Wednesday, May 27, 2009

We Can't Stop Carbon Emissions (redux)

. Wednesday, May 27, 2009

Ryan Avent takes Wilkinson to task for having the audacity to claim that there may be some collective action problems involved in getting the major world powers to credibly commit to reduce their carbon emissions, and in the process unwittingly stumbles onto the principle problem in international relations: in the absence of an over-arching international authority, cooperation among states is very difficult to achieve. Problem is, he has no idea that he's stumbled onto any problem at all. In fact, he denies that there even is a problem:

This seems almost deliberately dense. In particular, it makes no distinction between the world of billions of daily, anonymous transactions and the world in which a handful of great powers attempt to hammer out a diplomatic agreement. Unsurprisingly, it's very difficult to get millions of urban denizens to voluntarily come together to build and fund a road network or transit system in the absence of a coercive mechanism. The benefits are too broadly shared, and the incentive to free ride too great. But the smaller the number of players, the more concentrated the benefits, and the easier it is to find a mutually beneficial agreement. ...

In fact, there are fewer than ten relevant players, and only two really relevant players not already committed to reductions -- the US and China. Given that climate negotiations are part of a repeated game between the two great powers (that is, they're more or less constantly talking about one economic or political issue or another), it seems very likely indeed that an American pre-commitment to emission reductions would facilitate a similar Chinese commitment.

Ignore the rich condescension for a moment, and the irony as well. Avent is poking around at a game-theoretical argument: an over-arching international government isn't necessarily needed, because we only have to get a "handful of great powers" to agree to things, and coordination is easier with a smaller number of players. Well, that should be simple then. But there are all sorts of problems with this.

First, while it might be true that it is easier to coordinate among fewer players, it isn't necessarily true. In many game-theoretic models the number of players is irrelevant: coordination is just as difficult with 2 players as it is with 2,000. In other models, there may be lots of players, but only some have any ability to influence the final decision. In the picture Avent is trying to paint, this is actually the case: despite there being 200-odd states, really all that's necessary is securing the commitments of a relatively small number of those.

But even if Avent were completely correct about this, he's missing something else entirely: the salience of the issue. It's easier to tax millions of people a small amount and build roads with the proceeds than it is for neighbors in a property dispute to resolve the matter without litigation. If Avent's story were complete, this would not be the case. But the history of international relations is filled with instances of disputes between two states that lead to very costly wars or trade disputes. This occurs when the salience of the issue is high, or when the costs of some action are sufficiently high to dissuade a state from taking that action.

Finally, even if states could negotiate an agreement, there is no mechanism to prevent them from defecting. States say one thing and do another all the time, especially in violation of multi-lateral agreements like this. This is what Wilkinson was really driving at, and Avent doesn't even seem to consider the possibility. But this is a very, very big problem. In a nutshell, this is what the study of international relations is all about: finding the conditions under which actual cooperation occurs, and not just cheap talk. Turns out, it's harder then one might think, especially if the major powers are in major disagreement as in this case.

Even Avent's own examples speak to this: he says that of the "necessary" handful of countries needed to reach a meaningful agreement, only the U.S. and China have not committed to reducing emissions. Well, perhaps, but an agreement is not the same as an actual change in behavior. Among E.U. states, only the U.K. and Sweden are on track to meet their Kyoto obligations by 2010, and the U.S. has actually had fewer increases in emissions than many countries that ratified the treaty (including Canada and Australia, as well as a half-dozen or so E.U. states). The incentive will always be to defect, because compliance is costly and enforcement is essentially impossible. One simply cannot assume, as Avent does, that these collective action problems magically disappear if a formal agreement is made. In fact, Avent does worse than that by arguing that if the U.S. could only lead the way, China would surely follow. Really? He's never heard of the free-rider problem? When have the Chinese followed the U.S. on anything? China has given absolutely no indication that they will be willing to curb emissions if that means reducing economic growth by any calculable amount. Nor can the Chinese regime halt growth and still remain in power. And without China, the world's largest polluter (and the gap is growing amazingly fast), no overall progress can be made. And this says nothing of India or Brazil or Eastern Europe or Southeast Asia or Africa (if that continent ever starts achieving real growth). Getting real global cooperation on this issue would be the single greatest achievement in the history of international institutions. Which makes it very unlikely that it will happen.

In any case, Avent is arguing in this way to support the Waxman-Markey cap-and-trade legislation currently in the U.S. Congress. Strangely, Avent doesn't even like this bill, calling it "highly imperfect" and "totally inadequate". He also acknowledges that more progressive legislation is politically impossible despite an amazingly popular president who campaigned on this very issue and exceptionally large majorities in both houses of Congress. (Indeed, even this watered-down bill is unlikely to become law.) So Avent finds himself using deeply flawed logic to argue in favor of a wasteful, ineffectual bill that will make some people worse off by imposing a series of costs on them, but will not make anyone better off. He hopes that this bill will create better options in the future, but why? The electorate seems more likely to oppose future actions if present actions are wasteful and ineffective.

This is all odd, because Avent is usually much better than this, and he certainly knows enough economics to know better. So what's going on?

For even more difficulties, please see Tyler Cowen and Jim Manzi, and this previous post by me.


We Can't Stop Carbon Emissions (redux)




Add to Technorati Favorites