Please note that this is BBC copyright and may not be reproduced or copied for any other purpose. RADIO 4 CURRENT AFFAIRS ANALYSIS THE ECONOMY ON THE COUCH TRANSCRIPT OF A RECORDED DOCUMENTARY Presenter: Diane Coyle Producer: Michael Blastland Editor: Nicola Meyrick BBC White City 201 Wood Lane London W12 7TS 020 8752 7279 Broadcast Date: 02.12.04 Repeat Date: 05.12.04 Tape Number: Duration: Taking part in order of appearance: Professor Steffen Huck, Centre for Economic Learning and Social Evolution, University College, London. Professor Colin Camerer California Institute for Technology Vernon Smith Professor of Econmics and Law, Goerge Mason University, Nobel laureate 2002. Gordon Foxhall Professor of Marketing, Cardiff University Business School Victoria Chick Emeritus Professor of Economics, University College London Benoit Mandelbrot Mathematician Charles Bean Chief Economist, Bank of England. COYLE: Let’s play a game. The BBC has given me £100 and I’m going to share it with you. The catch is, you can either accept exactly what I offer, or you can turn it down and we both get nothing. It’s called the ultimatum game. But this isn’t frivolous – it’s cutting-edge economics. Professor Steffen Huck of University College London plays this game for real. HUCK: The central assumption of conventional economic theory is that we all like more money more than less money, and that we all know this; and in that case, as long as you offer me some positive amount, I will prefer this positive amount over getting nothing - hence, I will accept all positive offers. Now that means that you can just offer me £1 out of the £100 and then I will accept. So the conventional prediction of economic theory is you suggest to take £99 for yourself, offer me £1, and I will accept. COYLE: What would you do if I offered you £1? Reject it, I bet. If you’re typical, you’ll reject as much as £30 or £40 out of the hundred. The ultimatum game proves that people will sometimes turn down free money. A strong sense of fairness trumps the logic of profit. We’re all human, and all too well aware of the impact of fundamental human passions, love, anger, and envy, on our behaviour. If you’re not an economist, this won’t be news. But how emotion and reason combine is one of the subject’s most controversial questions. The conventional approach assumes we know what we want and set about getting it methodically. But a group of iconoclasts claims that experiments such as the ultimatum game pose a revolutionary challenge to conventional economic theory and policy. Colin Camerer, professor of economics at the California Institute for Technology or Caltech, is at the forefront of the flourishing new field known as behavioural economics. CAMERER: Economic theory really came of age in the 1930s to 1950s sort of in the spirit of physics where the goal was to build up a theory of an entire economy or the theory of an entire market based on kind of particles that interacted. And in order to do that mathematically, the decision was made to make the models of the particles very simple. They’re really not like us. So behavioural economics is an attempt to make these particles have emotions - they care about what other people are buying or wearing, they may get jealous, they succumb to temptation and so on – and it starts to look a lot more like say biology than physics. COYLE: So how do emotions affect the economy? Some aspects of our psychology turn out to make a big difference to how we take economic decisions. For example, we lack willpower. We usually prefer immediate gratification to long-term benefits, and even seem to make the decisions about present and future in different parts of the brain. Many people run up debt on their credit cards – and won’t dip into their savings to pay it off. We can be strikingly inconsistent. For instance, people who won’t hesitate to buy a £20 shirt when it’s 50% off will be underwhelmed by the fact that they can also save a tenner when there’s 1.3% off a £750 sofa. Not all ten pounds are equal, especially when you have to do sums. But do examples like these really add up to more than quirky exceptions to rational decisions about making and spending our money? According to Colin Camerer, the answer’s yes: The research is undermining some of the basics of economic theory, such as the principle that people will do more work for more pay – a principle he started to question after a stay in New York. CAMERER: We actually took a lot of cabs in New York and started to ask the cab drivers “how do you decide how many hours to work every day?” And this is a pretty useful question because a lot of labour economics is interested in the question of labour supply – how many hours do people work depending on how high the wage is. And in most studies, it’s actually very hard to get at this because most workers have essentially a fixed number of hours per year and a fixed annual wage, so you don’t really get much change day to day in hours. But cab drivers you do because they basically rent their cabs in New York for twelve hours, but if they want to quit early they can - so they have a lot of flexibility in how many hours to work. What we found was that a lot of them said, “I set a daily target, and when I hit that, I go home”. So this is what we call a heuristic. It’s kind of a short-cut rule that sounds pretty good, but it has the following implication which goes against economic theory as squarely as can be: which is on a very high wage day – it’s raining let’s say, lots of people want cabs in New York – they’re likely to earn more money per hour because they’re constantly picking people up and dropping them off, so they reach a certain target very fast. On a low wage day – say it’s sunny, people are walking or New York City is empty – they will take a lot more hours to reach their target. So a cab driver using this target rule is going to drive a lot of hours when the wage is very low and not drive very much when the wage is very high, which is exactly the opposite of the normal theory of labour supply: if someone offers you a really high wage, you work harder; at a low wage, you goof off. COYLE: How can that be rational when if you worked harder on high pay days, you might meet your target for the next two days as well? In fact, whether it’s going out to work, saving or shopping, our everyday experience suggests that economics works a lot better in theory than in practice. One of the pioneers of the behavioural challenge to conventional economics is Vernon Smith, joint winner of the 2002 Nobel Prize for his early experimental work. SMITH: You know you walk into a supermarket, there are hundreds of thousands of items for sale. In economics, we have a way of representing preferences where it’s assumed that the individual is able to order all the items that he or she might consider purchasing, order them in a scale of preferences, and then the idea is they maximize those preferences in allocating their budget among goods. Well this is not the way real people operate. You know you go into the grocery store and you have in mind maybe a fairly small number of things that you want to purchase and you’re not bogged down by all kinds of irrelevant considerations which might be relevant next week when you come to market because you have different needs. People are good at solving these problems by kind of a groping trial and error process and remembering the things they did that were unsatisfactory, and then they correct that behaviour in later periods. SEGUE FOXALL: By and large consumers have a small repertoire of brands and they buy among those tried and tested brands almost haphazardly. COYLE: Gordon Foxall is professor of marketing at Cardiff University Business School. FOXALL: For instance, consumers do maximize in the sense of buying the cheapest brand in their repertoire, but that may not be the cheapest brand available on the market. Even on the same shopping trip, they may buy the cheapest brand in their repertoire, but from time to time they will also buy another brand of the same product class which costs considerably more. We may buy the cheapest orange juice on the market for breakfast purposes, for instance, and a much more expensive one to take to work for lunch. The rationality, the maximization that consumers show seems to be hedged around with this limited range of opportunities that they take into consideration. COYLE: His research has found that purchases always have a symbolic value for consumers, as well as the obvious practical ones. People might buy a car for the image it helps them project as well as transportation from A to B, or shoes for glamour rather than comfort. Even basic groceries offer symbolic benefits. The psychological aspects don’t mean consumers’ choices are completely irrational – price still matters, and so does income. But human psychology complicates the economic calculation. Just as the urge for a piece of cake at teatime easily overwhelms that morning’s resolution to go on a diet, a glimpse of some object of desire in a shop window can set our credit cards throbbing uncontrollably, even though we do truly mean to save more. So it’s hard to know which of the new dress or the extra savings would be the self-interested choice. With its emphasis on individual psychology, behavioural economics has another advantage here.. It can experiment. Experiment on the economy by doubling interest rates, and you’ll ruin everyone. But you can test individuals in the lab. No need for alarm – experiments like the ultimatum game don’t hurt. Professor Vernon Smith, has been lab testing new arrangements for, say, financial auctions or electricity markets for decades. SMITH: The important thing that laboratory experiments bring to the table is that you can study things which are not. You see with econometric studies, you can only study what is because it’s the systems that exist now that are generating the data that is the input to econometric applications of economic theory. What you can do in the laboratory is think about completely different arrangements, entirely new institutions, entirely new kinds of markets that have never been tried before. COYLE: Experiments have become increasingly widely used in the design of new markets, such as carbon emissions trading, or auctions such as the auction of the mobile phone spectrum which earned the Treasury more than £22bn four years ago. The centre for economic learning and social evolution at University College London uses computer mock-ups of an industry or market. Participants sit in cubicles in front of a computer screen. Their decisions under different sets of rules – played for cash to make them more realistic - are analysed by the experimenter. But if this sounds unconventional, the latest branch of behavioural economics delves even deeper into the human psyche. Colin Camerer calls it neuroeconomics. CAMERER: Neuroeconomics I think is the way that we’re going to get economics back on a biological footing and away from the physics model, so I think that’s actually quite a profound paradigm shift. It hasn’t happened quite yet and hopefully by the time my kids are in college, they’ll be reading a textbook with some pictures of the brain. But the basic idea is to move away from the idea of a person as a kind of unified being that’s necessarily behaving perfectly consistently and instead think of the brain as it is – as a kind of bundle of, sometimes people use the word modules, that are evolved to do different things, but also collaborate and compete when an actual important decision is made. COYLE: It certainly sounds revolutionary. What does neuroeconomics involve in practice? CAMERER: I’ll give you an example of something we’ve been interested in we call fear of the economic unknown. So in some cases, the probabilities of an outcome like you know if you’re insuring car drivers of a certain age, you have a pretty good sample of statistics of how many are likely to have accidents; if you’re insuring a building in Los Angeles against a terrorist attack, who knows? The probability isn’t zero probably, although it’s never happened before, so really you can’t use historical frequencies as a basis. And we were interested in whether there was a separate brain area that was expressing a kind of fear of the economic unknown in these cases where probabilities really can’t be guessed from historical data. So a typical paradigm is what’s called subtraction; so what that means is in some cases they’re lying on their back in the scanner, they have a pair of goggles on. Inside the goggles is the computer screen exactly as you can see it on your computer screen outside, so they’re seeing the screen right in these pair of goggles. And in some of the trials they’re shown bets where they don’t know the probability of winning and they’re asked whether they want to bet or take some certain amount of money instead. In other trials, they’re shown a bet where they do know the probabilities and what we do is we just subtract the brain activity when they don’t know the odds from the brain activity when they do know the odds. And the difference is we think the special areas that are being active when they don’t know the odds. In our studies, it looks like an emotional region called the insula. The insula’s a small area about the size of a robin’s egg that’s kind of a way-station for body discomfort like squeamishness and pain, disgust, social phobia. All those areas activate the insula. So we think that kind of thing is evidence that when people don’t know the odds, compared to when they do, that there’s a literal body discomfort. When somebody says, “I really don’t know what to do. That kind of makes me nervous”, they mean that in a literal body sense, not just a metaphorical sense. COYLE: So you have these insurance executives lying in scanners wearing goggles? CAMERER: Well we haven’t recruited them. So far, we’ve just been using student subjects. Although there do appear to be people who have the opposite reaction, people who might be almost delusionally optimistic – you know "if I start up a biotech company, I’ll get really rich; and if I don’t do that, and I start up a software company, I’ll still get really rich." So those types of people may actually be very useful in the economy because they do things. They even do things that are bad bets for them but which generate occasionally huge wins for society. And whether they’re exhibiting an opposite brain reaction of normal people who have a fear of the unknown or they’re doing something completely different, we can find out by looking in the brain. COYLE: Neuroeconomics is such a novelty it’s not yet clear whether it will become as widely used as economic experiments in computer labs. But why do economists need to start collecting brain scans at all to gain insights about decision-making which might seem like simple common sense to anybody else? The reason may lie in the fact that actual human behaviour has played so little part in the so-called neo-classical, rational choice approach to economics, which has been dominant for generations. CHICK: Well I’ve been an economist for longer than I care to admit. I’ve been here at University College for forty years. And when I started in economics, neo-classical choice theory was already well settled in. It was the norm. COYLE: Victoria Chick, Emeritus Professor of Economics at UCL, is a long-time critic of mainstream economics, and so in a good position to judge the claims of behavioural economists to have a radical new challenge. She agrees that the neo-classical approach makes sweeping assumptions about our ability to assess information and calculate outcomes of different choices. It assumes we have fixed preferences for various goods and services, forever into the future; that we know what all these preferences are; and that at every moment we try to satisfy all of them them as much as possible, given our incomes. So it assumes that when we nip out to buy a pint of milk, we have our summer holiday ten years from now in mind as well. I asked Professor Chick the obvious question. COYLE: Why did conventional economics ever get to be so daft? CHICK: (Laughs) I don’t know, I wasn’t in on that one. I think it was an attempt to systematize, to mathematicise, to be as complete as possible, and I think – and this is important – also to dehumanize. There’s a strand in economics, which really takes the human being out of economics. You speak of 'agents'. And these agents behave in a very automatic way, and all the things that make us human – unlimited computational ability and basic facts of our existence, in particular the fact that we can’t know the future – you know “will you still need me, will you still feed me when I’m 64” – that uncertainty is there, and you don’t know what your preferences are going to be when you’re 64. These elements are knocked out in this neo-classical choice theory and the only advantage to the economist, if not the world at large, is theoretical simplicity, and that is an advantage that this theory has built on very strongly over a large number of years. COYLE: Conventional economics has always allowed for people to be motivated by something other than money; but whatever the motivation, it assumes people set about getting what they want. And surely this is more natural than the alternative assumption, that people will consistently act against their own best interests and desires. The behavioural alternative gives us many examples of such inconsistent choices, but can’t yet explain why they occur. But it may be that behavioural economics is part of a broader intellectual tide, a shift away from simple, mechanistic theory in favour of the complications of reality. Benoit Mandelbrot is an idiosyncratic genius, a celebrity in mathematical circles. He’s recently been looking at the difference between the smoothness of theory and the roughness of real life in the financial markets. MANDELBROT: Human experience has been very much influenced by the contrast between the smooth and the rough. Primitive man saw the moon, which is round, berries and so on, a few round shapes - almost everything was rough. Most of science had been created for the purpose of describing smooth phenomena. COYLE: His latest book is called The Misbehaviour of Markets. No, it’s not about cocaine and fraud on Wall Street, but rather the fact that financial markets disobey the predictions of economic theory based on full information and rational choice. MANDELBROT: One hopes for regularity but one lives in roughness. Astrologers have found that long ago, that you try to predict a person’s future by the stars, but it has not been very effective. It’s a kind of dream to simplify, and one of the great reasons for the immense success that the theorist Newton had in the 18th century was that Newton showed humans that some aspects of ordinary life can be smooth. That was the surprising thing. People assumed everything was very, very complicated and very rough. So in a way, roughness was assumed by everybody but was pushed aside under this idea that somehow things are smooth after all. COYLE: He holds out the hope that in time we’ll have the theoretical tools to make predictable the inconsistencies of human behaviour. We may be able to say when and how we’ll behave non-rationally. That would certainly turn behavioural economics into a serious challenge to today’s rational choice mainstream. But what will we gain from theories which incorporate the rougher edges of life? Will we be able to eliminate our unwise impulse buys? Or will it make it easier for others to manipulate our purchases? Gordon Foxall, professor of marketing at Cardiff University Business School. FOXALL: The idea that firms can manipulate consumers say for new products in that way, I think is superficial because for consumer goods something like eight or nine out of ten fail at the time that they hit the marketplace, at the point of consumer acceptance. Even though they’ve been through a long process of new product development, test marketing, market testing - still when the customer is given the opportunity do you want to buy this or not, a very large proportion fail. If marketers had the tools to manipulate consumers, we would see much lower failure rates for new products. COYLE: Let’s hope he’s right – it’s hard to see marketers passing up the new behavioural knowledge. But what about those clever people in the corridors of power? Should policy-makers be taking account of how we really behave? Charles Bean, the Bank of England’s chief economist, says that to some extent the Monetary Policy Committee already does think about the psychology of choice. BEAN: It matters significantly. It’s not something that is formally incorporated in a lot of the models we use since at this stage a lot of the thinking which has been coming into economics through psychology hasn’t permeated through into macroeconomic modeling; but in the sort of discussions that we often have in the committee room or in the discussions with the staff during the quarterly forecast round, we often do find ourselves talking about the psychological response of individuals. When you’re discussing things like events in the housing market or in financial markets where psychology’s important, inevitably committee members tend to stray into this sort of territory. COYLE: That makes it sound like they shouldn’t really be there. But then, he argues that conventional theory works well nine tenths of the time. A lot of what we do is rational. And even when it isn’t, your impulsive choices and mine may not matter. Many people will be shopping for ultra-fashionable iPods as Christmas presents, and others will opt for a pair of socks, but it’s the total spent the Bank of England cares about when its Monetary Policy Committee is setting interest rates. Old-fashioned economics applies to consumption decisions in aggregate. Still, there are other areas where the behavioural approach does change policy-making. BEAN: One of the obvious issues for us on the committee recently has been the behaviour of house prices and also the link between house prices and consumer spending because that’s ultimately what affects demand and inflation, which is the thing we’re charged with targeting. Now we believe there are good reasons why the ratio of house prices to earnings that can be sustained is higher now than it was in the past. They’re due to demographic changes in the economy, the fact that the rate of house building has been relatively low. The fact that it moved to a low inflation, a low interest rate environment means that it’s easier for people to make their initial mortgage payments. But we’ve had to struggle with the question of how much of the increase in house prices has been down to those economic fundamentals and how much of it might have been driven by psychological elements in the housing market – people getting in and buying houses because they think ooh you can’t go wrong with bricks and mortar, that house prices are going to continue rising at rapid rates indefinitely, which of course is the classic example of a bubble phenomenon. Now it’s very difficult to distinguish between the two elements how much is fundamentals, how much is being driven by expectations, but they’re the sort of things that the committee has to take a judgement on. And of course we look for evidence that might suggest it’s one way or the other. The reports that we get from our regional agents about what’s going on in estate agents out there in the country and so forth is a great help at helping us try and sort out the fundamentals from the psychological elements. COYLE: The problem in the housing market or financial markets is that nobody can predict what will happen in future. We simply can’t make a fully-informed and rational assessment, so the psychological elements determine our choice. There are other areas, though, where the issue isn’t uncertainty so much as human frailty, like the common tendency to opt for instant gratification. What are the implications when politicians and officials know, for example, that we should be saving more and spending less? Does behavioural economics justify paternalistic policies? Steffen Huck, the experimental economist from UCL. HUCK: If you look at problems of procrastination, absolutely this is people who want to quit smoking and always think oh it’s optimal for me to quit tomorrow; and then the next day comes and it’s still optimal for them to quit tomorrow. Of course they would prefer, they would be better off if somebody else would take the decision for them. This is just a fact. COYLE: I must say it makes me very uneasy, the thought that somebody should be telling me to save twice as much as I am. HUCK: Yes, Diane, but you’re an orthodox economist. COYLE: You make it sound like an insult. (Laughs) HUCK: No, I know. I mean this doesn’t necessarily imply that we would say yes somebody else should take the decision. What you can do is you can explain this to people, that they are prone to these fantasies, and find ways out of this by explaining how you can commit yourself to stick to the solution that you find is ex-ante optimal, so that doesn’t necessarily imply that we are for the nanny state. COYLE: But Victoria Chick is more inclined to favour the idea that people should sometimes be made to do what’s good for them. CHICK: That would be a conclusion that one could draw; that there might be a tendency for policy makers to assume that people are more rational than they really are in the sense that we’re using the technical sense, that we’re using the word rationality now. Or you could simply recognise that people will behave this way and let them go ahead and do it if that is the colour of the government in question. For example, the pension fund debate. A set of incentives may be set up and a set of behaviours is predicted to follow from those incentives and then they discover, which they would know if they were at all introspective, that people are very myopic about pensions – you know I’m not ever going to be that age, it’s never going to happen to me – and under-save. But you know if you go back to the whole psychology of that myopia, it even goes as far as our denial of our mortality. I’m tremendously glad that you know it’s part of being in the university sector that you’re part of a pension scheme which is reasonably well funded, and it’s just not questioned that you will join it and that’s that. COYLE: The decision is made for you. CHICK: Yes, and it’s far the best way. (Laughs) COYLE: But should we be thinking about designing new taxes based on the results of brain scans to see how people react to certain choices? CHICK: Ooh you’re going quite far out for me. Well... why not? If the results are quite systematic, quite universal? And I’d be very cautious about it, but as this knowledge accumulates it’s there to be used. COYLE: It is hard to defend basing policies on the assumption that people will behave in ways we know they won’t. That’s just what pensions policy has often done, and look at the mess it’s in. But it would be dangerous if the economist in the lab started claiming to know what people really want better than they do themselves. Behavioural research is injecting psychological realism back into mainstream economics, so it should help design better policies working with the grain of human nature. But understanding people’s decisions doesn’t justify overriding them. The essential insight of economics since the days of Adam Smith, a wise observer of humankind, is that we’re all better off if individuals are left to make their own choices – even if those choices aren’t as rational as they might be.