Policy Devolution and Cooperation Dilemmas

• In Event: Representation, Inequality and Geography in Environmental and Climate Governance

Abstract

Politics is a social activity. Citizens engage in political activities to influence others and secure outcomes that enhance their welfare. Such activities, however, often have unintended consequences or spillovers for non-participants. Spillovers can either be negative and positive. Negative spillovers generate costs for non-participants and have been well-studied by the political economy community. More unexplored are positive spillovers that not only generate benefits for non-participants but also induce a collective action dilemma for downstream participants. Given their beneficiary status, downstream public goods recipients have incentive to limit their contribution. Moreover, the incentive to shirk may be particularly stark depending on one’s relative location and relationship to others in a social network. In this research, we are interested in exploring how such public good spillovers influence agent effort in an interconnected environment.

The problem of public good spillovers generating collective action dilemmas is ubiquitous. Homeowner choices over upkeep and renovation generate benefits for neighboring communities. Education and increases in human capital can enhance economic productivity (Moretti 2004}, increase wages of other workers (Moretti 2004), and ease the educational tasks for university professors (Anderson et al. 1994). Upstream governments and private firms can take actions that reduce environmental pollution loads for downstream entities entities (Monogan et al. 2017, Revesz 1996). Cross community engagement may ease communal tensions across racial, linguistic or religious lines (Dasgupta and Kanbur 2006). Despite the wide availability of this problem, there have been relatively few attempts to model this process.

The general problem has received relatively little attention in formal models of politics or policy, but it has been studied in a recent line of work in economic theory. Consequently, our model has some unique features both in its differences with extent spillover models (i.e. on the theoretical side) and also in how the agents are rewarded for effort (i.e. on the policy side). The model most similar to ours is by Bramoulle (2007), who develop a network model of public good provision, which we generalize to the continuous case and allow for large spillovers and work exclusively with heterogeneous agents, which (it turns out) allows us to learn more about how size and spillovers interact. In these regards, our model is similar to that of Bloch (2007), who concern themselves with both local and realm-wide public good provision across a continuum of agents. Our focus, however, is on formal political actors, and so our reward structure differs substantially from theirs.

Our model suggests that the imposition of administrative boundaries on a policy domain has the potential to exacerbate public good delivery. Artificially created administrative boundaries shape the quality of an agent’s jurisdiction by dissecting natural environmental resources limiting the degree to which agents responsible for managing these resources can use their discretion to shift blame to those officers with which they share this responsibility.

We test our expectations using a newly-constructed dataset of U.S. state regional offices charged with water pollution control responsibilities. These regional administrative offices are created by state environmental agencies and are used to implement activities under the U.S. Clean Water Act, such as permitting and compliance assurance. Using GIS analysis, we generate a novel dataset that spatially delineates each regional office's regulatory jurisdiction, its overlap with watershed boundaries, and using elevation and stream flow data whether the portion of the watershed within their jurisdiction is upstream or downstream of adjacent regional offices. To test our hypotheses, we use multilevel mixed-effects regression analysis to investigate the degree to which higher levels of fragmentation of environmental resources boundaries (i.e., watersheds) by administrative boundaries results in less regulatory effort from regional agents. The subjects of the analysis are about 6,000 major water pollution sources regulated under the U.S. Clean Water Act for which we have detailed historical data on compliance and regulatory activity.

Our analysis finds evidence consistent with our expectations, across multiple model specifications. We believe these results make contributions to both theoretical and empirical literatures on institutions, policy coordination, and environmental politics. In addition, the research has significant implications for environmental governance by examining the effects of fragmented institutions on regulatory effort, and ultimately risks to public health and the environment.

Authors

• Robert J. Carroll, University of Illinois at Urbana-Champaign

• David Konisky, Indiana University, Bloomington

• Christopher M. Reenock, Florida State University