Network Structure and Cooperation
in Regulatory Enforcement
by
John T. Scholz
and
Cheng-Lung Wang
Department of Political Science
Florida State University
August 2003
Abstract
Networks that coordinate local policy decisions can enhance the value of policy
outcomes for the multiple users and agencies by overcoming adverse effects of policy
fragmentation in federalist systems. Our previous study provides empirical evidence that
effective local water policy networks increase both enforcement and compliance rates in the
permit system authorized by the Clean Water Act. Our proposed APSA presentation intended
to extend this study to consider the compliance of government permit holders, but data
limitations and the problem of getting more accurate measures of networks precluded that study.
In particular, we found that the theoretical foundations for analyzing the effectiveness of
networks was too weak to allow us to readily develop the appropriate measures for networks.
Consequently, this study probes the theoretical basis for the policy gains attributed to
networks, focusing in particular on the potential gains from cooperation in the regulatory
enforcement game and the role of network structure in supporting cooperation. We present
simulation results that compare the impact of regular, random, and “small-world” networks of
varying density on levels of cooperation in the regulatory enforcement game.
Our full project explores the evolution of cooperation in the complex situation
involving multiple regulatory agencies and multiple firms. We present here the first half of the
exploration involving a single agency playing a “tit-for-tat” strategy against multiple firms.
Since no strategy can do better than cooperate in this relatively simple case, we investigate the
impact of alternative network structures on the level of cooperation achieved and the speed with
which high levels of cooperation are achieved.
The first third of the paper reviews the general perspective for analyzing the impact of
local networks on federal regulatory enforcement. The second third reviews some promising
concepts in network theory and discusses how we use agent-based simulations of the iterated
prisoner’s dilemma-- specifically the iterated cooperative enforcement game—to explore the
impact of network characteristics on the evolution of cooperative solutions. The final third
presents the results of the simulations.
______________________________________________________________________
Paper prepared for the Annual Meeting of American Political Science Association, Philadelphia, August
27 – 31, 2003. We would like to thank Troy Tassier for helpful comments
.
Page 1 of 31 
Convention