Networks with spatially distributed externalities

Regulating spatially distributed pollution poses a big challenge as the standard regulating instruments such as cap-and-trade or command-and-control systems that usually implement an efficient allocation do not work facing spatially distributed pollution or an authority to enforce them is missing. The problem of optimal pollution control is further aggravated by the presence of asymmetric information as many governments face limitations in monitoring the mitigation costs or the pollution flows. In this thesis, I consider multi-polluter networks and analyse if and to what extent the efficient levels of pollution can be implemented by voluntary agreements or regulatory instruments when pollution is spatially distributed. The thesis consists of three chapters. In chapter 1, co-authored with Ralph Winkler, we consider the problem of efficient emission abatement in a multi-polluter setting, where agents are located along a river in which net emissions accumulate and induce negative externalities to downstream riparians. Assuming a cooperative transferable utility game, we seek welfare distributions that are in the non-cooperative core and satisfy a specific fairness constraint. Meaning, we search for welfare distributions that satisfy all agents’ participation constraints, in that each coalition is at least as well off as it were if acting on its own and that is perceived to be fair, in that no coalition is better off than it were if all non-members of the coalition do not pollute the river at all. We show that the downstream incremental distribution, as introduced by Ambec and Sprumont (2002), is the only welfare distribution satisfying both constraints. In addition, we show that this result holds true for numerous extensions of our model. In chapter 2, we analyse the contractual mitigation of a global public bad along a river in the presence of a federalist governance structure, where the lower tiers have private information about their abatement costs. We propose that the federal government nominates one of the lower tiers to be the principal, who is authorized to offer mitigation contracts to the other tiers sharing the river. The elected principal can do so either in a centralizedmanner, i.e. he offers contracts simultaneously to all other tiers, or in a delegated manner, i.e. he starts an upstream and downstream sequential contracting process by contracting with his up- and downriver neighbouring tiers, to which he then gives the authority to subcontract with their respective neighbouring tiers till all tiers received a contract. We showthat under certain conditions, a nominated principal can achieve the same abatement allocation with the delegated as with the centralized contracting method while matching his expected costs. As all potential principals implement a different abatement allocation, the choice of the prime principal matters for the total expected costs occurring in the river basin. We show that the tier located most downriver, which is subject to the same informational constraints as the federal government, is never the best choice to be nominated as the principal. In the third and final chapter, we propose a second-best optimal solution to the problem of pollution abatement in a multi-polluter network with heterogeneously dispersed pollution. Instead of taking an exogenously given and predetermined pollution cap in a cap-and-trade system, the pollution cap is endogenized so that it is determined by the total cost-minimizing equilibrium of a cap-and-trade system. We show that with quadratic abatement costs and linear damage costs, the first-best optimal pollution cap implements the second-best cost-minimizing equilibrium of the cap-and-trade system for any network. However, the second-best optimal abatement allocation differs from the first-best optimal abatement allocation, implying higher second-best optimal total costs than first-best optimal. In particular, second-best optimal total abatement costs fall short of first-best total abatement costs, while second-best optimal total damage costs exceed first-best optimal damage costs. These findings hold for two different cap-and-trade systems considered, the emission permit market and the ambient pollution market