie parties to specific policies, which gives voters a more credible
policy-oriented rationale for differentiating between candidates for office. Members of
stable coalitions can also more easily establish "policy brand names" that reduce citizens’
uncertainty about the policy consequences of voting for a particular candidate (Cox and
McCubbins 1993). Indeed, stable coalitions make it easier for voters to hold government
officials accountable for their actions than would free-floating majorities.
In sum, formalized, stable coalescence is a survival strategy – it provides parties with the
ability to influence government decisions, earn the trust of non-governmental actors, and
maintain good long-term relations with voters while paying relatively low transaction costs. To
the extent that such relations and cost savings are more important than the policy freedom that
free-floating coalitions could allow, stable coalitions will prevail, and indeed in west European
parliamentary democracies they do.
Having established why parties choose to coalesce, we now turn to the question of who
gets what in coalition bargaining. We begin with a simple scenario in which parties are certain
about the future. This simplification allows us to clarify important aspects of coalition
bargaining easily. We then move to more realistic settings where parties bargain in the shadow
of an uncertain future.
Walk-Away Values Affect Who Gets What
In the simplest economic bargaining models, everybody knows everything. When
applied to coalition governance, such models imply that every party leader knows everything
about the situation at hand -- such as the preferences, options, and constraints of every other
party leader – both at the time of negotiation and forever into the future. In such a world,
negotiators anticipate all consequences of their actions and bargainers can engage in complete
contracting. Complete contracts allow coalitions to specify in detail what each partner will do
in any circumstance. Of course, governing coalitions seldom know so much. However,
examining some properties of complete contracting is an efficient way to uncover key insights
about coalition governance. Consider, for example, the question of who gets what when
coalition members disagree.
In bargaining models, the answer to the question “Who gets what?” hinges on the
credibility of a bargainer’s negotiating positions. We call this factor a walk-away value. A
walk-away value is what a negotiator, such as a party leader, can gain without a new agreement;
i.e., what he or she secures by walking away from the bargaining table. Economists also use
the term reservation wage to describe this concept (see, e.g., McMillan 1992).
In a bargaining session, walk-away values function as implicit threats against other
parties. The more easily a party can walk away from the bargaining table and still end up in a
good situation, the more concerned the other parties have to be about satisfying that party’s
demands -- if they want that party in the coalition. Put another way, those who have little to lose
from disagreement often have much to gain from negotiation. In a market economy, walk-away
values are usually determined by the monetary values of other economic opportunities. For
example, a worker who can earn high wages from many employers has a higher walk-away
value than a similarly situated worker with no outside opportunities.
The source of walk-away values in parliamentary contexts often lies elsewhere. In
democracies, public support is the lynchpin of a political party’s power. Parties who lack
public support can make no promises that are of value to other potential coalition partners. But
while public support is necessary for a party to have a positive walk-away value in
coalition negotiations, it is not sufficient. For a party to have a positive walk-away value, it
must be able to benefit from its assets (including its public support) without tying itself to any
specific coalition agreement. For example, when one party has enough seats to govern on its
own, only it has a positive walk-away value (i.e., only it can govern without the help of others).
3 By walk-away value, we mean something related to, but different from, famous bargaining power indices such as
those of Banzhaf (1968) and Shapley-Shubik (1954). Our concept is, however, directly comparable to those two if
we assume zero transaction costs. In other words, the bargaining power indices account for the number of
alternative coalitions each bargainer can join, but under the assumption that such moves do not require costly
contracting. Since these indices are widely available, they will be used in some of the chapters that follow. When
When no party has resources (i.e., seats) sufficient to govern alone, walk-away values are
determined by which party combinations are capable of governing (i.e., which teams of parties
can control a number of seats sufficient to control parliamentary decisions). Parties that can be
members of more than one governing coalition can have increased walk-away values. For
example, party X, a member of the incumbent governing coalition, gets bargaining leverage
when other incumbent coalition members believe that failure to defer to it on a particular issue
will cause it to join a different majority. Indeed, if a party can maintain a credible exit threat
throughout a coalition’s term (i.e., if by defecting it can bring down a government and produce
a situation from which it benefits), then we expect it to have great political influence. The
religious bloc in the Israeli Knesset has such power – it can, and has, coalesced with each of the
two major parties (Labour and Likud) and it regularly gets its way on social policy -- though the
bloc never holds anywhere near a majority of seats.
Coalition decisions depend on how parties evaluate each other’s walk-away values.
Understanding this fact clarifies many aspects of coalition behavior – including the errors
inherent in the common belief that resources, such as number of seats held, transfer directly into
bargaining power. Many people believe that the larger a party is, the more powerful it is in
coalition negotiations. This belief is true only in cases where walk-away values and size are
positively and strongly correlated. In such cases, we expect allegiance to “parity norms” –
agreements where a coalition member’s share of portfolios corresponds to its size relative to
other coalition members. When walk-away value and size are less correlated, we expect parity
norms to be abandoned.
To see how size and walk-away value can have very little relation to one another,
consider a simple example. Three parties have seats in a 101-member parliament, which decides
by majority rule. Party A has 50 seats, Party B has 48 seats, and Party C has 3 seats. No party
has a majority. For policy reasons, all parties prefer certain coalitions to others. Suppose that all
parties know these preferences and all other important aspects of the bargaining situation.
Suppose further that any realized gains from coalescence are divisible among coalition
members, that all parties prefer a larger share of power in a coalition to a smaller share, and that
any party is free to negotiate with any other party. To simplify the example, we describe the
case where any coalition including C produces individual benefits for both coalition members,
they are used in conjunction with this chapter’s theoretical framework, they should be interpreted with this
footnote’s caveat in mind. 16
whereas a coalition including both A and B produces no such benefits. Party C may be a
centrist party whose members are not repulsed by the policy desires of other parties and vice
versa, while A and B have disagreements so fundamental that there is nothing that they that
could agree to do together.
The result of bargaining in this case (following Lupia and Strøm 1995) is as follows:
• C h a s t h e f
e s t s e a t s , b u t i s t h e o n l y o n e w
i t h a p o s i t i v e w
a l k - a w
a y v a l u e - - i t i s t h e o n l y
p a r t y t h a t c a n t h r e a t e n p o t e n t i a l s u i t o r s t h a t i t w
i l l s e e k t o b e p a r t o f a n o t h e r v a l u e -
g e n e r a t i n g c o a l i t i o n .
l t h o u g h A a n d B a r e m
u c h l a r g e r t h a n C
, t h e y h a v e n o p o s i t i v e w
a l k - a w
a y v a l u e s w
h e n
n e g o t i a t i n g w
i t h C
h e o n l y s u s t a i n a b l e o u t c o m
e i s a c o n t r a c t g i v i n g p a r t y C v i r t u a l l y a l l o f t h e b e n e f
i t s f
r o m
t h e g o v e r n i n g c o a l i t i o n .
In this simple example, size and bargaining power are not positively correlated. C, by
far the smallest party, is the most powerful. The lesson here is that walk-away values trump
size. Put another way, size is not always power. This example is also interesting because it
resembles West Germany in the 1970s and 1980s – where the Social Democrats and the
Christian Democrats were like parties A and B, and the Free Democrats were like party C.
Though smaller than the other two parties, the Free Democrats (FDP) wielded great bargaining
power. Our framework implies that the FDP should then have secured a disproportionate share
of cabinet portfolios, and indeed they did.
Walk-away values, however, do not affect coalition bargaining outcomes in isolation.
Politics often feature a variety of rules about who gets to bargain over what, and we broadly
refer to such rules as institutions. Institutions, which are the product of previous
bargaining, affect walk-away values. Institutions such as size or composition rules,
formateur/investiture rules (Huber and McCarty 2001), recognition rules (Bawn 1991, 1993),
the independence of the judiciary/civil service, cabinet operating procedures, and internal party
rules affect coalition bargaining by constraining the options of party leaders (Strøm, Budge, and
Laver 1994). For a few years in the 1990’s, for example, Israel’s Basic Law transferred the first
opportunity to form a coalition from the largest party in parliament to a separately elected Prime
Minister. This change affected party walk-away values and was sufficient to affect whether its
defense policies would be left or right of center (i.e., whether Labour or Likud would be in the
coalition). In Belgium, unique institutions have an analogous effect on party walk-away values.
There, the constitution requires linguistic parity in the cabinet and thus constrains coalitions to
contain both Flemish and Walloon parties. This constraint affects the set of coalition
agreements from which parties can credibly walk away. In sum, when institutional constraints
affect parties’ walk-away values, then they can affect who gets what in coalition bargaining.
Understanding the role of bargaining and, by implication, walk-away values, casts the
actions of coalitions in a new light. Under majority rule, size need not be power in coalition
governance. Unless a party earns more than one-half of the seats in parliament, its bargaining
power depends on walk-away values. The same logic implies that any institutional, resource-
based, preference-based, or country-specific factor affects coalition agreements only if it
affects a potential coalition partner’s walk-away value. In some nations, for example,
cultural taboos make certain coalition agreements untenable to voters and can make parties’
relative walk-away values different than they would otherwise be.
Another common belief about coalitions is that if parties share common preferences,
then they will necessarily coalesce. Focusing on the necessity of bargaining ultimately reveals
such conjectures to be false. For example, if the transaction costs of reaching agreement
with a particular partner are too high, then parties may cast agreements with that partner
aside – opting instead for an agreement entailing lower transaction costs with a partner
whose preferences are less common. Indeed, even between parties with otherwise common
interests, if there are too many items to negotiate or too many contingencies that might arise,
then we cannot simply assume that they will coalesce. As Fudenberg and Tirole (1991: 397)
“A bargaining situation involves players who must reach an agreement in order to realize
gains from trade. The standard example is the problem of sharing a pie. No player can
have any pie until they all agree about the shares each will receive. Negotiating about the
shares is costly, and the pie may decay or disappear if the negotiations go on for very
Such concerns certainly ring true in parliamentary democracies. Bargaining can take
considerable time. In the Netherlands, coalition formation negotiations often last at least three
months and in 1977 lasted for the better part of a year! In the meantime, decisions on major
political issues are suspended, the government is reduced to caretaker status, and parties do not
enjoy the perquisites of office. In such cases, there is no doubt that the “pie” decays as the
parliamentary term passes and the next election approaches.
Uncertainty and Discount Rates Make Patience a Virtue in Bargaining
High transaction costs are sufficient to derail an otherwise successful bargaining
session. Uncertainty is another factor with the same potential. An important source of
uncertainty in coalition politics is the two requirements of coalition politics (the need to form
and maintain agreements with other parties while simultaneously cultivating voter support).
The first of these requirements introduces uncertainty when party leaders lack information
about each other’s strategies. The second requirement introduces uncertainty if voters are fickle
or if their preferences are difficult to gauge.
We turn our attention now to coalition governance dynamics in cases where such
uncertainty is present. Coalition members may, for instance, lack information about what other
parties want or what sacrifices they are willing to make in exchange for coalition membership.
In such cases, parties may try to strengthen their bargaining position by overstating their walk-
away values (e.g., to insist that you will settle for nothing less than ministries X, Y, and Z,
when, in fact, you would be satisfied with two of the three). Therefore, a key to bargaining
under uncertainty is separating fact from fiction when assessing other parties’ walk-away
values. A complete account of what people believe about the walk-away values of others would
involve a fair amount of psychology. However, bargaining models built from the premises
described above produce important insights about these cognitions with less effort (see, e.g., the
review in Laffont and Tirole 1993). They show that credible commitment is the key to
bargaining under uncertainty.
While any party can claim a high walk-away value, such claims are more credible if
commitments back them. If, for example, a party leader can credibly commit to resigning his
post if a particular objective is not achieved, then his coalition partners may have a greater
incentive to give in (especially if they dislike his potential replacement). Thus, actors who can
+1 anno fa
Materiale didattico per il corso di Politica comparata del prof. Marco Giuliani. Trattasi del saggio di Arthur Lupia dal titolo "Bargaining, Transaction Costs, and Coalition Governance" all'interno del quale sono analizzati i rapporti di coalizione fra i partiti politici di governo e si fornisce un modello teorico per interpretarne le evoluzioni.
I contenuti di questa pagina costituiscono rielaborazioni personali del Publisher Atreyu di informazioni apprese con la frequenza delle lezioni di Politica comparata e studio autonomo di eventuali libri di riferimento in preparazione dell'esame finale o della tesi. Non devono intendersi come materiale ufficiale dell'università Milano - Unimi o del prof Giuliani Marco.
Acquista con carta o conto PayPal
Scarica il file tutte le volte che vuoi
Paga con un conto PayPal per usufruire della garanzia Soddisfatto o rimborsato