An interview with George Akerlof

Earlier this week I spoke with George Akerlof, a professor of economics at Berkeley and a winner of the 2001 Nobel Prize in economics, about his new book, Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism. The book, which Akerlof wrote with Robert Shiller of Yale, concerns departures from the full-employment economy: How do we explain the fluctuations of the business cycle, or the existence of involuntary unemployment?

According to Akerlof and Shiller, the traditional economic answers to these questions are tortured and unsatisfying. In search of a better answer, they turn to John Maynard Keynes's notion of the animal spirits: "the restless and inconsistent element in the economy" that is not easily explained by reference to rational actors with simple economic motivations.

I spoke with Akerlof about the book, Keynes, the place of psychology in economics, and the implications of all this for the current crisis. I recorded a similar conversation with Professor Shiller, which I will post tomorrow.

Conor Clarke: The book's introduction says you started this project in the spring of 2003, and now here we are, almost in the spring of 2009. Were you waiting for a major financial crisis before releasing the book?

George Akerlof: [laughs] No, we've always been worried about the type of crisis that was going to occur. And we wanted to get it out before the crisis. And so actually this is late. We wanted to tell people what could potentially be done before the crisis.

Your book is one of a number that have come out recently -- I'm thinking of Richard Thaler and Cass Sunstein's Nudge, and Daniel Ariely's Predictably Irrational -- that attempt to inject psychology into economics. Why all this interest?

I think it is the case that there's been a significant movement to bring psychology into economics over a long period of time. I think what Bob Shiller and I are doing is we're focusing on macroeconomics and the role of psychology in macroeconomics. And I guess both of us have also been working on this for a long period of time and we thought we should bring it together. We initially thought we would do something small -- we thought we would bring out a book of readings. And so we proposed that and we wrote a fairly long introduction.

Then we thought we should do something more significant -- that instead of a book of readings we should write something that somehow gave our opinions on the role of psychology in economics. So that's what we've come out with, hopefully.

It does seem like this book is easily distinguishable from a Nudge because it concerns that macroeconomy. And yet when I think of psychology it's hard not to think about microeconomic concerns -- what shapes the decisions of individuals and individual firms and so forth. Is it easy to move from that stuff to a consideration of the economy as a whole?

Well I don't necessarily think there's a conflict there, at least given my own personal history. I've always been a macroeconomist. That's what I teach. And I guess that's what I've been concerned with ever since I've been very young. I've always wanted to know what caused unemployment. So I think it's natural to combine psychology and macroeconomics. Actually, if you don't take psychology into account, I think it's fairly hard to give a model of the economy that explains a great deal of the economic fluctuations that are going on.

It turns out that the easiest and simplest theory of those fluctuations is the one we give in the book -- that there are these changes in animal spirits or in confidence.

If I were to go back to my macroeconomics textbook the explanation for fluctuations in the business cycle is short-run price stickiness. Is that story wrong?

I think sticky prices might be part of the explanation, but they are not the whole explanation. The real question is why, with the sticky prices that we have, do we have the degree of economic fluctuation that we have? And I think a major part of the explanation is that we have these cycles in confidence, and in the stories we tell about the economy.

So sometimes people are just more confident and more willing to invest then at other times. And sometimes they're more willing to trust other people, and there are stories being circulated about why they should do so and why the economy is doing well. And then people go out -- and it turns out what they do is they tend to binge. They tend to be -- as Bob Shiller would say -- over-exuberant. And this over-exuberance translates into bad investments. Lots of bad things happen. But they're only uncovered when somehow the bubble ends and the commonly accepted stories about the economy change. And then people understand that in fact we were over-exuberant and overenthusiastic, and then the economy falls, and we go into a new phase of the business cycle.

And is the point of this to say, "Macroeconomics has tried to create clean models out of things -- like human psychology -- that are not amenable to clean modeling"? Or is it something like, "There are these consistent and predictable non-economic features of human nature that the clean models have avoided"?

It seems to me that the standard macroeconomics -- where people only have economic motivations -- has a fairly hard time explaining why there should be such a thing as involuntary unemployment. And in fact there are a large number of economists at the moment who believe that involuntary unemployment is pretty minor. So in order to explain the fluctuations in the business cycle we have to go to motivations that they don't use in their models. You have to go beyond economic motivations.

So that's one reason for doing this -- with the standard models people are using, it's hard to explain the significant fluctuations that we're looking at.

And the second reason we're doing this project is this: my view of economics is that one of the most fundamental things it should do, when it's setting up the basics, is that you want to use realistic human motivations.

Economics has got to be an accurate description of human nature.

Yes. Economics needs to be an accurate description of human nature -- an accurate description of how people actually behave. That is one of my fundamental beliefs. You don't start with the idea that people have only economic motivations. You start with what are people actually are and how are they are actually behaving. And you take off from there.

So how discouraged should we be about the models that economics has developed thus far? I take it you think the rational expectations model is not accurate.

I don't see any reason why you can't add on the features of psychology to economics. And it seems to me that that's what Keynes originally had in mind. And I don't think that's inherently more difficult than what we're doing already.

Well on that note -- I saw that my colleague Clive Crook reviewed your book in the Financial Times. Did you see that review?


It seemed like one of the points he wanted to make was that even though you have an apt criticism of the standard model, he would be reluctant to abandon them because the oversimplified models have still done a lot of good. Do you sympathize at all with that?

Well, it depends on the oversimplified model. It seems to me that you'd do better with a model that actually explains how humans behave.

When you read the book, one of the things you see is background -- that we're using as background the standard Keynesian model. And we're basically adding these psychological features. And it seems to me that it's pretty easy to take the standard Keynesian economic model and add these psychological features. It's not very difficult. And you get the answers to the questions that we ask in the book -- questions about why there are things like involuntary unemployment and fluctuations in income.

So apply that to a question of public policy, like fiscal stimulus. In the book you say that in addition to being large enough to make a dent in the output gap, any fiscal stimulus needs to be large enough to affect the animal spirits. How does that work?

Well, I think what the book says is that there's certainly a need for a stimulus package. That's one of the policies we should undertake.

If we go back to the great depression, I think the problem was that people didn't have a proper theory of how the economy works. And so Hoover and Roosevelt at different times -- they vacillated on what they thought -- but at different times they had the right view as to what should be done. You know, new programs and some government spending and so forth. But the trouble was they didn't have a proper model of how the economy works. And because they didn't have the proper model of how the economy works, they were too unambitious about what they did. What both of them needed was the confidence that what they were doing -- at least at one time or another -- was a move the right direction.

So that's one of the aims of this book. To give that theory of how the economy works, so that people who pursue the policies know that they actually need to do something quite big at the moment. That's one. And the second thing is, this book is -- well, we actually think most people will accept most of our arguments.


Well, at a fairly simple level. And the second thing we wanted to do was give legitimacy to those people who think that the government has some responsibility for the economy, and who think the measures that should be taken be the right order of magnitude.

And what are the implications of your theory for the sort of fiscal policy we should be pursuing?

Well, one of the things is that one of the roles of the government is to offset the animal spirits. So that when animal spirits are high -- and people are too trusting and they engage in investment projects that they shouldn't engage in -- one of the roles of the government is to offset them. More should have been done to curb the over-exuberance and excesses in the housing market. That's one.

But at the same time, if the confidence then dries up, it's the role of the government to stimulate the demand that's fallen because of the lost confidence. So basically one of the roles of the government is to fulfill the role that was given it in the Employment Act of 1946 -- which says that the government should have the role of maintaining a full employment economy. So when demand is too high, or when there are securities markets misbehaving, then it's the role of the government to regulate them. But then if demand drops off, it's the role of the government to fill in the gap. And that's basically what Keynes and the Employment Act had in mind.

So now that there's been a drop in confidence, it's basically a question of what the government should do -- what the government should do to fill in the gap in demand because of the loss in confidence.

And how has that debate fared? What do you think about the way everyone is going back to draw from the Keynesian well?

Well, I think he got it right in the 1930s. And he got it right in a much more subtle way than was subsequently appreciated. And now we're coming back to it because it's been needed. Keynesian economics has always been needed.

I think one of the interesting facts the whole postwar period is that we haven't had a major very major downturn. And one of the reasons that we haven't had a major downturn is that for the most part policymakers have believed in Keynesian economics.

So what does that mean? It means that the government felt that it had a responsibility -- that if there was a downturn the government would step in and would maintain the economy at something like full employment. There've actually been relatively few fluctuations in demand, and there've been relatively few rocky spots. And I think it was the advent of Keynesian economics in the 1930s that has meant such spectacular economic growth both in the United States and around the world for the past 70 years.