Möller, KlausKlausMöllerEgle, StefanStefanEgle2023-04-132023-04-132014-06-04https://www.alexandria.unisg.ch/handle/20.500.14171/86824The purpose of this paper is to explain the interrelation between formal control and trust in dyadic buyer-supplier relationships when facing appropriation concerns. From a methodological perspective our starting point is not to take as given an often suggested humped-shaped relationship between theses control devises (Dekker, 2004, Tomkins, 2001), but to provide a dynamic model that generates endogenously such a pattern. This model involves trust generation mechanisms that interrelate with formal control in the contract and execution phase of a buyer-supplier relationship. The model structure involves feedback loops that contribute to diminishing causality concerns and is able to go beyond a comparative analysis. From a theoretical perspective, we reconsider the control problem of appropriation concerns when buyers face a performance risk. Our simulation results imply that a performance shock triggers a complementary (substitutive) relationship between goodwill (contractual) trust and monitoring frequency. Efficiency or effectivity gains are able to restore the humped-shaped relationships between the trust types and formal control. From a practical perspective and a transaction costs economics view, these results imply that giving a second chance to a supplier in favor of a long-term relationship may establish cost saving trust as a control devise despite the shock. Two directions determine the next steps for theory development induced by the model. First, the model is extended by asset specificity and partner selection. Asset specificity via idiosyncratic investments is regarded as a further antecedent of appropriation concerns; partner selection, is considered as an alternative to formal control. Second, the causal structure of the dynamic model may be tested empirically with a survey-based, non-recursive structural equation model. We consider these recommondations as invitations for future work. In conclusion, we consider our dynamic model that involves a humped-shaped relationship between the control devises as a candidate for extending existing theories regarding interfirm relationships.enconference paper