Game-theoretic analysis of international climate agreements: the design of transfer schemes and the role of technological change

Global warming is one of the crucial challenges that the world is facing now. The allocation of reduction efforts among regions has long been negotiated and it will not be an easy task to achieve a full cooperation with stringent targets. The thesis examines the formation of international climate agreements (ICAs) in a game-theoretic framework. I analyse strategic behaviour of a number of regions to reduce greenhouse gas (GHG) emissions. Game-theoretic approaches have been widely used to examine an interaction between countries in the negotiations on climate change, and have emphasised difficulties in designing such a voluntary agreement. This thesis provides a systematic approach to examine the impacts of designs of the ICAs on the success of ICAs. In Chapter 2, I present the basic structure of the STACO-2.1 model, composed of a game-theoretic framework and applied features, with specifications and calibrations of the functions used in the model. I analyse the results for (i) the All Singletons coalition structure, (ii) the Grand coalition structure, and (iii) all stable coalition structures. This case can serve as a suitable reference point for the analysis of the various mechanisms in the following chapters. The results show that a coalition of EU15 and Japan is stable. Both regions have an interest in cooperation, because of their higher marginal benefits from abatement. The results suggest that regions with relatively lower marginal abatement costs and lower marginal benefits would be worse off when they cooperate as they bear the largest burden of abatement but obtain the least benefits. This suggests that transfer schemes can be effective to stabilise larger coalitions. In Chapter 3, I analyse the impact of pragmatic and optimal transfer schemes on the incentives for regions to join international climate agreements. With an applied model that comprises twelve world regions I investigate: (i) a benchmark without transfers, (ii) scenarios with allocation-based rules where coalition members receive tradable emission permits proportional to initial or future emissions, (iii) scenarios with outcome-based rules where the coalition surplus is distributed proportional to the emissions, and (iv) a scenario based on an optimal sharing rule where the coalition surplus is distributed proportional to outside option payoffs. I find that when the transfer scheme is poorly designed in the sense that it increases incentives to free-ride, the best performing stable coalition may be worse than in the case of no transfers. In our applied setting this occurs for the initial-emissions-based tradable permit system (grandfathering). Improvements of the initial-emissions-based tradable permit system, such as a tradable permit system based on the full path of emissions or a surplus sharing scheme, do enhance stability of coalitions. For the optimal transfer scheme we find that larger coalitions, which include key players such as the United States and China, can be stable, but no transfer scheme is capable of stabilising the Grand Coalition. The results show that optimal transfers perform much better than more pragmatic transfer schemes. Such schemes, however, require detailed insight into the incentive structures of the regions. Therefore, there is a trade-off between more pragmatic schemes that may be easier to implement but are hardly effective and optimal transfers which may be hard to achieve in actual negotiations. In Chapter 4, I explore how different technology spillover mechanisms among regions can influence the incentives to join and stabilise an international agreement. Several theories on the impact of technology spillovers are evaluated by simulating a range of alternative specifications: (i) no spillovers, (ii) internal spillovers, (iii) global spillovers, (iv) coalitional spillovers, and (v) extended spillovers (all possible technology spillovers). I find that spillovers are a good instrument to increase the abatement efforts of coalitions and reduce the associated costs. In our setting, however, they cannot overcome the strong free-rider incentives that are present in larger coalitions. Therefore, technology spillovers do not substantially increase the success of international environmental agreements. This conclusion is robust with respect to the specification of technology spillovers. In Chapter 5, I relax the assumption of exogenous technological change analysed in the previous chapters and explore the impacts of induced technological change (ITC) on the stability of an international climate agreement. To examine the impacts of different specifications of technological change in reducing abatement costs on regional incentives, three scenarios are investigated: (i) no technological change (noTC), (ii) exogenous technological change (ExTC), and (iii) induced technological change (ITC). Technological change is induced by the abatement targets. It reduces emissions through regional R&D investments, which lowers abatement costs over time. The results reveal that the set of best-performing stable coalitions and the associated indicator of success hardly change between the scenario of noTC and ExTC, but ITC does produce a different set of stable coalitions. Coalitions that are stable in all three scenarios can achieve the highest NPV of payoffs in the case of ITC. The results indicate that coalition members increase their investments and abatement substantially when they cooperate in the case of ITC. As a result of increased global abatement, not only coalition members but also singletons obtain high benefits, which leads to decrease (increase) in the number of internally (externally) stable coalitions. Therefore, ITC might improve global payoffs, however, at the same time it tends to increase free-rider incentives due to the public good nature of global warming. I find that the indicator of success is quite robust with respect to the productivity of R&D. Furthermore, the number of internally (externally) stable coalitions decreases (increases) with the value of the productivity of R&D, as free-riding incentives increase. I find that stability is sensitive with respect to changes of the discount rate. The number of stable coalitions increases with the value of discount rate. In both cases, the dominating mechanism is that higher productivity of R&D or a lower discount rate increase the payoffs of regions, and thus increase the gains of cooperation, but also increase free-rider incentives. Overall, the main finding of this thesis is that well-designed mechanisms can facilitate successful formation of partial coalitions, although global cooperation is still hard to achieve. The reason lies in the public good nature of global warming and regional characteristics of benefits and costs. Following insights of the current literature on transfer schemes, our systematic analysis supports the fact that an incentive to join the agreement is highly sensitive to the design of transfers. For different designs of transfer schemes, there is a trade-off between feasibility and effectiveness. Allocation-based transfer schemes are easier to implement than an optimal transfer scheme which can achieve more successful coalition formation in the context of global payoffs and CO2 emissions. The role of technological change has received significant attentions to reduce a significant amount of emissions. Two types of sources of technological change are investigated in the thesis: (i) technology spillovers and (ii) R&D investment. If technology spillovers are treated as private goods, a country with higher abatement technology can be an attractive partner for other countries to cooperate with, as cooperation on abatement will lead to reduction in abatement costs without paying for technology improvement. Within the context of assumptions used here, quantitative results suggest that the spillovers between cooperating regions may not be effective enough to overcome the freeriding incentives for non-cooperative countries, as large emission reduction by cooperating countries will bring large benefits also to non-cooperating countries. In the case of induced technological change, regional R&D investments improve the stock of knowledge which leads to low abatement costs. When cooperating with other regions, signatories can obtain higher payoffs than in All Singletons, which is driven by increased investment after cooperation on abatement. ITC plays a significant role in increasing global payoffs, however, it also increases freerider incentives as non-signatories also benefit from a large reduction made by signatories. As long as R&D investments increase payoffs under cooperation on abatement and the gains from cooperation are large, the difference in the source of technological change will not provide any significant differences in terms of improvement in the success of a climate agreement. The Kyoto Protocol was the first significant step that provided stimulus for search for successful international cooperation on climate change policies, although there are issues to be improved. Now, negotiation on the post Kyoto framework has been taking place with an aim of large cooperation on tackling climate change among countries. Flexible measures which lead to a win-win cooperation for the countries involved will continue to play a crucial role in achieving a successful cooperation and the search for well-designed mechanisms will be further pursued.