Whereof Game Theory Cannot Speak….

Conor Clarke's interview with Thomas Schelling is fascinating. And also a bit confusing. And disappointing. Schelling is one of my intellectual heroes. But in this interview, he seems to bounce back and forth between the kind of reasoning I  learned from his work and a kind of aspirational moralizing that I learned from his work to distrust.
For example, when Schelling speaks as an economist, he makes a great deal sense to me:

I think the best hope for India is to grow its economy as fast as it can in order to outgrow its vulnerability to climate change.
If I were to come clean to the American public I would say that, except for a very low probability of a very bad result — which is the disintegration of the West Antarctic ice sheet, which would put Washington DC under water — we are probably going to outgrow any vulnerability we have to climate change. And in case we'll be able to afford to buy food or import it is necessary. You know, very little of the US economy is susceptible to climate. All of agriculture is less than 3% of our gross product. Forestry may be endangered. Fisheries may be endangered. But recreation might actually benefit!
So if we can double our GDP in the next 70 or 80 years, even if we lose some of our GDP from climate change — even if we lose 10% of our GDP from climate change — we're still ahead so much that the effect of climate change wouldn't be noticed. But it would be pretty disastrous in a lot of the less developed parts of the world. And that's why I think it's crucially important not to demand anything of China, India and so forth that will significantly impede their economic progress.

Right on.
But I feel like Schelling is not very enlightening on questions about policies that will affect future generations. For example:

[Clark] I wanted to ask more question, to go back to the moral issue here. It does seem to me that the strongest case for mitigating the effects of global climate change is a moral one. It is based on not on our own interest but on the interests of people in the developing world who don't yet exist. But it also seems to me that — while I don't know much about game theory — collective bargaining theories generally assume the participants are rational and self-interested. So how does one go about making sense of an arrangement where we must set our self-interest aside? How does one make the moral case in a situation like this? Or is my description of collective bargaining just totally idiotic?
Well, I think you have to realize that most people have very strong moral feelings. I think in a lot of cases they're misdirected. I wish moral feelings about a two-month old fetus were attached to hungry children in Africa. But I think people have very strong moral feelings. In fact, I'm always amazed by the number of people who at least pretend they're worried about the polar bears.

That's not wrong, but it's terrifically weak stuff. Which is not very surprising. There is little for a game theorist to do but moralize here, since game theory doesn't apply here. As readers of Derek Parfit know, deliberation about harm to future generations is doubly confusing because it's not even quite right to say that, although we can affect people in the future beyond our deaths, those people can't affect us. Who are “those people” we are talking about? There is a temptation to see future populations as determinate but simply yet-to-be-realized. That's wrong. Future populations are indeterminate. What we do now, the results of the games we are playing presently, determines who will and will not exist in the future.
So, for example, if we anticipate that the burdens of global warming will fall most heavily on future people in particular regions, we can mitigate those harms in any number ways. We could encourage a lower birthrate later by encouraging various techniques of birth control now. The fewer future Bangladeshis there are, the lower the possible total harm to future Bangladeshis. But… I'm not a fan of this idea. However, in some ways, economic growth is a form of birth control, since birthrates tend to fall with income growth. So, not only does greater wealth allow for greater adaptation, it lowers the expected total harm of current activities by reducing the expected population. Does anybody try to model this?
Or, to take another tack, if the negative consequences of today's actions will fall most heavily upon poorer people in certain regions a century hence, the regions most likely to benefit from warming, and those best equipped to adapt, can open themselves more fully to immigration from places where the greatest harm is predicted. We could increase the gains to warming by increasing the proportion of the world population residing in areas likely to benefit from it. Now, that sounds crazy. But it's just as much a possibility as massive wealth transfers from rich to poor countries, which Schelling recommends (and which, based on the record of development aid, I would anticipate to have many negative consequences).
The interesting question is how we coordinate now to facilitate any prevention or mitigation strategy. And Schelling is not even very helpful here. He is clearly worried. But, at the same time, he sounds pessimistic about the possibility of really effective coordination. It sounds like he thinks the best developed countries can do is to announce the intention to do something and start doing it, while basically hoping that it makes a difference. And the best advice he gives is not advice, but moral exhortation. And even then, it's pretty indirect. He exhorts churches to do more exhortation about climate change.

And one thing that I think ought to help but doesn't is that — and my impression is that maybe this is slightly changing — the organized churches in American don't take seriously preserving the heritage that God gave us. I've heard congressmen confess to being devote Episcopalians say that what god gave us ought to be preserved. But I get no impression that Protestants and Catholics are sermonizing on the importance of preserving the bounty of the earth, the richness of the species, or preserving the planet as we would like to know it. And I think that if someone could mobilize the church to be interested…
I spent a long time concerned with smoking behavior. And when I was a boy the churches were very adamant about smoking. And my grandfather, who was sort of devout, he wouldn't hire a boy to mow the lawn if he knew the boy smoked. And we know how potent the churches can be, because nobody smokes on campus at the University of Utah. And nobody smokes among the Seventh-Day Adventists or the Jehovah's Witnesses. And the Mormons aren't supposed to drink coke and coffee, let alone smoke.
And I think the churches don't realize that they could have a potent effect in not letting so much of gods legacy — in terms of flora and fauna — be destroyed by climate change.

More than a bit disappointing.

Author: Will Wilkinson

Vice President for Research at the Niskanen Center