American and Israeli Share Nobel Prize in Economics

NY Times ^ | October 11, 2005 | LOUIS UCHITELLE

[Sabramerican]

American and Israeli Share Nobel Prize in Economics

By LOUIS UCHITELLE Published: October 11, 2005

Robert J. Aumann and Thomas C. Schelling won the Nobel Memorial Prize in Economic Science yesterday for their work in game theory, which explains the choices that competitors make in situations that require strategic thinking. Two Win Nobel Prize for Discovering Bacterium Tied to Stomach Ailments (October 4, 2005)

Their work has helped to illuminate the dynamics in labor negotiations, business transactions and arms negotiations, among other situations. An article that Mr. Schelling wrote prompted the director Stanley Kubrick to make the movie "Dr. Strangelove," consulting with Mr. Schelling during the filming.

The Royal Swedish Academy of Science announced in Stockholm that Mr. Aumann, 75, an Israeli who teaches at the Hebrew University of Jerusalem, and Mr. Schelling, 84, an American who is a professor emeritus at the University of Maryland, had been honored "for having enhanced our understanding of conflict and cooperation through game-theory analysis."

The two men will share a $1.3 million prize.

....The Nobel judges said that the work of Mr. Schelling and Mr. Aumann "was essential in developing noncooperative game theory further and bringing it to bear on major questions in the social sciences."

.....Game theorists assume that in a given situation people are affected by what other people do or what they imagine others will do, particularly when their goals are conflicting.

From a game theorist's point of view, for example, the Oslo accords negotiated between the Palestinians and the Israelis were a success in part because the negotiations proceeded in small steps. Each side made small concessions and the other reciprocated, but neither would make a big concession. In the case of a big concession, there could be considerable damage if the other side did not reciprocate.

[Sabramerican]

From a game theorist's point of view, for example, the Oslo accords negotiated between the Palestinians and the Israelis were a success

Is there some prize I would win for recognizing that anyone who considers OSLO a successes in any way, shape or form is a total moron?

[Moonman62]

You'll get your prize the same day that someone gets a Nobel in Economics for figuring out that economic growth doesn't cause inflation.

[Paleo Conservative]

From a game theorist's point of view, for example, the Oslo accords negotiated between the Palestinians and the Israelis were a success in part because the negotiations proceeded in small steps.

By that definition of success, boiling a frog by putting it in a pan of cold water on a stove and gradually turing the heat up should also be considered a success.

[Enchante]

Incredible that they could cite the catastrophic failure of the "Oslo process" as an example of success. That should be a counter-example to the value of small incremental steps: that two or more parties to a negotiation can reach a pretend agreement while nothing of importance is actually settled. Oslo is a perfect example of media image over substance, of negotiating FRAUD, not an example of success.

Schelling is a brilliant man and may not be responsible for this foolish example (more likely the NY Slimes reporter thought of it), but what a #&$^$%## stupid example to give of successful negotiation for conflict resolution.

[Poopyhead]

Whether or not Oslo was a success has no bearing on their achievement in Game Theory. Game Theory is based on the notion that people/companies/organizations are rational and will eventually always follow the incentives. ... As opposed to following the wishful thinking of politicians, such as with welfare, tax policy, etc.

[Paleo Conservative]

Whether or not Oslo was a success has no bearing on their achievement in Game Theory.

I understand. I was just making fun of the lame example the journalist used.

[Sabramerican]

It was an example of the NY Times finding a way to interject their bias into the story- as usual.

What does think Aumann think?


Nobel winner: Conflict won't stop
AVI KRAWITZ, THE JERUSALEM POST Oct. 10, 2005

Hebrew University Professor of Mathematics Robert J. Aumann, who on Monday won the Nobel prize in economics for his work on using game theory to understand conflict resolution, says he sees no end to the Middle East conflict that claimed the life of his eldest son 23 years ago.

“This conflict has been going on for 80 years and to my sorrow I believe it will last for at least another 80,” Aumann said in a press conference at the Hebrew University of Jerusalem’s Givat Ram campus, where he resides as a professor emeritus in the Institute of Mathematics and as a member of Center for Rationality.

http://www.jpost.com/servlet/Satellite?cid=1128864032218&pagename=JPost%2FJPArticle%2FShowFull

[ExitPurgamentum]

From a game theorist's point of view, for example, the Oslo accords negotiated between the Palestinians and the Israelis were a success in part because the negotiations proceeded in small steps.

It's not a definition: it is a misunderstanding on the part of the author. There is no theorem in game theory that says people coordinate better, or do so in the first place, when they change their actions in small steps.

[Poopyhead]

Game theorists just describe reality with models. There is no theorem at all in Game Theory that states one approach is better than another. There are all types of games that people play, the Oslo "game" is probably the one known as "Tit for Tat". There are single step games, like the entrant/monopolist games, and then there are multi step games like "Tit for Tat".

Rupert Murdock used a multi-step (unknowingly) game to enforce tacit collusion. He raised the price of one of his newspapers (I think it was a NY paper) by X%. His only competition did not follow, so Rupert repriced by cutting the original price in half, forcing his competition to match the cut. He kept the price down for a while, then raised it back up to the X% over the original price. This time the competition followed. Perfectly legal.

[ExitPurgamentum]

There is no theorem at all in Game Theory that states one approach is better than another.

Isn't that what I said?

Rupert Murdock used a multi-step (unknowingly) game to enforce tacit collusion. He raised the price of one of his newspapers (I think it was a NY paper) by X%. His only competition did not follow, so Rupert repriced by cutting the original price in half, forcing his competition to match the cut. He kept the price down for a while, then raised it back up to the X% over the original price. This time the competition followed. Perfectly legal.

This is not tacit collusion: this is what is know as coordination, and that is why it is perfectly legal. the strategies of this multi-stage game simply converged to a Nash equilibrium.

[Poopyhead]

Not really. It's pretty much the definition of tacit collusion.

A good explanation: http://siepr.stanford.edu/papers/pdf/02-26.pdf

But I'm ok with whatever.

[ExitPurgamentum]

Re the terminology of coordination vs. collustion, from p. 4 of the paper you cited:

“Tacit collusion” need not involve any “collusion” in the legal sense, and in particular need involve no communication between the parties. It is referred to as tacit collusion only because the outcome (in terms of prices set or quantities produced, for example) may well resemble that of explicit collusion or even of an official cartel. A better term from a legal perspective might be “tacit coordination”. In the rest of this paper we shall continue to refer to tacit collusion as this better reflects the terminology in the economic literature, but at no point does our analysis presuppose that the collusion is explicit.

More importantly, your remark regarding Murdoch's actions did not demonstrate that the outcome was that of a cartel. If the stage game of a repeated game has a single Nash eqm, the convergence to that eqm would occur just as you described. There is no collusion when players converge to a single eqm, is there? To show that there is tacit collusion must therefore show that players coordinate on cartel-like eqm (if there is more than one). Am I missing something?

[ExitPurgamentum]

Here is another reason why both the terms and their use the terms must be spelled out in this area:

Despite an extensive literature on “tacit collusion” in economics, there is considerable ambiguity about the use of the term. Often, as e.g. in Tirole [1988], it is identified with collusion “in a purely noncooperative manner,” thus allowing for explicit coordination on how to interact in the game and making a large variety of behaviors supported by repeated interaction eligible. Others, such as Carlton and Perloff [1999], emphasize the absence of explicit communication as defining tacit collusion. The latter definition is in line with legal practice in the US, where any explicit communication between competitors coordinating behavior in the market place is considered illegal.^1

----- 1. See Ayres [1987] who writes that “[l]egal scholars have traditionally distinguished between explicit and tacit collusion. The law punishes the former, so that the act of communication is of central importance. For economists, however, this distinction has no meaning.”

The above quote is the very beginning of the body of Modeling Tacit Collusion in Auctions by Blume and Heidhues. You may view these names as less prestigious than the Toulouse group, but the observation these authors make appears to be correct.