How Can Integration Reduce Inefficiencies Due to Ex Post Adaptation?

How can integrated firms immediately settle ex post adaptations to unanticipated disturbances? While this question is crucial to the understanding of transaction cost economics (TCE), TCE has not provided any formal answer. This paper develops a model that explores this question by employing three behavioral assumptions: reference-dependent preference, self-serving bias, and shading. We present two reasons why integration can avoid costly renegotiations and realize immediate adaptations; these stem from the fact that while nonintegrated parties have to engage in negotiations for the adaptations, integrated firms can implement these by fiat. First, punishments for rejection of an order under integration are severer than those for rejection of an offer under non-integration. Second, under integration, the utility improvement for a subordinate from rejecting an order is not sufficient to offset the loss from a severe punishment. Furthermore, we point out a trade-off between immediate agreement and the aggregate sense of loss.