Network Learning, Trust, and Effectiveness in Collaborative Governance Networks: A Comparative Case Study of Social Impact Bonds

Abstract

At its core, this dissertation is about network effectiveness in a new type of collaborative governance network, one which introduces new stakeholders and new logics to the negotiating table. This new type of network is known as the Social Impact Bond (SIB). Network effectiveness is heavily theorized in the public management and administration literatures. Determinants of network effectiveness can be categorized into four categories: network structures, management strategies, process (trust, commitment to process/goals, power imbalances, legitimacy, network learning), and context (resource munificence, system stability, sociopolitical environment). I apply this theory to SIB networks, and identify one source of unexplained variation, the sequential introduction of stakeholders into the network. I argue that the first partnership to the negotiating table constitutes that “founding partnership”, which then embarks on an imprinting process impacting the structural, managerial, and process components of the network. The imprinting process in SIB networks is theorized to be particularly consequential given the wide variety of potential value conflicts in the network. While this challenge is universal, the solution is not. Instead, I theorize two ideal type paths (leveraging and inclusion) that consist of a differing set of cohesive set of structural, management practices, and process components leading to network effectiveness. I employ a comparative case study design and fuzzy set qualitative comparative approach to assess the coverage and consistency of the theorized paths based on the 16 cases for which there is no missing data. The results of this dissertation provide some support for the two theorized paths to network effectiveness for SIB networks, but also a few points of inconsistency. Future research directions stemming from this analysis are then proposed in the concluding chapter.