TOPICS > Economy

Game Theorists Awarded Nobel Prize in Economics

October 15, 2007 at 6:30 PM EDT
LISTEN SEE PODCASTS

TRANSCRIPT

RAY SUAREZ: This year’s winners were awarded the prize for their research in showing how big marketplaces work. The theory they won for is known as mechanism design.

And we turn to one of the winners to help us understand it. He’s Roger Myerson, a professor of economics at the University of Chicago.

Congratulations, Professor.

ROGER MYERSON, Nobel Prize, Economics: Thank you.

RAY SUAREZ: Well, let’s start as basically as we can. What is mechanism design theory?

ROGER MYERSON: You know, the term “mechanism,” mechanism is just something that does things. But what we’re doing, it’s come to be used in economics as a term to mean, when we think of the market or the economy as a communication medium, for communicating information from people about what they can produce to or what they need from other people in order to allocate resources.

So when we think about the economy as a way of using people’s information and communicating it from people who have information to other people who have to make decisions, we’ve come to call that a mechanism. Actually, I think it was Leo Hurwicz’s work that put that use of the word.

But it’s about communication; it’s about communication and incentives to share information that other people need for decisions.

Benefits of mediation

Roger Myerson
Nobel Prize winner in economics
The problem of sharing information in economic transactions is absolutely central to the problem of coordinating in society and in the economy in particular.

RAY SUAREZ: Now, economists normally extol when marketplace actors come together and are upfront about their information, but I understand your work also helps us understand times when people don't want to share all the information, if you're trying to sell a less-than-wonderful used car or don't want an insurance company to know that you've got a disease.

ROGER MYERSON: The problem of sharing information in economic transactions is absolutely central to the problem of coordinating in society and in the economy in particular. The breakthroughs in information economics in the late 1970s and the early 1980s tended to focus on the problem of carrying information between people who are separated.

And I'd like to think that one of the ways of getting a handle on mechanism design and the important concepts of it is to think of the problem of a mediator who's trying to help a buyer and a seller to agree on terms. And in that medication process, sometimes separating the people so that one can make some concessions without having the other person find out immediately about it, that's a benefit.

So we do tend to need to sometimes separate each other. International medication is often very -- it's often important to have shuttle diplomacy to achieve agreements. But the theory of how to optimally mediate in order to maximize the probability of a successful agreement, that's really one of the central questions of mechanism design.


Studying economy trends

Roger Myerson
Nobel Prize winner in economics
So we were trying to redefine, what is a well-functioning marketplace? That was the centerpiece of what we were working on, I think.

RAY SUAREZ: Now, is this research meant to come up with something that's predictive, so we understand how large numbers of people will react to a given situation in a marketplace, or more as a device to help us understand what they do when they're in this situation?

ROGER MYERSON: I think it's more in the latter. We often say that one of the central pieces is known as the revelation principle. Eric Maskin and I worked on it separately in papers in the late '70s, which in some sense argues that a mediator who's trying to help them, without loss of generality, could make honesty the best policy, should consider talking to people in such a way that, in these confidential sessions, they have no incentive to lie to him.

This is not so much because -- well, I think people being honest is a good idea -- but it's not so much trying to predict or normatively say within that theory that honesty is always the best policy, but as much as to get a handle on, when is our communication process dysfunctional? When is the communication process not as effective as it could be?

And so I think that's really the centerpiece of what we were working on, was trying to understand, how good can you make a communication process in getting people to reach agreements on mutually beneficial deals?

And it's subtle, because sometimes, in order to get the buyer to admit that she's willing to pay a little more or to get the seller to admit that he's willing to go for a lower price, sometimes a very tangible risk of disagreement is necessary, even among people who we know in some sense could both benefit from trading with each other.

That risk of disagreement sounds like a failure of the marketplace, but when we realize that incentives are a part of the economic problem, we realize the right question is not avoiding inefficient disagreements but minimizing their probability. So we were trying to redefine, what is a well-functioning marketplace? That was the centerpiece of what we were working on, I think.

Mechanism design and public policy

Roger Myerson
Nobel Prize winner in economics
I think mechanism design, per se, is about getting information from where people have needs -- and, yes, I've worked on applications in voting.

RAY SUAREZ: And very briefly, Professor, does this also have applications when it comes to designing public policy, voting or a social welfare program?

ROGER MYERSON: I think mechanism design, per se, is about getting information from where people have needs -- and, yes, I've worked on applications in voting. I think some of the applications in voting have a different dynamic.

I've spent a lot of time in the 1980s and early 1990s working on the advantages of multi-voting, which I think, for example, in presidential primaries, approval multi-voting, if people would consider it, probably they'd figure out that it might make a more effective democratic system.

RAY SUAREZ: Approval multi-voting, meaning that you rank the candidates in order of preference or...

ROGER MYERSON: No, you can indicate approval. If there are 10 candidates on the ballot running, say, in a Democratic primary, you can vote for several of them, but you can't give more than one vote to any one of them. And the person who gets the most votes, most approval votes, was approved by the largest majority, gets the bonus of winning the votes of the state.

RAY SUAREZ: Well, Professor, you'll be splitting your 10 million krona, $1.5 million prize, with Professor Hurwicz at the University of Minnesota and Professor Maskin at the University of Princeton.

Thank you. Congratulations again.

ROGER MYERSON: Thank you so much.

RAY SUAREZ: Good to talk to you.

ROGER MYERSON: It's been a pleasure to talk to you.