Thomas Schelling, new Nobel Laureate

Note my biases, Schelling was my mentor at Harvard.

Tom is an unassuming guy, who looks as if he sells Hush Puppies at the local mall.  But he is one of the sharpest people you will meet.  He delivers the killer point, argument, or anecdote with striking regularity.  Even in his eighties he is sharp as a tack.  He has a deeply philosophical and humanistic approach to economics.  What are his contributions?

1. The idea of precommitment.  You can be better off, either individually, or institutionally, if your choices are limited in advance.  This is a key idea in monetary policy (many governments seek to tie the hands of their central banks), the theory of bargaining (try buying a used car, and see if the salesman doesn’t talk about "the boss upstairs"), and industrial organization (firms may invest in capacity to precommit a market position and deter rivals).  You find the precommitment frequently in movies as well, especially where kidnapping is involved; what is that Mel Gibson flick again?  Here is an excellent Jon Elster piece on the ambiguities of precommitment.  Here is my piece on similar themes.

2. The paradox of nuclear deterrence.  Ever see Dr. Strangelove?  Tom developed the idea that deterrence is never fully credible (why retaliate once you are wiped out?).  The best deterrent might involve precommitment, some element of randomness, or a partly crazy leader.  I recall Tom telling me he was briefly an advisor to Kubrick.  Here is someone else’s essay on the paradox of deterrence.

3. Focal points.  People coordinate by directing their attention to commonly recognized points of importance.  If a meeting time for lunch is not specified, you might assume 12 noon.  If someone mentions "economics blog," of course comes to mind.  And so on.  Much social coordination occurs in this manner.  I once asked me class: "If you had to hide a one hundred dollar bill in a book, so that your friend would find it, but you could not announce the book, which volume should you choose?"  Many said The Bible but of course the game theorist picks Schelling’s The Strategy of Conflict.

4. Behavioral economics and the theory of self-constraint.  One of Tom’s best pieces is "The Mind as a Consuming Organ," American Economics Review, 1984. Here is a lecture of his on self control.  Will Wilkinson cites a bit of that essay.  Tom made it respectable for economists to talk once again about happiness.

Tom has been an underrecognized father of behavioral economics.  His work on addiction, memory, and personal control was pathbreaking and came nearly twenty years before the "behavioral revolution" in economics.  He analyses the tricks people use to control their wills.  For Tom, self-control is often a more important determinant of happiness than is wealth.  Tom once told me his work sprung from his own attempts to quit smoking, which he did finally manage.  Several times.

5. The economics of segregation. Tom showed how communities can end up segregated even when no single individual cares to live in a segregated neighborhood.  Under the right conditions, it only need be the case that the person does not want to live as a minority in the neighborhood, and will move to a neighborhood where the family can be in the majority.  Try playing this game with white and black chess pieces, I bet you will get to segregation pretty quickly.  Here is a demo model for playing the game.

6. Later in his life Tom turned his attention to issues of global warming.  He has been skeptical of the idea that global warming involves insuperably high economic costs.  Here is a short essay by Tom on the topic.  Here is his excellent AER piece on the same topic.

Tom is not well-represented on the web, here is one photo, but the associated links are mostly broken.  Here is Tom’s piece on Hiroshima.  Here is the Wikipedia bio, note Wikipedia already reports he won the prize.  Here is a great interview with Tom.  Here is Tom’s work on the Copenhagen Consensus.

A few bio facts: Like so many other prestigious American economists, Tom worked for the Marshall Plan in its early stages.  He spent most of his career at Harvard, first in the economics department, later in the Kennedy School.  He had close ties with the Rand Corporation.  He was an advisor to Kissinger during the Vietnam War but quit in disgust.  He is now emeritus at University of Maryland.  I have always interpreted Tom’s political views as those of a conservative Democrat.

Here is a piece I wrote with Dan Klein and Timur Kuran, Salute to Schelling: Keeping it Human.  In this piece, recently published, we asked the Committee to give the Prize to Tom.  Here are Tom’s books, they are all worth reading.

Comments are open, you can add more; Tom could have won more than one Nobel Prize for all his contributions.


Strategy of Conflict is one of the most important books ever written in international security. Schelling's ideas about pre-commitment and focal points are pervasive across the social sciences. This is an excellent choice and long overdue.

Where do you hide the C-note?
"Capital," of course. Or perhaps "The Economic Point of View."

For a long time I was hugely underwhelmed by the idea of the focal point. But one of the best pieces on it that I have read changed my mind. It is an essay by Roger Myerson at

Schelling seems to have done lots of interesting stuff, and is no doubt deserving of the prize, but the essay on global warming contains a stunningly stupid statement:


In the developed world hardly any component of the national income is affected by climate. Agriculture is practically the only sector of the economy affected by climate, and it contributes only a small percentage -- three percent in the United States -- of national income. If agricultural productivity were drastically reduced by climate change, the cost of living would rise by one or two percent, and at a time when per capita income will likely have doubled.


Of course, if agricultural productivity were drastically reduced by climate change, most of us would die of starvation!

"Of course, if agricultural productivity were drastically reduced by climate change, most of us would die of starvation!"

Not really. Productivity is a measure of output per unit of input. For example, labor productivity measures production per unit of labor (sometimes per worker, sometimes per hour worked). It's entirely possible for productivity to decline and still get the same output, in this case food, it simply means that you have to devote more resources to that sector. As I understand it, Schelling is arguing that we will in fact have so many more resources that this increase will seem small small.

"Of course, if agricultural productivity were drastically reduced by climate change, most of us would die of starvation!"

Steve Vickers responded:

"Not really. Productivity is a measure of output per unit of input. For example, labor productivity measures production per unit of labor (sometimes per worker, sometimes per hour worked). It's entirely possible for productivity to decline and still get the same output, in this case food, it simply means that you have to devote more resources to that sector. As I understand it, Schelling is arguing that we will in fact have so many more resources that this increase will seem small small."

It occurred to me that Schelling might have meant "productivity" in this sense, but if so, how did he conclude that "the cost of living would rise by one or two percent"? Any such calculation would require a detailed analysis of agricultural production, to see how many other resources would need to be deployed to compensate for poorer climate. Since agriculture is probably mostly limited by the amount of arable land, which is not expandable (on the assumption, of course, that climate change turns out to be bad, not good), it's hard to see how such an analysis, if done, would lead to such an optimistic outcome.

Difficult as it may be to believe, it seems that Schelling simply thought that eliminating food production would reduce economic activity by only a few percent of GDP, and so is no big deal. It's stunningly stupid. But it seems it's possible if your economic thinking becomes sufficiently detached from reality.

How exactly did Buchanan's and Smith's Nobels trivialize the prize? Public choice and experimental econ are very important, even if the popularity of the former is decreasing.

Even in the early 90s, the assumption that climate change would affect only that proportion of developed-country economic activity that explicitly took place outdoors was a pretty-well-debunked crock. It doesn't speak well for Schelling -- or the economics profession -- that he should have adopted it. He's obviously quite a smart guy, but (in part as a result of that intelligence coupled with superficially good writing skills) not one I'd trust anywhere near an actual policy decision.

Which is of course true of almost all of the best economists and their can-openers.

(By the early 1990s, in addition to the severe-weather predictions, climatologists had already produced detailed maps of coastline changes for various coastal centers, estimates of increases in required local generating capacity for HVAC, and so forth. Of course, for someone who had lived through a worldwide depression and a world war, any dislocations smaller than a complete fall of civilization might have seemed minor...)

Barkley Rosser writes:

"Well, actually Schelling is pretty much right about all this... There was a big debate between Nordhaus and Mendelsohn over ten years ago about the impact on US agriculture. Nordhaus got a several percent loss from warming by assuming that farmers kept on planting the same crops. Mendelsohn found very little loss because farmers would change their crops to adjust..."

This isn't the issue. Fears of global warming may be exaggerated, and if it does occur, it might well turn out to actually be beneficial to agriculture, or at least not very harmful. However, Schelling's statement ASSUMES that global warming occurs, and that it IS very harmful to agriculture. He then goes on to say that that wouldn't matter much, since agriculture accounts for only three percent of the economy.

The most sense I can make of his statement is that he's thinking that production with the current level of labour and other inputs might decline by, say, a factor of two, so all you need is to increase labour and other inputs by a factor of two to get back to the old level of production. This is naive in the extreme. Even less sensibly, perhaps he just hasn't remembered that unlike DVDs, toilet paper, and flashlights, food is actually essential to life.

Now, one could actually argue that North America could cope with a factor of two reduction in food production, by reducing meat consumption, which is a very inefficient way of feeding humans. That makes strong assumptions about North America managing to isolate itself from the rest of the world, however. In any case, it's not an argument that Schelling makes.

As a non-economist, I have little to add to certain discussions above. What I can say is that as a Harvard undergraduate I had Schelling as a professor for what I had (typically, pathetically) enrolled myself in as a gut course: his introduction to game theory.

It turned out to be the most mind-blowing course I took there...and it was a gut! The stuff wasn't hard to understand, but it completely changed my view of the world.

This is simply great news to me. Thos. Schelling was a great teacher and an obviously good man, and the scope of his imagination and the rigor with which he applied it were and continue to be an inspiration to me.

Barkley Rosser writes:

"Rather than making baseless statements about what Schelling might have said or might think I suggest you go read his address to the American Economic Association in the American Economic Review on the matter. It is publicly available. Naivete, sloppy statments, and inaccurate calculations are the last things in the world one will find associated with Thomas Schelling."

The article you mention does not contain any further technical arguments. In it, Schelling says:

"... even if agricultural productivity [in developed nations] declined by a third over the next half century, the per capita GNP we might have achieved by 2050 we would achive only in 2051."

He goes on to say that productivity might instead continue to increase, climate change might turn out to be beneficial, etc. But that's a different argument. The statement above differs from the one in the essay linked above only in that "drastically reduced" is specified to be by one third, and the time scale is made clear.

It's still a stupid statement. There is no reason to think that the effects of climate change, if they are indeed so bad, are going to be fixable by a proportionate increase in labour or capital inputs, since the constraint may well be the amount of arable land, which you can't just decide to increase. His argument comes down to "agriculture is a trivial part of current GNP, therefore the effect of climate change on agriculture can't be very important". The flaw in this argument should be obvious, but apparently it isn't obvious to everyone.

On "focal points", I thought unemployment, inflation, GDP, GDP/capita, and GDP growth were the main focal points on which to judge economic performance. Though interest rates and private home ownership rates seem pretty important, too.

I note this because, under Bush, such rates for the USA seem pretty good -- but the Leftist press seems want to change the points (goalposts), so as to make Bush look bad.

If coordination is going to be effective, there needs to be agreement on how to measure results. I also like the precommitment ideas; I'm wondering if there isn't more to Becker's marriage contract with additional precommitment and postcommitment advantages.

I had the privilege of taking a course and then have and independent resarch project with Thomas Schelling. Since I went to the first of his classes I knew he was going to get the Nobel prize sooner or later (and as I told everybody I met at Harvard and MIT, I have now dozens of witnesses). It took 23 years to fulfill my prophecy and this is just one of the best news I haver ever gotten. I admire Tom SDchelling and I can only say that this partly compensates for the frustration I have because Joan Robinson never got it.
Hernan Garrido-Lecca, Peru (Harvard MPA 84 and MIT SM 90).

Comments for this post are closed