Editor's Note: When George Akerlof and I worked on the "Yale Daily News" more than 40 years ago, he was widely respected by his would-be journalist colleagues. He had the curiosity and discipline of a reporter and almost always seemed to enjoy what he was doing. George observed, "I wanted to do two things with the newspaper. First, I found it too much of an official organ. A typical prime assignment was for a leading reporter to interview the President (of Yale University) and to report his views. I wanted the newspaper to do something different. I wanted it to have more stories about student issues, and also more features of human interest. I wanted it to be less solemn and more serious."

Now, despite winning the immensely prestigious 2001 Nobel Prize in Economics for his landmark study on the role of asymmetric information in the market for "lemon" used cars, George works with the same determination and humility. As he put it in explaining to me why he wanted to write for our web site about his work, rather than his fame, "I intentionally left the Nobel-ly stuff out. We might as well let it speak for how I am trying to live my life as the dog that did not bark in the night."

Akerlof is married to prominent economist Janet Yellen. Both teach at the University of California in Berkeley. Yellen served as chairperson of the Council of Economic Advisors in the second Clinton Administration and was recently named President of the San Francisco Federal Reserve Bank. Over the years, they have worked together on a variety of research projects. "He has real flashes of insight into human problems, into what may explain social phenomena," Yellen commented. "It's wonderful to work with him; he's so original." Who should know better?

Contrasting economic theorists with French chefs a few years ago, George noted that both "have developed stylized models whose ingredients are limited by some unwritten rules." George added, "I disagree with any rules that limit the nature of the ingredients in economic models." (Al Chambers)

"Changing Metaphors"

By George Akerlof
Berkeley, CA
June 16, 2004

I thought that it might be fun to say a few words for this column about what I do, as an economic theorist, for a living. I am sure that most of the public thinks that economists, especially macroeconomists, spend vast hours thinking about interest rates and the state of the economy, so that, for example, if you asked me the rate on AAA 30-year bonds in Detroit on May 17, I could tell you that it was 5.39 % and I could even give you a prediction what that rate will be on August 17, although I might need my computer to do that. To be honest, I am not sure that there is such a thing as AAA 30-year bonds, especially in Detroit, and, if there were, that the yield would be anything like 5.39 %, but I did give you the type of question that my dentist or doctor asks me when he wants to be friendly.

If economists (or economic theorists) do not sit around and think about such weighty questions as the level of interest rates and how they vary, what do they do? After all, it is now more than 40 years since I sat in that class room in SSS on the corner of College and Grove and learned that demand curves slope downwards, and supply curves slope upwards, and equilibrium occurs where the two curves cross, which is where demand equals supply.

Typically economic theorists would describe such a demand and supply system in terms of mathematics: for example, the supply curve is a function relating the amount that suppliers want to sell at any given price; the demand curve is a similar function relating how much demanders wish to buy. And the equilibrium, which is determined by the point where the two curves cross, is an equation. So what went onto the board at that early hour of freshman year as lines of chalk, can be thought of in terms of mathematics.

But the linguist George Lakoff has described mathematics as metaphor, and he views most theory in science, and, also in the social sciences as well, as, likewise, forms of metaphor. The job of an economic theorist is then to develop new mathematics, which will also entail developing new metaphors for how people behave in economic transactions. So that is how I have spent the 40-plus years since leaving Yale.

What does it mean to try to develop new metaphor to describe economic life? In that same course where I learned about supply and demand, I also learned about the wonders of free, competitive economic markets. Are there alternative metaphors for economic markets? Have you ever tried to buy, or to sell, a used car? If you have, you know that it is a messy business. As the buyer of a used car, you are suspicious of the motives of the seller, or, alternatively, if you are the seller, the buyer will be suspicious of you. So an alternative metaphor for markets is not that it is some pure situation where the buyers and the sellers have their respective demand and supply, and trade at something called the market price, but instead itís like the used car market where before a deal can be made, the buyers and the sellers must have some modus operandi for allaying their suspicions of each other. It is surprising how much of economics is changed, including our view of capitalism, and the need for regulation (for example, the beneficial role of the SEC) once one views economics through the metaphor of used cars.

I will give you another example of the type of work that I have done in developing metaphor. The standard view of the labor market, the one that I also learned in that first course in economics, is that the labor market can be viewed just like any other. Firms (employers) want to hire labor, and they will hire more labor, if the price of labor is lower. Workers want to work and they will work harder, and more of them will be in the labor market if their pay goes up. The operation of the labor market then works, again, according to the metaphor of supply and demand.

Young Dr. AkerlofBut an alternative metaphor takes into account the fact that people are human. They are ornery and just because they are paid to do something, they do not necessarily do it. In equestrian terms: "you can lead a horse to water but you canít make him drink." That is an alternative metaphor for the labor market. Again it is surprising how much of economics is changed by using the "horse-to-water" metaphor. Remarkably, it allows one to answer puzzles that had eluded economists for decades. In the old metaphor of supply and demand, it is impossible to explain involuntary unemployment. Those who are unemployed are those who are not willing to accept the market price for their labor. But the new metaphor suggests that employers may pay extra high wages not just to attract workers to their jobs, but also to get them to work (to make the horses drink) once they are at the work place. In that case wages will be higher than market-clearing (where supply equals demand). Some people willing to work at those wages will be unable to obtain jobs, so there will be involuntary unemployment.

For the past ten years or so, I have been working with Rachel Kranton (of the University of Maryland) on a new metaphor for why and how people behave in the work place. They care about who they are, and they also care about their interactions with others. Psychologists, sociologists, literary critics, and historians would say that they care about their identity. We believe that motivation of this sort, where people do what they do, largely because they care who they are — rather than just because they care about financial rewards — is essential to explain a whole range of economic phenomena: traditional gender and racial discrimination, motivation of minorities in the work place, the costs and benefits of schooling, and the role of leadership in organizations.

What do I do? It is not so very easy to get people — in this case professional economists, who are very proud and opinionated — to entertain the notion that their use of metaphor is invalid, or even incomplete. Indeed, I have always had a very hard time getting my papers published; it remains a struggle today. My goal in these papers is to write a few lines of math that succinctly characterize how the new metaphor differs from the old. I then try to supplement this characterization with special examples that are especially well described by the new metaphor, and also especially poorly described by the old. In perhaps the biggest compliment of my life, a friend of mine, Avinash Dixit, once introduced me with the memory from graduate school that I posed questions that "only a damned fool would ask." But then, Avinash continued, he often changed his mind. I do not want to confess how frequently that I really was a "damned fool." But I am continuing to look for questions like that, and to take the further step of trying to get others to change their minds. It is a game of competing metaphors.