Special Issue: Trust in an Organizational Context: Information Processing Moderators of the Effectiveness of Trust-Based Governance in Interfirm R&D Collaboration

It is generally assumed that improved outcomes accompany the use of trust as a governance mechanism in an interfirm relationship. Briefly, trust is a social lubricant that reduces the friction costs of existing trade and/or serves to increase the scope of trade. In contrast to this universalistic view, we posit that the performance of trust-based governance is contingent on the ability of trading partners to "read" each other and learn about counterpart behavior. These information-processing abilities allow firms to assess partner trustworthiness better, which reduces the risk of misplaced trust. The increased efficacy of communication and leaming from one another also enables them to better capitalize from the adaptation and revision possibilities uncovered through trust-based governance during the taskexecution phase. Given the central role of these cognitive requirements, we assess these contingent effects with data from a knowledge-intensive task setting. Using a sample of 129 firms that have engaged outside contractors on client-sponsored R&D projects, we find strong support for our thesis. Specifically, we find that trust-based governance has a larger positive impact on task performance when the client is more skilled at understanding the outsourced tasks at hand, the task itself requires skills that are relatively more readily taught (less tacit), and the task itself is organized in parallel with work being done at the contractor as well as the client. As a corollary, we also find that firms adopt trust-based governance to a greater extent as these information-processing abilities increase, as well as with the colocation of the contractor and client. Suggestions for engineering trust-based governance are offered. (Trust; Alliances; Collaboration/Partnership; Information Processing; Governance) Introduction In recent years, the prominence of collaborative economic relationships between firms has motivated a large literature on the role of trust in these ties. Its central theme is that trust may be an efficacious means of coordinating economic exchange, very much like contracting or ownership; indeed, it may well outperform the latter alternatives. This literature has made considerable progress in identifying the mechanisms of interfirm tmst (e.g., Gambetta 1988), but trust-performance outcomes have been dealt with in a largely undifferentiated manner. First, it is generally assumed that relatiotiships governed by tmst will unfold in a particular way—that firms which are tmsted will cooperate with and behave towards each other in a tmstworthy manner. Put another way, there is no systematic consideration of the costs and risks of misplaced tmst. Second, it is assumed that the increased fiexibility and superior information exchange associated with tmst will enable greater responsiveness to environmental change—ultimately resulting in enhanced performance. Indeed, although the tmst literature has often been criticized for its eclectic and often inconsistent underpinnings, these assumptions appear in both the economics stream (e.g., Kreps 1990) as well as the behavioral stream (e.g., Zaheer et al. 1998). We challenge these assumptions about tmst-based govemance of economic exchange. In reality, tmst-based govemance is quite complicated. Our core argument is simple. First, it is plausible that there might be instances 1047-7039/03/1401/0045/$05.0