Self-interest and fairness in coalition formation: A social utility approach to understanding partner selection and payoff allocations in groups

Abstract

In this chapter we present a social utility approach to coalition formation. The central tenet of our approach is that outcome allocations and partner selection in multiparty situations are affected by self-interest and fairness. Inspired by the social utility model we argue that the relative weight assigned to both components is affected by structural aspects of the situation and individual characteristics of the negotiators. We first investigate how coalition bargainers substantiate their coalition demands. We show that bargainers are self-serving in their choice of allocation rules, indicating that perceptions of fairness can be coloured by self-interest. Second, we investigate how the alignment of self-interest and fairness fosters the formation of coalitions that maximise the payoffs of its members. Finally, we present a series of experiments that expands the notion of being fair to those who are excluded from a coalition. We show that bargainers are reluctant to benefit themselves when this harms the outcomes of others and that this is dependent on personal factors (e.g., social value orientations), situational factors (e.g., the valence of outcomes), and whether bargainers negotiate in an interindividual or in an intergroup setting.

Full text

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Self-interest and fairness in coalition formation:
A social utility approach to understanding partner
selection and payoffallocations in groups
Ilja Van Beest and Eric Van Dijk
Leiden University Institute for Psychological Research, The Netherlands
In this chapter we present a social utility approach to coalition formation. The
central tenet of our approach is that outcome allocations and partner selection
in multiparty situations are affected by self-interest and fairness. Inspired by the
social utility model we argue that the relative weight assigned to both
components is affected by structural aspects of the situation and individual
characteristics of the negotiators. We first investigate how coalition bargainers
substantiate their coalition demands. We show that bargainers are self-serving
in their choice of allocation rules, indicating that perceptions of fairness can be
coloured by self-interest. Second, we investigate how the alignment of self-
interest and fairness fosters the formation of coalitions that maximise the
payoffs of its members. Finally, we present a series of experiments that expands
the notion of being fair to those who are excluded from a coalition. We show
that bargainers are reluctant to benefit themselves when this harms the
outcomes of others and that this is dependent on personal factors (e.g., social
value orientations), situational factors (e.g., the valence of outcomes), and
whether bargainers negotiate in an interindividual or in an intergroup setting.
Keywords:Coalition formation; Fairness; Multiparty negotiation; Social
exclusion
How individuals reach an agreement in a situation of conflict is a core topic
in social psychological literature on bargaining and negotiation (Bazerman,
Curhan, Moore, & Valley, 2000; De Dreu & Carnevale, 2003; Lewicki,
Saunders, & Barry, 2006). Interestingly, this line of research has tended to
Correspondence should be addressed to Ilja Van Beest, Leiden University Institute for
Psychological Research, P.O. Box 9555, 2300 RB Leiden, The Netherlands.
E-mail: vanbeest@fsw.leidenuniv.nl
This research was supported by a Veni grant NWO-451-04-069 from the Netherlands
Organisation for Scientific Research awarded to Ilja Van Beest. We would like to thank Colette
van Laar for proofreading our manuscript.
EUROPEAN REVIEW OF SOCIAL PSYCHOLOGY
2007, 18, 132 – 174
!2007 European Association of Experimental Social Psychology
http://www.psypress.com/ersp DOI: 10.1080/10463280701595354

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
ignore small group situations in which more than two individuals are in
conflict. This lack of attention to small group conflict is unfortunate because
just introducing a third person to the negotiation table increases the
complexity and social dynamics of the situation dramatically. When two
individuals perceive divergent interests they may either reach an agreement
or not. When three or more individuals negotiate, an agreement does not
necessarily include all. Indeed, a major difference between bilateral and
multilateral negotiations is that the latter allow the formation of coalitions.
The question to answer in multilateral negotiations is thus not only whether
individuals reach an agreement but also who reaches an agreement.
The question of who is included and excluded from a negotiated
agreement is the focus of coalition research. Coalitions are common and
examples can be found at every level of human organisation. A trade union
is a coalition of workers who have joined forces to obtain better working
conditions. Mergers are coalitions of companies that hope to increase their
market share. Political parties are coalitions of individuals who have joined
forces to influence the direction of a nation. NATO is a coalition of nations
that have pooled their economic and military resources to resist potential
aggressive acts of other nations. Several useful definitions of coalition
formation have been given to capture the common thread underlying these
examples. Polzer, Mannix, and Neale (1998) define a coalition as ‘‘two or
more parties who agree to cooperate in order to obtain a mutually desired
outcome that satisfies the interests of the coalition rather than those of the
entire group in which it is embedded’’ (p. 42). This definition is based on
Kahan and Rapoport’s (1984a) and Komorita and Kravitz’s (1983)
assumption that coalitions are formed when individuals cannot obtain
desired outcomes alone. Thibaut and Kelley (1959) take a more general
perspective. They defined a coalition as ‘‘two or more persons who act
jointly to affect the outcomes of one or more other persons’’ (p. 205). In
addition, researchers have stressed that coalition formation should be
viewed as a mixed-motive interaction (Gamson, 1964, Komorita & Parks,
1995; Schelling, 1960). On the one hand, parties need to cooperate to secure
a coalition. On the other hand, parties are in conflict because they need to
divide a desired outcome.
Social psychological research using a rational choice approach to
understand coalition formation prospered in the 1970s. Inspired by game
theory (Luce & Raiffa, 1959; Von Neumann & Morgenstern, 1947) the
basic assumption of this research was that participants are primarily
motivated to maximise some positive outcome often conceptualised as
money (Komorita & Parks, 1995). Participants usually control a number of
resources (e.g., votes in a political convention paradigm) and a joint control
over a majority of these resources is needed to obtain money. The method to
study coalition behaviour in this approach has been to fit the predicted
SELF-INTEREST AND FAIRNESS IN COALITIONS 133

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
outcomes of various theories to the actual outcome of a coalition game.
Social psychological research using this classic approach to study coalition
formation has provided many useful insights (for a thorough review of this
approach to study coalition formation in social psychology see, e.g.,
Blumberg, 1994; Kahan & Rapoport, 1984b; Komorita, 1984; Komorita &
Parks, 1995; Murnighan, 1978; Wilke, 1985). Yet studies adopting this
fitting approach came to a standstill. We agree with Komorita and Parks
(1995) that one possible reason why this approach ended is the saturation
achieved and by the insufficiency of this method to explain the underlying
dynamics and processes of coalition behaviour. Regrettably, it also means
that key insights in human decision making that have been developed since
this rational approach came to a standstill have not been incorporated into
our understanding of coalition formation.
In the current chapter we provide an overview of our recent line of
research that takes a different approach to studying coalition formation.
Unlike early theorising, which set out to test theories in which individuals
are assumed to maximise their own outcomes, we set out to test a social
utility model in which negotiators are also concerned as to whether and how
their actions affect the outcome of others. Extending the social utility model
that is usually tested in bilateral situations, we present five research lines that
aimed to test how self-interest and fairness affect the basis, process, and
outcome of coalition bargaining.
THEORETICAL BACKGROUND
‘‘People care about the outcomes of others.’’ With this opening line,
Loewenstein, Thompson, and Bazerman (1989, p. 426) introduced their
seminal article on their social utility model of decision making in
interpersonal contexts. Building on insights derived from interdependence
theory (Kelley, 1979; Rusbult & Van Lange, 2003) and equity theory (e.g.
Walster, Walster & Berscheid, 1978), Loewenstein and colleagues set out to
investigate how people in dispute relations derive utility from increasing
their own outcomes and from increasing the outcomes of others. In a
nutshell, they suggested that in dispute and bargaining relations, people’s
evaluations and behaviours can best be understood by a combination of a
concern for one’s own outcomes (i.e., self-interest) and a concern for other’s
outcomes (i.e., fairness).
The social utility model agrees with other models. For example, the dual
concern model of bargaining (Pruitt & Rubin, 1986) also postulates two
kinds of concerns—other concern and own concern—and also assumes that
both concerns may independently be high or low. There is a difference in
emphasis, however, in that the social utility model is more focused on
distributional aspects of allocations in interdependence. In more formal
134 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
terms, the social utility model states that the utility people derive from
interpersonal allocations (i.e., allocations between self and others) comprises
the summed utility assigned to one’s own outcomes, and utility assigned to
the difference between one’s own and others’ outcomes. Research on two-
person bargaining has further incorporated this social utility notion by
suggesting that in bargaining, self-interest and fairness concerns may
independently determine and explain bargaining behaviour (see e.g.,
Handgraaf, Van Dijk, & De Cremer, 2003).
The reason we were inspired by this model is that it provides a more
encompassing and differentiated account of what motivates bargainers than
does a pure self-interested approach. Although we do not wish to discount
the importance of self-interest, we do believe that the theorising on coalition
formation is served best by moving beyond self-interest. In this endeavour it
should of course be realised that the social utility model is a conceptual
model. The fact that on the conceptual level self-interest and fairness can be
seen as independent and additive components does not imply that this
distinction is as easy to make on the behavioural level.
In the present chapter we draw attention to five issues that deal with the
interplay between the self-interest component and the fairness component of
the social utility model. The first issue concerns the fact that coalition
bargainers may have egocentric perceptions of fairness.Equitytheoryhas
argued that negotiators demand a share of the payoffs that is in keeping with
input (Adam, 1963; Homans, 1974; Walster et al., 1978). Yet it does not
specify what type of input negotiators are likely to use. As the literature on
egocentric perceptions of fairness has shown (e.g., Loewenstein, Issacharoff,
Camerer, & Babcock, 1993; Loewenstein & Moore, 2004; Messick & Sentis,
1979, 1983; Thompson & Loewenstein, 1992), people may strongly differ on
what they consider to be a relevant input. What constitutes a fair allocation
of payoffs is therefore not a given, especially when asymmetries exist between
the negotiators involved. As a first step towards understanding coalition
formation it is therefore crucial to assess what type of information is used to
substantiate coalition demands and when these demands are self-serving.
The second issue concerns the alignment of self-interest and fairness.Inthe
domain of two-person ultimatum bargaining—a domain that has been
greatly inspired by the social utility model (see, e.g., the overview by
Handgraaf et al., 2003)—it is generally assumed that fair offers can be made
out of a concern for self-interest. For example, when bargaining over the
distribution of 100 euros, allocators in ultimatum bargaining may offer an
equal split, not necessarily because they want to be fair, but often because
they do not expect that their opponent will accept an unequal offer. This
implies that the making of fair offers may be instrumental in furthering one’s
own outcomes. For coalition bargaining this raises an interesting issue.
Because the selection of partners is essential in coalition bargaining, and
SELF-INTEREST AND FAIRNESS IN COALITIONS 135

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
selection is steered by expectations regarding what others may use as a
standard for allocation, it raises the possibility that people may be especially
likely to select partners with whom they think a fair allocation will be
possible. The reasoning also implies that if fairness cannot be reached (i.e., if
self-interest and fairness are not aligned), coalition bargaining may become
more difficult. As we will demonstrate, the issue of alignment of self-interest
and fairness may not only determine with whom people want to join forces,
but also with how many people they want to join forces, and how difficult it
will be to reach agreement.
The third issue concerns the scope of the fairness component of the social
utility model. In the literature on coalition formation, standards for
allocation are only expected to encompass the individuals who are included
in the winning coalition. However, standards of fairness may also extend
beyond the included members. A concern for fairness may, for example,
include a concern for what happens to the excluded players. In particular,
people may feel that outcomes should also be allocated to members who are
not part of the coalition, or even strive to include all in the negotiated
agreement (i.e., strive for the grand coalition). In most studies on coalition
bargaining these possibilities are explicitly prohibited. But what happens
with the bargaining process when these are allowed, and when people can
decide to include all members and therefore strive for a more encompassing
implementation of fairness concerns? When do people care about what
happens to excluded players?
The fourth issue we address is that the weights assigned to the self-interest
and fairness component of the social utility model may not be constant, but
may vary as a function of the situation at hand. In particular, we draw
attention to the fact that the differential weights of the model may be
dependent on the valence of the coalition outcomes.Itmaybenotedthat,
almost without exception, research on coalition bargaining has addressed
how coalitions are formed when positive outcomes are at stake. Yet
coalitions often have to deal with negative outcomes, and sometimes such
coalitions are even explicitly formed with this in mind (e.g., when parties
realise that they cannot bear the costs of a project all by themselves).
Inspired by Kahneman and Tversky’s (1979) prospect theory, decision-
making research has repeatedly shown that decisions in the negative domain
strongly differ from decisions in the positive domain. In an interpersonal
situation like coalition bargaining, it is not immediately clear how these
insights apply. Consider for example the concept of loss aversion. Loss
aversion refers to the observation that losses loom larger than gains—that is,
that losses are weighted more heavily than gains. Although in individual
contexts this notion has received substantive support (but see Harinck, Van
Dijk, Van Beest, & Mersmann, in press), one might wonder how this would
affect the components of the social utility model in coalition bargaining.
136 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
On the one hand, moving from the gain to the loss domain could result in
more self-interested behaviour when the effect runs through the weight of
the self-interest component. On the other hand, if it impacts the weight of
the fairness component it could also result in a process that is more
characterised by fairness concerns when negative outcomes are at stake.
The fifth and final issue we address is that the weights of self-interest and
fairness may be affected by the size and composition of the parties involved.
Although much of the initial theorising on coalition formation was inspired
by the political example of government formation, social psychological
theories on coalition formation have taken an individual perspective by
examining how individuals bargain to join forces. Whereas this individual
focus facilitates the translation and adaptation of insights on two-person
bargaining to coalition bargaining, it also provides a rather incomplete
picture of the more complex situation where parties comprise more
individuals. Insights on the interindividual – intergroup discontinuity effect
(for reviews see Schopler & Insko, 1992; Wildschut, Pinter, Vevea, Insko, &
Schopler, 2003) suggest that groups are more self-interested than
individuals. In terms of the social utility model this implies that the weight
of the self-interest component in coalition bargaining can increase when the
parties are groups rather than individuals, and thus for example that the
willingness to exclude others would increase as well. As we will show,
however, this analysis may fail because it does not acknowledge the
assumption underlying the social utility model—that is, that people’s goals
may differ and that people may also value fairness. In particular, we will
show that groups that comprise people who value fairness (i.e., prosocials,
see e.g., Van Lange, De Cremer, Van Dijk, & Van Vugt, 2007) may remain
motivated to include all.
1. EGOCENTRIC INTERPRETATIONS OF
STANDARDS FOR ALLOCATION IN COALITION
FORMATION
Although self-interest and fairness can be distinguished conceptually as
comprising two distinct parameters of the social utility function, the
literature on egocentric fairness suggests that the two can be closely linked
and that people may suffer from a biased perception of what is fair. When
multiple standards of fairness may apply, people are likely to prefer the
standard that best suits their own interests. This bias does not have to be the
result of conscious strategic behaviour in which people deliberately advocate
the rule that suits them best. The biased perception will often be the result
of unintentional and unconscious processes, such that people truly feel
that their coloured view of reality is correct and in fact unbiased (see
e.g., Loewenstein et al., 1993).
SELF-INTEREST AND FAIRNESS IN COALITIONS 137

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Biased interpretations of reality are usually the result of some sort of
ambiguity. For the context of coalition bargaining, two distinct features
may provide such ambiguity. The first feature concerns the presence of
multiple standards; the second—and related—feature concerns the presence
of asymmetry. Research on two-person bargaining has indicated that
egocentric interpretations of fairness become larger in situations of
ambiguity, and that asymmetry breeds ambiguity. As Babcock and
Loewenstein (1997, p. 120) put it: ‘‘As soon as asymmetries are introduced
between parties (. . .) both parties’ notions of fairness will tend to gravitate
toward settlements that favor themselves. They will not only view these
settlements as fair, but believe that their personal conception of fairness is
impartial.’’ It seems plausible that these insights may also apply to coalition
bargaining.
Previous research on payoffallocations in coalition formation has yielded
three relevant findings. One finding is that coalition bargainers may base
their demands on the resources that they contribute to a coalition. In this
line of research it is often observed that participants demand a payoffshare
that is directly proportional to the number of resources they possess (e.g.,
Gamson, 1961; Komorita & Chertkoff, 1973; Kravitz, 1981; Miller, 1980;
Miller & Wong, 1986). For example, political parties may base their
demands on the number of voters they represent. A second finding is that
participants may base their demands on power (e.g., Chertkoff& Braden,
1974; Gamson, 1964; Kelley & Arrowood, 1960; Komorita, 1974; Komorita
& Moore, 1976). A central aspect of power is that the power holder has the
capacity to influence the outcome of others (French & Raven, 1959; Lawler,
1992; Thibaut & Kelley, 1959), which in coalition formation is often
conceptualised as the a priori chance that participants are included in a
winning coalition (Shaply, 1953). Participants who can be a member of
many coalitions (i.e., who have many alternatives to form a coalition) have
more chances of obtaining payoffs than participants who can only be a
member of few winning coalitions (i.e., who have few alternatives to form a
coalition). Participants with an advantage in alternatives should therefore be
able to secure a better agreement than participants with a disadvantage in
alternatives. A third finding is that not all members of a coalition adhere to a
proportional allocation of the payoffs. Research has shown that participants
are self-serving in their choice of allocation rule. Participants only adhere to
aproportionalallocationwhenthismaximisestheirpayoffshare.For
example, in one of the most influential coalition theories—bargaining theory
(Komorita & Chertkoff, 1973)—it is argued that participants will demand a
proportional payoffshare when they have an advantage in resources, but an
equal payoffshare when they have a disadvantage in resources.
Unfortunately, previous research has almost exclusively focused on
coalition situations in which resources also determine power. A greater
138 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
number of resources often went hand in hand with a greater number of
alternatives, leaving unanswered how both factors affect coalition beha-
viour independently. We are, of course, not the first to note this relation
between resources and power (e.g., Komorita & Nagao, 1983; Kravitz, 1981;
Kravitz & Iwaniszek, 1984; Michener, Fleishman, Vaske, & Statza, 1975).
For some researchers it was indeed the reason why such situations were
studied in the first place. Many real-life situations are characterised by this
co-dependency of resources and power. For example, political parties that
represent many voters have by definition more opportunities to form a
majority coalition than political parties that represent fewer voters.
Moreover, the fact that resources and power are correlated does not mean
that predictions that are based on either resources or power are
interchangeable. It only means that predictions that are based on resources
only, or on power only, often point in the same direction.
Yet we believe that it is equally important to acknowledge that not all
real-life coalition situations are necessarily characterised by this co-
dependency. In many real-life coalition situations resources and power are
less strongly connected. To continue with the original example of
government formation, it may be noted, for instance, that some coalitions
cannot be realised because of ideological differences. Although the
distribution of votes between the political parties determines which
coalitions represent a majority of the votes, it is the position of the political
parties in the ideological landscape that determines whether coalitions are
viable. Irrespective of the number of votes, parties on the extreme ends of a
political continuum may find it harder to reach an agreement than parties
that are more in the centre of a political continuum. Teasing apart the effects
of resources and power may therefore also contribute to a better
understanding of such real-life situations.
To study the separate effects of resources and power on coalition
behaviour we developed the ‘‘landowner paradigm’’ (see Van Beest, Van
Dijk, & Wilke, 2004a). The essence of this paradigm is that each participant
owns a parcel of land. This parcel is characterised by its size and position
relative to the parcels that are owned by the other participants. After the
parcel assignment, participants are told that a project developer wants to
buy adjacent parcels that total a specific minimum of hectares. Thus, the
participants learn that both the size and the position of a parcel are
important factors to meet the offer of the project developer. The size of a
parcel is important because the project developer is only interested in buying
parcels that at least a specific minimum of hectares. The position of a parcel
is important because the project developer is only interested in buying
parcels that are adjacent. Thus, owners of parcels that are not adjacent
cannot form a coalition even though the combined parcels may satisfy the
demand in terms of size. As a result, we were able to generate all
SELF-INTEREST AND FAIRNESS IN COALITIONS 139

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
combinations of resources and power that are of interest (with power being
operationalised as the number of alternative winning coalitions of which a
participant can be a member). For example, a participant can have a big
parcel that is not adjacent to many other parcels (i.e., advantage in resources
but a disadvantage in alternatives), or a small parcel that is adjacent to many
other parcels (i.e., disadvantage in resources but an advantage in
alternatives).
In a first test of the landowner paradigm we focused on the opening offers
of participants (Van Beest et al., 2004a, Experiment 1). Participants were led
to believe that they would negotiate with three other participants and
assigned to the conditions of a 3 (resources: advantage, equal, disadvan-
tage) 63(alternatives:advantage,equal,disadvantage)design.Thecondi-
tion was described using one of the pictures that are presented in Figure 1.
Participants were always assigned to the position of player A, and we tested
with whom they wanted to form a coalition and how they wanted to
allocated the payoffs. They were instructed that only a coalition of three
parcels could satisfy the demands of the project developer and that these
parcels needed to be adjacent. This allowed us to investigate whether
participants base their payoffpreference on both resources and alternatives
or on one of these exclusively. In addition, it allowed us to test whether
participants are self-serving when they make an opening offer (Komorita,
1979; Komorita & Chertkoff, 1973; Messick & Sentis, 1979, 1983).
1
Analyses of coalition preferences yielded several interesting findings. One
finding was that participants chose to minimise the total size of parcels that
were offered to the project developer. This is consistent with a basic axiom of
coalition theory that one should not offer more than necessary (we consider
this aspect in more detail in the next section on the scope of fairness). The
more relevant finding—for the current purpose—concerned the way
participants allocated the payoffs among the members of a preferred
coalition. As can be seen in Figure 2, results yielded a main effect of
resources, a main effect of power, and no interaction effect of resources and
1
In all studies participants were informed that their experimental pay was based on
bargaining performance. In many experiments we instructed participants that obtaining a good
bargaining outcome maximised their chances of winning 50 euros in a lottery that would be held
after the experiment was completed (Van Beest, Andeweg, Koning, & Van Lange, in press;
Van Beest et al., 2004a, Experiment 2; Van Beest et al., 2004b, Experiment 1 to 3; Van Beest,
Wilke, & Van Dijk, 2003, Experiment 2). In some studies we did not use a lottery, and after the
experiment ended we informed participants that they would get a flat fee instead (Van Beest,
Van Dijk, De Dreu, & Wilke, 2005, Experiment 1; Van Beest et al., 2004a, Experiment 1; Van
Beest et al., 2003, Experiment 1). Finally in Experiment 2 of Van Beest et al. (2005) participants
got exactly what they obtained during bargaining. Given that these differences in incentive
schemes did not alter the way participants negotiated, we feel safe in assuming that participants
do not bargain differently when they bargain about points that are translated into chances of
winning a bonus, or when they bargain about real money.
140 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
power. This shows that both factors have an independent effect on payoff
allocations in coalition bargaining.
In addition, we analysed whether the allocations were self-serving. Based
on Messick and Sentis (1979, 1983) we operationalised self-serving
behaviour as the difference between what individuals demand in a specific
bargaining position and what individuals would offer to that specific
bargaining position. Using this method we observed that all participants
were indeed self-serving. For example, participants with a disadvantage in
resources and alternatives (i.e., what player A gives to him/herself in
condition 1) demanded a larger payoffshare than they were willing to give to
this position (i.e., what A gives to player B in condition 9). The more
important finding was that this varied as a function of resources and
alternatives. As can be seen in Figure 3, participants showed stronger self-
serving tendencies as a function of alternatives than as a function of
resources.
In a subsequent study (Van Beest et al., 2004a, Experiment 2) participants
negotiated with other participants. This allowed us to further investigate the
bargaining process. Research on bilateral situations has shown that
bargaining length—e.g. the number of rounds needed to reach an
agreement—is dependent on the initial disagreement about payoffallocation
(Carnevale & Pruitt, 1992; Druckman, 1994; Rubin, Pruitt, & Kim, 1994).
Moreover, research on asymmetries in bargaining has indicated that
Figure 1. Graphical picture of the different Landowner games in which player A has relatively
fewer, equal, or more resources and/or alternatives than some other players (adapted from Van
Beest et al., 2004a, Experiment 1).
SELF-INTEREST AND FAIRNESS IN COALITIONS 141

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Figure 3. Self-serving behaviour as measured by differences in percentage of payoffs demanded
by and offered to a specific bargaining position in an unequal resources and unequal alternatives
game (adapted from Van Beest et al., 2004a, Experiment 1).
Figure 2. Percentage of payoffdemands as a function of resources and alternatives (adapted
from Van Beest et al., 2004a, Experiment 1).
142 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
egocentric judgements would lead to longer bargaining (Babcock,
Loewenstein, Issacharoff, & Camerer, 1995; Loewenstein et al., 1993).
Given that self-serving behaviour was more apparent as a function of
alternatives than as a function of resources, we therefore predicted that
power differences between individuals would lead to longer bargaining, as
compared to resources differences.
Depending on condition, participants were either player A, B, C, or D
and assigned to the conditions of a 2 (alternatives: equal, unequal) 62
(resources: equal, unequal) design. We again explained the situation by
means of a graphical representation (see Figure 4). Results on opening
demands and obtained payoffs replicated the findings of the first study.
Again we observed only main effects and no interaction effects. Both payoff
allocations in the opening offer and payoffallocations in the final agreement
were based on a combination of resources and alternatives. This provides
further confidence that both factors have an independent effect on payoff
allocation. In addition, we found that resources and alternatives had a
different impact on the bargaining process. Differences in the position of a
parcel (alternatives) led to more rounds of bargaining, whereas differences in
the size of a parcel (resources) led to more concessions. This is consistent
with the observation of our first study that self-serving allocation are based
more on differences in alternatives than on differences in resources, and
shows that self-serving allocations may cause more conflict during
bargaining.
Figure 4. Graphical picture of the landowner games in which resources and alternative are either
equal or unequal (adapted from Van Beest et al., 2004a, Experiment 2).
SELF-INTEREST AND FAIRNESS IN COALITIONS 143

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
The combined results of these studies thus showed that resources and
power (operationalised by alternatives) have an independent effect on
allocation preferences, allocation outcomes, and even the duration of a
negotiation. This can be contrasted with previous insights suggesting that
reward allocation can be explained solely in terms of resources (e.g.,
minimum resource theory, Gamson, 1964) or solely in terms of power
(e.g., minimum power theory, Gamson, 1964; the weighted probability
model, Komorita, 1974).
2
It also helps to advance one of the most
influential social psychological theories in coalition formation. Bargaining
theory (Komorita & Chertkoff, 1973) assumes that individuals with an
advantage in resources will demand a share of the reward that is
based on resources, whereas individuals with a disadvantage in resources
will demand an equal share of the reward. Note, however, that
bargaining theory was usually tested in situations in which resources
and alternatives went hand in hand. The present results show that
when both factors are teased apart, individuals will only demand an
equal allocation when they have a disadvantage in power. In terms of the
social utility model, these findings corroborate the notion that the self-
interest and fairness components are closely related in coalition
bargaining. Biased perceptions, especially in situations with asymmetric
power distributions may lead to biased judgements and result in more
conflict.
2
Minimum resource theory (Gamson, 1961) is based on two assumptions. The first
assumption is that players want to maximise their share of the reward. The second assumption
is that players will allocate the reward in proportion to resources. Combining both assumptions,
the theory predicts the formation of a coalition that minimises the total number of resources of
its members, because it maximises the reward of its members. Minimum power theory
(Gamson, 1964) is based on an extension of Shapley and Shubik’s (1954) index of pivotal power.
A player’s pivotal power is defined as the relative frequency with which his or her resources,
when added to the resources of other players, convert a non-winning coalition (a non-winning
coalition is a coalition that does not control at least a majority of the resources) into a winning
coalition. This measure equals P/N!, where N! is the total number of permutations (all possible
orders of entry of players into a coalition) and P is the number of permutations in which a
player is pivotal. Similar to minimum resource theory, players are assumed to maximise their
share of the reward, but this time it is assumed that players allocate the reward in proportion to
their pivotal power. As a result, the coalition that minimises the sum of pivotal power indices is
most likely to form. The weighted probability model is a refinement of minimum power
(Komorita, 1974). The basic assumption of the weighted probability model is that large
coalitions (in terms of number of players) are more difficult to form than small coalitions. The
theory assumes that the probability of a coalition is an inverse function of its size, and
predictions of reward allocation among the coalition members are based on the probability of
an individual being included in a coalition. A player’s probability of being included is equal to
the number of times he or she can be a member of a coalition.
144 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
2. THE ALIGNMENT OF SELF-INTEREST AND
FAIRNESS
The second aspect of social utility model that we set out to investigate
pertains to the alignment of self-interest and fairness in coalition formation.
In the previous section we discussed how self-interest and fairness are used
to substantiate allocation demands. In this section we discuss how both
aspects of the utility model are used to select a coalition partner.
This analysis was inspired by Kahan and Rapoport’s (1984a) observation
that rational choice theories of coalition formation are based on a similar set
of assumptions. The first assumption is that participants want to maximise
their own payoff. To stress this assumption, participants are often explicitly
told to obtain as much of the payoffs as possible. The second assumption is
that participants use fairness norms to allocate rewards in a coalition.
Consistent with insights that are based on equity theory (Walster et al.,
1978) and distributive justice (Homans, 1974), it is assumed that participants
expect a share of the payoffs that is in direct proportion to some source of
entitlement, such as power, resources, or number of people in a coalition.
Finally, it is assumed that both these aspects of coalition formation are
intertwined. Every participant is expected to estimate on the basis of equity
what he or she is likely to obtain in all possible coalitions, and based on
these expectancies each participant is expected to prefer a coalition that
renders him or her the largest payoffs. Although coalition theories may
differ on what information participants use to estimate their prospective
coalition payoff, they all agree that the predicted outcome of a coalition
situation is a coalition that maximises the payoffof its members and
provides each member with an equitable payoffshare.
So why do individuals use fairness rules? The bargaining literature
suggests that—apart from a true concern for fairness—one of the main
advantages of norms in bargaining situations is that they may serve as a
prominent solution for the conflict that bargainers face (Pruitt & Carnevale,
1993; Schelling, 1960). Equity concerns facilitate agreements between
negotiators. Interestingly, for the field of coalition formation, this may
imply that bargainers will seek to negotiate with coalition partners with
whom they will be able to reach an equitable distribution of the coalition
payoff. Put differently, the outcome of coalition formation may strongly
depend on the possibility to obtain an equitable share of the coalition
payoff.
An interesting question to ask, then, is what happens when negotiators
cannot align self-interest and fairness. To our knowledge, this issue has not
been the focus of research before. That is, previous research on coalition
formation has concentrated on situations in which it is possible to satisfy
both motivations but has not investigated situations in which coalition
SELF-INTEREST AND FAIRNESS IN COALITIONS 145

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
members may need to make trade-offs between maximising one’s payoffand
getting an equitable payoffshare. Or, more specifically, previous research
has focused on situations in which some players can tempt other players into
joining forces by offering them the opportunity to simultaneously maximise
their payoffshare and obtain an equitable payoff. This leaves unanswered
whether that specific coalition is formed because of self-interest, equity, or a
combination of the two. To understand why both these factors are
important one has to study situations in which it is more difficult to
simultaneously satisfy self-interest and fairness. Moreover, many real-life
situations do not enable a joint pursuit of self-interest and fairness. For
example, political parties that negotiate about the allocation of seats in
cabinet may learn that some proportional allocations are not possible
because of the available number of seats.
3
Disentangling self-interest and
fairness may thus also help to understand such real-life situations.
The above observations resulted in a series of studies that investigated the
alignment of self-interest and fairness in coalition formation (see Van Beest
et al., 2004b). In all these experiments we investigated this relation by
manipulating the divisibility of the reward (see Allison & Messick, 1990;
Allison, McQueen, & Schaerfl, 1992; Mclean Parks et al., 1996; Nitz &
Phillips, 1969; Young, 1995). We reasoned that if equity concerns determine
which coalitions will be formed, then manipulating the number of payoff
units would facilitate the formation of some coalitions, yet inhibit the
formation of others.
In a first study we tested our theoretical model in the classic 5(4 3 2)
setting. In this setting there are three players who are labelled by capital
letter in descending order of the number of resources that they possess. The
combined total of all resources is nine and a total of five resources is needed
to form a majority coalition. In terms of the landowner paradigm it
describes a situation in which participant A has a parcel of four acres, player
Baparcelofthreeacres,andplayerCaparceloftwoacres.Thus,
participants learn that every two-player coalition (AB, AC, or BC coalition)
and the grand three-player coalition (ABC coalition) is able to meet the
demands of the project developer.
We started our investigations in this specific situation because it is one of
the most investigated situations in coalition research (for an overview of this
specific coalition situation, see Murnighan, 1991). This enabled us to predict
what coalition will be formed when participants are able to simultaneously
3
Amorespecificexampleofsuchanallocationproblemisthe1998generalelectionsinthe
Netherlands. In this year three parties wanted to form a government but learned that the
number of available seats in cabinet was not enough to satisfy the demands of all the parties.
After some haggling, the smallest party threatened to break up the negotiation if they did not
get the number of seats they felt they were entitled to. In the end the problem was solved by
creating a new seat in cabinet.
146 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
satisfy self-interest and fairness. Previous research has repeatedly shown that
when participants can align self-interest and fairness they are likely to form
coalitions that minimise the total number of resources. This is called the
strength-is-weakness effect (Murnighan, 1991; Vinacke & Arkoff, 1957). If
all participants demand a proportional payoffshare, then each is better off
in a coalition with a weak partner than in a coalition with a strong partner.
In the 5(4 3 2) setting, players B and C realise that they obtain a greater
payoffshare in a BC coalition than in a coalition with A. Consequently, the
outcome of this setting is likely to be a BC coalition in which player B
obtains 60% of the payoffs and player C obtains 40% of the payoffs.
This is an important finding because it shows that specific players benefit
when the alignment of self-interest and fairness is not constrained. In the
5(4 3 2) setting it is often observed that both player B and C maximise their
payoffby using an allocation norm that is based on resources, but what if
they were not able to obtain an equitable payoffshare in the BC coalition?
Will the outcome still be a coalition that minimises the total number of
resources? Or will it enhance the likelihood that other coalitions will be
formed? In this respect it is useful to consider that the power distribution
underlying this example would yield another prediction. That is, in the 5(4 3
2) game, every participant can be a member of two minimal winning
coalitions (i.e., participants have an equal number of alternatives to form a
coalition). If participants were to use this power distribution, then every
minimal winning coalition should be as likely to form and payoffs should be
equally divided among the members. For example, Kelley and Arrowood
(1960) demonstrated that after repeated play, each two-player coalition in
the 5(4 3 2) setting becomes equally likely, showing that participants may
also base their demands on power.
To study how the alignment of self-interest and fairness affects partner
selection in this classic coalition situation we made the reward available in
different units (Van Beest et al., 2004b, Experiment 1). In the control
condition the reward was paid in five units. This enabled participants to
divide the reward in proportion to resources in a BC coalition. In the
experimental condition the reward was paid in three units. This inhibited the
possibility of dividing the reward in proportion to resources in a BC
coalition. Results showed that such a simple manipulation had a huge
impact on formed coalitions and thus on the size of the parcel that was
offered to the project developer (see Figure 5). In the control condition we
replicated the strength-is-weakness effect. Consistent with earlier research
using relatively naı
¨ve coalition bargainers (i.e., the typical psychology
student who does not have a strong education in game theory), the outcome
was most often a BC coalition in which B got three reward units and player
C two reward units. However, when the reward was made available in three
units, participants did not minimise the size of the parcel that was offered to
SELF-INTEREST AND FAIRNESS IN COALITIONS 147

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
the project developer, nor did every two-player coalition become equally
likely. Instead, participants continued to use a rule that was based on
resources. In the three-unit condition the outcome was most often an AC
coalition. This shows that participants are willing to offer six acres of land to
the project developer, while he only needed five acres. Furthermore, in the
AC coalition, A got two reward units and C one reward unit.
These results warranted two conclusions. The first conclusion was that
players base their allocations on resources. For example, when a BC
coalition was formed, player B obtained a larger payoffshare than player C.
Moreover, when the reward was paid in three units, participants did not
form grand coalition in which payoffs were allocated in direct proportion to
power (i.e., each participant got one reward unit). Instead, they formed AC
coalitions in which payoffs were allocated in direct proportion to resources.
The second, more important, conclusion is that the alignment of self-interest
and fairness affected partner selection. As predicted, the number in which
the payoffwas made available affected the composition of a coalition
because participants seemed to strive for a coalition that enabled its
members to obtain a fair payoffshare.
In Studies 2 and 3 we went a step further. The general aim of these studies
was to investigate how the alignment of self-interest and fairness may induce
players to include unnecessary members. The set-up of these studies was a
Figure 5. Percentage of formed BC coalitions and AC coalitions in a 5(4 3 2) game as a function
of units in which the reward is made available (adapted from Van Beest et al., 2003b,
Experiment 1).
148 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
2(1 1 1) setting. The reason for choosing this setting lies in its simplicity. The
players are equal on all possible level of comparison and it is therefore not
possible to distinguish between explanations based on power or resources.
Nor is it possible to distinguish equality from proportionality. In this setting
the only equitable allocation is one in which all coalition members obtain
the same payoffshare. The main distinction that can be made in this setting
is whether players will form a small coalition that excludes one of the players
or a grand coalition that includes all.
The payoffs were made available in three, six, or nine units. This
manipulation did not affect the possibility of allocating the payoffequitably
in the grand coalition. In the three-units condition each member could
obtain one unit; in the six-units condition each member could obtain two
units; and in the nine-units condition each member could obtain three units.
However, the manipulation did affect whether it was possible to allocate the
payoffequitably in a small coalition. In the six-units condition it was
possible to obtain an equitable payoffshare in a small coalition. In both the
three- and the nine-units conditions it was not possible to obtain an
equitable payoffshare in a small coalition. The question to answer is thus
whether participants will form unnecessary large coalitions when the reward
is paid in three or in nine units. For example, when the reward is paid in nine
units, will participants rather obtain four or five reward units in a small
coalition or three units in a grand coalition?
As predicted, the alignment of self-interest and fairness affected the size
of the coalition formed. When the reward was paid in six units, individuals
formed small two-player coalition (80%) in which each member obtained
three reward units. This shows that when members of small coalitions could
allocate the reward in proportion to their source of entitlement (i.e., could
obtain a fair share of the reward), small coalitions were formed more often
than large coalitions. However, when the number of units in which the
reward was paid did not enable members of small coalitions to obtain a fair
share of the reward, large coalitions (in which it was possible to obtain a fair
share of the reward) were formed instead. For example, when the reward
was paid in three units, almost no small two-player coalitions were formed
(23%). Instead, participants mostly ended up forming grand three-player
coalitions in which each participant obtained one reward unit.
Asecondfindingwasthatthealignmentofself-interestandfairness
affected the negotiation process (Van Beest et al., 2004a, Experiment 3).
Results revealed that it took longer to form a coalition when the reward was
paid in a number of units that did not allow members of small coalitions to
allocate fairly. For example, it took longer to form a coalition when the
reward was paid in nine units than when it was paid in six units. This
demonstrates that when prospective members of a small coalition cannot
obtain a fair share of the reward, the negotiation process becomes more
SELF-INTEREST AND FAIRNESS IN COALITIONS 149

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
problematic. It shows that fairness only facilitates the formation of an
agreement when this agreement is also in the best interest of the participants.
The above findings show that the number of units in which the reward is
paid may affect coalition behaviour. The results even suggest that the reward
in itself may be used to predict coalition behaviour. This approach differs from
other approaches in which differences between individuals have been used to
predict coalition behaviour (e.g., the resource versus alternatives experiments
presented above). Or, in terms of equity theory (Adams, 1963; Homans, 1974;
Walster et al., 1978), former research varied the input side of the equity
equation, whereas the divisibility experiments varied the output of the equity
equation. The reward affected the way individuals negotiated because it
constrained or facilitated the possibility to form a specific coalition.
To conclude, the results of the alignment studies showed that the number
of units in which the reward was paid moderated the outcome of a coalition
setting. This lends support to the notion of the social utility model that
fairness and self-interest are indeed important in coalition formation. When
it is possible to allocate fairly in a small coalition, small coalitions are
formed without any difficulty. However, when it is not possible to
simultaneously maximise one’s share of the reward by obtaining a fair
share of the reward (i.e., be a member of a fair small coalition), the
negotiation process becomes more problematic, and larger coalitions in
which it is possible to obtain a fair share of the reward are formed instead.
3. THE SCOPE OF FAIRNESS
We have already mentioned that previous theorising on coalition formation
has its roots in game theory. The essence of this approach is that individuals
aim to maximise their own payoffs. Evidence in favour of this prediction can
be found in many experiments (e.g., Gamson, 1964; Komorita, 1974:
Komorita & Meek, 1978; Michener et al., 1975; Willis, 1962) and to some
extent also in the experiments that were discussed in the above sections of
this chapter. The results of these prior experiments emphasise that
individuals want to maximise their own outcome, and seem to illustrate
that individuals do not hesitate to exclude other individuals in order to
achieve this goal. We believe, however, that the above findings may in part
be attributed to the payoffstructure of many coalition experiments. In most
coalition experiments, members of a coalition obtain a share of the reward
and excluded individuals do not. Or, more specifically, members of a
coalition increase their payoffand excluded individuals do not. Participants
enter the laboratory and after a negotiation some participants go home with
againandothersgohomewithout again.
We concur that some real-life situation are characterised by this simple
structure in which only coalition members benefit. Yet we also want to stress
150 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
that most real-life situations are more complex. In many situations the
excluded player’s payoffis affected beyond the fact that he or she does not
obtain a share of the reward. In some situations, the formation of a coalition
may decrease the excluded player’s payoff. For example, the creation of a
free-trade area may hurt countries that are left out. In other situations, the
formation of a coalition may increase the excluded player’s payoff. When
two companies combine forces to market a product, other companies that
sell a similar product may benefit. Moreover, political parties that form
governments may sometimes implement policies which also benefit the
political parties that are not a member of the government. Hence, to provide
a more balanced account of how self-interest and fairness affect coalition
bargaining, one should also investigate coalition situations in which
excluded individuals obtain more or obtain less than they initially started
out with.
To test the scope of the fairness component of the social utility model in
the context of coalition formation we conducted two experiments (see Van
Beest et al., 2003). The theoretical perspective of these experiments was
based on a combination of research on the ‘‘do-no-harm principle’’ (Baron,
1993, 1994), and social exclusion (Baumeister & Leary, 1994; Williams,
2007). According to the ‘‘do-no-harm’’ principle, individuals are reluctant to
harm members of one group in order to favour another group. According to
research on social exclusion (e.g., Leary, 1990; Van Beest & Williams, 2006;
Williams, Cheung, & Choi, 2000), the act of exclusion and being excluded
are associated with negative feelings. Based on these insights we reasoned
that the excluded player’s payoffmay affect partner selection because
individuals may perceive exclusion as harmful. In terms of the social utility
model, we expected that the weight of the fairness component of the model
would increase when the formation of a coalition decreased the outcomes of
the excluded players.
4
In a first study (Van Beest et al., 2003, Experiment 1) we tested our
reasoning in a scenario of a three-player landowner game. Participants
learned that each player owned a parcel of one acre and that a project
developer was interested in buying at least two acres for a fixed prize of
24,000 euros. Next participants were informed that forming a small coalition
in which only two acres was offered to the project developer would affect the
value of the unsold parcel. Depending on condition, participants learned
that an unsold parcel would decrease in value by 50%, maintain the same
4
We are aware that coalitions are sometimes formed with the explicit purpose of reducing the
outcomes of others. For insights on these situations we refer the reader to research on
revolutionary coalitions that investigates how low-status members may form revolutionary
coalitions to hurt the payoffs of high-status members (e.g., Lawler, 1995; Michener & Lawler,
1971).
SELF-INTEREST AND FAIRNESS IN COALITIONS 151

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
value, or increase in value by 50%. More specifically, the initial value of
each parcel was 6000 euros in the decrease condition, 3000 euros in the equal
condition, and 2000 euros in the increase conditions. And the final value of
an unsold parcel was 3000 euros in all conditions. Finally, participants read
that they had to choose between forming a small coalition and forming a
grand coalition. In the small coalition it was made clear that they would
then earn 12,000 euros. In the grand coalition it was made clear that they
would then earn 8000 euros. Hence it was more beneficial to be a member of
a small coalition than to be a member of a grand coalition, which was in
turn more beneficial than being excluded.
Results corroborated our reasoning that participants care about what
happens to excluded players (see Figure 6). As expected, participants did not
hesitate to maximise their own payoffs when the value of an unsold parcel
was not or was positively affected. In these conditions the majority of the
participants chose to form a small coalition in which they obtained 12,000
euros. However, they did lower their own payoffs when the value of an
unsold parcel decreased. Instead of choosing the self-benefiting small
coalition, they rather formed a grand coalition that provided them a payoff
of 8000 euros.
In a second study (Van Beest et al., 2003, Experiment 2) we went a step
further and argued that not all individuals might be equally sensitive to what
happens to an excluded individual. Inspired by research on social value
Figure 6. Percentage of formed small and grand coalitions in a 2(1 1 1) game as a function of the
value of the unsold parcel (adapted from Van Beest et al., 2003, Experiment 1).
152 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
orientations (Messick & McClintock, 1968; Van Lange, 2000), we assumed
that participants with a prosocial orientation (i.e., individuals who are
motivated to maximise the gains of self and other, and are motivated to
minimise the differences between the gains of self and other) would be more
affected by the excluded player’s payoffthan individuals with a proself
orientation (i.e., individuals who are motivated to maximise their own gains
either in an absolute or relative manner). Social value orientation was
assessed with the nine-item decomposed games measure of social value
orientation (Messick & McClintock, 1968, Van Lange & Kuhlman, 1994).
Another difference was that participants were not presented with a scenario.
This time they negotiated with two other participants who were also present
during the experiment. As predicted, social value orientations moderated the
effect of the unsold parcel (see Figure 7). Prosocials lowered their preference
for being members of a small coalition when the excluded player’s payoff
decreased. Proselfs preferred being members of small coalitions regardless of
the excluded player’s payoff. Thus, compared to proselfs, prosocials were
more willing to lower their own outcome in order not to decrease the
outcome of an excluded player.
Taken together, these studies corroborate the notion derived from the
social utility model that parties in coalition bargaining may not only be
concerned with maximising their own outcome (i.e., self-interest). The
Figure 7. Percentage of proposed small coalitions in a 2(1 1 1) game as a function of value of the
unsold parcel and social value orientation of the participant (adapted from Van Beest et al.,
2003, Experiment 2).
SELF-INTEREST AND FAIRNESS IN COALITIONS 153

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
results clearly show that people may also be concerned with what happens to
those who suffer from such decisions to maximise own payoffs. In line with
prior research on social exclusion (Williams, 2007) and the do-no-harm
principle (Baron, 1993, 1994), we thus demonstrated that whether one
proposes forming a small coalition (i.e., to exclude someone) depends on
whether one perceives the situation as harmful to the excluded individual.
The current findings showed that this perception is shaped by the outcomes
structure of the situation and the disposition of bargainers involved.
4. THE VALENCE OF COALITION OUTCOMES AND
THE WEIGHTS OF SELF-INTEREST AND FAIRNESS
The fact that coalition negotiators may be persuaded to forgo self-interest so
as not to harm the outcome of other members of a group can also be studied
using a different structural variable. To this end it is worthwhile to consider
that previous studies on coalition formation have focused exclusively on
situations in which participants negotiate about positive outcomes. Similar
to experiments described in the previous sections, participants are told that
they can gain money if they form a coalition. But what about losses?For
example, how are governments formed in times of economic recession when
parties need to decide who will suffer from cuts in funding? Or, when a
company is downsizing, how do decision makers agree which departments
are going to fire people?
If one takes the stand that individuals are primarily motivated to
maximise their own outcome, one could argue that the valence of the payoff
is inconsequential. Individuals should be as likely to minimise their
individual loss as they are to maximise their individual gain. Yet, based
on our social utility model, it can be argued that payoffvalence is important
because it may affect the relative weight of the self-interest and fairness
components.
For our predictions we invoked prospect theory (Kahneman & Tversky,
1979) which asserts that people compare their payoffto a reference point,
and consider payoffs as losses when they fall below their reference point and
as gains when they fall above their reference point. A key assumption of
prospect theory is that individuals put a greater value on losses than on
gains (i.e., losses loom larger than gains). However, prospect theory is
designed especially for individual decision making. To apply it to a situation
of interdependence, we thus formulated a self-oriented interpretation of loss
aversion and a more other-oriented interpretation of loss aversion. The self-
oriented interpretation of loss aversion assumes that individuals will
emphasise the self-interest component of the social utility model. This
interpretation of loss aversion implies that participants would be more
likely to minimise their own loss than to maximise their own gain. The
154 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
other-oriented interpretation of loss aversion assumes that individuals will
emphasise the fairness component of the social utility model. The other-
oriented interpretation of loss aversion implies that individuals should be
more likely to minimise the loss of others than to maximise the gain of
others in coalition formation.
Interestingly, one can find experimental evidence for both these
contrasting hypotheses if one turns to other domains than coalition
formation. Supporting the self-oriented interpretation of loss aversion,
research on bilateral negotiations has shown that loss-framed bargainers
may demand more and impasse more frequently than gain-framed
bargainers (e.g., Bottom & Studt, 1993; De Dreu, 1996; Neale & Bazerman,
1985; Ohtsubo & Kameda, 1998). Likewise, research on social dilemmas has
shown that individuals become less cooperative when losses are involved
than when gains are involved (Brewer & Kramer, 1986; McCusker &
Carnevale, 1995). Yet, supporting the other-oriented interpretation of loss
aversion, research on distributive fairness has shown that individuals tend to
allocate losses more equally than profits (To
¨rnblom & Jonsson, 1985, 1987).
Futhermore, categorising individuals into arbitrary social groups elicits
ingroup favouritism when positively valued stimuli are used, but not when
negatively valued stimuli are used (Mummendy & Otten, 1998; Mummendey
et al., 1992).
To test these contrasting hypotheses we conducted two experiments
(Van Beest et al., 2005). In a first test of these hypotheses participants
negotiated with two other participants (Van Beest et al., 2005, Experiment
1). In the gain conditions, participants were endowed with 0 euros and told
that they had to negotiate about a gain of 60 euros. In the loss conditions,
participants were endowed with 40 euros and told that they had to
negotiate about a loss of 60 euros. It was stressed that coalition members
had the power to decide any allocation they wanted. In the gain domain
they could therefore decide to give gains to the included coalition members
and not to the excluded individuals. In the loss conditions they could
decide to allocate all the losses to the excluded player and not to the
included coalition members. In both conditions it was therefore more
profitable to form a small two-player coalition than a grand three-player
coalition.
Results of Experiment 1 showed that payoffvalence had an effect on
formed coalition (see Figure 8). Participants did not hesitate to maximise
their own positive payoffs. In the gain condition, small coalitions in which
both members obtained 20 euros were formed more often than grand
coalitions in which each member obtained 10 euros. However, participants
were reluctant to minimise their own negative payoffs. In the loss condition,
grand coalitions in which each member paid 20 euros were formed
more often than small coalition in which each member paid 10 euros. The
SELF-INTEREST AND FAIRNESS IN COALITIONS 155

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
other-oriented interpretation of loss aversion was further supported by
additional analyses of motivations. Participants indicated whether their
actions were motivated by fairness and/or by self-interest. Results showed
that both these questions were extremes of the same dimension and that the
relative weight given to these aspects of the utility model were affected by
payoffvalence. In the loss condition participants were more motivated by
fairness. In the gain condition participants were more motivated by self-
interest. An additional mediation analyses showed that this mediated the
effect of valence on formed coalitions.
The more other-oriented interpretation of loss aversion was further
supported in a subsequent study (Van Beest et al., 2005, Experiment 2). In
this study participants negotiated with two preprogrammed computer
players about a gain or loss of 12 euros. In the gain condition participants
were endowed with 0 euros and informed that forming a small coalition
implied that the excluded player would then obtain no payoffs. In the loss
condition participants were endowed with 8 euros and informed that
forming a small coalition implied that the excluded player would then lose 8
euros. Unknown to the participant, we preprogrammed the two simulated
players to make offers that included the participant. In the gain condition,
one simulated player invited the participant to become a member of a small
coalition in which both members would obtain 6 euros. The other simulated
Figure 8. Percentage of formed small and grand coalitions in a 2(1 1 1) game as a function
payoffvalence (adapted from Van Beest et al., 2005, Experiment 1).
156 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
player invited the participant to become a member of a grand coalition in
which each member would obtain 4 euros. In the loss conditions participants
could choose between being a member of a small coalition in which both
members would pay 2 euros and being a member of a grand coalition in
which each member would pay 4 euros. The difference between the offers
was thus that one simulated player always proposed forming a small
coalition and that the other simulated player always proposed forming a
grand coalition.
In addition to our manipulation of valence, we also measured perspective
taking. Perspective taking was measured by the subscale of the Interpersonal
Reactivity Index (Davis, 1983, 1994). The perspective-taking scale consists
of seven items which measure the extent to which people spontaneously
adapt to the psychological viewpoint of others. A typical item from this
scale is ‘‘I try to look at everybody’s side of a disagreement before I make a
decision’’. Research on perspective taking has argued that the ability to
consider the viewpoint of others is a prerequisite for moral action (Batson
et al., 1997, 2003), and one could therefore maintain that it is a prerequisite
for assigning weight to the fairness component of the social utility model.
Based on this argument we reasoned that the willingness to maximise
positive payoffs (or minimise negative payoffs) at the expense of others
would thus be moderated by the extent to which individuals are able to
consider the viewpoint of others.
Consistent with the other-oriented interpretation of loss aversion and the
notion of social utility, the results indeed showed that perspective taking
moderated self-serving behaviour (see Figure 9). Low perspective takers
formed small coalition regardless of payoffvalence. In both the gain and the
loss domain they maximised their outcome at the expense of others. High
perspective takers, on the other hand, were influenced by payoffvalence.
They were less likely to form small coalitions in the negative domain than in
the positive domain.
To further investigate how behaviour was related to the fairness
component of the social utility model, we also measured the cognitive
accessibility of the fairness construct. Using a word-fragment completion
task that was developed by Van Prooijen, Van den Bos, and Wilke (2002) we
assessed accessibility by having participants complete word fragments that
could or could not be related to fairness construct. For example,
participants had to construct a Dutch word of two syllables that ended
with nrecht,whichcanbecompletedasonrecht (unjust) or as aanrecht
(kitchen sink). The frequency of completing the word in a fairness-related
way is then taken as a measure for the accessibility of the fairness construct.
Interestingly, this analysis yielded only a main effect of payoffvalence,
showing that fairness was more accessible in the loss condition than in the
gain condition. Although these data are consistent with the other-oriented
SELF-INTEREST AND FAIRNESS IN COALITIONS 157

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
version of loss aversion, it may be noted that the accessibility data did not
reveal the interaction of payoffvalence and perspective taking that we did
observe on the behavioural data regarding the formation of small coalitions.
This suggests that accessibility of fairness does not necessarily imply that
one does not exclude others. Indeed, as demonstrated in Study 1, for this to
occur one needs to be motivated by fairness.
Taken together, these studies further illustrate the usefulness of the social
utility model for coalition formation. In the previous section on the scope of
fairness we noted that the relative weight assigned to the fairness component
increases when the value of an unsold parcel decreases. In this section we
extended this line of reasoning by using another structural variable that has
been overlooked in previous coalition research. Moving from the gain
domain to the loss domain affects the relative weights given to self-interest
and fairness. Moreover, a possible limitation of the studies presented in the
previous sections is that we did not assess motivations underlying coalition
formation. Similar to most other coalition studies we only reported
behavioural choices. Another contribution of the current studies is therefore
that we demonstrated that the behavioural choice to form a small coalition
coincided with the motivation to maximise one’s own payoffs, whereas the
behavioural choice to form a grand coalition coincided with the motivation
to be fair.
Figure 9. Percentage of formed small coalitions in a 2(1 1 1) game as a function payoffvalence
and perspective taking (adapted from Van Beest et al., 2005, Experiment 2).
158 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
5. THE SIZE OF THE COALITION PARTIES AND THE
WEIGHTS OF SELF-INTEREST AND FAIRNESS
In the previous experiments we sometimes used political examples to
illustrate our reasoning. In the first section we noted that political parties
may use their votes or the number of coalitions of which they can be a
member to substantiate demands. In the second section we noted that the
number of available seats in cabinet may facilitate or inhibit the formation
of specific coalitions. In the previous section we noted that political parties
may be reluctant to benefit their own constituents in times of economic
recession. A core theme in all these examples is that we provided answers
that may be used to explain why under specific conditions participants form
unnecessary large coalitions and thus reduce their own payoffs.
In this sense it is relevant to note that political scientists have debated for
quite some time now that the simple rational choice models of government
formation do not provide all the answers (Martin & Stevenson, 2001). One
of the problems is the relatively high incidence of ‘‘over-sized’’ or ‘‘surplus-
majority’’ governments. In fact, 39% of all governing coalitions in western
European countries between 1945 and 1998 included political parties that
were not necessary to obtain a parliamentary majority (Gallagher, Laver, &
Mair, 2001). However, political scientists may be reluctant to consider some
of the reasons that we presented to understand why such surplus majority
governments are formed, because our experiments were based on individuals
and not on groups.
To investigate whether and how interindividual and intergroup coalitions
would be different we used a set-up that we also used when we tested the
scope of the fairness component of the social utility model. In that research
line we observed that individuals with a prosocial orientation were less likely
to form small coalition than individuals with a proself orientation when an
unsold parcel decreased in value. In the current section we set out to test
whether this would generalise to a situation in which groups would interact.
Would groups of prosocials also be reluctant to increase the payoffs of their
own subgroup at the expense of the payoffs of other subgroups?
Our hypotheses were inspired by research on the interindividual –
intergroup discontinuity effect (for reviews see Schopler & Insko, 1992;
Wildschut et al., 2003). Numerous experiments by Insko, Schopler, and
colleagues on the prisoner’s dilemma game have shown that conflicts
between groups are more competitive than conflicts between individuals.
Wildschut and colleagues (2003) provide three reasons that are based on a
combination of fear and greed to explain this effect. First, individuals in
groups provide mutual support for the pursuit of self-interest. Second,
groups provide a shield of anonymity facilitating the pursuit of self-interest.
Finally, the anticipation of interacting with another group activates an
SELF-INTEREST AND FAIRNESS IN COALITIONS 159

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
outgroup schema, consisting of learned beliefs and expectations that
intergroup interactions are aggressive and competitive. Given these reasons
we predicted that proselfs might indeed be more likely to pursue self-interest
in an intergroup setting than in an interindividual setting. Thus, for proselfs
the weight of the self-interest component of the social utility model should
increase in an intergroup setting. However, given that prosocials are
characterised by their disposition to value the fairness component of the
social utility model (e.g., Van Lange et al., 2007), we argued that for
prosocials, variables associated with intergroup interaction (e.g., social
support, schema-based distrust) should have a lesser impact. Thus, we
anticipated that proselfs would not necessarily become more self-interested
in an intergroup setting than in an interindividual setting.
To test our reasoning participants were assigned to a 2 (coalition setting:
interpersonal vs intergroup) 62(socialvalueorientation:prosocialvs
proself) design. In the intergroup settings participants negotiated as dyads
with two other dyads.
5
In the interpersonal settings participants negotiated
as individuals with two other individuals. Social value orientation was
measured 2 weeks in advance via the Internet. Depending on condition, but
unknown to the participant, we only invited participants with a proself or
with a prosocial orientation to the lab. We did not investigate conditions
with heterogeneous groups. As in the excluded player studies (Van Beest
et al., 2003) we told participants that they owned one parcel of land with an
assessed value of 20,000 euros. Next, we informed participants that a project
developer was interested in buying at least two parcels for a fixed prize of
60,000 euros and that unsold parcels would lose all their value. Provided
that coalition members obtain an equal payoffshare, this implied that
participants could only obtain more money than the assessed value if they
formed a small coalition. Using this design we tested our hypothesis on
different levels. We analysed what participants wanted to form, what they
eventually ended up forming, and how difficult it was to reach an agreement.
Consistent with our reasoning the results showed that moving from an
interpersonal coalition setting to an intergroup coalition setting had a
different impact on coalition preference of proselfs than on those of
5
Similar to other research comparing individual and group behaviour, we used dyads to
manipulate groups and not triplets or more. One reason was that previous research on the
discontinuity effects established that competitiveness increases greatly as one moves from one-
on-one interaction to two-on-two interactions, but slightly as one moves from two-on-two
interaction to three-on-three interaction (Wildschut et al., 2003). Another reason was that we
did not want to complicate our research by using a nested coalition design. Using triplets or
more as a group manipulation has the disadvantage that party members may form within-party
coalitions to establish what course of action to take. That is, if more than two party members
are involved one has to establish a majority decision rule, whereas in the current version all
members had to agree about what decision they wanted to make.
160 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
prosocials (see Figure 10). Proselfs in an intergroup coalition setting were
more likely to opt for small coalitions than proselfs in an interpersonal
coalition setting. Prosocials in contrast, preferred to include all and were as
likely to form small coalitions in both the intergroup and interpersonal
setting. These results show that intergroup coalition behaviour should not
only be understood in terms of self-interest but also in terms of more other-
oriented concerns. In fact, given that differences in behaviour between
proselfs and prosocials were more pronounced at the group level than at the
individual level, it may even be argued that the relative importance of these
motivations is augmented at the group level.
We did not observe an effect of coalition setting on the number of rounds
that participants needed to come to an agreement. This illustrates that there
is not a one-on-one relation between the outcomes of bargaining (e.g., the
formation of a small versus grand coalition) and the process that leads up to
these outcomes. The formation of small coalitions, for example, need not
necessarily be accompanied by more conflict (i.e., a longer bargaining
process), as two parties may even instantly agree to form such a small
coalition. What we did find was that social value orientation affected conflict
during negotiations (see Figure 11). Results indicated that proselfs had more
conflict during negotiations than prosocials, and especially when they ended
up forming a grand coalition. This is interesting because from a
coordination perspective it has been argued that reaching an agreement
Figure 10. Percentage of proposed small coalitions in a 2(1 1 1) game as a function social value
orientation and game type (adapted from Van Beest et al., in press).
SELF-INTEREST AND FAIRNESS IN COALITIONS 161

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
between three parties (i.e., forming a grand coalition), should be more
difficult to coordinate than reaching an agreement between two parties (i.e.,
forming a small coalition) (Gamson, 1964; Komorita & Parks, 1995;
Murnighan, 1978). We concur, but add that one has to consider what parties
value. Forming a grand coalition may indeed be a coordination problem for
parties that aim to maximise their own payoffs (i.e., those who assign great
weight to the self-interest component of the social utility model). It seems to
be less problematic for parties that value obtaining equality in payoffs (i.e.,
those who assign great weight to the fairness component of the model).
GENERAL DISCUSSION
In the previous sections of this paper, we summarised the main findings of
five lines of research. We demonstrated the advantages of applying the
conceptual frame of the social utility model to coalition bargaining. The
notion that self-interest and fairness comprise distinct motivations in
coalition bargaining had predictive value for a range of findings that cannot
be explained by self-interest alone. In our programmatic line of research we
investigated how the two main components of the model—self-interest and
fairness—determine the selection of partners in bargaining, and how they
affect the distribution of outcomes in and beyond the coalition. We showed
Figure 11. Number of potential partner switches prior a final agreement as a function of social
value orientation and formed coalition (adapted from Van Beest et al., in press).
162 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
that the possibility of obtaining a fair share of the reward facilitates the
formation of specific coalitions. We also revealed that even when coalition
members of a small coalition can obtain a fair share of the reward,
individuals can be induced to form a coalition of all individuals, and that
this finding may even be more pronounced in coalitions between groups. In
this section we now relate these findings to the different functions of fairness
in coalition bargaining. Next, we propose that coalition formation can be
viewed as a game of inclusion and as a game of exclusion. To conclude, we
propose some avenues of further research.
The function of fairness in coalition bargaining
In the majority of coalition experiments that are presented in this paper,
fairness is considered to be a means or mechanism to reach an agreement
about the allocation of the reward in a coalition. In our studies on the
egocentric selection of standards (i.e., resources versus alternatives), the
results indicated that coalition members are likely to demand and obtain a
share of the reward that is in keeping with their source of entitlement, and
that people are egocentric in their selection of standards. In the experiments
on the alignment of self-interest and fairness, the results even indicated that
when coalition members cannot obtain a fair share of the reward they might
form an alternative coalition in which they can obtain a fair share of the
reward.
Fairness as a means to reach an agreement is closely linked to self-
interest. In line with the notion that social norms are used to govern the
conflict that arises when people want to maximise their own outcome
(Pruitt & Carnevale, 1993), fairness can be used to maximise one’s own
outcome in a coalition situation. In experiments on the egocentric selection
of standards, prospective coalition members chose a fairness rule that
maximised their own outcome. In the studies on the alignment of self-
interest and fairness, we demonstrated that if bargainers could reach a fair
distribution within a small coalition, they did not hesitate to exclude others
to maximise their own payoffshare.
The studies on the scope of fairness and the payoffvalence experiments,
however, revealed that fairness can also be related to a true concern for
others. The scope of fairness experiments demonstrated that the possibility
of obtaining a fair payoffshare in a small coalition did not persuade
individuals to maximise their payoffs when this lowered the positive payoff
of excluded individuals. The payoffvalence experiments revealed that
individuals were more reluctant to form small coalitions that minimised
their own loss at the expense of others than to form small coalitions that
maximised their own gain at the expense of others. Both these lines of
research show that fairness is not merely a useful mechanism for the
SELF-INTEREST AND FAIRNESS IN COALITIONS 163

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
formation of a coalition, but also related to a moral obligation not to harm
others.
In sum, fairness can serve two purposes in coalition formation. Fairness
can be conceived as an instrumental means to maximise one’s own
outcomes, but it can also be defined in terms of a true concern for others
(see also Van Dijk & De Cremer, 2006). Fairness can thus be related to a
concern for self and to a concern for others. The reason why we believe this
is an interesting distinction is that it illustrates the various ways in which
fairness considerations may play a role in coalition formation. When one
defines fairness in terms of concern for one’s self, one tends to view fairness
as an instrumental mechanism to be included in a coalition. This is the most
dominant perspective in research on coalition formation and is closely
linked to self-interest. When one defines fairness in terms of concern for
others, fairness is not primarily viewed as a mechanism to form a coalition.
This perspective has a more moral connotation: It is not fair to exclude
others from obtaining a share of the reward.
It is interesting to see that the issues we raise here resemble a discussion in
the context of two-person bargaining. Especially in the context of ultimatum
bargaining, researchers have addressed the question of whether the
predominance of fair offers in ultimatum bargaining (for reviews see e.g.,
Camerer & Thaler, 1995; Gu
¨th & Tietz, 1990; Handgraaf et al., 2003; Thaler,
1988) reflects a true concern for fairness or that bargainers make fair offers
because they fear that unfair offers will be turned down. This latter use of
fairness has been referred to as the ‘‘strategic use of fairness’’ (e.g., Van Dijk &
Vermunt, 2000; Van Dijk, De Cremer, & Handgraaf, 2004; see also Camerer &
Thaler, 1990). Whereas initially researchers interpreted the making of fair
offers primarily as evidence for a true concern of fairness, subsequent
theorising and research suggests that in (ultimatum) bargaining both true
fairness and strategic fairness play an important role. Often these two types of
fairness may lead to similar behaviour (e.g., when bargainers make fair offers
because they fear that an unfair offer would be rejected). On some occasions,
however, both types can lead to different behaviour (e.g., when the
consequences of rejection are low; see e.g. Van Dijk & Vermunt, 2000).
In a similar vein, the two types of fairness we distinguish here—
instrumental fairness (comparable to strategic fairness) and true fairness—
may give rise to the same effect in coalition formation. In the divisibility of
the reward experiments, grand coalitions were formed, because self-
interested individuals could not reach an agreement based on fairness in a
small coalition. However, in the excluded player and payoffvalence
experiments, grand coalitions were proposed because an excluded player
would suffer if a small coalition were formed. Hence, although in both
experiments grand coalitions were formed, the reason for forming a grand
coalition was quite different.
164 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Inclusion and exclusion
Traditional social psychological coalition theories predict who is likely to
join forces with another person to obtain positive payoffs that cannot be
obtained alone. To investigate this, outcomes are usually only available for
the ones who become included in the winning coalition, with the set-up
being that individuals cannot allocate resources to excluded individuals and
consequences to the excluded individuals are essentially kept out of the
picture. In fact, not only are individuals usually forbidden to form grand
coalitions, they are also specifically instructed that side-payments are not
allowed. This approach—which is predominantly focused on being included
in the coalition—is in agreement with the conceptualisation of coalitions as
to or more parties agreeing to cooperate to obtain a mutually desired
outcome (Kahan & Rapoport, 1984; Komorita & Kravitz, 1983; Polzer
et al., 1998). We argue here that this focus on inclusion may facilitate the
motivation to maximise own outcomes. For example, when negotiating
about 60 euros in a three-player coalition setting, it is easy to approach one
of the other individuals with the statement that you want to split the reward
if you do not have to attend to what happens to the excluded individual. It is
easy because you do not have to mention (and therefore worry about) the
fact that one other player will then get nothing.
In the scope of fairness, payoffvalence, and discontinuity experiments we
took another approach and argued that inclusion is often associated with
exclusion. In these lines of research we showed that when exclusion is salient
individuals may in fact be reluctant to maximise their own financial payoff.
This suggests that there is some disutility associated with exclusion. When
coalition individuals are forced to think about the consequences of exclusion
they may be persuaded to forgo self-interest in terms of financial outcome,
and be persuaded to also be concerned about social outcomes. Following
our results we would like to propose a more general definition of coalition
formation. Stressing the relevance of inclusion and exclusion we propose
that coalition formation should be defined as ‘‘the process in which two or
more parties negotiate about the decision to allocate payoffs to those that are
included and to those that are excluded in a coalition.’’ This definition accords
with Thibaut and Kelley’s (1959) general statement that coalitions affect the
outcomes of one or more other persons, but is more specific in that it
specifies the nature of these other persons. Our definition captures the fact
that the formation of a coalition affects the included individuals as well as
the excluded individuals. Moreover by not restricting the definition to
situations in which coalitions are formed to obtain desired outcomes, it also
acknowledges that coalition formation is not only aimed at acquiring
positive (desired) outcomes, but often also about deciding how to allocate
negative outcomes. As our studies have shown, this broader definition, and
SELF-INTEREST AND FAIRNESS IN COALITIONS 165

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
the experiments that are derived from it, allow for a more comprehensive,
and more social account of the bargaining processes and outcomes in
coalition formation.
Future research
Many experiments that were presented in this paper went beyond outcome
variables such as formed coalition and reward allocation. In various
experiments we included process variables to measure the amount of
experienced conflict. For example, the experiments on egocentric selection of
standards demonstrated that differences in alternatives led to longer
bargaining, whereas differences in resources led to more concessions.
Moreover, the experiments on the alignment of self-interest and fairness
demonstrated that when individuals could not obtain a fair share of the
reward in a small coalition, it took longer to reach an agreement and
coalition members were more easily persuaded to form an alternative
coalition than when individuals could obtain a fair share of the reward in a
small coalition. This illustrates that the negotiation process may become
more problematic when it is not possible to simultaneously satisfy self-
interest and fairness. Finally, the discontinuity study showed that prosocials
experienced more conflict when the negotiations ended in a small coalition,
whereas proselfs experienced more conflict when the negotiations ended in a
grand coalition. This intriguing result takes a different spin on the usual
observation that including unnecessary members in a coalition has to be
more difficult than forming a small coalition.
Apossiblelineoffutureresearchcouldaddresstheconsequencesofbeing
confronted with demands that do not accord with one’s own preferred
principle of fairness. What would happen when individuals use different
sources of entitlement to demand a share of the reward? Suppose, for
example, that individuals who base their demands on resources are
confronted with individuals who base their demands on alternatives. Again,
it seems likely that the negotiation process becomes more problematic (cf.
Loewenstein & Moore, 2004; Loewenstein et al., 1993; Thompson &
Loewenstein, 1992). This highlights that it is not fairness per se that
facilitates coalition formation, but the universal applicability of one fairness
principle. Indeed, before one can reach an agreement about the allocation of
the reward, one has to reach an agreement about which fairness principle
will be applied.
Asecondlineofresearchcouldfocusmoreontheconsequencesof
forming a coalition. What happens to a coalition once it is formed? In this
paper we provided some insights about the negotiation process before a
coalition is being formed. We showed that when individuals can
simultaneously satisfy self-interest and fairness the bargaining process prior
166 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
to forming a coalition is associated with fewer difficulties than when
individuals cannot simultaneously satisfy self-interest and fairness. This
finding can be used to generate predictions about the stability of a coalition
after it has been formed (see also Mannix, 1994; Polzer et al., 1998). Given
the results of this paper, it seems that coalitions are most stable when
members have obtained a fair share of the reward and have maximised their
share of the reward. In such a coalition, members should not have an
incentive to leave the coalition. In contrast, in situations where both
interests are not simultaneously met bargainers could have an incentive to
leave the coalition.
Third, coalition research could include more dependent variables to
assess the different aspects of partner selection and payoffallocation. In the
current paper we mainly reported behavioural data to support the relative
importance of self-interest and fairness. In addition, we also included some
data on motivations and accessibility of fairness. A specific group of
variables that we did not include, and which should get more attention, is
emotions (for a review of current insights on emotions and bargaining see
Van Kleef, Van Dijk, Steinel, Harinck, & Van Beest, in press). How do
individuals feel when they are included or excluded from a negotiation deal?
And are these feelings more related to partner selection or to payoff
allocations?
To this end it may be worthwhile to consider a recent study by Van Beest
and Williams (2006). They showed that being excluded from a game of ball-
toss decreased well-being on a host of measurements and that this decrease
in well-being was not moderated by financial incentives. Participants were
either included or excluded from a game of ball-toss and, depending on
condition, participants either got money for receiving a ball toss or had to
pay for receiving a ball toss. Hence, in financial terms it was more beneficial
to be excluded from this game than it was to be included in this game. Yet
results showed that participants who were financially compensated for being
excluded felt worse than participants who were financially punished for
being included even though the first group left the experiment with more
money in their pocket than the latter group. This suggests that being
punished in social terms is more detrimental than being punished in financial
terms. A question that remains to be answered, however, is whether this
finding would also generalise to a coalition setting in which exclusion and
incentives often go hand in hand.
Finally, we would like to mention that most of the experiments discussed
in the current chapter used situations in which every possible coalition
yielded a similar payoff. These situations are the most studied setting in
social psychology (Komorita & Parks, 1995) and with their simple structure
they provide an excellent environment to test the comparative weight of self-
interest and fairness of our social utility approach. It should be noted,
SELF-INTEREST AND FAIRNESS IN COALITIONS 167

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
however, that research on coalition bargaining also studies more complex
settings. A promising line of future research would thus be to test our
assumptions in more complex situations. One interesting situation would be
to focus on coalition setting in which each coalition yields a different payoff
(e.g., Crott & Albers, 1981; Komorita, 1979; Rapoport & Kahan, 1982).
Another interesting situation would be to focus on settings in which
negotiators can reach agreements on a range of different issues (e.g.,
Beersma & De Dreu, 2002; Ten Velden, Beersma, & De Dreu, 2007).
Finally, research could expand our reasoning to situations in which some
coalitions (usually called unions) may block the realisation of other
coalitions (e.g., Michener, 1992, Michener & Myers, 1998). We predict that
increasing the level of complexity will introduce new bases on which
negotiators can differ and thus more room for self-serving interpretations,
more possibilities to align self-interest and fairness, and finally more
dimensions on which some individuals can be either included or excluded.
CONCLUSION
To conclude, we turn to one of the most influential social psychologists in
the field of coalition formation: Sam Komorita. In one of the last review
papers on coalition formation, he and his colleague Craig D. Parks noted
that interest in research on coalition formation has declined (Komorita &
Parks, 1995). They attributed this decline to the absence of process-oriented
approaches that clarify what factors underlie coalition formation (see also
Cook & Gillmore, 1984; Kahan & Rapoport, 1984a; Komorita, 1984;
Levine & Moreland, 1990). In addition to this argument, we argued in this
paper that this decline could also be attributed to the dominant research
method used in most coalition experiments. In answer to these criticisms we
presented five lines of research in which we took a different approach to
study coalition formation. Inspired by the social utility model we presented
five lines of research that showed how moving from a self-interested account
of coalition formation to a more balanced account which also acknowledges
that negotiations are guided by fairness, furthers the understanding of
coalition bargaining. It provided a more comprehensive account of when
and why some individuals are more likely to be excluded from a negotiated
agreement than others.
REFERENCES
Adams, J. S. (1963). Towards an understanding of inequity. Journal of Abnormal and Social
Psychology,67,422–436.
Allison, S. R., & Messick, D. M. (1990). Social decision heuristics in the use of shared resources.
Journal of Behavioural Decision Making,3,195–204.
168 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Allison, S. T., McQueen, L. R., & Schaerfl, L. M. (1992). Social decision making and the
equal partitionment of shared resources. Journal of Experimental Social Psychology,28,
23 – 42.
Babcock, L., & Loewenstein, G. (1997). Explaining bargaining impasse: The role of self-serving
biases. Journal of Economic Perspectives,11,109–126.
Babcock, L., Loewenstein, G., Issacharoff, S., & Camerer, C. (1995). Biased judgements of
fairness in bargaining. American Economic Review,85,1337–1334.
Baron, J. (1993). Heuristics and biases in equity judgements: A utilitarian approach. In
B. A. Mellers & J. Baron (Eds.), Psychological perspectives on justice: Theory and
applications (pp. 109 – 137). New York: Cambridge University Press.
Baron, J. (1994). Blind justice: Fairness to groups and the do-no-harm principle. Journal of
Behavioural Decision Making,8,71–83.
Batson, C. D., Early, S., & Salvarani, G. (1997). Perspective taking: Imagining how another
feels versus imagining how you would feel. Personality and Social Psychology Bulletin,23,
751 – 758.
Batson, C. D., Lishner, D. A., Carpenter, A., Dulin, L., Harjusola-Webb, S., Stocks, E. L., et al.
(2003). ‘‘. . . As you would have them do unto you’’: Does imagining yourself in the other’s
place stimulate moral action? Personality and Social Psychology Bulletin,29,1190–1201.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin,117,497–529.
Bazerman, M. H., Curhan, J. R., Moore, D. A., & Valley, K. L. (2000). Negotiation. Annual
Review of Psychology,51,279–314.
Beersma, B., & De Dreu, C. K. W. (2002). Integrative and distributive negotiation in small
groups: Effects of task structure, decision rule and social motive. Organisational Behaviour
and Human Decision Processes,87,227–252.
Blumberg, H. H. (1994). Bargaining, coalitions, and games. In A. P. Hare, H. H. Blumberg,
M. F. Davies, & M. V. Kent (Eds.), Small group research: A handbook. Norwood, NJ: Ablex
Publishing Corporation.
Bottom, W., & Studt, A. (1993). Framing effects and the distributive aspects of integrative
bargaining. Organisational Behaviour and Human Decision Processes,56,459–474.
Brewer, M., & Kramer, R. (1986). Choice behaviour in social dilemmas: Effects of social
identity, group size, and decision framing. Journal of Personality and Social Psychology,50,
543 – 549.
Camerer, C. F., & Thaler, R. H. (1995). Ultimatums, dictators and manners. Journal of
Economic Perspectives,9,209–219.
Carnevale, P. J., & Pruitt, D. G. (1992). Negotiation and mediation. Annual Review of
Psychology,43,531–582.
Chertkoff, J. M., & Braden, J. L. (1974). Effects of experience and bargaining restrictions on
coalition formation. Journal of Personality and Social Psychology,30(1), 169 – 177.
Cook, K. S., & Gillmore, M. R. (1984). Power, dependence, and coalitions. Advances in Group
Processes,1,27–58.
Crott, H. W., & Albers, W. (1981). The equal division kernel: An equity approach to coalition
formation and payoffdistribution in N-person games. European Journal of Social
Psychology,11,285–305.
Davis, M. H. (1983). Measuring individual-differences in empathy: Evidence for a multi-
dimensional approach. Journal of Personality and Social Psychology,44,113–126.
Davis, M. H. (1994). Empathy: A social psychological approach. Madison, WI: Brown &
Benchmark.
De Dreu, C. K. W. (1996). Gain-loss-frame in outcome-interdependence: Does it
influence equality or equity considerations? European Journal of Social Psychology,26,
315 – 324.
SELF-INTEREST AND FAIRNESS IN COALITIONS 169

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
De Dreu, C. K. W., & Carnevale, P. (2003). Motivational bases of information processing and
strategy in conflict and negotiation. In M. P. Zanna (Ed.), Advances in experimental social
psychology (Vol. 35, pp. 235 – 291). San Diego, CA: Academic Press.
Druckman, D. (1994). Determinants of compromising behaviour in negotiation. Journal of
Conflict Resolution,38,507–556.
French, J. R. P., & Raven, B. (1959). The bases of social power. In D. Cartwright (Ed.), Studies
in social power. Ann Arbor, MI: Institute for Social Research.
Gallagher, M., Laver, M., & Mair, P. (2001). Representative government in modern Europe, 3th
ed. New York: McGraw Hill.
Gamson, W. A. (1961). An experimental test of a theory of coalition formation. American
Sociological Review,26,565–573.
Gamson, W. A. (1964). Experimental studies of coalition formation. In L. Berkowitz (Ed.),
Advances in experimental social psychology (Vol. 1, pp. 81 – 110). New York: Academic Press.
Gu
¨th, W., & Tietz, R. (1990). Ultimatum bargaining behaviour: A survey and comparison of
experimental results. Journal of Economic Psychology,11,417–449.
Handgraaf, M. J. J., Van Dijk, E., & De Cremer, D. (2003). Social utility in ultimatum
bargaining. Social Justice Research,16,263–283.
Harinck, F., Van Dijk, E., Van Beest, I., & Mersmann, P. (in press). When gains loom larger
than losses: Loss aversion and small amounts of money. Psychological Science.
Homans, G. C. (1974). Social behaviour: Its elementary forms. New York: Harcourt Brace
Jovanovitch, Inc.
Kahan, J. P., & Helwig, R. A. (1971). Coalitions: A system of programs for computer-
controlled bargaining games. General Systems,16,31–41.
Kahan, J. P., & Rapoport, A. (1984a). Social psychology and the theory of games: A mixed-
motive relationship. Representative Research in Social Psychology,14,65–71.
Kahan, J. P., & Rapoport, A. (1984b). Theories of coalition formation. Hillsdale, NJ: Lawrence
Erlbaum Associates Inc.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk.
Econometrica,47,263–291.
Kelley, H. H. (1979). Personal relationships. Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Kelley, H. H., & Arrowood, A. J. (1960). Coalitions in the triad: Critique and experiment.
Sociometry,23,231–244.
Komorita, S. S. (1974). A weighted probability model of coalition formation. Psychological
Review,81,242–256.
Komorita, S. S. (1979). An equal excess model of coalition formation. Behavioural Science,24,
369 – 381.
Komorita, S. S. (1984). Coalition bargaining. Advances in Experimental Social Psychology,18,
183 – 245.
Komorita, S. S., & Chertkoff, J. M. (1973). A bargaining theory of coalition formation.
Psychological Review,80,149–162.
Komorita, S. S., & Kravitz, D. (1983). Coalition formation: A social psychological approach. In
P. B. Paulus (Ed.), Basic group processes (pp. 179 – 203). New York: Springer-Verlag.
Komorita, S. S., & Meek, D. D. (1978). Generality and validity of some theories of coalition
formation. Journal of Personality and Social Psychology,36,392–404.
Komorita, S. S., & Moore, D. (1976). Theories and processes of coalition formation. Journal of
Personality and Social Psychology,33,371–381.
Komorita, S. S., & Nagao, D. (1983). The functions of resources in coalition bargaining.
Journal of Personality and Social Psychology,44,95–106.
Komorita, S. S., & Parks, C. D. (1995). Interpersonal relations: Mixed-motive interaction. In
J. T. Spence, J. M. Darley, & D. J. Foss (Eds.), Annual review of psychology (Vol. 46,
pp. 183 – 207). Palo Alto, CA: Annual Reviews.
170 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Kravitz, D. A. (1981). Effects of resources and alternatives on coalition formation. Journal of
Personality and Social Psychology,41,87–98.
Kravitz, D. A., & Iwaniszek, J. (1984). Number of coalitions and resources as sources of power
in coalition bargaining. Journal of Personality and Social Psychology,47,534–548.
Lawler, E. J. (1975). An experimental study of factors affecting the mobilization of
revolutionary coalitions. Sociometry,38,163–179.
Lawler, E. J. (1992). Power processes in bargaining. The Sociological Quarterly,33,17–34.
Leary, M. R. (1990). Responses to social exclusion: social anxiety, jealousy, loneliness,
depression, and low self-esteem. Journal of Social and Clinical Psychology,9,221–229.
Levine, J. M., & Moreland, R. L. (1990). Progress in small group research. Annual Review of
Psychology,41,585–634.
Lewicki, R. J., Saunders, D. M., & Barry, B. (2006). Negotiation (5th ed.). Singapore: McGraw-
Hill.
Loewenstein, G., Issacharoff, S., Camerer, C., & Babcock, L. (1993). Self-serving assessments of
fairness and pretrial bargaining. Journal of Legal Studies,22,135–159.
Loewenstein, G., & Moore, D. A. (2004). When ignorance is bliss: Information exchange and
ineffeciency in bargaining. Journal of Legal Studies,33,37–58.
Loewenstein, G., Thompson, L., Bazerman, M. H. (1989). Social utility and decision making in
interpersonal contexts. Journal of Personality and Social Psychology,57,426–441.
Luce, R. D., & Raiffa, H. (1959). Games and decisions. New York: Wiley.
Mannix, E. A. (1994). Will we meet again? Effects of power, distribution norms and scope of
future interaction in small group negotiation. Journal of Conflict Management,5,343–368.
Martin, L. W., & Stevenson, R. T. (2001). Government formation in parliamentary
democracies. American Journal of Political Science,45,33–50.
McCusker, C., & Carnevale, P. (1995). Frame and loss aversion in social dilemmas.
Organisational Behaviour and Human Decision Processes,61,190–201.
Mclean Parks, J., Gatewood, W., Nekkich, J. C., Boles, T. L., Gibson, K., Straub, P., et al.
(1996). Distributing adventitious outcomes: Social norms, egocentric martyrs, and the effects
on future relationships. Organisational Behaviour and Human Decision Processes,67,181–
200.
Messick, D. M., & McClintock, C. G. (1968). Motivational bases of choice in experimental
games. Journal of Experimental and Social Psychology,4,1–25.
Messick, D. M., & Sentis, K. P. (1979). Fairness and preference. Journal of Experimental Social
Psychology,15,418–434.
Messick, D. M., & Sentis, K. P. (1983). Fairness, preference, and fairness biases. In
D. M. Messick & K. S. Cook (Eds.), Equity theory (pp. 61 – 94). New York: Praeger.
Michener, H. A. (1992). Coalition anomalies in light of the central-union theory. Advances in
Group Processes,9,65–88.
Michener, H. A., Fleishman, J. A., Vaske, J. J., & Statza, G. R. (1975). Minimum resource and
pivotal power theories: A competitive test in four-person coalitional situations. Journal of
Conflict Resolution,19(1), 89 – 107.
Michener, H. A., & Lawler, E. J. (1971). Revolutionary coalition strength and collective failure
as determinants of status reallocation. Journal of Experimental Social Psychology,8,180–
195.
Michener, H. A., & Myers, D. (1998). Probabilistic coalition structure theories: An empirical
comparison in four-person superadditive sidepayment games. Journal of Conflict Resolution,
42,830–860.
Miller, C. E. (1980). A test of four theories of coalition formation: Effects of payoffs and
resources. Journal of Personality and Social Psychology,38,153–164.
Miller, C. E., & Wong, J. (1986). Coalition behaviour: Effects of earned versus unearned
resources. Organisational Behaviour and Human Decision Processes,38,257–277.
SELF-INTEREST AND FAIRNESS IN COALITIONS 171

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Mummendey, A., & Otten, S. (1998). Positive – negative asymmetry in social discrimination. In
W. Stroebe & M. Hewstone (Eds.), European review of social psychology (Vol. 9, pp. 107 –
143). Chichester, UK: John Wiley & Sons.
Mummendey, A., Simon, B., Dietze, C., Grunert, M., Haeger, G., Kessler, S., et al. (1992).
Categorization is not enough: Intergroup discrimination in negative outcome allocation.
Journal of Experimental Social Psychology,28,125–144.
Murnighan, J. K. (1978). Models of coalition behaviour: Game theoretic, social psychological,
and political perspectives. Psychological Bulletin,85,1130–1153.
Murnighan, J. K. (1991). The dynamics of bargaining games. Englewood Cliffs, NJ: Prentice
Hall.
Neale, M. A., & Bazerman, M. H. (1985). The effect of framing and negotiator overcon-
fidence on bagaining behaviour and outcomes. Academy of Management Journal,28,
34 – 49.
Nitz, L. H., & Phillips, J. L. (1969). The effects of divisibility of payoffon confederative
behaviour. Journal of Conflict Resolution,13,381–387.
Ohtsubo, Y., & Kameda, T. (1998). The function of equality heuristic in distributive bargaining:
Negotiated allocation of costs and benefits in a demand revelation context. Journal of
Experimental Social Psychology,34,90–108.
Polzer, J. T., Mannix, E. A., & Neale, M. A. (1998). Interest alignment and coalitions in
multiparty negotiation. Academy of Management Journal,41,42–54.
Pruitt, D. G., & Carnevale, P. J. (1993). Negotiation in social conflict. Buckingham, UK: Open
University Press.
Pruitt, D. G., & Rubin, J. Z. (1986). Social conflict: Escalation, stalemate, and settlement. New
York: Random House.
Rapoport, A., & Kahan, J. P. (1982). The power of a coalition and payoffdisbursement in
three-person negotiable conflicts. Journal of Mathematical Sociology,8,193–225.
Rubin, J. Z., Pruitt, D. G., & Kim, S. H. (1994). Social conflict: Escalation, stalemate, and
settlement. New York: McGraw-Hill, Inc.
Rusbult, C. E., & Van Lange, P. A. M. (2003). Interdependence, interaction and relationships.
Annual Review of Psychology,54,351–375.
Schelling, T. C. (1960). The strategy of conflict. Cambridge, MA: Harvard University Press.
Schopler, J., & Insko, C. A. (1992). The discontinuity effect in interpersonal and intergroup
relations: Generality and mediation. In W. Stroebe & M. Hewstone (Eds.), European review
of social psychology (Vol. 3, pp. 121 – 151). Chichester, UK: Wiley.
Shaply, L. S. (1953). A value for n-person games. In H. W. Kuhn & A. W. Tucker (Eds.),
Contributions to the theory of games (Vol. 2, pp. 307 – 317). Princeton, NJ: Princeton
University Press.
Ten Velden, F. S., Beersma, B., & De Dreu, C. K. W. (2007). Majority and minority influence in
group negotiation: The moderating effects of social motivation and decision rules. Journal of
Applied Psychology,92,259–268.
Thaler, R. H. (1988). The ultimatum game. Journal of Economic Perspectives,2,195–206.
Thibaut, J., & Kelley, H. H. (1959). The social psychology of groups. New York: Wiley.
Thompson, L., & Loewenstein, G. (1992). Egocentric interpretations of fairness and
interpersonal conflict. Organisational Behaviour and Human Decision Processes,51,176–
197.
To
¨rnblom, K. Y., & Jonsson, D. R. (1985). Subrules of the equality and contribution principles:
Their perceived fairness in distribution and retribution. Social Psychology Quarterly,48,
249 – 261.
To
¨rnblom, K. Y., & Jonsson, D. R. (1987). Distribution vs. retribution: The perceived justice of
the contribution and equality principles for cooperative and competitive relationships. Acta
Sociologica,30,25–52.
172 VAN BEEST AND VAN DIJK

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Van Beest, I., Andeweg, R., Koning, L., & Van Lange, P. A. M. (in press). Do groups exclude
others more readily than individuals in coalition formation? Group Processes and Intergroup
Relations.
Van Beest, I., Van Dijk, E., De Dreu, C. K. W., & Wilke, H. A. M. (2005). Do-no-harm in
coalition formation: Why losses inhibit exclusion and promote fairness cognitions. Journal
of Experimental Social Psychology,41,609–617.
Van Beest, I., Van Dijk, E., & Wilke, H. (2004a). Resources and alternatives in coalition
formation: The effects on payoff, self-serving behaviour, and bargaining length. European of
Journal of Social Psychology,34,713–728.
Van Beest, I., Van Dijk, E., & Wilke, H. (2004b). The interplay of self-interest and equity in
coalition formation. European of Journal of Social Psychology,34,547–565.
Van Beest, I., Wilke, H., & Van Dijk, E. (2003). The excluded player in coalition formation.
Personality and Social Psychology Bulletin,29,237–247.
Van Beest, I., & Williams, K. D. (2006). When inclusion costs and ostracism pays, ostracism still
hurts. Journal of Personality and Social Psychology,91,918–928.
Van Dijk, E., & De Cremer, D. (2006). Tacit coordination and social dilemmas: On
the importance of self-interest and fairness. In D. De Cremer, M. Zeelenberg, &
J. K. Murnighan (Eds.), Social psychology and economics (pp. 141 – 154). Mahwah, NJ:
Lawrence Erlbaum Associates Inc.
Van Dijk, E., De Cremer, D., & Handgraaf, M. J. J. (2004). Social value orientations and the
strategic use of fairness in ultimatum bargaining. Journal of Experimental Social Psychology,
40,697–707.
Van Dijk, E., & Vermunt, R. (2000). Strategy and fairness in social decision making:
Sometimes it pays to be powerless. Journal of Experimental Social Psychology,36,
1–25.
Van Kleef, G. A., Van Dijk, E., Steinel, W., Harinck, F., & Van Beest, I. (in press). Anger in
social conflict: Cross-situational comparisons and suggestions for the future. Group Decision
and Negotiation.
Van Lange, P. A. M. (2000). Beyond self-interest: A set of propositions relevant to interpersonal
orientations. In W. Stroebe & M. Hewstone (Eds.), European review of social psychology
(Vol. 11, pp. 297 – 331). Chichester, UK: John Wiley & Sons Ltd.
Van Lange, P. A. M., De Cremer, D., Van Dijk, E., & Van Vugt, M., (2007). Self-interest and
beyond: Basic principles of social interaction. In A. W. Kruglanski & E. T. Higgins (Eds.),
Social psychology: Handbook of basic principles (2nd ed., pp. 540 – 561). New York: Guilford
Press.
Van Lange, P. A. M., & Kuhlman, D. M. (1994). Social value orientations and impressions of
partner’s honesty and intelligence: A test of the might versus morality effect. Journal of
Personality and Social Psychology,67,126–141.
Van Prooijen, J.-W., Van den Bos, K., & Wilke, H. A. M. (2002). Procedural justice and status:
status salience as antecedents of procedural fairness effects. Journal of Personality and Social
Psychology,83,1353–1361.
Vinacke, W. E., & Arkoff, A. (1957). An experimental study of coalitions in the triad. American
Sociological Review,22(1), 406 – 414.
Von Neumann, J., & Morgenstern, O. (1947). Theory of games and economic behaviour
(2nd ed.). Princeton, NJ: Princeton University Press.
Walster, E., & Walster, G. W. (1975). Equity and social justice. Journal of Social Issues,31,
21 – 43.
Walster, E., Walster, G. W., & Berscheid, E. (1978). Equity: Theory and research. Boston: Allyn
&Bacon.
Ward, H. (1990). Three men in a boat, two must row: An analysis of a three-person chicken
pregame. Journal of Conflict Resolution,34,371–400.
SELF-INTEREST AND FAIRNESS IN COALITIONS 173

Downloaded By: [Van Beest, Ilja] At: 10:01 31 October 2007
Wildschut, T., Pinter, B., Vevea, J. L., Insko, C. A., & Schopler, J. (2003). Beyond the group
mind: A quantitative review of the interindividual – intergroup discontinuity effect.
Psychological Bulletin,129,698–722.
Wilke, H. A. M. (1985). Coalition formation. New York: Elsevier.
Williams, K. D. (2001). Ostracism: The power of silence. New York: Guilford Press.
Williams, K. D. (2007). Ostracism. Annual Review of Psychology,58,425–452.
Williams, K. D., Cheung, C. K. T., & Choi, W. (2000). Cyberostracism: Effects of being ignored
over the Internet. Journal of Personality and Social Psychology,79,748–762.
Willis, R. H. (1962). Coalitions in the triad. Sociometry,25,358–376.
Young, H. P. (1995). Dividing the Indivisible. American Behavioural Scientist,38,904–920.
174 VAN BEEST AND VAN DIJK