The Telegraph
Since 1st March, 1999
Email This Page
- What game theory says about terrorism and global warming

You may be surprised to see me turning to Thomas Schelling's work so soon after Bhaskar Dutta wrote so elegantly on both Robert Aumann and Schelling, the joint winners of this year's Nobel Memorial Prize in economic science ('Not a zero sum', Oct 20). Let me explain why I want to talk about Schelling again. To be frank, Aumann would have been a bit too much for me to handle. He is an uncompromising mathematical theorist who had probably deserved the Nobel along with John Nash the first time the prize went to game theory. He works within the very core of that theory, not worrying too much about what levels of treatment mere mortals might feel comfortable with.

Schelling's many contributions to game theory are of an altogether different genre, for it is not only possible but also necessary for the concerned non-specialist to understand what he is saying. Throughout his writing life, Schelling has directly addressed the concerned people, taking great pains to explain how in all their concerns they have to listen to experts in diverse disciplines because they have to come to rational decisions of their own. The labour to be undertaken is herculean and not the kind of the easy 'interdisciplinary' research that is fashionable. One feels thankful that Schelling did not shut out the concerned outsider from this venture.

After his famous The Strategy of Conflict appeared in 1960 there has been a spate of writings asking what is game theory, and on the economics of segregation, 'self-command' as a new operational concept in the theory of rational choice, economics of global warning and the economic sociology of addictive drugs.

These might seem a disjoint set of themes until one sees the two unifying factors: Schelling's vision of the game-theoretic approach to public choice and his patent social concern that alone determines his choice of the themes. In each essay, Schelling stresses the interfaces of a steadily increasing number of disciplines. Let me quote from the one on global warming to give you an idea of the kind of interfaces he works on: 'Demography, economics, biology, and the technology sciences are needed to project emissions, atmospheric chemistry, oceanography, biology, and meteorology are needed to translate emissions into climates; biology, agronomy, health sciences, economics, sociology, and glaciology are needed to identify and assess impacts on human societies and natural ecosystems.

'And those are not all.'

Schelling's work invites the use of the game-theoretic approach in behavioural economics that came into vogue in the early Fifties. His analysis of strategic commitments has explained a wide range of behavioural phenomena, from the competitive strategies of firms to the delegation of political decisionmaking power. On a personal note, I can remember how excited I had been as a young researcher in choice theory 50 years ago, by the insights game theory had begun to provide. In that early enthusiasm, with no more than a rudimentary knowledge in the area, I myself published in 1956 my only game-theoretic paper 'Choice and Revealed Preference' in Econometrica to demonstrate that strategic voting will often hide and not reveal actual voter preference. I had felt quietly proud to see my modest exercise included in, for example, the melting of the polar ice cap, Games and Decisions by R.D. Luce and H. Raiffa in 1957.

Schelling's words have always merited close attention from economists and sociologists. Those working in development and environmental economics will now perhaps pay even greater heed to his writings after the announcement of the Nobel prize. It is in this context that I wish to share with you a few misgivings. Let me touch briefly on my anxiety with Schelling's treatments of two major concerns.

Dealing with the nuclear arms race in the late Fifties, Schelling's book, The Strategy of Conflict, had first set forth his vision of game theory as a unifying framework for analysis. Schelling showed a party in confrontation can sometimes actually strengthen its relative position by taking a weaker posture. Instead of showing off its defensive capability it can more usefully announce what might happen if it is defeated, and the bulk of its population destroyed by a nuclear first strike: for example, there may be a credible possibility of an unstoppable retaliation mode activated, leading to a nuclear, chemical or biological strike back that nobody may be in a position to contain. This is, what I may call, the 'Strategy of the Wounded Tiger'.

Schelling showed that the opponent's capability of automatic retaliation can be more frightening than the certainty of its resistance, and that the threat of uncertainty of events and consequences unleashed in the retaliation mode is more efficient. These insights may have been of great relevance for the efforts to avoid nuclear war, and the efforts towards conflict resolution in recent years in our own subcontinent. But a simple question has remained unanswered: suppose the aggressor begins the game in the role of the wounded tiger in the first place. Can you deter him by assuring the certainty of your retaliation, or frighten him with the dark uncertainties about the extent of unstated but dire consequences' Israel has tried it in Palestine and failed. Other countries, in a manner, are trying it now in a much wider context to eliminate international terrorism. What chances do they have of success' My question, in short, then is: can a rational behavioural gambit, that deters rationally self-serving homicides, be expected to deter a bunch of irrational suicide bombers who just decided they have found an instant passage to paradise for themselves'

Take the next one in the 'greenhouse effect' scenario ' a complicated process by which the earth is becoming progressively warmer. Let me quote Schelling again: 'This greenhouse phenomenon is truly the result of a 'global common'. Because no one owns the atmosphere, no one has a sufficient incentive to take account of the change to the atmosphere caused by his or her emission of carbon. Also, carbon emitted has the same effect no matter where on earth it happens. The expected change in global average temperature for a doubling of CO2 is 1.5 to 4.5 degrees Centigrade.'

Schelling, to put it in a nutshell, makes two points. First, climates will change more in the next 100 years than in the last 10,000, but it is still slow change. People have been migrating great distances for thousands of years, experiencing greater changes in the temperatures of their habitats. Secondly, other changes will also be gradual. For example, the melting of the polar ice cap will affect mostly the low-lying areas of the continents and only the third world economies might be seriously damaged, not being able to migrate inland. Therefore, there will be no economic reason for actual and serious concern in the developed world, notwithstanding the statements that may ensue showing such concern. As for the developing countries, they better not get into this at all. There is no point in sacrificing the clear gains from rapid development or frittering some of it away paying for environmental protection now. Think about it again when you have become a developed country yourselves. This is a startling position to take, but an extremely well-argued one and may even be liked by the Planning Commission.

But again a question remains. Has the developed world got its sums right' No matter what the records said about the last few thousand years of relatively gradual changes and about the earth deteriorating as human habitat only part by part, the devil taking the hindmost, what about the signs of the times ' our times' The South Pacific tsunami, the south Asian earthquake, the variously named hurricanes devastating the American coastline, the looming fear of the bird flu pandemic coinciding with a cyclical peaking of human influenza ' all this and more can happen before the first decade of the 21st century is through ' none needing to apply for visas to cross frontiers.

Is it not time to go back and tend the 'global common' regardless of where everybody came from'

Email This Page