Global environmental problems need regional solutions

Richard Tol, Dritan Osmany 23 June 2010



In 1994, Scott Barrett predicted that international environmental agreements were either broad but shallow, or deep but narrow (Barrett 1994). That is, international agreements had either many signatories who promised to do very little for the environment; or a few signatories who promised to do a lot. Phrased differently, an international environmental agreement will do little to improve welfare and to reduce emissions. Since then, international negotiators of a climate treaty have confirmed Barrett’s predictions with depressing regularity.

Barrett defines an international environmental problem as one in which there are private benefits from emissions and social disbenefits. In the opposite case, emission reduction has private costs and social benefits. A rational actor would prefer to freeride on other actors’ emission reductions. An international environmental agreement is an agreement between sovereign actors to provide a public good.

Barrett’s paper is based on d’Aspremont’s notion of cartel formation (d'Aspremont et al. 1983). An international environmental agreement is said to be stable if and only if it is:

  • internally stable – no member wants to leave the coalition
  • externally stable – no non-member wants to join the coalition
  • and profitable – coalition members are better off cooperating with each other than not at all.

Testing the theory

Barrett made a number of simplifying assumptions, all of which have stood the test of other scholars trying to overturn his basic result (Finus 2008). In a forthcoming paper (Osmani and Tol 2010), we test another assumption. Barrett assumes that there is one international environmental agreement only. What if there are more?

The definition of cartel stability is incomplete for multiple coalitions (Carraro 1999). An agreement is stable if it is profitable, internally stable, externally stable, and intercoalitionally stable – no member of one coalition wants to switch to another coalition. With one coalition, the equilibrium has to satisfy three conditions. With two coalitions, there are the same number of conditions per coalitions plus two intercoalition conditions, i.e. a total of eight. With n coalitions, there are 3n+n(n-1)=n2+2n conditions. This shows that the number of coalitions is necessarily small relative to the number of players. In our paper, we therefore only consider the case with two international environmental agreements.

We find the following. If there are a large number of players, Barrett’s result stands if there are two international environmental agreements instead of one. Emissions and welfare improve only slightly with respect to the situation in which each country acts unilaterally and selfishly. However, Barrett’s result does not stand if there are a small number of players. In that case, a second international environmental agreement does have a noticeable impact.

The intuition behind this result becomes clear if we pretend that the two agreements are negotiated in sequence (rather than in parallel as in the paper). If we start with a large number of non-cooperative players, and a small group forms the first coalition, then we are left with an almost as large number of non-cooperative players – a small group of which would form the second coalition. That is, the two coalitions are formed under virtually identical circumstances. On the other hand, if we start with a small number of players, and a small group forms the first coalition, then we are left with an even smaller number of non-cooperative players – a large fraction of which would form the second coalition. The majority of actors thus play cooperatively.

We therefore have the paradoxical result that multiple coalitions would help to solve regional environmental problems (e.g., acidification, eutrophication), but not global environmental problems (e.g., climate change). That, at least, is one interpretation of our result.

Another interpretation is as follows. International climate negotiations have been conducted through the UN. Every member state takes part, even those with puniest emissions. Our result shows that it does not help to break up the UN into smaller bands – say, the EU and the Copenhagen Five (Brazil, China, India, South Africa, US). It would help, however, to negotiate with the twenty or so largest emitters only. If such negotiations break down into two or three competing treaties, then emissions would still fall and welfare still rise.


Barrett, S (1994), "Self-Enforcing International Environmental Agreements", Oxford Economic Papers, 46:878-894.
Carraro, C (1999), "The Structure of International Environmental Agreements", in International Environmental Agreements on Climate Change, C Carraro (ed.), Kluwer Academic Publishers, Dordrecht:9-26.
d'Aspremont, C, A Jacquemin, JJ Gabszewicz, and JA Weymark (1983), "On the Stability of Collusive Price Leadership", Canadian Journal of Economics, 16(1): 17-25.
Finus, M (2008), "Game theoretic research on the design of international environmental agreements: Insights, critical remarks, and future challenges", International Review of Environmental and Resource Economics, 2(1):29-67.
Osmani, D and RSJ Tol (2010), "The Case of Two Self-Enforcing International Agreements for Environmental Protection with Asymmetric Countries", Computational Economics (forthcoming).



Topics:  Environment

Tags:  climate change, global warming, Cartels, coalitions

Professor or Economics, University of Sussex; and Professor of the Economics of Climate Change, Vrije Universiteit Amsterdam

Post-doctoral Researcher at CliSAP


CEPR Policy Research