Climate games: Who’s on first? What’s on second?

  • Margaret Insley, Department of Economics, University of Waterloo
  • Peter A. Forsyth, David R. Cheriton School of Computer Science, University of Waterloo
A depiction of the earth sitting on a chessboard,

Executive Summary

In view of the recent turmoil caused by international trade negotiations, we can view climate change policy in a game theoretic setting. In other words, although the major players (e.g. major economic zones such as the U.S., EU and China) may benefit from global emission reduction, there is a natural tendency to engage in climate policy as a strategic game.

We study three different climate change games and compare with the outcome of choices by a Social Planner. In a dynamic setting, two players choose levels of carbon emissions. Rising atmospheric carbon stocks increase average global temperature which damages player utilities. Temperature is modeled as a stochastic differential equation. We contrast the results of a Stackelberg game with a game in which both players as leaders (a Leader-Leader or Trumpian game). We also examine a game, called an Interleaved game, where there is a significant time interval between player decisions. One or both players may be better of in these alternative games compared to the Stackelberg game, depending on state variables. We conclude that it is important to consider alternate game structures in examining strategic interactions in pollution games. We also demonstrate that the Stackelberg game is the limit of the Interleaved game as the time between decisions goes to zero.

In the case of two symmetric players (i.e. players with similar sized economies and expected damages due to increased global temperature), we have the following advice to
international emission reduction negotiators

  • In some cases, local politics may demand immediate response to a competing players actions (witness the current tit for tat tariffs in trade negotiations). In this case, there is little advantage to being a Stackelberg follower compared to the Trumpian strategy reacting as if they are the leader. This results in a very small reduction in utility (compared with the optimal follower choice) but causes a severe reduction in the the leader’s utility. Hence this strategy constitutes a credible threat, assuming that the political rewards from causing damage to the Trump player outweigh the small reduction in own utility. This rule can be summarized as \When faced with Trump, play Trump.”

  • If possible, a better strategy, when faced with a Trump player, is to interleave the game, i.e. delay reacting. This is a superior strategy to either the Stackelberg follower or leader-leader policy.

  • Of course, it is better (in terms of global utility) if everyone cooperates (the Social Planner policy). However, in today’s international environment, this seems unlikely.

This Research Report is part of the larger GRI Funded Research Project: Multi-Period MV Approach to Risk & Return in Climate Change Policy.