The economist Dani Rodrik, a professor at Harvard, recently spent a couple of years at Princeton's Institute for Advanced Study. In his new book, "Economics Rules: Why Economics Works, When it Fails, and How to Tell the Difference," he recalls just what a "mind-stretching experience" that sojourn was. He found that many of the visitors to the Institute's School of Social Sciences, prominent academics from other disciplines, harboured a deep "suspicion toward economists." Those visitors seemed to believe, he writes, that "economists either stated the obvious or greatly overreached by applying simple frameworks to complex social phenomena." It felt, Rodrik says, as if economists were being cast as the "idiots savants of social science: good with math and statistics, but not much use otherwise."
Part of the problem, Rodrik thinks, is "misinformation" about what it is economists do, exactly. "Economics Rules" is in part, therefore, an attempt to set the record straight—and to rebut some fairly widespread criticisms of economics in the process. But it's also aimed at his colleagues in the economics profession, who he thinks have made a sorry fist of "presenting their science to the world." When I spoke to him on the phone from the United States this week, I asked about that assumption he'd encountered at Princeton—that economists are "good with math and statistics" and not much else.
DR: Often we take it [mathematics] too far. There’s a certain fetishism that comes along with the use of math. And that shows up in two ways: one is that arguments which are relatively straightforward, that can be put in a directly literary form, we feel we have mathematise them. Sometimes, there’s undue mathematisation or formalisation. We get so enamoured of the math that the mathematical structure of models becomes an object of analysis. And that’s one of the problems with economic theory—that it often becomes applied mathematics, where the point is the mathematical properties of the models. And so it becomes more and more peripheral to what economics should be about, which is to look at social phenomena. But there's a much better appreciation today of the role and also the limits of math in economics than there was 30 years ago.
JD: Right. You say in the book that one of the most significant developments in economics over the past three decades or so has been the increasingly widespread use of empirical methods.
Yes, that has definitely been a [sign of] great progress and has forced us to be much more grounded. And it has squeezed out the space for mindless, abstract theorising or modelling for the sake of modelling. But there’s a tendency in parts of the profession [today] to believe that if you’re just doing empirical work, then you can do away with theory or with thinking about the models that lie behind the particular empirical application. The point that is important to realise—and I’m not sure if I make this sufficiently strongly in the book itself—is that it’s impossible to interpret any empirical evidence without either an implicit or, better still, an explicit model behind it. So every time we make a causal assertion about the real world using data we are implicitly using a model.
The idea of the economic model is one of the central concepts in the book. Where does the explanatory power of economic models come from?
Models are stylised abstractions that lay bare the relationship between cause and effect. I liken [models] to lab experiments. When you conduct an experiment in a lab, you're trying to isolate the thing you’re looking at.
You draw an interesting comparison between models and fables. You say that models, like fables, leave out or abstract from certain aspects of the world as it is. And that, in your view, is a strength, a feature, as you put it, rather than a bug.
As long as we don’t forget that [the model we're using] is a model, not the model. An immediate implication of what I just said, of the way I defined the usefulness of the model, is that the model captures only one of many different causal effects. And it’s going to be most useful when we apply it to a real world setting where, in some sense, that causal effect is the dominant one. But [we should not] forget that there will be other settings where other causal effects or other models are more relevant.
A charge often made against economics, and which you try to rebut in the book, is that many of its assumptions, particularly about the rationality of economic actors, are unrealistic. To what extent does behavioural economics, which injects the insights of psychology into formal economic modelling, take that kind of criticism for granted? Or, to put it another way, does behavioural economics overturn or invalidate what you call the “garden-variety perfectly competitive market model”?
Yes, but it’s only the latest a stream of models that have all had the effect of overturning the central implication of the perfectly competitive model. We’ve known since the 19th century that a market with a few firms would not produce the efficiency consequences of the perfectly competitive model. Then, of course, in the 1970s there was the imperfect competition revolution, where it turns out that, in the presence of asymmetric information, all kinds of consequences follow. So the behavioural revolution isn’t new in the sense of generating results that overturn the basic implications of the perfectly competitive model. It’s new in that it directly removes an assumption that had been at the core of neoclassical theorising—the notion of individual rationality.
There’s a tendency now to interpret the behavioural models as implying that we can now forget about “rational man,” that we can forget about all these optimising frameworks. And again I think that’s wrong. There are going to be settings in which the behavioural model provides important insights. But it would be wrong to discard models in which rational behaviour plays an important role. The trick is to know when to apply a behavioural approach and when to apply a rational approach.
You have a chapter entitled “When Economists Go Wrong" in which you argue that economists’ biggest mistake concerns the claims they often make for the general validity of certain assumptions and models. The danger, in other words, is that of confusing a model with the model.
Right. In policy, that’s where we fall on our faces repeatedly. When we are called on for policy advice our biggest mistake is not drawing the links between the critical assumptions of a model and the real world context with the same kind of rigour and systematic thinking that we exercise when we’re operating within a model.
You give a couple of examples in the book of the way theoretical errors can lead to policy errors. The first example you give concerns the “efficient markets hypothesis”. What role did an overestimation of the scope and explanatory power of that hypothesis play in the run-up to the global financial crisis of 2007-08?
If we take as our central model one under which the efficient markets hypothesis is correct—and that’s a model where there are a number of critical assumptions: one is rationality (we rule out behavioural aspects like bandwagons, excessive optimism and so on); second, we rule out externalities and agency problems—there’s a natural tendency in the policy world to liberalise as many markets as possible and to make regulation as light as possible. In the run-up to the financial crisis, if you'd looked at the steady increase in house prices or the growth of the shadow banking system from the perspective of the efficient markets hypothesis, they wouldn't have bothered you at all. You'd tell a story about how wonderful financial liberalisation and innovation are—so many people, who didn’t have access before to mortgages, were now able to afford houses; here was a supreme example of free markets providing social benefits.
But if you took the same [set of] facts, and applied the kind of models that people who had been looking at sovereign debt crises in emerging markets had been developing—boom and bust cycles, behavioural biases, agency problems, externalities, too-big-to-fail problems—if you applied those tools to the same facts, you’d get a very different kind of story. I wish we’d put greater weight on stories of the second kind rather than the first. We’d have been better off if we’d done so.
You also have a chapter on “Economics and Its Critics”. To what extent does your point about economists’ tendency to overestimate the scope and power of their models neutralise some fairly common criticisms of the discipline made by non-economists? Your point being, as I understand it, that the problem is not so much with the models themselves as with economists’ expectations of what those models will yield.
What I’m claiming is that if economists were actually truer to their discipline and were to project their discipline to the rest of the world as a collection of models, to a large extent it would help neutralise the criticism that economists are [wedded to] one model in particular. You don’t get a reputation as a successful researcher by demonstrating that Adam Smith was right! You get a reputation by showing that there are very circumstances in which he might have been wrong. But this richness, this willingness to countenance non-free-market outcomes, is somehow rarely revealed to the outside world.
Dani Rodrik's "Economics Rules: Why Economics Works, When It Fails, And How to Tell the Difference" is published by Oxford University Press (£16.99)
Part of the problem, Rodrik thinks, is "misinformation" about what it is economists do, exactly. "Economics Rules" is in part, therefore, an attempt to set the record straight—and to rebut some fairly widespread criticisms of economics in the process. But it's also aimed at his colleagues in the economics profession, who he thinks have made a sorry fist of "presenting their science to the world." When I spoke to him on the phone from the United States this week, I asked about that assumption he'd encountered at Princeton—that economists are "good with math and statistics" and not much else.
DR: Often we take it [mathematics] too far. There’s a certain fetishism that comes along with the use of math. And that shows up in two ways: one is that arguments which are relatively straightforward, that can be put in a directly literary form, we feel we have mathematise them. Sometimes, there’s undue mathematisation or formalisation. We get so enamoured of the math that the mathematical structure of models becomes an object of analysis. And that’s one of the problems with economic theory—that it often becomes applied mathematics, where the point is the mathematical properties of the models. And so it becomes more and more peripheral to what economics should be about, which is to look at social phenomena. But there's a much better appreciation today of the role and also the limits of math in economics than there was 30 years ago.
JD: Right. You say in the book that one of the most significant developments in economics over the past three decades or so has been the increasingly widespread use of empirical methods.
Yes, that has definitely been a [sign of] great progress and has forced us to be much more grounded. And it has squeezed out the space for mindless, abstract theorising or modelling for the sake of modelling. But there’s a tendency in parts of the profession [today] to believe that if you’re just doing empirical work, then you can do away with theory or with thinking about the models that lie behind the particular empirical application. The point that is important to realise—and I’m not sure if I make this sufficiently strongly in the book itself—is that it’s impossible to interpret any empirical evidence without either an implicit or, better still, an explicit model behind it. So every time we make a causal assertion about the real world using data we are implicitly using a model.
The idea of the economic model is one of the central concepts in the book. Where does the explanatory power of economic models come from?
Models are stylised abstractions that lay bare the relationship between cause and effect. I liken [models] to lab experiments. When you conduct an experiment in a lab, you're trying to isolate the thing you’re looking at.
You draw an interesting comparison between models and fables. You say that models, like fables, leave out or abstract from certain aspects of the world as it is. And that, in your view, is a strength, a feature, as you put it, rather than a bug.
As long as we don’t forget that [the model we're using] is a model, not the model. An immediate implication of what I just said, of the way I defined the usefulness of the model, is that the model captures only one of many different causal effects. And it’s going to be most useful when we apply it to a real world setting where, in some sense, that causal effect is the dominant one. But [we should not] forget that there will be other settings where other causal effects or other models are more relevant.
A charge often made against economics, and which you try to rebut in the book, is that many of its assumptions, particularly about the rationality of economic actors, are unrealistic. To what extent does behavioural economics, which injects the insights of psychology into formal economic modelling, take that kind of criticism for granted? Or, to put it another way, does behavioural economics overturn or invalidate what you call the “garden-variety perfectly competitive market model”?
Yes, but it’s only the latest a stream of models that have all had the effect of overturning the central implication of the perfectly competitive model. We’ve known since the 19th century that a market with a few firms would not produce the efficiency consequences of the perfectly competitive model. Then, of course, in the 1970s there was the imperfect competition revolution, where it turns out that, in the presence of asymmetric information, all kinds of consequences follow. So the behavioural revolution isn’t new in the sense of generating results that overturn the basic implications of the perfectly competitive model. It’s new in that it directly removes an assumption that had been at the core of neoclassical theorising—the notion of individual rationality.
There’s a tendency now to interpret the behavioural models as implying that we can now forget about “rational man,” that we can forget about all these optimising frameworks. And again I think that’s wrong. There are going to be settings in which the behavioural model provides important insights. But it would be wrong to discard models in which rational behaviour plays an important role. The trick is to know when to apply a behavioural approach and when to apply a rational approach.
You have a chapter entitled “When Economists Go Wrong" in which you argue that economists’ biggest mistake concerns the claims they often make for the general validity of certain assumptions and models. The danger, in other words, is that of confusing a model with the model.
Right. In policy, that’s where we fall on our faces repeatedly. When we are called on for policy advice our biggest mistake is not drawing the links between the critical assumptions of a model and the real world context with the same kind of rigour and systematic thinking that we exercise when we’re operating within a model.
You give a couple of examples in the book of the way theoretical errors can lead to policy errors. The first example you give concerns the “efficient markets hypothesis”. What role did an overestimation of the scope and explanatory power of that hypothesis play in the run-up to the global financial crisis of 2007-08?
If we take as our central model one under which the efficient markets hypothesis is correct—and that’s a model where there are a number of critical assumptions: one is rationality (we rule out behavioural aspects like bandwagons, excessive optimism and so on); second, we rule out externalities and agency problems—there’s a natural tendency in the policy world to liberalise as many markets as possible and to make regulation as light as possible. In the run-up to the financial crisis, if you'd looked at the steady increase in house prices or the growth of the shadow banking system from the perspective of the efficient markets hypothesis, they wouldn't have bothered you at all. You'd tell a story about how wonderful financial liberalisation and innovation are—so many people, who didn’t have access before to mortgages, were now able to afford houses; here was a supreme example of free markets providing social benefits.
But if you took the same [set of] facts, and applied the kind of models that people who had been looking at sovereign debt crises in emerging markets had been developing—boom and bust cycles, behavioural biases, agency problems, externalities, too-big-to-fail problems—if you applied those tools to the same facts, you’d get a very different kind of story. I wish we’d put greater weight on stories of the second kind rather than the first. We’d have been better off if we’d done so.
You also have a chapter on “Economics and Its Critics”. To what extent does your point about economists’ tendency to overestimate the scope and power of their models neutralise some fairly common criticisms of the discipline made by non-economists? Your point being, as I understand it, that the problem is not so much with the models themselves as with economists’ expectations of what those models will yield.
What I’m claiming is that if economists were actually truer to their discipline and were to project their discipline to the rest of the world as a collection of models, to a large extent it would help neutralise the criticism that economists are [wedded to] one model in particular. You don’t get a reputation as a successful researcher by demonstrating that Adam Smith was right! You get a reputation by showing that there are very circumstances in which he might have been wrong. But this richness, this willingness to countenance non-free-market outcomes, is somehow rarely revealed to the outside world.
Dani Rodrik's "Economics Rules: Why Economics Works, When It Fails, And How to Tell the Difference" is published by Oxford University Press (£16.99)
