When does the network organizational form fail? Examining the impact of project characteristics on organizational structure and performance

This dissertation integrates economic and sociological approaches to network organizing to explain the structure and performance of network organizational forms. Previous theorizing from economics and sociology linked network organizational structure to “pairwise” or dyadic assessments of transaction efficiency and relational efficacy. Research based on these theories offered only partial understanding of network organizational performance because this work ignores the impact of multiple dyads interacting simultaneously, which occurs at the network level of analysis. This study integrates economic and sociological theories, treating them as interdependent explanations of network structure and performance. Theory is developed at the network level of analysis, which is necessary to explain the structure and performance of network organizations. Taking a network governance perspective, I formulate a theoretical model predicting the impact of exchange conditions upon the structure and performance of network organizations. I focus upon a specific variant of network organizations, “temporary interorganizational networks” (TINs), and develop and test hypotheses derived from transaction cost economics and from the sociological perspective focusing on relational embeddedness. I test these hypotheses by constructing a unique dataset containing comprehensive financial, organizational, and performance information regarding a population of network organizations during the years 2000–2007. Each observation in this dataset constitutes a network form designed to address a specific project, and these observations include both those networks that succeeded and those that failed. The study's design overcomes a limitation of prior cross-sectional analyses: Most prior analyses treat network ties as durable and assume that all ties add value to an organization. This assumption is challenged by empirical findings suggesting that the value of a relational tie decays rapidly with time. In contrast, the transient relationships common in TINS repeatedly form and dissolve over time. By observing both the formation and dissolution of ties and both successful and unsuccessful interorganizational networks, this study is among the first to test the full range of network organizational performance. My results indicate that exchange conditions significantly affect both the structure and performance of the network organizational form. Additionally, analyses reliably predict failure of the network form, which amends and extends prior theory.