Designing a climate agreement

Now, in the run-up Paris, countries are trying again to address these same two issues. This time around, they have solved the first problem by asking countries to submit their "intended nationally determined contributions"-in plain language, voluntary pledges. They initially tried to solve the second problem by casting these pledges within a framework of "assessment and review"-meaning, a process by which pledges would be reviewed by other countries for their adequacy, with follow-up to see if countries actually fulfilled their pledges. As the agreement eschews material sanctions, this approach amounts to enforcement by "naming and shaming".

This framework may not survive the negotiation process. The latest version of the agreement, incorporated within a "non-paper" by the co-chairs of the Paris conference and not negotiated by the parties themselves, makes little mention of a review procedure, referring only to "tracking of progress" and "stocktaking." However, even if the Paris Agreement does not explicitly incorporate a review process, it also can do nothing to stop such a process from unfolding informally.

Can a process of assessment and review really change behavior? It will be hard to know. First, it will take years to know whether countries fulfill their pledges. More fundamentally, we'll never be sure whether the process got countries to pledge and actually to do more than they would have done in the absence of the process. To see this we'd need a "counterfactual"-evidence for what these countries would have done in the absence of the review process.

In my presentation, I shall report the results of a laboratory experiment, done jointly with Astrid Dannenberg of the University of Kassel, on the effect of a review process on voluntary contributions to supply a public good. Our lab experiment is valuable because it provides the needed counterfactual.

Our results show that a review process: (1) increases group targets directly; (2) increases pledges by individual players indirectly; and (3) increases contributions in an even more roundabout way, but that the effect is small and statistically insignificant. Our results also show that the timing at which the review is done (an important feature of an earlier draft of the Paris agreement) affects targets and pledges but not contributions. Around these tendencies, we observe enormous variation in group-level contributions. Group composition, we show, matters much more than whether or not there is a review process. The groups that contribute a lot are comprised of "conditional cooperators." Groups that contribute little have too many "free riders."

This observation might seem to suggest that the way to address climate change is for the conditional cooperators to cooperate amongst themselves and to shun the "free riders." When groups form "clubs" they can choose their members, and there is now much interest in eschewing the UN process in favor of a club approach to addressing climate change. However, to work, the non-members of a club must be excluded from enjoying the benefits created by the club, and as emission reductions are a global public good, no country can be excluded from benefiting from limits on climate change.

To succeed in addressing climate change, clubs must therefore focus not on emission reductions but on something like cooperation in the development of a new technology or special trade arrangements. Having done this, clubs must then leverage their supply of the "good" for the purpose of getting all countries to limit emissions. William Nordhaus has recently analyzed one way of doing this-forming a trade club, in which members adopt an emission reduction policy and impose tariffs on countries that refuse to join in their effort to limit climate change. However, as this arrangement implies coercion, it could provoke retaliation, stimulating conflict rather than cooperation.

More than a decade ago, I proposed a related but somewhat different approach, emphasizing the need for an agreement to focus on choices that facilitate coordination. With this approach, conditional cooperators would offer a combination of sticks and carrots to broaden participation and increase contributions. This could involve collective adoption of a strategic trade measure, as was done in the Montreal Protocol, an agreement adopted to protect the stratospheric ozone layer, or the setting of technology standards (the adoption of which would involve trade restrictions implicitly), as was done in the MARPOL agreement to limit the release of oil into the sea by tankers. In contrast to the proposal by Nordhaus, this one would focus on individual gases and sectors. Rather than entail a single overarching agreement, this approach would comprise a family of separate but related agreements.

The past 25 years of climate negotiations have shown that the world cannot rely on the goodwill of a small number of conditional cooperators. The best way to limit global emissions is to devise mechanisms that allow the behavior of a core group of conditional cooperators to be leveraged to bring about meaningful global collective action. The Paris Agreement is unlikely to do this. However, it is also unlikely to stand in the way of such additional efforts. This is where I think attention should focus in the years after Paris-not only on bringing into force and implementing what has already been agreed, but also on pursuing complementary efforts, targeted to limit the emissions of individual gases and sectors.