Self-Organizing Policy Networks: Risk, Partner Selection, and Cooperation in Estuaries


  • Research was partly funded by NSF grant SES-0519459. We thank Mark Schneider and Mark Lubell for use of data from NSF grant SBR97295905 and Tom Snijders and Christian Steglich for suggestions at several stages of the research. Replication materials are available at

Ramiro Berardo is Assistant Research Professor of Government and Public Policy, University of Arizona, Tucson, AZ 85721 ( John T. Scholz is Professor of Political Science, Florida State University, Tallahassee, FL 32306-2230 (


Policy actors seek network contacts to improve individual payoffs in the institutional collective action dilemmas endemic to fragmented policy arenas. The risk hypothesis argues that actors seek bridging relationships (well-connected, popular partners that maximize their access to information) when cooperation involves low risks, but seek bonding relationships (transitive, reciprocal relationships that maximize credibility) when risks of defection increase. We test this hypothesis in newly developing policy arenas expected to favor relationships that resolve low-risk dilemmas. A stochastic actor-based model for network evolution estimated with survey data from 1999 and 2001 in 10 U.S. estuaries finds that actors do tend to select popular actors as partners, which presumably creates a centralized bridging structure capable of efficient information transmission for coordinating policies even without any government mandate. Actors also seek reciprocal bonding relationships supportive of small joint projects and quickly learn whether or not to trust their partners.