Here is my interview with Yale's Robert Shiller about his new book, Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism. The book, which Shiller wrote with George Akerlof of Berkeley, concerns departures from the full-employment economy: How do we explain the fluctuations of the business cycle, or the existence of involuntary unemployment? In answering these questions, Shiller and Akerlof 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.
Shiller talked about the legacy of Keynes, what Milton Friedman got wrong, and getting the stimulus package right. I posted my interview with Professor Akerlof yesterday.
Conor Clarke: The first question I have is one I asked George Akerlof. If I were to look through my textbook for why we have a fluctuating business cycle, the explanation would be something like, "because prices are sticky in the short-run." And I take it from reading Animal Spirits that you think that this explanation is insufficient.
Robert Shiller: Well that's a big, complicated question. It goes back to Keynes, really. Before Keynes people thought recessions were very simple: that wages were sticky and prices were falling -- this goes back a hundred years, the idea that wages are sticky even when prices are falling. It means that real wages are rising in a depression, so employers can't afford to hire additional people. Keynes thought that was oversimplified, but especially...ah, it's so complicated, you asked me an economics theory question.
You're allowed to give a complicated answer.
Well, one way to say it in simple terms is: Yes, there is some price stickiness. But the whole explanation of what goes on can't be found just there. We have factors that are driving the economy. These are the animal spirits that we named our book after: people are inconstant. They change. And if that's the case then sticky prices are not going to be the only effects on the economy.
Another broad question, but hopefully not so theoretical. Animal Spirits, and I guess your book Subprime Solution have a fair amount of psychology in them. So do some other recent books like Nudge and Predictably Irrational. Why this sudden interest in psychology?
Well, we have a problem in the University in that we divide scholars up into departments. Then we are asked to be at "the frontier of research," which is really hard to do. How do you be at the frontier of research? Have you ever tried it?
Nope, I have never tried it.
I'll tell you: the problem is, there are so many millions of people doing things. And every time you get an idea, you find someone else has already done it. So the incentive is to specialize really narrowly and get to know some narrow field really well. And that's how you know you're on the frontier, because you know the field, everything that's been written in this field.
But in many ways it's a bad incentive because it makes you too specialized. One way of specializing is to say, "alright I don't have time to read this psychology stuff, I'm just going to focus in on mathematical economics." This is a career route that many people choose and it yields some important results. But it is a narrow specialization, and many of the people that go down this route are not capable of understanding big events like the crisis we are going through now. You can't do that with just mathematical economics.
So the failure of economics as applied math is a reason why we are now looking toward things like psychology for more robust answers to questions about the business cycle?
There has always been a tendency for some people in the economics profession to want it to be a very pure discipline. It seems to push you back to thinking that everyone is being really smart. I think it is a little bit like Shakespeare. Shakespeare wrote these plays, and just about everybody seems so smart. Sure, you have a fool, but even the fool talks in verse. It's all perfect. And that's an appealing story and it becomes ego-involving for some economists. They really want to think the economy is perfect.
So how did this split come about? It sort of seems like there was a fork in the road, following Keynes's General Theory. Keynes wanted some of the psychological exploration that you and George Akerlof are exploring now, and then Milton Friedman came in and offered the natural rate of unemployment and rational expectations model -- and things like that sort of diverted economics from what might have been a discipline that included a lot more psychology.
I should add that -- and I don't know if I remember this right -- but I think Milton Friedman at some point in his career said he thought that rational expectations was overblown. That's not a quote exactly, but he was a complex man, as was Samuelson. It's a little bit hard to summarize the history of economic thought in simple terms.
But, I think it is true that I wouldn't blame it on... well I would probably blame it on Friedman. He did write Essays In Positive Economics in the nineteen fifties. It was short book, it was very influential, and it was used to justify the assumption that everyone is rational. And so, he wrote it, so I guess he is to blame for a lot of the rational expectations stuff. Not that I think that he necessarily believed all of what he said. But he had the famous example of the billiard player. Do you know this?
It comes from the book Essays in Positive Economics. Suppose we were asked to predict the behavior of a skilled billiards player. How would you do it? So we put balls on the table, and then your job is going to be to predict what his next shot will be. Well would you use psychology to do that? Probably not. Friedman said what you probably have to do is just assume that this guy is rational, and then you can bring in a physicist who understands how these balls behave. And the physicist could calculate the optimal shot, and then that would be your best prediction.
And Friedman says that this doesn't look right superficially: we are trying to predict the behavior of this billiard player who knows no mathematics. The physicist is producing this big mathematical model to describe his behavior. But Friedman still says, 'well that is the right way because there is no better way.' He says you can't ask the billiard player to explain how he arrived at his shot because he is not articulate -- he just knows. And this was used to justify representing people as if they were solving complicated mathematical problems. It's really ridiculous, actually, when you get down to it.
And your book is explaining why that story is unsatisfactory.
Realizing that we're not all skilled billiard players is the first step.
Sure -- maybe even the skilled billiard player has an inefficient sense of fairness, or has been told inaccurate stories about what optimal billiard playing looks like.
And also a sense of what the other player is like. That's the thing the physicist will never get: that billiards is a game of strategy and you are aware of who you are playing against. And life is like that too. It's very complicated, and hard to model with mathematical models. That does not mean that I am opposed to these models. I just think that they just have their place. And they tend to get overblown.
Is it that human behavior is predictable, but described inaccurately by classical economics? Or is it that human nature is just inherently unpredictable and no style of economics could describe it?
Well part of it is inherently unpredictable, but we have to try and find ways of predicting it. And I think that psychologists have important insights for that, and we can't ignore them; otherwise we won't have a complete theory. But the problem with psychology is that what psychologists have to say turns out to be very complicated: people are not easy to understand, and they don't follow simple rules. Economics theorists don't like that because it doesn't give them a very elegant model. Its just too many details.
But the way I view it is that human nature has evolved for millions of years of evolution and every little desire, demand or impulse that we have served some purpose some time in history. And who knows -- it may seem like a crazy quilt of emotions that we have, and we'll never completely understand where they came from. But they drive us.
So is evolutionary biology going to become a greater part of economics -- based on the simple story you just described there?
I think evolutionary biology informs our broad model of the world. And it's not new: there are some economists who emphasis this. In fact there's even an association that meets every year at the American Economics Association convention. I think it is an association that emphasizes that evolution.
But I guess the problem with thinking about human evolution is that it's not tightly linked to economics, in that it doesn't gives predictions about what to expect of people. It says that we evolve for our environment. So what does that mean? What predictions does that give us for economics? It is not a tight link. It's not easy to think about what to do with these insights, so they don't get emphasized.
I want to ask a couple questions about current events. You wrote a piece for the WSJ recently, kind of riffing off the book and the stimulus bill, saying that in addition to closing an output gap the stimulus bill also needs to close a confidence gap -- confidence was one of the Animal Spirits that you had in mind in the book. And one question is, how do you measure something like confidence, how does that inform the size of the stimulus bill? And did the current stimulus bill get it right? Is it going to close the confidence gap?
Well yes, the problem is that confidence is hard to measure. We do have confidence indices that some people have made, and they do go partway toward measuring it. (And by the way they are saying that confidence is very low right now.) But I think confidence does have a number of dimensions. In the book we talk about perceptions of fairness and trust in others. They are also qualitative, and are not necessarily something to attach a number to.
So confidence is hard to measure. But the stimulus bill is only an indirect way of getting at confidence: we don't have a direct route into the brains of citizens. But it does seem that, if you look at history, the kinds of things that get us out of bad economic situations are some sort of stimulus -- whether by the government or some natural, outside event. You know, the end of the Great Depression was World War II. But we don't want to rely on that sort of event. So we have to try to create something. It is hard to do politically, though. And I am worried that we won't do enough.
Is manipulating confidence something akin to a placebo effect? You know, fiscal stimulus will work as long as we convince people that fiscal stimulus will work, because that restores some consumer and investor confidence independent of the effects the stimulus has on the real economy?
Yes that's right, but the thing is that people are really savvy about someone trying to instill confidence in them. And the classic example of that is Herbert Hoover, who was President when the Depression started. He thought that the best thing for him to do was to keep talking up the economy, saying that there's nothing wrong. In fact he went so far as to give predictions, that the recovery will start in two months. And then two months passed and things were even worse, so he just extended his deadline for the recovery. And eventually he became the laughingstock. See that's the problem, it's more subtle than that. My wife is a psychologist and sometimes I will say something like 'You've got to get control of yourself,' or 'Pull yourself together' and she will say 'That's not an enlightened way to talk. We don't say that as psychologists.' You can't tell people to pull themselves together, it doesn't work. So, I guess the best thing we have is a stimulus passage.
One thought I had reading the book and the Wall Street Journal op-ed is, there is some amount of polling evidence indicating that people are not terribly optimistic about the effects the stimulus will have on the economy. Might that undermine the case that a stimulus is necessary to induce a resurgence in confidence, if in fact it is true that Americans are not confident that the stimulus will work?
Well, I think the stimulus has to actually happen. I don't think the stimulus package is a panacea at all, because what is going to happen is that it is going to take a long time for the stimulus to take effect, it won't be big enough, we will see unemployment continuing to rise, and it will be seen as confirming the general public view that stimulus isn't going to work.
Are you confident about the stimulus?
I think it will create some jobs, as they say. If the government hires people, they'll get jobs. But the problem is that the private sector outside may be worsening faster than they are fixing things.
Your piece speculates to some extent that any stimulus will need to be larger, I think you say something to the effect of, 'It has got be bigger than it is to instill confidence and restore trust.' Knowing what you know now, what would an optimally designed stimulus package look like in broad, rough strokes?
Well what I emphasized in Subprime Solution was that it is not just the stimulus package. We should be setting a course toward a better economy. That's why I have these things like, subsidizing financial advice. Nobody else picks that up but...
But I think you wrote a New York Times article about it.
I did write about that. We have to give people the impression, the valid impression that things are being fixed and are getting better. And I have these ideas for new mortgage institutions and new kinds of financial markets and home equity insurance. These are the kinds of things that might give an additional boost by giving us the sense that we have a government that is creative and is responsive to the people's needs. We have a serious problem. It's not easy for the government to fix, and I am giving the best ideas that I think that we have. But it's not easy.
We want to hear what you think about this article. Submit a letter to the editor or write to firstname.lastname@example.org.