Association for Public Policy Analysis & Management

How does policy learning affect collaborative patterns before and after natural disasters? The policy literature suggests that stakeholders with similar policy beliefs often work together to achieve shared policy goals. Furthermore, it is theorized that such beliefs change marginally over time in light of new information and experience, manifesting as stable patterns of collaboration. In the context of urban stormwater management, the rate at which policy learning occurs and the collaborative network configuration can both have important implications for planning and policy decisions to address problems stemming from rapid urbanization, aging infrastructure, and effects of climate change. Theory suggests that shocks are necessary to influence major belief change which can, in turn, motivate policy stakeholders to rewire their ties in the collaborative network. However, empirical evidence shows variation in when and whether policy learning occurs following shocks. To address this puzzle, our study examines cases of collaborative stormwater management networks in two North Carolina cities—Fayetteville and Wilmington—where Hurricanes Matthew and Florence caused extensive flooding and damage in 2016 and 2018. Using automated text analysis techniques, we leverage data from news reports and public meeting minutes. We then use inferential network modeling in a longitudinal analysis of policy learning and how it affects collaborative patterns before and after the hurricanes. Through this study, we build on current work examining the role of beliefs and learning in the policy process, as well as research on hazard mitigation.