Wald W5D1
TRUST AND COLLABORATION IN THE AFTERMATH OF CONFLICT: THE EFFECTS OF CONTRACT STRUCTURE
DEEPAK MALHOTRA Harvard University
FABRICE LUMINEAU University of Technology Sydney
Leveraging a longitudinal data set concerning 102 interfirm disputes, we evaluate the effects of contract structure on trust and on the likelihood of continued collaboration. We theoretically refine and empirically extend prior research by (1) distinguishing between the control and coordination functions of contracts, (2) separating goodwill- based and competence-based trust, and (3) evaluating the effects of contract structure on relational outcomes in the context of disputes. We find that control provisions increase competence-based trust but reduce goodwill-based trust, resulting in a net decrease in the likelihood of continued collaboration. Coordination provisions in- crease competence-based trust, leading to an increased likelihood of continued collaboration.
Interfirm relationships allow firms to create value and build competitive advantage (Agarwal, Croson, & Mahoney, 2010), but cooperation in such relationships is neither automatic nor easily fos- tered. Two key impediments to cooperation are the threat of exploitation by an opportunistic exchange partner (Williamson, 1985) and the possibility of coordination failures that can derail the efforts of even well-intentioned parties (Gulati, Lawrence, & Puranam, 2005; Knez & Camerer, 2000). Recogniz- ing that mixed motives underlie most exchange relationships (e.g., Kogut, 1988) and that coordinat- ing partners’ expectations and actions is inherently difficult (Camerer, 2003), firms rely on contracts to mitigate risks, facilitate coordination, and promote cooperation (Ring & Van de Ven, 1992).
Despite the use of contracts to facilitate coordi- nation and control, however, interfirm disputes can emerge. This raises questions regarding the kind of relationship that will emerge and the viability of continued collaboration after parties have been un- successful in preventing conflict. In this study, we
examine these dynamics by evaluating how con- tract structure affects trust, and subsequently, con- tracting parties’ intent to continue collaboration, in the context of interfirm disputes. We extend prior research on the effects of contracts on trust by (1) distinguishing between the control and coordina- tion functions of contracts, (2) distinguishing be- tween the goodwill and competence dimensions of trust judgments, and (3) evaluating these relation- ships in the context of interfirm conflict.
We argue that a more nuanced approach that encompasses the different functions of contracts (coordination versus control) and the different di- mensions of trust judgments (goodwill versus com- petence) may provide a more complete assessment of the effects of contracts on trust and collabora- tion. Prior research has often focused narrowly on a subset of these distinctions and in some cases over- looked these distinctions altogether (Puranam & Vanneste, 2009). Furthermore, our analysis of firms’ willingness to continue relationships after suffering costly disputes allows us to evaluate the mechanisms underlying an important but rarely studied aspect of interfirm exchange: relationship repair (Zaheer, Lofstrom, & George, 2002).
To test our hypotheses, we use a rich data set comprising more than 150,000 pages of details re- garding 102 business disputes arising in vertical exchange relationships in the years 1991–2005. The data include a wide range of contractual and exchange characteristics for each relationship, along with thousands of pages of communication between the disputants. The contracts enabled us to codify the degree to which control and coordi-
We thank our associate editor, Wenpin Tsai, and three anonymous reviewers for their constructive feedback throughout the review process. We also thank Jean Beuve, Marc Fréchet, Michael Jensen, and Kathleen McGinn for their comments on earlier versions of the paper. We are grateful to the law firm that gave us access to data and to the lawyers and law professors for sharing their time and expertise with us.
Editor’s note: The manuscript for this article was ac- cepted for publication during the term of AMJ’s previous editor, Duane Ireland.
� Academy of Management Journal 2011, Vol. 54, No. 5, 981–998. http://dx.doi.org/10.5465/amj.2009.0683
981
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.
nation provisions were incorporated into agree- ments, and the communications allowed us to code for statements that reveal goodwill-based and com- petence-based trust in the relationships. This is notable because, to our knowledge, the current data set is the first to provide this level of detail on interfirm conflict, and the first to allow such a fine-grained analysis of trust in interorganizational relationships.
THEORETICAL BACKGROUND AND HYPOTHESES
Contracts as Instruments of Control and Coordination
Seminal works in organization studies (Barnard, 1938; Burns & Stalker, 1961), as well as work by legal scholars (e.g., Baird, Gertner, & Picker, 1994; McAdams, 2009), decision theorists (e.g., Luce & Raiffa, 1957; Schelling, 1963), and economists (e.g., Camerer, 2003; Knez & Camerer, 2000), have previ- ously addressed the distinction between control problems (those stemming from misaligned incen- tives) and coordination problems (those stemming from misaligned expectations and behavior) in ex- change relationships (Gulati et al., 2005). Although control and coordination have often been tackled separately in the literature (Kretschmer & Puranam, 2008), some recent studies have suggested that or- ganizational structures—and, in particular, inter- firm contracts—serve both functions (e.g., Argyres, Bercovitz, & Mayer, 2007; Mellewigt, Madhok, & Weibel, 2007; Reuer & Ariño, 2007; Salbu, 1997).
Organizational scholars have long considered contracts as instruments of control (Macneil, 1978; Williamson, 1985, 1991). Interfirm collaborations have the potential to create value, but parties to such relationships must contend with the risk of exploitation by their partners (Walker & Weber, 1984; Williamson, 1985). The legal underpinnings of contracts give firms the option of sanctioning an exchange partner who is unable or unwilling to abide by agreed-upon terms (Joskow, 1987). The coordination function of contracts has received less attention (Argyres et al., 2007; Mayer & Argyres, 2004). A contract—and the contracting process— help parties make explicit their assumptions and expectations regarding their transaction and each side’s role (Beatty & Samuelson, 2001; Smitka, 1994). Coordination-oriented provisions in a con- tract are aimed at mitigating the risk that misunder- standings will disrupt collaboration among (presum- ably) well-intentioned parties (Macaulay, 1963).
In this article, we build on the work of those who have called for a broader perspective on contractual
complexity and a more nuanced approach to study- ing the effects of contracts on relational attitudes and exchange outcomes (Puranam & Vanneste, 2009). We evaluate contracts at the level of individ- ual provisions, distinguishing between provisions aimed primarily at exerting control and those aimed primarily at facilitating coordination.
Two Dimensions of Trust Judgments: Goodwill and Competence
Following Rousseau, Sitkin, Burt, and Camerer (1998), and in keeping with other influential con- ceptualizations of trust (Lewicki, McAllister, & Bies, 1998; Mayer, Davis, & Schoorman, 1995), we define trust as the willingness of a party to be vulnerable to the actions of another party based on positive expectations regarding the other party’s motivation and/or behavior. Trust, so defined, can be distinguished from the underlying dimensions of trust judgments, which entail attributions of an- other party’s trustworthiness along relevant charac- teristics (e.g., integrity). Following prior work on the attributional basis of trust (Ferrin & Dirks, 2003; Mayer et al., 1995; Weber, Malhotra, & Murnighan, 2005), we posit that attributions along relevant di- mensions are what create in a truster a willingness to accept vulnerability.
We follow the lead of Nooteboom (1996) and Das and Teng (2001), who focused on two dimensions of trust judgments: goodwill and competence. Per- ceptions of goodwill entail attributions regarding the intention of another party to behave in a trust- worthy manner; perceptions of competence entail attributions regarding the other party’s ability to behave or perform as expected (Nooteboom, 1996).1
The Effect of Contracts on Trust
Contracts and trust represent alternative means by which parties can manage risk in exchange re- lationships, but in interfirm relationships, firms typically use contracts while simultaneously at- tempting to build trust (Poppo & Zenger, 2002; Wicks, Berman, & Jones, 1999). The seemingly modal preference regarding interfirm gover-
1 This distinction captures all three dimensions in Mayer and colleagues’ (1995) trust framework: ability, benevolence, and integrity. Competence captures attribu- tions of ability; goodwill captures benevolence and in- tegrity. Combining benevolence and integrity as “good- will” is useful because many of the statements in our data are hard to categorize as either benevolence or integrity attributions; many are ambiguous or suggestive of a dual attribution.
982 OctoberAcademy of Management Journal
nance—to use contracts and build trust— has sparked a debate regarding the viability of this strategy. Some have argued that contracts and trust are often incompatible (Malhotra & Murnighan, 2002; Molm, Takahashi, & Peterson, 2000; Sitkin & Roth, 1993). Others have suggested that contracts and trust are not only compatible, but mutually reinforcing (Gulati & Nickerson, 2008; Poppo & Zenger, 2002). This divergence makes it difficult to predict whether an emphasis on contracts in a re- lationship between firms will enhance or inhibit the prospects for the firms’ continued collaboration after a dispute. Our goal is not to reconcile the vast amount of prior research on this topic, nor do we align ourselves completely with either side. Rather, we borrow from the literatures forwarding each of these viewpoints on the (in)compatibility of con- tracts and trust to expound a more comprehensive (and nuanced) perspective on the effect of contracts on trust and collaboration in the aftermath of con- flict. Our review of prior research suggests that both positive and negative effects of contracts could be better understood when we separately consider the effects of control versus coordination provisions on goodwill-versus competence-based trust.
The effect of control provisions on trust. Those who have posited a negative relationship between contracts and trust have largely focused on the control function of contracts and on the goodwill dimension of trust (Malhotra & Murnighan, 2002; Sitkin & Roth, 1993; Tenbrunsel & Messick, 1999). Macaulay (1963) and Ghoshal and Moran (1996) suggested that the mere suggestion or introduction of contracts may signal distrust of another party’s intentions, thereby disrupting the process of trust development (Pillutla, Malhotra, & Murnighan, 2003). Tenbrunsel and Messick (1999) argued that an overreliance on control mechanisms changes the “decision frames” of exchange partners; including too many control provisions may, ironically, pro- mote opportunistic behavior by inducing a “busi- ness” rather than “ethical” framing of the interac- tion. Finally, Malhotra and Murnighan (2002) argued that overly controlling contracts, which leave little room for discretion, crowd out trust development because they lead to situational rather than personal attributions for the cooperativeness of partners. This crowding out may be especially likely during conflict, because parties are less likely to make generous attributions of each other’s behavior when their relationship has turned antag- onistic (Ross & Stillinger, 1991). These mechanisms suggest:
Hypothesis 1. The higher the number of control provisions in an interfirm contract, the lower
the subsequent level of goodwill-based trust between them.
Prior research has not directly examined the ef- fect of control provisions on competence-based trust. However, there are two reasons to expect that control provisions will enhance perceptions of competence in the context of disputes. First, by eliminating incentives for cheating and reneging, control provisions may force parties to focus more time and effort on their roles and responsibilities. This “substitution effect,” by limiting nefarious conduct, may promote exactly the types of behavior (e.g., attention to detail, timeliness, etc.) that en- hance competence attributions. Another possibility is that the time spent on drafting contractual lan- guage, even for control provisions, leads to a clari- fication of expectations and assumptions, which in turn facilitates competence attributions (Argyres et al., 2007; Mayer & Argyres, 2004). We therefore hypothesize the following:
Hypothesis 2. The higher the number of control provisions in an interfirm contract, the higher the subsequent level of competence-based trust between them.
The effect of coordination provisions on trust. In addition to serving a control function, contracts provide a means by which parties can coordinate their expectations and efforts (Gulati et al., 2005; Mayer & Argyres, 2004; Reuer & Ariño, 2007). As a result, common knowledge structures such as shared language and routinized interactions emerge that make it easier for the parties to com- municate their ability to meet each other’s needs (Morgan & Hunt, 1994; Puranam, Singh, & Zollo, 2006). The process of coordination can thus facili- tate competence-based trust development. In their analysis of 11 contracts signed between two firms, Mayer and Argyres provided a relevant example: “HW Inc. had expressed frustration in the first two projects over the length of time it took Softstar to complete what HW, Inc. perceived to be minor changes. . . . Softstar added a system architecture section to the third [contract]. This section allowed both firms to better understand how the entire product fit together and the impact to Softstar if HW, Inc. made a late hardware change” (2004: 400). In this incident, the revised contract was aimed at aligning expectations regarding the link between change requests and delays, lest HW Inc. attribute delays to Softstar’s incompetence. Coordination structures may be especially important for compe- tence perceptions after a conflict has arisen, be- cause disputing parties are otherwise unlikely to engage in the kinds of spontaneous communication
2011 983Malhotra and Lumineau
that mitigate conflict and promote positive attribu- tions (Hinds & Mortensen, 2005). This suggests:
Hypothesis 3. The higher the number of coor- dination provisions in an interfirm contract, the higher the subsequent level of competence- based trust between them.
Coordination provisions are also expected to in- crease goodwill-based trust in the context of dis- putes. By creating channels through which differ- ences in perspective will be resolved, coordination provisions help mitigate misunderstandings of the kind that raise questions about the intent of another party; this promotes— or at least minimizes damage to—attributions of goodwill during a conflict (Vlaar, Van den Bosch, & Volberda, 2007). More- over, parties that establish norms and procedures allowing them to coordinate on when and how to expend effort in their relationship are less likely to face situations in which one party feels overworked or exploited, or is concerned that the other side is not meeting its reciprocal obligations (Malhotra, 2004). Laboratory evidence has supported this argument. Ahn, Ostrom, Schmidt, Shupp, and Walker (2001) found that prior experience in a pure coordination game (with no incentives for nonco- operation) helps parties to move toward a mutually cooperative outcome, even in subsequent interac- tions that provide incentives for noncooperation. This suggests:
Hypothesis 4. The higher the number of coor- dination provisions in an interfirm contract, the higher the subsequent level of goodwill- based trust between them.
Trust and consequences. The considerable re- search attention devoted to the effect of contracts on trust reveals the extent to which trust is seen as crucial for interfirm collaboration and value cre- ation (Arrow, 1974; Uzzi, 1997) and as a source of competitive advantage for organizations (Barney & Hansen, 1994; Nahapiet & Ghoshal, 1998). Here, we consider a consequence of trust that has received little direct attention but that is critical to the value- creating potential of interfirm relationships: firms’ willingness to continue such relationships after disputes have arisen. We expect that contractual provisions aimed at coordination and control will influence competence- and goodwill-based trust, which will in turn influence parties’ decisions ei- ther to stay together or to end a relationship that has experienced conflict. Thus, our data allow us to begin the process of linking contract choices with relational outcomes, with trust serving as a mediator.
Zand (1972) suggested that goodwill-based trust increases the likelihood of continued collaboration because it leads parties to share accurate and timely information and to be more willing to accept de- pendence on each other even when formal control mechanisms cannot be applied. Zand (1972) also found that a high degree of goodwill-based trust increases motivation to implement agreements and makes parties less likely to switch partners. Re- search on the role of “psychological contracts” has also shown the positive effect of goodwill-based trust on relationship continuance (Morrison & Rob- inson, 1997; Robinson, 1996). Uzzi (1997) argued that, in interfirm contexts, (goodwill-based) trust is a crucial predictor of future exchange. This suggests:
Hypothesis 5. The higher the level of goodwill- based trust in an interfirm relationship, the higher their willingness to continue the rela- tionship after a dispute has arisen.
Competence-based trust should also increase the likelihood of relationship continuance. It is per- haps axiomatic that parties prefer to do business with those they consider to be competent. In the aftermath of conflict, the importance of compe- tence-based trust should be even greater. Das and Teng argued that “the lower the level of acceptable performance risk level, the higher the needed com- petence trust level” (2001: 266). Parties exiting a dispute will be especially sensitive to the degree of performance risk involved in continuing their rela- tionship. This suggests:
Hypothesis 6. The higher the level of compe- tence-based trust in an interfirm relationship, the higher their willingness to continue the relationship after a dispute has arisen.
Contracts and the Continuation of Relationships after a Dispute
We expect that control provisions, by reducing goodwill-based trust, should lessen the likelihood that disputing parties will agree to continue work- ing together. If a dispute arises despite reliance on provisions designed to protect against opportun- ism, the parties are likely to seriously question the viability of future exchange. The prospects for con- tinued collaboration are even dimmer if control provisions do not simply substitute for goodwill- based trust, but actually undermine it (Malhotra & Murnighan, 2002). By contrast, control provisions may increase the desire for continued exchange because of an increase in competence-based trust. If control provisions crowd out self-serving behavior
984 OctoberAcademy of Management Journal
in favor of task-oriented activity, competence per- ceptions should increase and make future collabo- ration more attractive.
The net effect of enhanced control on future col- laboration depends on whether the effect of good- will-based trust or competence-based trust domi- nates. We expect that the negative effect will dominate. Prior research has suggested that strained relationships are more difficult to repair when there has been a breach of goodwill-based trust rather than of competence-based trust (Kim, Dirks, Cooper, & Ferrin, 2006). Moreover, after a dispute, goodwill-based trust is likely to be more important than competence-based trust for resur- recting the scarred relationship: even if concerns remain about a partner’s ability to meet all of its obligations, the scope of the relationship can be redefined to focus on areas where competence is not in question; in contrast, concerns about the other’s goodwill are unlikely to be limited to a single domain, making it difficult to rekindle the relationship (Kim, Ferrin, Cooper, & Dirks, 2004). We therefore suggest:
Hypothesis 7a. The higher the number of con- trol provisions in an interfirm contract, the lower the firms’ willingness to continue their relationship after a dispute has arisen.
Hypothesis 7b. The effect of control provisions on relationship continuance is mediated by the level of goodwill-based trust.
Hypothesis 7c. The effect of control provisions on relationship continuance is mediated by the level of competence-based trust.
Meanwhile, coordination provisions, by increas- ing goodwill-based trust, will facilitate continued exchange. Recent case studies have pointed to a mutually reinforcing relationship between coordi- nation, goodwill-based trust, and relationship de- velopment (Faems, Janssens, Madhok, & Van Looy, 2008; Mayer & Argyres, 2004; Ness, 2009): relation- ship development facilitates more efficient con- tracting aimed at better coordination; better coordi- nation facilitates cooperation; and increased cooperation facilitates goodwill-based trust and re- lationship development. Coordination provisions should be especially important in the context of disputes: parties may question the usefulness of coordination provisions that failed to prevent a dis- pute, but if coordination provisions have facilitated goodwill-based trust in the interim, a basis on which to rebuild the relationship exists. Coordina- tion provisions should also facilitate relationship continuance via an increase in competence-based trust. As the relationship continues, the parties
learn more about the types of contingencies that can arise, which leads to the development of addi- tional coordination provisions aimed at improving the working arrangement (cf. Zaheer & Venkatra- man, 1994). Thus, as time goes on, coordination provisions not only allow parties to make fewer mistakes and appear more competent (as in Hy- pothesis 3), but also allow them to increase the domain of tasks in which they can demonstrate competency (Argyres et al., 2007). Both of these effects should facilitate continued collaboration. This suggests:
Hypothesis 8a. The higher the number of coor- dination provisions in an interfirm contract, the greater the firms’ willingness to continue their relationship after a dispute has arisen.
Hypothesis 8b. The effect of coordination pro- visions on relationship continuance is medi- ated by the level of goodwill-based trust.
Hypothesis 8c. The effect of coordination pro- visions on relationship continuance is medi- ated by the level of competence-based trust.
RESEARCH METHODS
Data
We were granted access to all legal files concern- ing contract disputes handled by one law firm in Western Europe between 1991 and 2005. This mid- sized law firm is a generalist in the field of corpo- rate law; its clients include small, midsize, and large firms from a variety of industries. We re- stricted our sample to all two-party disputes in- volving vertical relationships; these represented 80 percent of all two-party disputes the firm handled. Our sample consisted of 102 cases (i.e., disputes), 99 of which involved only European firms; each of the other 3 involved at least one non-European firm. Because some companies were repeat clients and involved in more than one dispute, the sample contained 178 different firms.2 To check for selec- tion bias, we examined differences between in- cluded and excluded files (i.e., those involving nonvertical relationships). We found no significant differences on any observable dimension (contrac- tual complexity, firm size, etc.).3
2 The results were unchanged when a “repeat client” control variable was included in the analyses.
3 Because all of the studied relationships involved legal disputes, we evaluated the representativeness of our sample—at the contract level and relationship lev- el—relative to the broader universe of interfirm relation-
2011 985Malhotra and Lumineau
Each legal file contained between 800 and 5,000 pages and included (1) the original contract, along with any contract revisions that were made prior to the dispute and (2) all documents exchanged dur- ing the dispute resolution process. In addition, the lawyers in each case obtained from the clients all potentially relevant information related to the ini- tial context of the relationship, the origins of the conflict, and its progression over time. In total, over 150,000 pages of documents were collected and analyzed for this study. Data collection took place over four months. The law firm did not allow us to contact the disputing firms directly.
The firms in our sample came from a variety of industries: manufacturing (52%), services (32%), retail (15%), and construction (2%). There were four types of contracts: distribution (35.3%), pro- duction supply (29.4%), information technology (IT; 26.5%), and consulting and other services (8.8%). Forty-six percent of the cases involved cross-border relationships; 65.7 percent of the con- tracts were time-bound in that they stipulated a prespecified an end to the relationship; and 32.4 percent of the cases included exchange part- ners that had interacted with each other previously.
Dependent Variable: Intent to Continue
We analyzed the intent of parties to continue a relationship after a dispute had been resolved. For multiple reasons, this performance variable is es- pecially relevant when one is considering the ef- fects of contract structure on trust in the context of conflict. First, strategic alliance scholars have tra- ditionally sought to investigate factors that contrib- ute to alliance survival (Dhanaraj & Beamish, 2004; Park & Russo, 1996) and stability (Blodgett, 1992; Inkpen & Beamish, 1997). Second, an analysis of the parties’ intent to continue collaborating after a dispute has arisen responds to the call by trust researchers to examine when and how damaged relationships can be repaired (Bottom, Gibson, Daniels, & Murnighan, 2002; Dirks, Lewicki, & Za- heer, 2009). Finally, this measure provides a more
direct measure of the consequences of trust than would be provided by some (eventual) financial measure of performance.
We examined the messages that were exchanged between the parties to each dispute, as well as the terms of the settlement/judgment recorded in the legal files, to look for indications of a willingness to continue with the relationship. Such intent was sometimes manifested in a direct communication between the firms (e.g., “I hope we have clarified and overcome this “misunderstanding” and we can now continue our fruitful collaboration on a sound basis”). In other cases, it was made evident through the crafting of a new agreement by the parties. For example, in one case, the following clause was added to the contract: “Addendum to Clause 14: The Parties thereby agree that [Firm A] and [Firm B] shall now each conduct by the end of each month review of the progress made [...] The Agree- ment is thereby extended for a 2 (two)-year period.” We coded as “no intent to continue” those cases in which either or both parties explicitly stated no willingness to continue the relationship (e.g., “You should perfectly understand that we have put an end to our collaboration”) and those cases in which no indication of an intent to continue the relation- ship was present. Intent to continue the relation- ship was indicated in 29 cases out of 102.
Independent Variables Control versus coordination provisions. Our
codification of contract provisions as control-ver- sus coordination– oriented was based on existing research (e.g., Parkhe, 1993; Reuer & Ariño, 2007) and supplemented with extensive interviews with legal experts specializing in contract law. We con- ducted 17 interviews with three practicing lawyers and seven professors in contract law. Interviews lasted between 1.5 and 3 hours. The experts exam- ined both the codification scheme we had prepared drawing on the literature and a sample of contracts from our data set. The experts then offered an eval- uation of the coding scheme and proposed some changes, which we implemented. Here, more pre- cisely, is the method we followed in coding our contract provisions:
Step 1: Codification based on prior research. We relied upon a set of indicators developed by Parkhe (1993: 829) that are designed to evaluate various provisions in formal contracts. Parkhe identified the following eight key provisions that might be included in a contract: (1) the exchange of periodic written reports of all relevant transactions, (2) prompt written notice of any departures from the agreement, (3) the right to examine and audit all
ships. On Parkhe’s (1993: 829) unweighted index of con- tractual complexity, which tabulates the presence of up to eight key contractual clause categories, our sample’s score (4.36) was situated comfortably between the score (3.69) for Reuer and Ariño’s (2002) sample and that (5.05) for Reuer, Ariño, and Mellewigt’s (2006) sample. At the relationship level, the percentage of prior ties among firms in our sample (32.4%) was within the range ob- served in prior research: 12 percent in Gulati (1995b); 20 percent in Reuer and Ariño (2002); and 53 percent in Hagedoorn and Hesen (2009).
986 OctoberAcademy of Management Journal
relevant records through a firm of CPAs, (4) desig- nation of certain information as proprietary and subject to confidentiality provisions of the contract, (5) nonuse of proprietary information even after termination of agreement, (6) termination of agree- ment clauses, (7) arbitration clauses, and (8) law- suit provisions. Reuer and Ariño (2007) factor- analyzed the inclusion of Parkhe’s eight provisions in an analysis of 88 strategic alliances and found that the first three provisions of Parkhe’s eight-item index related primarily to coordination, whereas the remaining five related primarily to enforcement (what we call control). They therefore defined co- ordination provisions as the number of coordina- tion-related clauses included in a contract (i.e., clauses 1, 2, and/or 3 from Parkhe [1993]), a mea- sure yielding an integer variable ranging from 0 to 3; control provisions, defined as the number of control-related clauses in a contract (i.e., clauses 4, 5, 6, 7, and/or 8 from Parkhe), yielded a score ranging from 0 to 5.
The coding of our contracts for control versus coordination provisions was done by (1) one of the authors and (2) a faculty member with a degree in law who was unaware of the hypotheses or the purpose of the study. To further eliminate the pos- sibility of bias, all coding of contracts was done prior to any analysis of dispute-related data (e.g., trust messages, intent to continue, etc.). Pairwise correlation among raters for the coding of control- related provisions (r � .91; p � .001) and for coor- dination-related provisions (r � .92; p � .001), along with high Cronbach’s alpha coefficients (0.95 and 0.96, respectively), confirmed the reliability of the coding. Any disagreements on coding were re- solved by discussion.
Step 2: Revisions based on expert advice. To evaluate and refine our coding scheme, we pre- sented the scheme, as well as a sample of real contracts, to a set of legal experts. These interviews yielded two primary results. First, the legal experts supported our general approach to codification, stating that contract provisions could be meaning- fully distinguished as being focused primarily on control versus coordination. Second, the legal ex- perts expressed concerns with two clause catego- ries from Reuer and Ariño’s (2007) coding frame- work. Specifically, the experts argued that clause category 3 (regarding the right to examine and audit all relevant records through a firm of CPAs) did not clearly represent a coordination function. In addi- tion, the legal experts suggested that clause cate- gory 7 (regarding arbitration clauses) was ambigu- ous because such provisions may not serve a clear control function. On the basis of this advice, we revised our categorization scheme so that our mea-
sure of coordination provisions would be based on clauses 1 and 2 from Parkhe (1993) and our mea- sure of control provisions would be based on clauses 4, 5, 6, and 8 from Parkhe (1993).4, 5
Robustness checks. The results below are based on the measures of coordination and control provi- sions derived by the two-step process described above. In addition, we conducted two robustness checks of our results. In the first set of analyses, we used the initial (step 1) Reuer and Ariño (2007) coding framework for our measures of coordination and control provisions. The findings based on this eight-clause analysis strongly converge with the results reported below (results available on re- quest). In the second set of analyses, we added an interpretive (coding) step to the step 1 categoriza- tion, in which a rater evaluated each provision in every contract of the data set for seeming ambiguity of intent (Hagedoorn & Hesen, 2009). All clauses that seemed ambiguous as to whether their func- tion was coordination or control were then elimi- nated. To test the reliability of this elimination procedure, a second rater evaluated ten randomly selected clauses for each of the eight types of pro- visions; the level of agreement was 91.25 percent. This process yielded more conservative measures in which we deleted 5.96 percent of the coordina- tion provisions and 12.42 percent of the control provisions. The results based on these measures were also consistent with our primary (reported) analyses, with no differences in the tests of our hypotheses. Together, these two robustness checks provide confidence in our reported analyses.
Mediator Variables: Competence-Based Trust and Goodwill-Based Trust
To assess the level of trust between the parties during the conflict, we analyzed every communi- cation (paper or electronic) exchanged between dis- puting firms during the entire resolution process for a focal dispute. The choice not to reply to a communication by the other party was also coded as a (“no reply”) message. In total, 2,293 messages were studied (of which only 132 were coded as “no reply”). A scheme for categorizing statements as relating to competence- and/or goodwill-based trust (or neither) was constructed to evaluate each message (see below for more details). We allowed
4 A confirmatory factor analysis confirmed that the revised scheme improves on Parkhe’s (1993) eight-clause coding.
5 Following Lui and Ngo (2004) and Barthélemy and Quélin (2006), we used unweighted composite indexes.
2011 987Malhotra and Lumineau
each message to be coded as signaling neither, one, or both types of trust. After an evaluation of all documents in a given dispute, the ratio of compe- tence-based trust messages to total messages was calculated, and it served as a measure of the degree of competence-based trust. Likewise, the ratio of goodwill-based trust messages to total messages served as a measure of the degree of goodwill-based trust. Thus, the score on each variable could vary between 0 (i.e., a complete absence of this type of trust) and 1 (i.e., all the messages exchanged be- tween the partners conveyed this type of trust).
A team of two researchers coded the messages: a coauthor of this paper and a colleague (with a law degree) who was unaware of the hypotheses or the topic of research. We followed the coding proce- dure developed by Weber (1990: 21–24), which includes the following steps: defining the message as our unit of analysis; developing a list of relevant preliminary response categories; applying the cod- ing scheme to a subsample (four cases); assessing and revising the coding rules as a result; and having both raters independently read and code each mes- sage in the data set.6 The percentage of agreement between raters (97 percent for competence-based trust and 95 percent for goodwill-based trust) and the pairwise correlation between raters (r � .94, p � .01 for competence-based trust, and r � .93, p � .01 for goodwill-based trust) were high. Any disagreements were resolved through discussion. (See Appendix A for examples of statements that were coded as competence-based trust and good- will-based trust.)
Control Variables Asymmetry of revenues. We controlled for
power asymmetry between the parties using firm revenues as a proxy for firm strength. Asymmetry is necessary to control because it may affect the like- lihood that the parties include coordination and/or control provision in the contract. Asymmetry was measured as ln(ABS[revenue of firm A – revenue of firm B]). Revenue was measured in thousands of inflation-adjusted euros for the year when the con- tract was signed. These data were obtained from the Bureau van Dijk’s ORBIS database.
Type of dispute. We evaluated the nature of each dispute as described by the disputants (Weaver & Dickson, 1998) at the outset. Disputes could be meaningfully distinguished as either disagree- ments regarding the nature of the transaction in-
volved (30.4%) or as a perceived failure on the part of one party to meet payment, delivery, or other clear objectives (69.6%). Type of dispute took a value of 1 in disputes regarding the nature of the transaction and 0 otherwise.7
Type of settlement. We controlled for the type of resolution that was eventually pursued—litigation versus private settlement— because the anticipa- tion of this eventuality may have influenced a par- ty’s willingness to make statements that admit to the other party’s trustworthiness (or lack thereof). Type of settlement took a value of 0 if the dispute was eventually settled through litigation and the value of 1 if the dispute was eventually settled via private negotiation.
Prior ties. To mitigate endogeneity concerns re- garding the relationship between contract provi- sions and trust, we controlled for preexisting trust. Following Gulati (1995a), we used the existence of prior ties between the parties to a dispute as a proxy. We then improved on this measure by cod- ing whether the parties viewed the previous trans- actions (if any) positively or negatively. Following prior research on satisfaction with trading partners (Heide & John, 1992; Jap & Ganesan, 2000), we coded files with messages that explicitly referenced norms of flexibility, participation, and/or solidarity as positive prior ties; files referencing inflexibility, nonparticipation, and/or individualism in prior in- teractions were coded as negative prior ties. Two dummy variables were created: 16.67 percent of relationships were coded as having positive prior ties, and 12.74 percent had negative prior ties. When there was no prior tie (67.65%), or if there was no reference to positive or negative percep- tions (2.94%), both variables took a value of 0.
Prior relationship length. We controlled for the length of the prior relationship between the parties to a dispute because a lengthy interaction history may help parties build trust (Kramer, 1999). As such, controlling for prior relationship length helps to further mitigate endogeneity concerns. Also, or- ganizations interacting repeatedly may learn from prior experiences, allowing contracts to be speci- fied in greater detail (Mayer & Argyres, 2004; Van- neste & Puranam, 2010). We measured the amount
6 We based coding on the item selection and classifi- cation process outlined by Jauch, Osborn, and Martin (1980).
7 It is arguable that disagreements regarding the na- ture of the transaction are coordination problems and that failure to meet objectives is a control problem. We pursued this intriguing possibility with supplemental analyses to test whether (1) our control or coordination independent variables predicted type of dispute and (2) type of dispute interacted with either independent vari- able to predict intent to continue. Neither test produced significant results.
988 OctoberAcademy of Management Journal
of time for which the firms had transacted prior to engaging in the transaction that led to the current conflict (Argyres et al., 2007; Dekker, 2008) as ln(number of days � 1). The mean prior relation- ship length was 942 days for the 33 cases in which firms had prior ties.
Revisions to the initial contract. In 10.8 percent of cases, the contract in place at the outset of the dispute was not identical to the original contract that was signed by the parties. Amendments over time may indicate recurrent conflict— or, to the contrary, the ability of parties to cooperatively re- engage to improve their relationship (Mayer & Teece, 2008; Reuer & Ariño, 2002). Thus, we con- trolled for whether a contract had been revised previously, prior to the current conflict.8 (We did not observe any revisions during a dispute period.)
Asymmetry of alternatives. Even parties that have low levels of trust may decide to continue collaborating if they have few viable alternatives. Given our dyad-level outcome variables, we con- trolled for outside alternatives by evaluating the degree of asymmetry between the parties’ alterna- tives, as well as the sum of their alternatives. To approximate each party’s number of alternatives to dealing with the other, we content-analyzed the communications to look for mentions of alternative options and/or partners (e.g., “You know that if you continue to deny the facts, we will turn to [firm X] to supply this part”; “If we aren’t able to put this relationship on the right track, we will produce the [part] ourselves”). Because the texts did not allow us to calculate the precise number of each party’s alternatives, we estimated the strength of alterna- tives in terms of the frequency with which a party mentioned alternatives. Asymmetry of alternatives was measured as ABS [(number of references to alternatives by firm A) – (number of references to alternatives by firm B)].
Sum of alternatives. We evaluated mutual de- pendence by calculating the sum of each party’s alternatives. A higher value indicated that the par- ties had strong alternatives to dealing with each other (i.e., a lower degree of mutual dependence).
Dropped variables. The following variables were tested in a supplemental set of tests for our hypotheses but dropped from the analyses reported below because they did not have any significant
effects in any of the analyses: industry (e.g., man- ufacturing), type of transaction (e.g., distribution contract), international (i.e., whether the transac- tion entailed a cross-border relationship), time bound (i.e., a dichotomous variable capturing whether an initial contract had a specified end time), technical detail (i.e., the level of complexity of a transaction), stakes (i.e., the amount of money involved in the contract), and geographic distance between contracting firms.
Analyses
Regression analyses were used to test the impact of contractual provisions on the level of each type of trust (Hypotheses 1– 4). As some companies in our sample were repeat clients (i.e., were involved in multiple disputes), correlated residuals across observations were possible; we therefore report re- sults with robust standard errors clustered on firms (76 clusters). When intent to continue the relation- ship, a binary variable, served as the dependent measure (Hypotheses 5– 8), we used probit models. As a robustness check, we also used logit models for these analyses; results were identical. Hypoth- eses 7b, 7c, 8b, and 8c predict mediated relation- ships. To test for mediation, we followed the pro- cedure outlined by Baron and Kenny (1986). Mediation is supported if (1) the independent vari- able (IV) significantly predicts the dependent vari- able (DV), (2) the IV significantly predicts the me- diator (MV), (3) the MV significantly predicts the DV, and (4) when the IV and MV are simultane- ously included in the analysis, the MV is a signif- icant predictor, but the IV is less (or no longer) significant.
It is worth noting that although regression anal- ysis cannot evaluate temporal causality, our data suggest that a temporal sequence is in play. The data on contract provisions and transaction attri- butes are based on information that predates the onset of the disputes (T � 1). The data on goodwill- and competence-based trust are based on messages exchanged at a later time (T � 2), after the onset of the conflict. Finally, the intent to continue the re- lationship is manifested at the end of the dispute resolution process (T � 3). In addition, we tried to address this issue of alternative relationships by including a host of control variables (described above) that might influence initial contract struc- ture, most notably the existence and influence of preexisting trust (measured both by the length of prior ties and with measures of the quality of prior ties). Finally, as an additional robustness check, we conducted a supplemental analysis aimed at miti- gating, to the degree possible, concerns regarding
8 We conducted an additional analysis at the provi- sion level and found that 96.1 percent of the control provisions and 94.4 percent of the coordination provi- sions were in the contracts from the beginning of the transactions. Including controls for changes made at the provision level did not change any of the results.
2011 989Malhotra and Lumineau
whether prior relationships (and prior trust, specif- ically) influenced the types of contractual clauses. In this analysis, we eliminated all 32.35 percent of relationships that contained any prior ties. The re- sults are consistent with those of our core analyses, again suggesting that endogeneity is perhaps not problematic for our results.
RESULTS
Interfirm relationships varied in the degree to which their associated contracts included coordi- nation versus control provisions. The mean num- ber of coordination provisions was 0.99 (out of 2), and the mean number of control provisions was 2.30 (out of 4). Table 1 provides summary statistics and Pearson correlations for the variables in our analysis. Because some variables were significantly correlated, we checked for multicollinearity prob- lems. The observed variance inflation factor (VIF) ranged from 2.47 to 4.70, diminishing this concern (Chatterjee & Price, 1991).
Table 2 displays the first set of results in which goodwill-based trust and competence-based trust are regressed on contractual provisions (control vs. coordination). As Hypothesis 1 predicts (Table 2, model 1d), when attributes of transaction, of dis- pute, and of relationship are controlled for, the higher the level of control provisions, the lower the level of goodwill-based trust (� � �0.05, p � .001). We also found (Table 2, model 2d) that control provisions positively influence competence-based trust (� � 0.06, p � .001), supporting Hypothesis 2. As Hypothesis 3 predicts, increasing the coordina- tion provisions of a contract results in an increase in competence-based trust (� � 0.16, p � .001; Table 2, model 2d). However, Hypothesis 4, which
predicts that an increase in coordination provisions will result in an increase in goodwill-based trust, was not supported (Table 2, model 1d).
Table 3 shows the results of binomial probit re- gressions in which the dichotomous dependent variable is intent to continue. A positive coefficient indicates an increased likelihood of continuing a relationship. In keeping with our predictions, we found (Table 3, model 3d) that both goodwill-based trust (� � 17.07, p � .001) and competence-based trust (� � 9.45, p � .01) positively impact the intent to continue, supporting Hypotheses 5 and 6, re- spectively.
We also predicted that the greater the level of control provisions in a contract, the less likely it would be that the parties to the contract would intend to continue the relationship (Hypothesis 7a) and that both goodwill-based (Hypothesis 7b) and competence-based (Hypothesis 7c) trust would me- diate this relationship. Supporting Hypothesis 7a (Table 3, model 4c), the higher the number of con- trol provisions, the lower the likelihood of continu- ing the relationship (� � �0.62, p � .01).
The mediation predictions of Hypotheses 7b and 7c required multiple tests (Baron & Kenny, 1986). We have already shown that control provisions pre- dict goodwill-based and competence-based trust. We have also shown that goodwill-based trust and competence-based trust are positively related to the intent to continue. Finally, in keeping with Hy- pothesis 7b, when we simultaneously included control provisions (the independent variable) and goodwill-based trust (the mediator variable) as pre- dictors of intent to continue (Table 3, model 5a), goodwill-based trust was still a significant predic- tor (� � 12.93, p � .001), but number of control provisions was no longer significant (� � �0.15,
TABLE 1 Descriptive Statistics and Correlations
Variables Mean Min. Max. s.d. 1 2 3 4 5 6 7 8 9 10 11 12 13
1. Intent to continue 0.28 0 1 0.45 2. Control provisions 2.30 0 4 0.90 �.13 3. Coordination provisions 0.99 0 2 0.89 .39* �.11 4. Goodwill-based trust 0.22 0 0.66 0.16 .78* �.16 .21* 5. Competence-based trust 0.26 0 0.8 0.22 .60* .24* .68* .47* 6. Asymmetry 7.66 5.17 10.28 0.96 .00 .04 �.02 .03 �.08 7. Type of conflict 0.32 0 1 0.47 .02 �.02 .14 .01 .16 .00 8. Settlement type 0.40 0 1 0.49 .45* .12 .09 .57* .36* .14 �.05 9. Negative prior ties 0.12 0 1 0.33 �.11 �.19 .07 �.12 �.20* .22* �.07 �.25*
10. Positive prior ties 0.16 0 1 0.37 .00 �.00 .00 .02 .03 .03 �.08 .16 �.17 11. Prior relationship length 1.80 0 8.63 2.94 �.08 �.14 .01 �.07 �.16 .15 �.13 �.07 .64* .59* 12. Revisions 0.22 0 4 0.72 .13 .32* .15 .13 .20 .12 �.09 .21* �.03 .11 .07 13. Asymmetry of alternatives 1.18 0 5 1.27 �.07 .10 �.10 �.06 .00 .09 �.08 �.13 .05 �.10 �.04 �.02 14. Sum of alternatives 1.51 0 8 1.65 .02 .09 �.04 �.05 .33 .03 �.00 �.14 .04 �.01 �.00 �.04 .82*
* p � .05
990 OctoberAcademy of Management Journal
n.s.), suggesting full mediation. To evaluate the mediation effect, we conducted a Sobel test (Sobel, 1982), computing a Z-value to assess whether the indirect effect of the independent variable on the dependent variable through the mediator is signif- icantly different from zero. The Sobel test con- firmed the mediating effect of goodwill-based trust between control provisions and the intent to con- tinue (Z � �2.71, p � .01). In contrast, when we simultaneously included control provisions (the IV) and competence-based trust (as the MV) as pre- dictors of intent to continue (Table 3, model 5b), both variables— competence-based trust (� � 11.90, p � .001) and control provisions (� � �2.11, p � .001)—remained highly significant, suggesting no support for the mediation predicted in Hypoth- esis 7c. Thus, the (negative) effect of control provi- sions on intent to continue a relationship is medi- ated by goodwill-based (but not competence- based) trust.
Hypothesis 8a predicts that coordination provi- sions will positively influence intent to continue. Hypotheses 8b and 8c predict that goodwill-based trust (8b) and competence-based trust (8c) will me- diate this effect. Supporting Hypothesis 8a, we found (Table 3, model 4c) that the higher the level of coordination provisions in the contract for a relationship between firms, the higher the likeli- hood of the firms’ continuing the relationship after
a dispute (� � 0.77, p � .01). Since we had already established, in failing to support Hypothesis 4, that coordination provisions do not affect goodwill- based trust, we did not need to assess the role of goodwill-based trust as a mediator; Hypothesis 8b was not supported. The mediation prediction re- garding competence-based trust (Hypothesis 8c), however, was supported: we have already shown that competence-based trust positively influences intent to continue a relationship. When we simul- taneously included coordination provisions and competence-based trust as predictors of intent to continue (Table 3, model 5b), competence-based trust remained a significant predictor (� � 11.90, p � .001), but coordination provisions did not (� � �0.61, n.s.). This pattern of results suggests full mediation, which a follow-up Sobel test con- firmed (Z � 3.32, p � .001). Thus, the (positive) effect of coordination provisions on the intent to continue collaboration is mediated by competence- based (but not goodwill-based) trust.
For completeness, as a final analysis (Table 3, model 5c), we simultaneously included both inde- pendent variables (coordination and control provi- sions) and both mediators (competence- and good- will-based trust) as predictors of intent to continue. Both mediators remained marginally significant predictors of intent to continue (� � 19.38, p � .01, and � � 14.96, p � .01). Also consistently with
TABLE 2 Effects of Control vs. Coordination Provisions on Goodwill- and Competence-Based Trusta
Variables
Goodwill-Based Trust Competence-Based Trust
Model 1a Model 1b Model 1c Model 1d Model 2a Model 2b Model 2c Model 2d
Independent Control provisions –.05*** (.01) –.05*** (.01) .03† (.02) .06*** (.01) Coordination provisions .03† (.01) .02 (.01) .15*** (.01) .16*** (.01)
Control Asymmetry �.00 (.01) �.00 (.01) �.00 (.01) �.00 (.01) �.03 (.02) -.03† (.02) �.02 (.01) �.02 (.01) Type of conflict .01 (.02) .01 (.02) .00 (.02) .00 (.02) .08† (.04) .08† (.04) .03 (.03) .03 (.03) Settlement type .20*** (.03) .20*** (.03) .20*** (.03) .20*** (.03) .16** (.04) .16** (.04) .13*** (.02) .12*** (.02) Negative prior ties �.01 (.10) �.05 (.10) �.06 (.10) �.08 (.10) .22 (.28) .24 (.27) �.02 (.18) �.00 (.16) Positive prior ties �.06 (.08) �.07 (.08) �.09 (.07) �.09 (.08) .18 (.23) .19 (.23) .03 (.15) .04 (.13) Prior relationship length .00 (.01) .00 (.01) .01 (.01) .00 (.01) �.03 (.03) �.03 (.03) �.01 (.02) �.01 (.02) Revisions .00 (.02) .02 (.01) �.00 (.01) .02 (.01) .05* (.02) .04† (.02) .02 (.01) �.00 (.01) Asymmetry of alternatives �.00 (.01) �.00 (.01) �.00 (.01) �.00 (.01) �.02 (.02) �.02 (.02) �.01 (.01) �.01 (.01) Sum of alternatives .00 (.01) .01 (.01) .00 (.01) .01 (.01) .03† (.01) .03† (.01) .03** (.01) .02* (.01)
Constant .19† (.11) .29* (.11) .15 (.11) .25* (.11) .42* (.17) .36† (.16) .18 (.12) .04 (.12) R2 .34 .41 .36 .42 .27 .29 .64 .70
a n � 102. Standard errors are in parentheses; clustering (n � 76) is on firms. † p � .10 * p � .05
** p � .01 *** p � .001
2011 991Malhotra and Lumineau
T A
B L
E 3
E ff
ec ts
o f
C o n
tr o l
a n
d C
o o rd
in a ti
o n
P ro
v is
io n
s o n
th e
In te
n t
to C
o n
ti n
u e
in a
R el
a ti
o n
sh ip
a
V a ri
a b le
s M
o d
el 3 a
M o d
el 3 b
M o d
el 3 c
M o d
el 3 d
M o d
el 4 a
M o d
el 4 b
M o
d el
4 c
M o
d el
5 a
M o
d el
5 b
M o
d el
5 c
In d
ep en
d en
t C
o n
tr o l
p ro
v is
io n
s – 0 .6
2 *
(0 .2
4 )
– 0 .6
2 *
* (0
.2 2
) �
0 .1
5 (0
.2 8
) �
2 .1
1 *
* *
(0 .5
1 )
� 1
.4 5
† (0
.7 5
) C
o o rd
in at
io n
p ro
v is
io n
s 0 .7
8 * *
(0 .2
3 )
0 .7
7 *
* (0
.2 4
) 1
.1 3
* (0
.4 5
) �
0 .6
1 (0
.3 7
) 0
.4 6
(0 .6
0 )
G o o d
w il
l- b as
ed tr
u st
1 0 .5
9 * * *
(1 .4
7 )
1 7 .0
7 * * *
(4 .4
4 )
1 2
.9 3
* *
* (1
.9 0
) 1
9 .3
8 *
* (6
.0 3
) C
o m
p et
en ce
-b as
ed tr
u st
4 .6
4 * * *
(0 .8
9 )
9 .4
5 * *
(2 .7
3 )
1 1
.9 0
* *
* (3
.0 1
) 1
4 .9
6 *
* (5
.5 4
)
C o n
tr o l
A sy
m m
et ry
� 0 .1
3 (0
.1 4 )
� 0 .1
1 (0
.2 4 )
� 0 .0
2 (0
.1 6 )
� 0 .0
0 (0
.2 5 )
� 0 .1
4 (0
.1 6 )
� 0 .0
9 (0
.1 4 )
� 0 .1
1 (0
.1 6
) �
0 .1
1 (0
.2 0
) �
0 .1
4 (0
.2 0
) �
0 .0
2 (0
.2 3
) T
y p
e o f
co n
fl ic
t 0 .0
9 (0
.3 0 )
� 0 .0
8 (0
.4 3 )
� 0 .3
6 (0
.3 5 )
� 0 .8
4 (0
.5 1 )
0 .1
6 (0
.3 6 )
� 0 .1
2 (0
.3 3 )
� 0 .0
4 (0
.3 3
) �
0 .2
3 (0
.4 5
) �
0 .4
1 (0
.4 5
) �
1 .2
7 (0
.9 4
) S
et tl
em en
t ty
p e
1 .5
3 * * *
(0 .3
2 )
0 .3
0 (0
.5 3 )
1 .2
1 * *
(0 .3
8 )
� 0 .2
2 (0
.7 3 )
1 .7
9 * * *
(0 .3
7 )
1 .6
8 * * *
(0 .4
4 )
1 .9
3 *
* *
(0 .4
9 )
0 .2
1 (0
.6 8
) 2
.0 8
* *
(0 .6
1 )
0 .3
4 (1
.1 6
) N
eg at
iv e
p ri
o r
ti es
0 .2
7 (1
.4 3 )
1 .0
2 (1
.4 9 )
� 0 .1
2 (1
.1 6 )
1 .0
7 (1
.5 4 )
� 0 .0
0 (1
.6 5 )
� 0 .6
7 (1
.2 8 )
� 0
.8 2
(1 .4
3 )
0 .1
0 (1
.4 8
) 1
.5 2
(1 .4
7 )
1 .7
4 (1
.7 3
) P
o si
ti v e
p ri
o r
ti es
� 0 .3
1 (1
.1 5 )
0 .4
2 (1
.1 9 )
� 0 .9
2 (1
.0 1 )
0 .1
6 (1
.2 5 )
� 0 .4
5 (1
.3 5 )
� 1 .0
5 (1
.0 7 )
� 1
.1 2
(1 .2
0 )
� 0
.1 1
(1 .1
9 )
� 0
.6 0
(1 .0
6 )
0 .1
3 (1
.4 7
) P
ri o r
re la
ti o n
sh ip
le n
gt h
� 0 .0
2 (0
.1 8 )
� 0 .1
2 (0
.2 0 )
0 .1
0 (0
.1 6 )
0 .0
1 (0
.2 2 )
� 0 .0
4 (0
.2 2 )
0 .0
9 (0
.1 7 )
0 .0
5 (0
.1 9
) �
0 .0
5 (0
.2 0
) �
0 .1
8 (0
.2 1
) �
0 .1
5 (0
.2 5
) R
ev is
io n
s 0 .1
1 (0
.1 9 )
0 .0
5 (0
.2 0 )
� 0 .1
6 (0
.1 8 )
� 1 .3
5 *
(0 .5
5 )
0 .4
4 (0
.2 2 )
0 .0
4 (0
.1 7 )
0 .3
1 (0
.2 0
) �
0 .2
1 (0
.2 7
) 0
.6 7
* (0
.2 6
) �
1 .1
4 †
(0 .6
6 )
A sy
m m
et ry
o f
al te
rn at
iv es
– 0 .3
9 †
(0 .2
2 )
– 0 .7
4 * *
(0 .2
6 )
� 0 .3
6 (0
.2 6 )
� 0 .4
5 (0
.3 4 )
� 0 .3
5 (0
.2 2 )
� 0 .3
9 (0
.2 7 )
� 0
.3 2
(0 .2
7 )
– 0
.6 2
† (0
.3 1
) �
0 .1
9 (0
.3 1
) �
0 .2
5 (0
.3 7
) S
u m
o f
al te
rn at
iv es
0 .3
4 *
(0 .1
6 )
0 .5
5 * *
(0 .1
8 )
0 .2
4 (0
.2 1 )
0 .2
5 (0
.2 0 )
0 .3
7 *
(0 .1
5 )
0 .3
8 †
(0 .2
1 )
0 .3
9 †
(0 .2
0 )
0 .5
1 †
(0 .2
4 )
0 .1
1 (0
.2 4
) 0
.2 0
(0 .2
8 )
C o n
st an
t �
0 .4
0 (1
.1 4 )
� 2 .6
1 (1
.8 3 )
� 2 .1
8 (1
.4 7 )
� 6 .5
2 * *
(2 .4
8 )
0 .7
7 (1
.3 9 )
� 1 .6
0 (1
.2 8 )
� 0 .2
2 (1
.5 1
) �
3 .8
8 *
(1 .8
9 )
1 .3
3 (1
.8 1
) �
6 .5
4 *
(2 .7
8 )
P se
u d
o -R
2 .2
3 .6
5 .4
3 .7
6 .3
1 .3
6 .4
2 .7
5 .6
5 .8
3 W
al d
� 2
2 6 .9
0 8 1 .7
3 4 3 .6
4 3 9 .8
2 3 3 .1
1 2 3 .6
0 3 1 .3
7 9
6 .7
3 2
8 .8
5 5
8 .7
3
a n
� 1 0 2 .
S ta
n d
ar d
er ro
rs ar
e in
p ar
en th
es es
; cl
u st
er in
g (n
� 7 6 )
is o n
fi rm
s. †
p �
.1 0
* p
� .0
5 * *
p �
.0 1
* * *
p �
.0 0 1
predictions, coordination provisions was no longer significant (� � 0.46, n.s.). Meanwhile, control pro- visions continued to have a marginally significant, direct negative effect on intent to continue (� � �1.45, p � .10), which suggests that control provi- sions may negatively influence the intent to con- tinue in ways that are not fully accounted for by the effects of control provisions on trust.
DISCUSSION
We sought to extend theoretical and empirical work by distinguishing between the control and coordination functions of contracts, and between goodwill-based and competence-based dimensions of trust judgments. Prior research on the effects of contracts on trust has focused primarily on the control function of contracts and the goodwill di- mension of trust judgments. Less attention has been paid to the coordination function of contracts, and the competence dimension of trust has been largely ignored. We found, as have those who have argued that contracts crowd out trust (e.g., Malhotra & Murnighan, 2002), that the greater the number of control-oriented provisions in a contract, the lower the subsequent level of goodwill-based trust in the associated relationship. In addition, in keeping with those who have perceived a more complemen- tary relationship between contracts and trust (e.g., Lazzarini, Miller, & Zenger, 2004; Poppo & Zenger, 2002), we found that increases in control provi- sions as well as increases in coordination provi- sions lead to higher levels of competence-based trust. Thus, our more nuanced approach to analyz- ing contracts and trust reveals that seemingly di- vergent conclusions in prior research are not nec- essarily incompatible.
The current investigation also addresses the scar- city of research on the effects of contractual gover- nance on performance and relational outcomes (e.g., Argyres et al., 2007; Lumineau, Fréchet, & Puthod, 2011). Our findings indicate that contract design affects the degree of trust that exists after a conflict has arisen, and, through this effect, the likelihood of relationship continuance. In particu- lar, control provisions have a negative effect on the willingness to continue a damaged relationship, and goodwill-based trust mediates this effect. Al- though control provisions enhance perceptions of competence and high levels of competence-based trust increase the likelihood of continued collabo- ration, competence-based trust does not act as a mediator in the relationship between control pro- visions and the intent to continue collaboration. Whereas the argument that control mechanisms can diminish goodwill-based trust has been made
in prior research (e.g., Malhotra & Murnighan, 2002), ours is the first study to document (1) the effects of control mechanisms on the willingness to continue a relationship and (2) the mediating role of goodwill-based trust in this relationship. Fur- thermore, although perceptions of competence do not mediate the effects of control on collabora- tion, ours is the first study to empirically document a positive relationship between (control provisions in) contracts and competence-based trust.
We also find that coordination provisions in- crease the likelihood of continued collaboration after a dispute and that perceptions of competence mediate this effect. Although prior research has predicted a relationship between coordination and continued collaboration (Argyres et al., 2007; Mayer & Argyres, 2004), ours is the first empirical study to identify a mechanism— enhanced percep- tions of competence— underlying this relationship. Contrary to predictions, we did not find an effect of coordination provisions on goodwill-based trust.
Theoretical Implications
The results yield a number of theoretical impli- cations that build upon and clarify prior research. Our study extends research based on transaction cost economics (e.g., Barthélemy & Quélin, 2006; Reuer & Ariño, 2002) by showing that firm deci- sions regarding contractual governance structures should embrace not only (1) transaction attributes (as per the transaction costs approach [Sampson, 2004; Williamson, 1985]) and (2) existing levels of trust (e.g., Gulati, 1995a), but also (3) the effect of contract choices on subsequent trust and commit- ment (cf. Puranam & Vanneste, 2009).
Another important implication of our results concerns the simultaneity with which contracts can produce positive and negative effects. Previ- ously, Vlaar argued:
The relationship between contracting and interor- ganizational performance is likely to follow a curve- linear path, where too little contracting gives rise to chaos and destructive or opportunistic behaviour and where too much contracting causes rigidity and curbs creativeness and entrepreneurial activities (Foss, Foss, & Klein, 2007; Luo, 2002; Mintzberg, 1994; Sampson, 2004a). In this respect, Mintzberg (1994: 386) notes that “formalization is a double- edged sword, easily reaching the point where help becomes hindrance.” (2008: 18)
Although this logic suggests that there may be an optimal level of contracting, our results suggest otherwise, at least as far as control provisions are concerned. We find that an increase in control pro-
2011 993Malhotra and Lumineau
visions decreases goodwill-based trust and in- creases competence-based trust, suggesting that an optimal contract will not be found by discovering the “point where help becomes hindrance,” but by appreciating inherent trade-offs and evaluating the priorities of current relationships. Future research that studies whether it is possible to avoid such trade-offs—for example, by mitigating the effects of control on goodwill-based trust—would be of sig- nificant value.
Finally, the current research suggests that future work on the effects of contractual governance should include outcome measures as well as medi- ator variables. In our complete model, for example, we find that our mediator, goodwill-based trust, accounts for some, but not all, of the effects of control provisions on the intent to continue collab- oration, thus raising additional questions regarding the mechanisms underlying these relationships.
Managerial Implications
The current investigation also has implications for managers who are tasked with the responsibility of mitigating relationship risks. By distinguishing between control and coordination provisions (rather than relying on standard measures of con- tractual complexity, such as length or detail [Jos- kow, 1987; Pirrong, 1993]), we are able to advocate for an increase in coordination provisions as a means of building competence-based trust in antic- ipation of conflict. In keeping with this view, a South Asian executive recently explained to one of us that he refuses to do business with any U.S. firm unless the firm contractually agrees to sending each new project manager to his city and on a car ride from the airport to his manufacturing facility; the facility is 18 kilometers from the airport, but re- quires three hours of travel. “Because if the manag- ers have not done that, they do not understand how things work here—and the next time something goes wrong, they think it is because we are incom- petent.”
Our results regarding control give us pause and suggest that optimal decisions regarding contract structure require an assessment of the key sources of vulnerability in a relationship. If the relationship is likely to evolve, and it is difficult to predict the kinds of vulnerabilities that will emerge over time—as is often the case when negotiating con- tracts at the outset of a long-term joint venture, or among partners in a start-up environment— good- will-based trust is likely to be critical, and manag- ers may choose to reduce the emphasis on control and increase the emphasis on coordination. If competence-based trust is critical, as it is in re-
lationships in which one or both of the parties are providing technically or operationally complex services, managers may increase reliance not only on coordination provisions (an obvious ini- tiative), but also on control provisions.
Managers might also take note that although 29 percent of the disputes in our sample—all of which had escalated to the point that law firms were in- volved—were resolved with the parties intending to continue collaborating, contract structure influ- enced whether the relationships could be revived. They were more likely to survive when they con- tained fewer control provisions and more coordi- nation provisions. This finding suggests that if parties anticipate future conflict (for example, in cross-cultural relationships) they might choose contracting structures that, even if they are subop- timal in some ways (e.g., for minimizing risk via control), are better able to encourage trust develop- ment (through an increase in coordination provi- sions). This suggestion highlights another insight for managers to appreciate: the types of contracts that are best at avoiding conflict may not be the most helpful in situations in which conflict has not been avoided.
Limitations and Directions for Future Research
A number of limitations of the current research can be identified. First, we focus on only one type of outcome variable: relationship continuance. Re- search on other performance variables (e.g., profits) and relational variables (e.g., partner satisfaction) would be of clear benefit. Second, as with any such analysis, it is impossible to fully ensure that the results of our analyses precisely support the causal relationships we have predicted. Our longitudinal data, control variables, and numerous robustness checks help mitigate these concerns, but future re- search using experimental designs would nicely complement the current investigation. Third, al- though our empirical approach suggests a clear dis- tinction between control and coordination provi- sions, we acknowledge that some provisions may simultaneously accomplish both objectives. To ad- dress this concern, we (1) revised the coding scheme that was derived from existing research with the help of legal experts and (2) conducted two separate robustness checks. Future research could extend our approach by seeking other meth- ods of evaluating the coordination versus control functions of contracts.
In this study, we sought to conceptually refine and empirically extend previous work on the effect of contracts on trust and trust-related outcomes in interorganizational relationships. The findings pro-
994 OctoberAcademy of Management Journal
vide a more nuanced understanding of these issues, as well as unique and actionable theoretical and managerial insights. We hope our investigation aids future research that builds on the strengths of our approach and overcomes the weaknesses herein admitted.
REFERENCES
Agarwal, R., Croson, R., & Mahoney, J. T. 2010. The role of incentives and communication in strategic alli- ances: An experimental investigation. Strategic Management Journal, 31: 413– 437.
Ahn, T. K., Ostrom, E., Schmidt, D., Shupp, R., & Walker, J. 2001. Cooperation in PD games: Fear, greed, and history of play. Public Choice, 106: 137–155.
Argyres, N. S., Bercovitz, J., & Mayer, K. J. 2007. Comple- mentarity and evolution of contractual provisions: An empirical study of IT service contracts. Organi- zation Science, 18: 3–19.
Arrow, K. J. 1974. The limits of organization. New York: Norton.
Baird, D., Gertner, R., & Picker, R. 1994. Game theory and the law. Cambridge, MA: Harvard University Press.
Barnard, C. 1938. The functions of the executive. Cam- bridge, MA: Harvard University Press.
Barney, J. B., & Hansen, M. H. 1994. Trustworthiness as a source of competitive advantage. Strategic Manage- ment Journal, 15: 175–190.
Baron, R. M., & Kenny, D. A. 1986. The mediator-moder- ator variables distinction in social psychological research: Conceptual, strategic and statistical con- siderations. Journal of Personality and Social Psy- chology, 51: 1173–1182.
Barthélemy, J., & Quélin, B. V. 2006. Complexity of out- sourcing contracts and ex post transaction costs: An empirical investigation. Journal of Management Studies, 43: 1775–1797.
Beatty, J., & Samuelson, S. 2001. Business law for a new century. Cincinnati: West Legal Studies in Business.
Blodgett, L. L. 1992. Factors in instability of international joint venture: An event history analysis. Strategic Management Journal, 13: 475– 481.
Bottom, W. P., Gibson, K., Daniels, S. E., & Murnighan, J. K. 2002. When talk is not cheap: Substantive pen- ance and expressions of intent in rebuilding cooper- ation. Organization Science, 13: 497–513.
Burns, T., & Stalker, G. M. 1961. The management of innovation. London: Tavistock.
Camerer, C. F. 2003. Behavioral game theory: Experi- ments in strategic interaction. Princeton, NJ: Prince- ton University Press.
Chatterjee, S., & Price, B. 1991. Regression analysis by example (2nd ed.). New York: Wiley.
Das, T. K., & Teng, B. S. 2001. Trust, control, and risk in strategic alliances: An integrated framework. Orga- nization Studies, 22: 251–283.
Davis, J. H., Schoorman, F. D., Mayer, R. C., & Tan, H. H. 2000. The trusted general manager and business unit performance: Empirical evidence of a competitive advantage. Strategic Management Journal, 21: 563– 576.
Dekker, H. C. 2008. Partner selection and governance design. Accounting, Organizations and Society, 33: 915–941.
Dhanaraj, C., & Beamish, P. W. 2004. Effect of equity ownership on the survival of international joint ven- tures. Strategic Management Journal, 25: 295–305.
Dirks, K. T., Lewicki, R. J., & Zaheer, A. 2009. Repairing relationships within and between organizations: Building a conceptual foundation. Academy of Management Review, 34: 401– 422.
Faems, D., Janssens, M., Madhok, A., & Van Looy, B. 2008. Towards an integrative perspective on alliance governance: Connecting contract design, contract ap- plication, and trust dynamics. Academy of Manage- ment Journal, 51: 1053–1078.
Ferrin, D. L., & Dirks, K. T. 2003. The use of rewards to increase and decrease trust: Mediating processes and differential effects. Organization Science, 14: 18 –31.
Foss, K., Foss, N. J., & Klein, P. G. 2007. Peripheral vision. Organization Studies, 28: 1893–1912.
Ghoshal, S., & Moran, P. 1996. Bad for practice: A critic of the transaction cost theory. Academy of Manage- ment Review, 21: 13– 47.
Gulati, R. 1995a. Does familiarity breed trust? The impli- cations of repeated ties for contractual choice in alliances. Academy of Management Journal, 38: 85–112.
Gulati, R. 1995b. Social structure and alliance formation patterns: A longitudinal analysis. Administrative Science Quarterly, 40: 619 – 652.
Gulati, R., Lawrence, P. R., & Puranam, P. 2005. Adapta- tion in vertical relationships: beyond incentive con- flict. Strategic Management Journal, 26: 415– 440.
Gulati, R., & Nickerson, J. 2008. Interorganizational trust, governance choice, and exchange performance. Or- ganization Science, 19: 688 –708.
Hagedoorn, J., & Hesen, G. 2009. Contractual complexity and the cognitive load of R&D alliance contracts. Journal of Empirical Legal Studies, 6: 867–903.
Heide, J. B., & John, G. 1992. Do norms matter in market- ing relationships? Journal of Marketing, 56(2): 32– 44.
2011 995Malhotra and Lumineau
Hinds, P. J., & Mortensen, M. 2005. Understanding con- flict in geographically distributed teams: The mod- erating effects of shared identity, shared context, and spontaneous communication. Organization Science, 16: 290 –307.
Inkpen, A. C., & Beamish, W. P. 1997. Knowledge, bar- gaining power and the instability of international joint venture. Academy of Management Review, 22: 177–202.
Jap, S. D., & Ganesan, S. 2000. Control mechanisms and the relationship life cycle: Implication for safeguard- ing specific investments and developing commit- ment. Journal of Marketing Research, 37: 227–245.
Jauch, L. R., Osborn, R. N., & Martin, T. N. 1980. Struc- tured content analysis of cases: A complimentary method for organizational research. Academy of Management Journal, 5: 517–526.
Joskow, P. L. 1987. Contract duration and relationship- specific investments: Empirical evidence from coal markets. American Economic Review, 77: 168 –185.
Kim, P. H., Dirks, K. T., Cooper, C. D., & Ferrin, D. L. 2006. When more blame is better than less: The implications of internal vs. external attributions for the repair of trust after a competence- vs. integrity- based trust violation. Organizational Behavior and Human Decision Processes, 99: 49 – 65.
Kim, P. H., Ferrin, D. L., Cooper, C. D., & Dirks, K. T. 2004. Removing the shadow of suspicion: The effects of apology versus denial for repairing competence- versus integrity-based trust violations. Journal of Applied Psychology, 89: 104 –118.
Knez, M., & Camerer, C. F. 2000. Increasing cooperation in prisoner’s dilemmas by establishing a precedent of efficiency in coordination games. Organizational Behavior and Human Decision Processes, 82: 194 – 216.
Kogut, B. 1988. Joint ventures: Theoretical and empirical perspectives. Strategic Management Journal, 9: 319 –332.
Kramer, R. M. 1999. Trust and distrust in organizations: Emerging perspectives, enduring questions. In J. T. Spence (Ed.), Annual review of psychology, vol. 50: 569 –598. Palo Alto, CA: Annual Reviews.
Kretschmer, T., & Puranam, P. 2008. Integration through incentives in differentiated organizations. Organi- zation Science, 19: 860 – 875.
Lazzarini, S. G., Miller, G. J., & Zenger, T. R. 2004. Order with some law: Complementarity versus substitution of formal and informal arrangements. Journal of Law, Economics and Organization, 20: 261–298.
Lewicki, R. J., McAllister, D. J., & Bies, R. J. 1998. Trust and distrust: New relationships and realities. Acad- emy of Management Review, 23: 438 – 458.
Luce, R. D., & Raiffa, H. 1957. Games and decisions. New York: Wiley.
Lui, S. S., & Ngo, H. Y. 2004. The role of trust and contractual safeguards on cooperation in non-equity alliances. Journal of Management, 30: 471– 485.
Lumineau, F., Fréchet, M., & Puthod, D. 2011. An organ- izational learning perspective on contract design. Strategic Organization, 9: 8 –32.
Luo, Y. 2002. Contract, cooperation, and performance in international joint ventures. Strategic Management Journal, 23: 903–919.
Macaulay, S. 1963. Non-contractual relations in busi- ness. American Sociological Review, 28: 55–70.
Macneil, I. 1978. Contracts: Adjustment of long-term eco- nomic relations under classical, neoclassical, and relational contract law. Northwestern University Law Review, 72: 854 –902.
Malhotra, D. 2004. Trust and reciprocity decisions: The differing perspectives of trusters and trusted parties. Organizational Behavior and Human Decision Processes, 94: 61–73.
Malhotra, D., & Murnighan, J. K. 2002. The effects of contracts on interpersonal trust. Administrative Sci- ence Quarterly, 47: 534 –559.
Mayer, K. J., & Argyres, N. S. 2004. Learning to contract: Evidence from the personal computer industry. Or- ganization Science, 15: 394 – 410.
Mayer, K. J., & Teece, D. J. 2008. Unpacking strategic alliances: The structure and purpose of alliance ver- sus supplier relationships. Journal of Economic Be- havior and Organization, 66: 106 –127.
Mayer, R. C., & Davis, J. H. 1999. The effect of the per- formance appraisal system on trust for management: A field quasi-experiment. Journal of Applied Psy- chology, 84: 123–136.
Mayer, R. C., Davis, J. H., & Schoorman, F. D. 1995. An integrative model of organizational trust. Academy of Management Review, 20: 709 –734.
McAdams, R. H. 2009. Beyond the prisoners’ dilemma: Coordination, game theory, and the law. Southern California Law Review, 82: 209 –258.
Mellewigt, T., Madhok, A., & Weibel, A. 2007. Trust and formal contracts in interorganizational relationships: Substitutes and complements. Managerial and De- cision Economics, 28: 833– 847.
Mintzberg, H. 1994. The rise and fall of strategic plan- ning. Hemel Hempstead, U.K.: Prentice Hall.
Molm, L. D., Takahashi, N., & Peterson, G. 2000. Risk and trust in social exchange: An experimental test of a classical proposition. American Journal of Sociol- ogy, 105: 1396 –1427.
Morgan, R. M., & Hunt, S. D. 1994. The commitment-trust
996 OctoberAcademy of Management Journal
theory of relationship marketing. Journal of Market- ing, 58: 20 –38.
Morrison, E. W., & Robinson, S. L. 1997. When employ- ees feel betrayed: A model of how psychological contract violation develops. Academy of Manage- ment Review, 22: 226 –256.
Nahapiet, J., & Ghoshal, S. 1998. Social capital, intellec- tual capital, and the organizational advantage. Acad- emy of Management Review, 23: 242–266.
Ness, H. 2009. Governance, negotiations, and alliance dynamics: Explaining the evolution of relational practice. Journal of Management Studies, 46: 451– 480.
Nooteboom, B. 1996. Trust, opportunism and gover- nance: A process and control model. Organization Studies, 17: 985–1010.
Park, S. H., & Russo, M. 1996. When competition eclipses cooperation: An event history analysis of alliance failure. Management Science, 42: 875– 890.
Parkhe, A. 1993. Strategic alliance structuring: A game theoretic and transaction cost examination of inter- firm cooperation. Academy of Management Jour- nal, 36: 794 – 829.
Pillutla, M. M., Malhotra, D., & Murnighan, J. K. 2003. Attributions of trust and the calculus of reciprocity. Journal of Experimental Social Psychology, 39: 448 – 455.
Pirrong, S. C. 1993. Contracting practices in bulk ship- ping markets: A transactions cost explanation. Jour- nal of Law and Economics, 36: 937–976.
Poppo, L., & Zenger, T. 2002. Do formal contracts and relational governance function as substitutes or com- plements? Strategic Management Journal, 23: 707– 726.
Puranam, P., Singh, H., & Zollo, M. 2006. Organizing for innovation: Managing the coordination-autonomy dilemma in technology acquisitions. Academy of Management Journal, 49: 263–280.
Puranam, P., & Vanneste, B. S. 2009. Trust and gover- nance: Untangling a tangled web. Academy of Man- agement Review, 34: 11–31.
Reuer, J., & Ariño, A. 2002. Contractual renegotiations in strategic alliances. Journal of Management, 28: 47– 68.
Reuer, J., & Ariño, A. 2007. Strategic alliance contracts: Dimensions and determinants of contractual com- plexity. Strategic Management Journal, 28: 313– 330.
Reuer, J., Ariño, A., & Mellewigt, T. 2006. Entrepreneur- ial alliances as contractual forms. Journal of Busi- ness Venturing, 21: 306 –325.
Ring, P. S., & Van de Ven, A. H. 1992. Structuring coop-
erative relationships between organizations. Strate- gic Management Journal, 13: 483– 498.
Robinson, S. L. 1996. Trust and breach of the psycholog- ical contract. Administrative Science Quarterly, 41: 574 –599.
Ross, L., & Stillinger, C. 1991. Barriers to conflict resolu- tion. Negotiation Journal, 7: 389 – 404.
Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. 1998. Not so different after all: A cross-discipline view of trust. Academy of Management Review, 23: 393– 404.
Salbu, S. R. 1997. Evolving contract as a device for flex- ible coordination and control. American Business Law Journal, 34: 329 –384.
Sampson, R. C. 2004. The cost of misaligned governance in R&D alliances. Journal of Law, Economics and Organization, 20: 484 –526.
Schelling, T. 1963. The strategy of conflict. New York: Oxford University Press.
Schoorman, F. D., Mayer, R. C., & Davis, J. H. 2007. An integrative model of organizational trust: Past, pres- ent, and future. Academy of Management Review, 32: 344 –354.
Sitkin, S. B., & Roth, N. L. 1993. Explaining the limited effectiveness of legalistic “remedies” for trust/dis- trust. Organization Science, 4: 367–392.
Smitka, M. J. 1994. Contracting without contracts: How the Japanese manage organizational transactions. In S. B. Sitkin, & R. J. Flies (Eds.), The legalistic organ- ization: 91–109. Thousand Oaks, CA: Sage.
Sobel, M. E. 1982. Asymptotic confidence intervals for indirect effects in structural models. In S. Leinhardt (Ed.), Sociological methodology: 290 –312. San Francisco: Jossey-Bass.
Tenbrunsel, A. E., & Messick, D. M. 1999. Sanctioning systems, decision frames, and cooperation. Admin- istrative Science Quarterly, 44: 684 –707.
Uzzi, B. 1997. Social structure and competition in inter- firm networks: The paradox of embeddedness. Ad- ministrative Science Quarterly, 42: 35– 67.
Vanneste, B. S., & Puranam, P. 2010. Repeated interac- tions and contractual detail: Identifying the learning effect. Organization Science, 21: 186 –201.
Vlaar, P. W. L. 2008. Contracts and trust in alliances. Discovering, creating, and appropriating value. Cheltenham: Edward Elgar.
Vlaar, P. W. L., Van den Bosch, F. A. J., & Volberda, H. W. 2007. On the evolution of trust, distrust, and formal coordination and control in interorganizational rela- tionships. Group and Organization Management, 32: 407– 429.
Walker, G., & Weber, D. 1984. A transaction cost ap-
2011 997Malhotra and Lumineau
proach to make-or-buy decisions. Administrative Science Quarterly, 29: 373–391.
Weaver, K. M., & Dickson, P. H. 1998. Outcome quality of small-to medium-sized enterprise-based alliances: The role of perceived partner behaviors. Journal of Business Venturing, 13: 505–522.
Weber, J. M., Malhotra, D., & Murnighan, J. K. 2005. Normal acts of irrational trust: Motivated attribu- tions and the trust development process. In R. M. Kramer (Ed.), Research in organizational behavior, vol. 27: 75–101. New York: Elsevier/JAI.
Weber, R. P. 1990. Basic content analysis. Thousand Oaks: Sage.
Wicks, A. C., Berman, S. L., & Jones, T. M. 1999. The structure of optimal trust: Moral and strategic impli- cations. Academy of Management Review, 24: 99 – 116.
Williamson, O. E. 1985. The economic institutions of capitalism. New York: Free Press.
Williamson, O. E. 1991. Comparative economic Organiza- tion: The analysis of discrete structural alternatives. Administrative Science Quarterly, 36: 269 –296.
Zaheer, A., Lofstrom, S., & George, V. 2002. Interpersonal and organizational trust in alliances. In F. Contractor & P. Lorange (Eds.), Cooperative strategies and al- liances: What we know 15 years later: 347–377. London: Elsevier Science.
Zaheer, A., & Venkatraman, N. 1994. Determinants of electronic integration in the insurance industry: An empirical test. Management Science, 40: 549 –566.
Zand, D. E. 1972. Trust and managerial problem solving. Administrative Science Quarterly, 17: 229 –239.
APPENDIX A
Examples of Statements Coded as Competence- or Goodwill-Based Trust
Response categories were derived from definitions of trust dimensions in Davis, Schoorman, Mayer, and Tan (2000), Mayer and Davis (1999), Mayer and colleagues (1995), and Schoorman, Mayer, and Davis (2007).
Competence-Based Trust Messages were coded for references to skills, com-
petencies, aptitude, training, and/or experience. Examples: “We know that you are able to do it properly.” “Usually you deliver it on time.” “My engineers told me that they are confident about
your experience in the [...] field.” “You manifested your high level of competence during
Phase 1 of the Project.”
Goodwill-Based Trust Messages were coded for references to benevolence
and/or integrity. Benevolence examples: “We know you want the success of this Project.” “We really appreciated your technicians’ efforts to re-
pair the damage during the night.” “Your employees have been kind and friendly to help
[Firm A] to face this issue.” Integrity examples: “So far, you have been fair and honest.” “[Firm B] is well known for respecting its employees.
It is what gives you a great reputation!” “You have moral principles and I like that.”
Deepak Malhotra (dmalhotra@hbs.edu) is a professor at Harvard Business School. He received his Ph.D. from the Kellogg School of Management at Northwestern Univer- sity. His research interests include issues related to trust development, competitive escalation, negotiation, and international and ethnic conflict.
Fabrice Lumineau (fabrice.lumineau@uts.edu.au) is an assistant professor at the University of Technology, Syd- ney. He received his Ph.D. in strategic management from HEC Paris. His research interests include interorgani- zational partnerships, governance design, contractual structures, and dispute negotiation strategies.
998 OctoberAcademy of Management Journal
Copyright of Academy of Management Journal is the property of Academy of Management and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.