TESTING THE WATERS: USING COLLECTIVE REAL OPTIONS TO MANAGE THE SOCIAL DILEMMA OF STRATEGIC ALLIANCES
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
姝 Academy of Management Review
2011, Vol. 36, No. 4, 621–640.
http://dx.doi.org/10.5465/amr.2009.0481
TESTING THE WATERS: USING COLLECTIVE
REAL OPTIONS TO MANAGE THE SOCIAL
DILEMMA OF STRATEGIC ALLIANCES
MATTHEW W. MCCARTER
Chapman University
JOSEPH T. MAHONEY
GREGORY B. NORTHCRAFT
University of Illinois at Urbana-Champaign
We extend real options research by introducing the concept of collective real options
and model how collective real options provide strategic alliances a mechanism to
manage social uncertainty. Collective real options manage social uncertainty by
producing relational small wins that develop trust. The amount of trust developed by
acquiring a collective real option depends on the exposure of alliance partners.
Alliance partner reputation also plays an important role in the impact of collective
real options.
Strategic alliances are “voluntary arrange- from a lack of information about the intentions
ments [among two or more organizations] in- of alliance partners, and it impedes interfirm
volving the exchange, sharing, or co-develop- cooperation because alliance partners may fear
ment of products, technologies, or services” they cannot trust each other to do what is best
(Gulati, 1998: 293) and are a means for organiza- for the alliance (Messick, Allison, & Samuelson,
tions to achieve goals they could not attain in- 1988; Park & Ungson, 2001). This lack of trust
dependently. Alliances create value through the pushes alliance partners to either underinvest
pooling of resources to provide alliance partners defensively or not enter into the alliance alto-
with competitive advantages over rivals (Das & gether, thereby missing opportunities to capital-
Teng, 2000; Hardy, Phillips, & Lawrence, 2003). ize fully on alliance potential (McCarter & North-
However, alliances often fail to create their an- craft, 2007).
ticipated benefits (Gottschalg & Zollo, 2007; For the past two decades, real options re-
Kale, Dyer, & Singh, 2002). This article examines search on uncertainty in alliance governance
how the use of real options can influence the (Kogut, 1991) has focused on understanding how
likelihood of alliance success. alliance partners use real options unilaterally to
Two critical factors influencing alliance suc- manage environmental uncertainty (Folta, 1998;
cess are environmental uncertainty and social Reuer & Tong, 2005). We develop a model of how
uncertainty. Environmental uncertainty stems real options can be used collectively to manage
from a lack of information about the market and social uncertainty among alliance partners,
task environment (Bowman & Hurry, 1993; Das & thereby enhancing the likelihood of alliance
Teng, 1996). As a consequence of environmental success. The core logic of our model comes in
uncertainty, alliance partners may fear that five parts: (1) large-scale alliance cooperation
their pooled investments will not yield a suffi- represents a social dilemma requiring trust, (2)
ciently profitable return (McCarter, Rockmann, collective real options offer alliance partners
& Northcraft, 2010), and they may therefore pro- the opportunity to achieve relational small wins
tect their own interests at the expense of the that increase trust, (3) increased trust decreases
alliance (Luo, 2007). Social uncertainty stems the perceived vulnerability among alliance
partners and, thus, fuels larger-scale alliance
cooperation, (4) exposure (the amount of re-
We express our appreciation to associate editor Jean
sources put at risk by alliance partners to ac-
Philippe-Bonardi and three anonymous reviewers for their quire the collective real option) moderates the
input and suggestions. effect of a relational small win on trust develop-
621
Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright
holder’s express written permission. Users may print, download, or email articles for individual use only.622 Academy of Management Review October
ment, and (5) the reputations of alliance partners contributes private resources to further the alli-
also play an important role in the impact of ance’s interest—such as contributing people or
collective real options. capital to identify more efficient recycling pro-
Examining how collective real options influ- cesses—this action is termed cooperation
ence social uncertainty in alliances helps (Dawes, 1980). Defection occurs when an alli-
bridge the (often independently studied) struc- ance partner fails to contribute (or underinvests)
tural and motivational approaches to alliance private resources toward alliance initiatives
governance. In doing so our conceptualization of (Dawes, 1980).
real options highlights the importance of social An alliance partner might underinvest in alli-
uncertainty—and its management via the psy- ance initiatives for two reasons. First, an alli-
chology of relational small wins—to alliance ance partner may attempt to “free ride”— enjoy
success. the created public good (e.g., more efficient re-
We begin by first reviewing essential ele- cycling processes) without contributing toward
ments of the social dilemma paradigm. Next, we its creation. In other words, offensive defection
introduce the concept of collective real options occurs when an alliance partner attempts to en-
as a hybrid approach to managing social uncer- joy the shared benefit without incurring much (if
tainty in alliances. We then propose a model any) cost (Zeng & Chen, 2003). Alternatively, an
capturing how collective real options influence alliance partner may defensively defect—
trust formation in alliances. We conclude by dis- underinvest or, in the extreme, not contribute at
cussing the theoretical and managerial implica- all—to avoid wasting resources if the alliance
tions of our model. partner believes others also will underinvest
(Rockmann & Northcraft, 2008). Unfortunately, if
enough alliance partners underinvest, regard-
STRATEGIC ALLIANCE SOCIAL DILEMMAS:
less of whether offensively or defensively, the
A BRIEF REVIEW
public good is not produced and no alliance
The social dilemma paradigm has proven partners benefit—for example, poor air quality
very useful in helping understand how social abounds and alliance resources are wasted
uncertainty influences cooperation among alli- (Messick & Brewer, 1983). We focus on defensive
ance partners (Zeng & Chen, 2003). A social di- defection here because even if measures are
lemma, broadly defined, is a situation in which taken to make offensive defection impossible,
it appears that being uncooperative can be per- an alliance partner may remain suspicious of
sonally beneficial at the collective’s expense— other alliance partners’ intentions and thereby
unless everyone else in the collective also defect defensively to avoid wasting personal re-
chooses to be uncooperative, in which case no sources (McCarter & Northcraft, 2007).2
one benefits (Liebrand, 1983). In strategic alli-
ances partners’ resources can be pooled to cre-
ate a public good—that is, a resource that, once this second instance the private public good is public to a
produced, is enjoyed by all alliance partners, smaller specified collective (e.g., alliance partners) while
whether they contributed or not (Olson, 1965). being privatized from a larger specified collective (e.g., the
entire populace of a city). While our propositions are not
For instance, a group of firms may pool re-
limited to either form of public good, in this article we refer
sources to identify more efficient recycling pro- to examples of private public goods, and we use the abbre-
cesses (Tirole, 1996).1 When an alliance partner viated term public good to remain consistent with the types
of public goods discussed in the literature on strategic alli-
ances (Agarwal, Croson, & Mahoney, 2010; McCarter &
1
Public goods come in two forms (Cornes & Sandler, 1996). Northcraft, 2007; Zeng & Chen, 2003).
2
Pure public goods are resources available to anyone in a The social dilemma paradigm is complementary to other
society or similar large-scale collective—for example, a na- game-theoretic–based paradigms, such as co-opetition and
tion, community, or city. Examples of pure public goods integrative negotiation. The similarity between the social
include public radio, national defense, and lighthouses. Pri- dilemma paradigm and other game-theoretic paradigms is
vate public goods (or club goods) are resources available to that value creation occurs through cooperative, collective
anyone within a smaller-scale collective, such as a strategic action. The difference resides in the degree of exclusivity
alliance. Examples of private public goods include new between cooperative and competitive actions. In social di-
technologies and processes developed in strategic alliances lemmas cooperation and defection are interdependent: an
and generic advertising campaigns (e.g., “Got milk?” and alliance partner cannot defect (or cooperate) without affect-
“Beef, it’s what for dinner!”) used in a particular industry. In ing the other alliance partners in negative (or positive) ways2011 McCarter, Mahoney, and Northcraft 623
Structural and Motivational Approaches to encouraging alliance partners to consider each
Strategic Alliance Social Dilemmas other’s needs (Spekman, Isabella, MacAvoy, &
Forbes, 1996). For instance, future alliance part-
The social dilemma paradigm offers two ap-
ners may meet frequently to plan, discuss goals,
proaches to managing social uncertainty in al-
and communicate intentions before contributing
liances: structural and motivational (Kollock,
resources toward alliance initiatives (Kanter,
1998; Zeng & Chen, 2003).3 Both approaches at-
1994; Murray & Mahon, 1993).
tempt to mitigate social uncertainty by increas-
ing trust among alliance partners, because trust
is a central critical factor determining partner Limitations of Structural and
investment in alliance initiatives (McCarter & Motivational Approaches
Northcraft, 2007).
Both structural and motivational solutions, in
The structural approach seeks to raise alli-
isolation, are limited in their ability to encour-
ance partner trust by making defection more
age cooperation among alliance partners (Lich-
costly than cooperation (Hennart, 1988) so that
bach, 1996). Structural solutions are unable to
alliance partners will believe no partner has an
cover every possible free-riding loophole (Wil-
incentive to defect. Classic structural solutions
liamson, 1979), are only effective “if [they] can be
involve sanctions, such as threatening monetary
done cheaply enough” (Telser, 1980: 28), and can
fines or the spread of bad reputations for defect-
negatively affect how alliance partners think
ing alliance partners (Das & Teng, 2002; Zeng &
about the alliance. Regarding this last point,
Chen, 2003), and contracting, such as relation-
contracting (for example) leads alliance part-
ship-specific investments (Parkhe, 1993), credi-
ners to trust the contract rather than each other,
ble commitments (Williamson, 1983), and self-
leaving the relationship vulnerable when the
enforcing agreements (Telser, 1980).
contract is repealed or a weakness in the con-
The motivational approach attempts to in-
tract is found (Malhotra & Murnighan, 2002; Sit-
crease trust by altering how alliance partners
kin & Roth, 1993). Along these lines, Malhotra
perceive each other (Dyer & Singh, 1998; Zaheer
and Lumineau’s (in press) field study on inter-
& Venkatraman, 1995) so that they will believe
firm contracting shows that when disputes
no partner is likely to defect, even if there ap-
arise, goodwill trust and willingness to continue
pears to be an incentive to do so. Traditional
collaboration among alliance partners decrease
motivational solutions entail alliance partners’
as the number of control provisions in the con-
communicating cooperative intent before joint
tract increase. As for sanctions, threatening
investments are made (Monge et al., 1998) and
punishment not only alters the payoffs associ-
ated with cooperation and defection but also
reframes how alliance partners perceive such
(Kaufman & Kerr, 1993; Kollock, 1998). The co-opetition para-
digm (Brandenburger & Nalebuff, 1996) and integrative ne-
behaviors. Tenbrunsel and Messick (1999) found
gotiation paradigm (Hoffman et al., 1999) assume that coop- that using sanctions to enforce cooperation di-
eration (or creating value) is independent of defection (or verts alliance partner attention from the moral
claiming value): individuals acting in their own self- consequences of defection to the economic con-
interests can lead to an outcome that is in everyone’s best sequences of being caught defecting. This shift
interests. Thus, whereas co-opetition and integrative nego-
tiation paradigms may be modeled using an invisible-hand of consideration actually can increase defection.
game (Miller, 1993), social dilemmas are modeled using n- Motivational solutions alone have limitations
prisoners’ (and assurance) dilemma games (Dawes, 1980; as well because of their reliance on verbal re-
Liebrand, 1983). assurances of alliance partner trustworthiness.
3
Other scholarship has used alternative typologies for Absent action, such verbal reassurance may be
grouping social dilemma solutions. For example, social psy-
chology and operations management research codify social viewed as “cheap talk,” leaving partners still
dilemma solutions as a function of whether solutions are suspicious of each other’s actual intent (Craw-
unilateral or jointly implemented (McCarter & Northcraft, ford, 1998; Farrell & Rabin, 1996). Further, alli-
2007; Messick & Brewer, 1983). To remain consistent with ance partners typically believe themselves to be
previous research applying the social dilemma paradigm to
more honest and more cooperative (i.e., more
alliance management, we delineate solutions as to whether
they change how alliance partners perceive either the struc- trustworthy) than others, which also fuels suspi-
ture of the alliance or the other exchange partners in the cion of others and, hence, defensive defection
alliance. (Epley, Caruso, & Bazerman, 2006). Indeed, re-624 Academy of Management Review October
search on international alliances shows that about how the phone system works, along with
trust is fostered only after alliance partners con- the compatibility and benefits of the system.
firm their intentions through investment in the Real options change the structure of an invest-
alliance (Robson, Katsikeas, & Bello, 2008). ment by separating resource allocation into a
These findings highlight that neither a struc- multistage process, with an option to exit at the
tural nor a motivational solution, indepen- conclusion of any stage (Adner & Levinthal,
dently, is sufficient to alleviate social uncer- 2004; Mu, 2006). While real options take a variety
tainty in strategic alliance social dilemmas, of forms (Trigeorgis, 1999), the focus here is on
suggesting the need to identify a governance “time-to-build” real options (also called
mechanism that incorporates elements of both “growth” or “compound” options), in which an
solutions (Foddy, Smithson, Schneider, & Hogg, individual firm’s investments come in multiple
1999). Research in strategic management sup- stages (Sing, 2002; Smit & Trigeorgis, 2004). In
ports this conclusion (Agarwal et al., 2010): a the first stage the firm—facing high environ-
laboratory simulation of cooperation in alli- mental uncertainty—acquires the real option by
ances found that mutual cooperation was high- investing a small amount of resources, which
est when both economic incentives were provides the firm an opportunity to uncover ad-
aligned (a structural solution) and communica- ditional information about the environment so
tion was allowed among alliance partners (a as to make more informed later-stage invest-
motivational solution). In this article we intro- ment decisions. In the later stages the firm—
duce collective real options as a hybrid solution having learned more about the environment—
that provides a more effective alternative than exercises the option of either investing
existing alliance governance mechanisms. additional resources or abandoning the venture.
A pilot project represents a time-to-build real
option (Ziedonis, 2007). A firm makes an initial
investment in a small version of the venture (a
REAL OPTIONS IN STRATEGIC ALLIANCES
pilot project) to see whether further investments
A real option is a right—without an obliga- might hold promise; the firm then can exercise
tion—to invest resources (e.g., labor, money, the real option created by the pilot project by
time) toward a course of action at a future point making subsequent larger investments in the
in time (Dixit & Pindyck, 1994). Because of the venture if the results from the pilot project are
environmental uncertainty inherent in making favorable (Fawcett, Magnan, & McCarter, 2008;
any investment (Hobfoll, 1998), real options pro- Kim & Sanders, 2002). In making the initial
vide an attractive “psychological hedge” in the smaller investment, the time-to-build real option
minds of investors (Bazerman & Gillespie, 1999; is acquired. In making the subsequent larger
Malhotra, 2005). Real options allow investors to investments, the time-to-build real option is ex-
feel less vulnerable to the negative effects of ercised (Li, James, Madhaven, & Mahoney, 2007).
investment failure (Meyerson, Weick, & Kramer,
1996). Environmental uncertainty is mitigated
Collective Real Options
because real options delay full investment until
critical information about the environment (such A collective real option is an action under-
as market conditions, profitability of a potential taken jointly by alliance partners when all part-
investment, and entry of new competitors) can ners agree to make a small initial investment of
be revealed over time (Bowman & Hurry, 1993). resources to uncover environmental and social
An example of how real options reduce environ- information about the possible success of a sub-
mental uncertainty is a university that is consid- sequent larger-scale alliance initiative. Envi-
ering adoption of a new technology, such as a ronmental information refers to factors exoge-
campus phone system (Ziedonis, 2007). Prior to nous to the alliance (such as market forces) that
investing large amounts of resources to roll out influence the anticipated benefits of collective
the new phone system campus wide, the univer- action (such as revenues and enhanced reputa-
sity can take the real option of field testing the tion). Social information refers to factors endog-
system in one campus building. After the field enous to the alliance that influence the antici-
test, environmental uncertainty is reduced be- pated benefits of collective action—specifically,
cause the university has additional information the perceived trustworthiness of the alliance2011 McCarter, Mahoney, and Northcraft 625
partners (Gulati, 1998), including their integrity, and it is this coordinated action that provides
reliability, and commitment. Acquiring the col- the opportunity to reduce social uncertainty in
lective real option occurs when the alliance the alliance.
partners jointly invest in the small-scale (e.g., The second difference between collective real
pilot project) initiative. This definition high- options and traditional real option concepts re-
lights two key elements that separate collective lates to the timing and benefits provided. Previ-
real options from traditional conceptualizations ous theorizing on time-to-build real options pos-
of real options and related constructs. its that their benefits are not realized until after
First, traditional research on real options typ- the pilot venture is completed and that these
ically has examined how individuals (or individ- benefits entail mitigating environmental uncer-
ual firms) acquire real options to reduce uncer- tainty (Majd & Pindyck, 1987). However, as sug-
tainty when making investment decisions gested by Dixit and Pindyck (1994) and Li et al.
(Bowman & Moskowitz, 2001; Folta & Miller, (2007), collective real options provide social in-
2002). However, as Li et al. (2007) suggest, real formation as well as environmental information.
options may be acquired and exercised by col- Social information can surface prior to the com-
lectives, such as an alliance. Although collec- pletion of the pilot venture. It is the uncovering
tive real options are created by the collective of information about the intentions, reliability,
(e.g., by an alliance), later they can be exercised and commitment of alliance partners during the
independently by each member of the collective pilot project that enables collective real options
(Pape & Schmidt-Tank, 2004)—that is, each mem- to change how alliance partners perceive each
ber of the collective can decide independently other. For example, empirical work on technol-
whether to fulfill or decline the subsequent ogy outsourcing shows that alliance partners
larger-scale investment. For example, the alli- jointly may make small initial investments to-
ance partners might collectively choose to pref- ward alliance initiatives to uncover more infor-
ace a large-scale alliance initiative with a pilot mation about each other’s intentions before
project (Fawcett et al., 2008), such as developing committing large amounts of resources to the
a new recycling process and pilot testing its alliance (van de Vrande, Vanhaverbeke, & Duys-
effects on a select region prior to agreeing to ters, 2009). This refocusing of the benefits of real
jointly develop a national network of recycling options from managing environmental uncer-
plants. The focus on joint (rather than unilateral) tainty to managing social uncertainty distin-
implementation distinguishes collective real guishes collective real options from similar con-
options from similar concepts, such as incre- cepts, such as contingency strategies (Nielsen,
mental risk taking (Andriopoulos & Lowe, 2000) 1988) and contingent contracts (Bazerman & Gil-
and reciprocal exchange (Gouldner, 1960; Molm, lespie, 1999), both of which are concerned with
Takahashi, & Petersen, 2000). While both are managing environmental uncertainty (Laak-
additional paths to managing social uncer- sonen, Jarimo, & Kulmala, 2009).
tainty, incremental risk taking (Dunbar, Neufeld, Collective real options bridge structural and
Libow, Cohen, & Foley, 1997) and reciprocal ex- motivational approaches to create a hybrid so-
change (Komorita, Chan, & Parks, 1993) are ac-
tions undertaken by an individual independent
social uncertainty (or risk of nonreciprocity; Lawler, 2001;
of other alliance partners. The acquisition of Molm, Collett, & Schaefer, 2007). Similar to other traditional
collective real options, in contrast, represents structural solutions (such as contracting and self-enforcing
coordinated action among alliance partners,4 commitments), negotiated exchange involves “agreements
that are strictly binding . . . they automatically produce the
benefits agreed upon” (Molm et al., 2007: 209), resulting in
“the risk [or perceived vulnerability] of non-reciprocity [be-
4
Reciprocial exchange can involve negotiated exchange, ing] eliminated” (2007: 212). Without social uncertainty, be-
where exchange partners create an agreement that “speci- liefs about another’s motives and intentions (i.e., trustwor-
fies the benefits that each actor will receive from the ex- thiness) are difficult to form, leaving room for suspicion
change” (Molm, 2010: 122), and these benefits can flow bilat- when the ability to enforce agreement ends or loopholes are
erally (Molm, 2003). Negotiated exchange is complementary discovered (Malhotra & Lumineau, 2011; Malhotra & Mur-
to collective real options. Resources are jointly invested into nighan, 2002). Thus, the collective real option framework
the relationship, strengthening the level of trust among the complements negotiated exchange (and other structural so-
partners. However, what distinguishes collective real op- lutions) and offers an alternative avenue for alliance part-
tions from a negotiated-exchange strategy is the presence of ners to increase mutual cooperation.626 Academy of Management Review October
lution to alliance governance. As a structural small win demonstrating the viability of the
solution, collective real options change the in- larger-scale cooperative effort (i.e., “proof-of-
centives of alliance participation by lowering concept”; McAdam, McAdam, Galbraith, &
the costs of cooperation (rather than by increas- Miller, 2010) and a small win demonstrating the
ing the costs of defecting, as other structural trustworthiness of alliance partners. Exploring
solutions do): large-scale alliance participation the viability of a collective venture reduces en-
is buffered by an initial less-costly (pilot project) vironmental uncertainty, while exploring the
initiative. As a motivational solution, collective trustworthiness of alliance partners reduces so-
real options reveal—through pilot project in- cial uncertainty. Because the success of the
vestment behaviors—alliance partners’ integ- small initial investment lowers the environmen-
rity, reliability, and commitment. By credibly tal and social uncertainty barriers to subse-
signaling intent in the pilot project, alliance quent larger-scale alliance investments, the
partners influence how others perceive them achievement of these small wins propels the
through actions and not just words. exchange parties forward to invest in larger-
scale collective efforts (Reay, Golden-Biddle, &
Germann, 2006). Whereas traditional research
Collective Real Options and Small Wins
on real options focuses on achieving small wins
Using collective real options to manage social to mitigate environmental uncertainty, this arti-
uncertainty can best be understood as an impor- cle shifts the focus to understanding how collec-
tant opportunity for a relational small win. tive real options reduce social uncertainty by
Small wins are “concrete, complete, imple- influencing the development of trust.
mented outcome[s] of moderate importance”
(Weick, 1984: 43), and they represent an opportu-
Relational Small Wins and Trust
nity for alliance partners to see (preview) the
potential value of successful cooperation. Relational small wins encourage subsequent
The small wins framework derives from the larger-scale cooperation in alliances by foster-
observation that most organizing efforts—for ex- ing trust and thereby reducing social uncer-
ample, alliance initiatives—are high in com- tainty. Trust is an individual’s “expectations, as-
plexity and uncertainty about both the inten- sumptions or beliefs about the likelihood that
tions of others and the outcomes of mutual another’s future actions will be beneficial, favor-
cooperation (Weick, 1984). For instance, the com- able or at least not detrimental” to the individ-
plexity of alliance partner coordination and un- ual (Robinson, 1996: 576), and it is formed
certainty about partner follow-through on prom- through social interaction. When individuals in-
ised material and intellectual contributions are teract, they bring with them values about what
ever-present barriers to developing successful behaviors are appropriate in a relationship,
interfirm initiatives (Wilson & Douglas, 2007). along with expectations about behaviors they
The high complexity and uncertainty surround- want from others, and they experience emotions
ing large-scale collective investment can lead based on how other individuals treat them
an alliance partner to be fearful about whether (Jones & George, 1998). Through social interac-
the other partners will cooperate. This fear often tion each individual attempts to (1) assess the
leads individuals to forgo cooperative initia- perceived values of the other (“Does the individ-
tives—to defensively defect (Hobfoll, 1989). The ual have integrity?”), (2) know whether others
small wins framework suggests that breaking a have met his or her expectations (“Did they co-
large-scale collective effort up into multiple operate or not?”), and (3) use his or her current
smaller collective efforts reduces the task’s com- emotions and moods as indicators to assess the
plexity and perceived uncertainty (Weick, 1984). quality of the relationships (“How do I feel
This, in turn, lowers the barrier for alliance part- based on how others treated me?”).
ners to invest resources in an initial smaller Research on trust in alliances references the
collective effort. The smaller collective effort real options research literature as holding
then provides the opportunity to achieve small promise in understanding how alliance partners
wins. build trust (Das & Teng, 1998). Trust evolves
The smaller (pilot project) collective effort, in through interaction among individuals when
fact, can provide two kinds of small wins: a values are perceived as congruent, positive ex-2011 McCarter, Mahoney, and Northcraft 627
pectations are met, and positive emotions and effort experiences relatively low exposure (Lu-
moods are experienced based on how an indi- bell, 2004). Thus, whereas perceived vulnerabil-
vidual has been treated by others (Jones & ity is about the probability of failure of mutual
George, 1998). Applying this logic to strategic cooperation in an alliance, exposure is con-
alliances, collective real options enable alli- cerned with how badly an alliance partner will
ance partners to uncover information about the be damaged if such a failure occurs. Exposure
reliability, integrity, and commitment of one an- may be central to understanding how collective
other. When the relational small wins are real- real options can impact trust formation even
ized, alliance partners know that everyone else when mutual cooperation occurs in its acquisi-
is reliable and committed to alliance success. tion.
As a consequence, the alliance partners become In summary, collective real options enable al-
more trusting of one another. liance partners to find out more information
Lewicki, McAllister, and Bies (1998) suggest about the social and task environments before
that trust should develop among alliance part- undertaking full investment in an alliance.
ners as a function of the nature of their experi- When mutual cooperation is successful in the
ence in acquiring the collective real option. How acquisition of the collective real option, the al-
the alliance experience, when acquiring a col- liance partners experience a relational small
lective real option, influences future cooperation win. The achievement of a relational small win
can best be seen through the lens of vulnerabil- reduces the uncertainty inherent in how an alli-
ity. ance partner perceives its vulnerability in the
alliance.
Trust and Perceived Vulnerability
MODELING REAL OPTIONS EFFECTS
Vulnerability has been defined variously as
First we try, then we trust! (Sean Connery in the
an individual’s biased assessment of a risky film Entrapment).
situation (Cho & Lee, 2006), composed of the de-
gree of variability in the outcome of the pur- As suggested by the above quotation, collec-
chase (or investment; Cox, 1967) and the amount tive real options enable alliance partners to
of resources that can be lost (Cunningham, “first try” to see if they can trust one another,
1967); as an individual’s assessed probability of prior to making full investments in the alliance.
an investment’s success and the confidence in In this section we articulate a series of proposi-
that assessment (Sitkin & Pablo, 1992; Sitkin & tions modeling how collective real options influ-
Weingart, 1995); and as a function of the social ence trust formation in alliance relationships
environment and the extent to which an individ- and, as a consequence, subsequent large-scale
ual believes his or her resources are in danger alliance investment. Propositions 1 through 3
of being exploited by others (Meyerson et al., constitute the base model. Successful mutual
1996). The recurring theme in these conceptual- cooperation when the alliance acquires the real
izations is that vulnerability is an individual’s option provides a relational small win, thereby
perception of risk subject to situational and so- increasing trust among the alliance partners.
cial influences. Here we define perceived vul- The increase in trust among the alliance part-
nerability as an individual’s assessment of un- ners reduces the perceived vulnerability from
certainty about the future behavior of others social uncertainty, so (as a consequence) alli-
and/or the environment. ance partners should be more willing to cooper-
Exposure is the amount an individual alliance ate by fully investing in the alliance. Proposi-
partner must risk losing (in proportion to that tions 4 through 7 extend the core model by
partner’s total wealth) to achieve the benefits considering how characteristics of the collective
from the investment (Cho & Lee, 2006; Cunning- real option (i.e., exposure) and the alliance part-
ham, 1967). For example, an alliance partner ners (i.e., reputation) influence trust formation.
risking 90 percent of its total wealth toward the Figure 1 provides a conceptual roadmap outlin-
collective effort experiences relatively high ex- ing these relationships.
posure, while an alliance partner risking 10 per- Two boundary conditions are necessary to un-
cent of its total wealth toward the collective derstand the impact of collective real options628 Academy of Management Review October
FIGURE 1
A Conceptual Model of a Collective Real Options Approach to the
Social Dilemma of Strategic Alliances
and the relational small wins they can produce. trust each other on some matters but not others.
First, the attainment of a small win requires the As suggested by Hardin (1993), individual A can
external environment both to be promising and trust individual B to do action X but not action Y,
to provide the alliance partners a favorable re- because either action Y is different from action X
turn on the real option (assuming mutual coop- or because previous experience has shown that
eration occurs). Otherwise, the firms have an individual B cannot (for whatever reason) do
incentive to exit the venture by not exercising action Y. This insight is illustrated in Larson’s
the option of further investment (Adner & (1992) case study of the formation of interfirm
Levinthal, 2004). This first assumption stems ties: although some alliance partners had con-
from recent social dilemma research showing siderable positive history with each other out-
that social uncertainty and environmental influ- side the alliance context, they nevertheless un-
ence are distinct determinants of alliance coop- dertook a “trial phase” to prove each other’s
eration: individuals will defect in the presence trustworthiness related to alliance activities.
of environmental uncertainty irrespective of Drawing from this approach, our theorizing pos-
how much they trust their alliance partners (Mc- its that, in relation to elements specific to the
Carter et al., 2010). strategic alliance, prospective alliance partners
Second, our model is based on the psycholog- begin with low trust. This assumption does not
ical approach to trust (see Lewicki, Tomlinson, & remove the possibility that the alliance partners
Gillespie, 2006, for a review), which assumes share a positive history outside the alliance (as
that trust among individuals begins low (at the will be discussed later); it leaves room for the
conceptual level of zero) and the first encounter alliance partners to enlarge the breadth of their
among individuals acts as a validation opportu- relationship through small wins germane to al-
nity regarding trustworthiness (Jones & George, liance initiatives.
1998). If the encounter is positive, then trust in-
creases as a function of the amount of the risk
Relational Small Wins of Mutual Cooperation
embedded in the encounter. The psychological
and Trust
approach also assumes that trust is not mono-
lithic across the breadth of the relationship Successful mutual cooperation increases trust
among alliance partners. Lewicki and col- among alliance partners. When alliance part-
leagues (1998) observed that relationships are ners invest in the small-scale alliance initiative,
complex and multifaceted. Thus, existing rela- they acquire an opportunity to experience suc-
tionships may be composed of individuals who cessful mutual cooperation in the form of a col-2011 McCarter, Mahoney, and Northcraft 629
lective real option. Significantly, acquiring a Proposition 2a: As trust among alli-
collective real option is not necessarily a one- ance partners increases, perceived
time event. Agreeing to invest in an initial vulnerability experienced by those al-
small-scale project must be followed up with liance partners decreases.
fulfillment of promised capital, personnel, and
Proposition 2b: Trust mediates the re-
commitment throughout the time period of the
lationship between relational small
pilot project (Robson et al., 2008).
wins and perceived vulnerability of
The real options research literature holds
alliance partners.
promise for understanding how “incremental re-
source commitments” could increase trust
among alliance partners (Das & Teng, 1998: 504). Vulnerability and Investment in
An alliance’s successful mutual cooperation in Strategic Alliances
acquiring a collective real option creates an ex-
change history among the alliance partners Perceived vulnerability negatively affects an
(Gulati, 1995a). This exchange history entails al- alliance partner’s willingness to contribute re-
liance partners’ uncovering one another’s val- sources toward the alliance. In social dilemmas
ues, expectations, and feelings about each other an individual experiences high perceived vul-
(such as integrity, reliability, and commitment nerability from uncertainty (about others or the
to the alliance)—all of which influence the for- value of cooperation) and the fact that resources
mation of trust (Jones & George, 1998; Poppo, must be risked. Resource conservation theories
Zhou, & Sungmin, 2008). As a result, the “predict- maintain that when individuals perceive that
ability” of alliance partners’ future behavior in- their resources are threatened with a potentially
creases (Weick, 1984) and trust (as a conse- irreversible loss, they experience stress and
quence) is enhanced (Sniezek, May, & Sawyer, therefore seek to avoid the threat by selecting a
1990; Vangen & Huxham, 2003): course of action that minimizes the likelihood
their resources will be lost (Hobfoll, 1998). Con-
Proposition 1: Alliances that experi- sumer purchasing theories make a similar pre-
ence the relational small wins of mu- diction about individuals facing risky invest-
tual cooperation by successfully ac- ments: high perceived risk (vulnerability) in an
quiring collective real options will investment motivates individuals to engage in
experience higher trust compared to “risk reduction strategies”— one of which is
alliances that do not experience rela- avoiding the investment (Cox, 1967). In social
tional small wins of mutual coopera- dilemma terms, risk reduction likely equates to
tion by successfully acquiring collec- defecting (or undercontributing) toward the alli-
tive real options. ance public good.
Proposition 3: An alliance partner will
be less likely to contribute toward the
Trust and Perceived Vulnerability alliance’s initiatives when perceived
Trust is a critical psychological factor in de- vulnerability is higher.
termining the level of perceived vulnerability
that an exchange partner experiences in an al-
liance. When trust is high, social uncertainty Exposure and Trust Development
about the future behavior of others is minimized Recent research on trust development in alli-
(Kollock, 1994). A decrease in social uncertainty ances suggests that it is not the length of the
makes alliance partners less fearful that their history among alliance partners that develops
investment will be lost because of free riding or trust but, rather, the quality of that history (Gu-
defensive defection (Schnake, 1991; Sniezek et lati & Sytch, 2008; Lewicki et al., 2006). Exposure
al., 1990). Perceived vulnerability is therefore re- when acquiring the collective real option may
duced among alliance partners. Trust will me- have a direct effect on whether a relational
diate the relationship between relational small small win is achieved and what impact that
wins of mutual cooperation in acquiring the col- relational small win has on trust development
lective real option and perceived vulnerability. among alliance partners.630 Academy of Management Review October
High exposure in creating the collective real relational small wins and trust can be found in
option may legislate against ever attempting a attribution theory, which suggests that causal
relational small win. Recent work on real op- attributions for an individual’s compliance to a
tions and organizational learning maintains request are a function of the request’s perceived
that if trial-and-error learning (e.g., via a pilot difficulty (Heider, 1958; Kelley, 1973). When alli-
project) is too risky (i.e., exposure is high), firms ance partners invest small amounts of resources
will be reluctant to invest in pilot projects (Fich- toward the acquisition of a collective real op-
man, 2004). Similarly, if acquiring a collective tion, that cooperation is more likely to be attrib-
real option requires high levels of exposure, al- uted to the low exposure of the request rather
liance partners may be particularly likely to ex- than to the commitment (or trustworthiness) of
perience higher vulnerability from social uncer- the alliance partners (Malhotra & Murnighan,
tainty (Cho & Lee, 2006; Hobfoll, 1998). 2002).
Experiencing heightened perceived vulnerabil- Empirical results from experimental social
ity, alliance partners will be inclined to under- psychology support this explanation of trust de-
invest (or not invest at all) in the acquisition of velopment. In the “trust game” Pillutla, Malho-
the collective real option. As a consequence of tra, and Murnighan (2003) found that large offers
this underinvestment, the opportunity for from the sender are attributed by receivers as
achieving the relational small win of mutual signals of high trust while low offers are not. In
cooperation is less likely. The unsuccessful other words, the extent to which an alliance
achievement of a relational small win when ac- partner risks exposure influences how commit-
quiring the collective real option will keep trust ted to the alliance (and, thus, trustworthy) that
low among alliance partners. Trust will be low alliance partner is perceived to be.
because the unsuccessful achievement of mu-
Proposition 4a: As exposure in acquir-
tual cooperation will create a trust breach, de-
ing the collective real option de-
creasing an alliance partner’s trust in the other
creases, the likelihood of investment
partners’ intentions to support the alliance
by alliance partners to realize a rela-
(Lount, Zhong, Sivanathan, & Murnighan, 2008;
tional small win of mutual coopera-
Stout, 2008).
tion increases.
Considering the potential negative outcome
of high exposure, an initial strategy to elicit Proposition 4b: Exposure in acquiring
cooperation among alliance partners may be to the collective real option will moder-
ask for small investments (minimizing exposure) ate the relationship between rela-
when acquiring the collective real option. How- tional small wins and trust such that
ever, low exposure when acquiring the collec- relational small wins of mutual coop-
tive real option may undermine the develop- eration will have a less beneficial ef-
ment of trust, even when the relational small fect on trust among alliance partners
win of mutual cooperation is realized in acquir- when exposure is low than when ex-
ing the collective real option. Molm et al. note posure is high.
that “both structural risk and trustworthy behav-
iors are necessary for the development of trust”
Partner Reputation and Trust Development
(2007: 212). Indeed, extant trust research pro-
poses that without “enough vulnerability” The previous section considered how a char-
among exchange parties, trust cannot success- acteristic of the collective real option— exposure
fully form (Mayer, Davis, & Schoorman, 1995; required of alliance partners—influences an al-
Ribstein, 2000). Empirical research on the “foot- liance partner’s cooperation, when acquiring
in-the-door” technique supports this claim: get- the collective real option, and trust formation. A
ting individuals to make a small initial invest- characteristic of the alliance partners them-
ment increases the likelihood they will comply selves—reputation—also may impact both col-
with larger requests in the future, and future lective real option acquisition and its effects on
compliance increases when the initial request is trust development. Reputation refers to informa-
moderate rather than low (Segilman, Bush, & tion about how an alliance partner performed
Kirsch, 1976). An explanation for this moderating previously in separate ventures (Shane & Cable,
effect of exposure on the relationship between 2002). Reputational information is a powerful so-2011 McCarter, Mahoney, and Northcraft 631
cial-structural element of alliance relationships achievement of relational small wins will more
(Gulati, 1995b) and can be acquired either first- likely be treated as confirmation of the reliabil-
hand, through direct social encounters (Ander- ity and trustworthiness of the alliance partners.
son & Shirako, 2008), or secondhand, through This should enhance trust more than if partners
social networks (Kogut, Shan, & Walker, 1992). initially have little or no reputational informa-
Reputation should directly influence coopera- tion to draw on.
tion in alliance initiatives because it provides
Proposition 6: Achieving relational
alliance partners with information about one
small wins will have more of an effect
another’s values, expectations, and feelings
on trust among alliance partners when
(Jones & George, 1998) and thereby signals alli-
reputations are positive than when
ance partners about the predictability of each
there is no reputation information.
other’s future behavior (Bolton, Katok, & Ock-
enfels, 2005; Kollock, 1994), thus reducing so- What happens when alliance partners fail to
cial uncertainty. Individuals who perceive one achieve mutual cooperation? Alliance partner
another as having a positive (cooperative) rep- reputation may act as a buffer between collec-
utation are more likely to cooperate in situa- tive failure and potential distrust among alli-
tions requiring trust (Weber, Malhotra, & Mur- ance partners. If an alliance partner has a pos-
nighan, 2005). Bohnet and Huck (2004) found itive reputation (history of being cooperative),
that allowing participants to know their cur- defection would likely be attributed to an exter-
rent partners’ history of cooperation with pre- nal cause, such as the riskiness or difficulty of
vious partners (and that the history was posi- cooperating, and the defecting alliance partner
tive) increased their willingness to initially would less likely be viewed as untrustworthy by
cooperate. This empirical finding suggests the other partner(s). In support of this position,
that other partners’ reputations should influ- Lount et al. (2008) found that participants in an
ence an alliance partner’s willingness to ac- iterated prisoner’s dilemma who experienced
quire a collective real option. partner defection after a history of partner coop-
eration were more willing to continue to trust
Proposition 5: The positive reputations
their partner, suggesting they attributed that de-
of other alliance partners increase the
fection to something external.
likelihood of investment by alliance
partners to realize a relational small Proposition 7: Failure to achieve small
win of mutual cooperation. wins will have less of an effect on trust
among alliance partners when repu-
Reputation may also indirectly influence trust
tations are positive than when there is
formation, when attempting relational small
no reputation information.
wins, by buffering the impact of real option suc-
cess or failure on subsequent trust development In summary, a collective real option creates
and alliance partner investments in alliance ac- the opportunity for alliances to achieve rela-
tivities. Kelley and Michela’s (1980) review of tional small wins of mutual cooperation. These
attribution theory is one of the first to suggest relational small wins build trust, and high trust
that an individual’s reputation can influence increases an alliance partner’s willingness to
how others attribute the cause of the individu- contribute toward the alliance by lowering that
al’s current behavior. Alliance partners may be partner’s perceived vulnerability. Exposure,
buffered (positively or negatively) by their rep- when acquiring the collective real option, di-
utations, which may impact how they are per- rectly affects the likelihood of mutual coopera-
ceived after small wins are attempted (Jones, tion and also, when a small win is achieved,
Jones, & Little, 2000). Similar to our discussion moderates the relationship between mutual co-
about the moderating effect of exposure, collec- operation and the development of trust among
tive members often seek causal explanations for alliance partners. Reputation of the alliance
the outcomes of their collective efforts (Stouten, partners directly affects the likelihood of mutual
De Cremer, & van Dijk, 2006; Weber, Camerer, cooperation and also moderates the relation-
Rottenstreich, & Knez, 2001). If reputation among ship between mutual cooperation success or
alliance partners is positive (from a previous failure and the development of trust among al-
history of cooperative behavior), then successful liance partners.632 Academy of Management Review October
DISCUSSION AND CONCLUSIONS it means to fulfill one’s obligations to the alli-
ance. Tenbrunsel and Northcraft (2010) note that
Past research on real options has focused pri-
not all defection in alliances need be inten-
marily on individual strategies for mitigating
tional. Any partner brings to an alliance idio-
environmental uncertainty. Here we draw from
syncratic views of what it means to be a “good
the small wins literature and social dilemma
partner” and may be fulfilling its own views
literature to introduce the concept of collective
even while not fulfilling what other alliance
real options, and we explore how collective real
partners think is fair. Furthermore, although a
options may impact the management of social
“logic of appropriateness” typically governs so-
uncertainty in strategic alliance social dilem-
cial exchanges (e.g., March, 1994; McCarter, Bu-
mas.
descu, & Scheffran, 2011), which logic of appro-
priateness applies in a particular situation may
differ among partners, depending on how they
Theoretical Implications
frame the alliance activity (e.g., as investing
A real options approach to navigating strate- versus cooperating). For these reasons talk not
gic alliance social dilemmas provides an under- only may be cheap but may also insufficiently
standing of how an alliance may be structured reveal what is in the minds of alliance partners.
to manage not only how exchange partners Happily, collective real options may provide
think about the alliance but also how they think both an opportunity to reveal differences of
about each other (Kollock, 1998; Zeng & Chen, opinion and perspective along these lines and a
2003). Collective real options enable alliance forum for reconciling those differences to form a
partners to hedge risk from the market environ- more perfect union during any later, larger col-
ment and to hedge risk from the social environ- laboration.
ment by allowing them to “first try, then trust” Given these concerns, it should not be surpris-
each other on larger alliance initiatives. In try- ing that alliance research has discovered a pe-
ing, alliance partners credibly reveal their coop- riod of “ambivalence” existing among partners
erative intentions through actions (i.e., investing in the early stages of an alliance (Gulati &
in the collective real option), and these actions Sytch, 2008). This period of ambivalence may
affect how other alliance partners perceive well reflect a tension within alliance partners to
them. approach the value creation opportunities alli-
Using collective real options to “test the wa- ances promise but to avoid the vulnerability
ters” of cooperating with alliance partners may that alliance cooperation inevitability entails.
be particularly useful in cases where intangible The attribution perspective on exposure and
contributions will have a lot to say in determin- trust development explored here suggests that
ing whether the alliance succeeds or fails. Re- erring on the side of too much caution— by
lated governance approaches, such as negoti- avoiding any vulnerability in early-stage alli-
ated exchange (e.g., Molm, 2010), may be good at ance initiatives—may doom an alliance to never
creating enforcement mechanisms for concrete develop enough trust. Alliances that use collec-
alliance contributions that can be monitored tive real options as a means to develop trust,
and confirmed, but it is difficult to legislate try- therefore, must search out the happy medium
ing hard and thinking hard. As suggested by whereby alliance partners are exposed enough
Sesil, when it is “difficult to monitor effort,” a to experience the development of trust, but not
mechanism that signals cooperative intent is so much as to discourage investment in the col-
helpful to increase trust and cooperation among lective real option. Because collective real op-
alliance partners (2006: 593). For such resourc- tions reveal the true character of alliance part-
es—and particularly when those resources will ners, the very willingness of partners to acquire
partly determine whether an alliance suc- a collective real option—and thus risk revealing
ceeds—alliance partners may need the proof of their character—suggests something about their
a relational small win to feel comfortable going intent to cooperate. This is not to suggest that a
the distance with other alliance partners. partner could not fake collaborative posture dur-
Collective real options may also prove partic- ing a pilot venture with the intention of behav-
ularly useful in revealing whether all alliance ing more opportunistically later. However, the
partners are on the same page in terms of what larger the exposure an alliance partner has in2011 McCarter, Mahoney, and Northcraft 633
the pilot venture, perhaps the more likely that ever, McCarter and Northcraft (2007) suggest
partner’s true character will be revealed. that even if an alliance is structured to remove
Our analysis of the impact of collective real offensive defection, such as by making all part-
options also offers a new perspective on the role ners irreplaceable, an alliance partner may still
of reputations in alliances. Previous research on doubt others’ ability and willingness to cooper-
trust and cooperation has highlighted the main ate and defect defensively. This insight reminds
effects of reputations on trust (Bohnet & Huck, us that alliance management is not only about
2004; Das & Teng, 2001; Weber et al., 2005). Our reducing opportunism but also about a partner
arguments shift the focus of reputations to their reducing other partners’ fear they will be made
indirect effects on partners’ interpretations of the “sucker” (Schnake, 1991). Whereas recent al-
success and failure of pilot ventures. Specifi- liance research considers perceptions when ex-
cally, a positive reputation may serve to en- amining alliance governance (McCarter &
hance the impact of cooperative behaviors on Northcraft, 2007), our model extends this think-
trust development when relational small wins ing by using real options to show one way to
are achieved, and it may buffer an alliance part- manage defensive perceptions among alliance
ner from being viewed as untrustworthy when partners. Rather than avoiding opportunism by
small wins are not achieved. Significantly, any others, the emphasis of this approach is avoid-
discussion of reputations in alliances begs ing missed opportunities to create value by low-
some interesting levels-of-analysis issues, such ering the barriers to collective cooperation.
as whether alliances can trust, or whether trust Solutions to managing social uncertainty
is really the province of individual decision might be applicable to strategic alliances where
makers. This becomes particularly important
offensive defection is either possible or not. Our
when we contemplate the possibility that repu-
model applies to both cases. If the alliance is
tation within a multiple-actor partner may not
structured to allow free riding, collective real
be homogeneous or that any partner’s reputa-
options enable familiarity among alliance part-
tion may not be homogeneously perceived
ners and the achievement of shared benefits,
within an alliance, or even within other partners
thereby reducing the temptation to opportunis-
(Arend, 2009). Although beyond the scope of this
tically hurt others and potentially kill the golden
article, such questions bring numbers into the
goose (Pruitt & Kimmel, 1977). If an alliance
discussion— both the number of partners in the
structure makes free riding impossible, then col-
alliance and the number of actors within any
“partner”—since trust becomes a harder nut to lective real options enable alliance partners to
crack as group size increases (e.g., Huck & Lun- garner trust and show both the motivation and
ser, 2010), probably on both dimensions. the capability to cooperate as the stakes grow
Another critical contribution of this article is (Chen, 1996).
that it redirects scholarly conversation about al- This article also provides several contribu-
liance governance, from managing opportunism tions to the study of real options. First, previous
to managing defensive perceptions. Previous re- research has examined how real options man-
search, such as game-theoretic and transaction age external market environments and mitigate
cost approaches to alliance management, has environmental uncertainty to encourage alli-
emphasized governing alliances to avoid offen- ance success (Kogut, 1991). Here we show how
sive (opportunistic) defection (Heide & Miner, real options can manage and reduce social un-
1992; Parkhe, 1993). Recent research on applying certainty among alliance partners, thereby an-
the social dilemma paradigm to alliances has swering the call from scholars to understand
echoed this emphasis by focusing on structural how real options affect the social, as well as the
and motivational strategies that help protect exogenous, environment (Busby & Pitts, 1997;
“an overtrusting partner [from being an] easy Foote & Folta, 2002). In acquiring the collective
target for exploitation by its greedy partners” real option, alliance partners may uncover each
(Zeng & Chen, 2003: 588). This does not suggest other’s intentions before significant investments
that research on alliance social dilemmas ig- are made. This insight redirects discussion of
nores defensive defection. Indeed, Zeng and real options from managing the external market
Chen (2003) encouraged trust building as a environment to navigating the internal social
means of alleviating fear of opportunism. How- environment.You can also read
