Published July 12, 2008
Psychology's ambassador to economics
The father of behavioural economics Daniel Kahneman talks to VIKRAM KHANNA about cognitive illusions, investor irrationality and measures of well-being
MANY stories about the Nobel prize winning psychologist Daniel Kahneman start with a puzzle. This is one of the favourites:
A bat and a ball together cost $1.10. The bat costs $1 more than the ball. How much does the ball cost?
Did you say 10 cents?
More than half of a Princeton University economics class gave the same answer (as did most of my friends), and it is wrong.
The thought process that produced this wrong answer is an example of what Prof Kahneman calls 'System 1 thinking'. It is, he says, 'fast, effortless, associative.' We just impulsively pluck out the most intuitively obvious-sounding answer. We don't stop to deliberate, especially when the question seems trivial.
Less frequently, we practise what he calls 'System 2 thinking', which he describes as 'conscious, deliberate, slower, serial, effortful'.
(The right answer to the above question is 5 cents - try 'System 2 thinking'.)
'System 1 works well almost all the time,' he says. 'The problem is when it comes to big decisions. You need to know when you must slow down and recognise situations where it's better to calculate rather than follow your intuition.'
Mr Kahneman, who won the Nobel prize for economics in 2002 and teaches at Princeton University, is widely credited - together with his collaborator for some 30 years, the late Amos Tversky of Stanford - as being the founder of behavioural economics.
One of the insights of this relatively new discipline is that when it comes to making decisions, people are often not the selfish, rational, far-sighted, 'profit maximisers' that mainstream economists assume them to be - and which forms the basis for many policy prescriptions.
'There are domains in which rationality provides a good approximation of behaviour, but there are other domains in which it is not even close,' explains Prof Kahneman. 'For example, individuals' attitudes to savings - at least in the US - are completely irrational, and there is a need for policy intervention to enhance rationality.'
Many mainstream economists still view behavioural economics with a mixture of curiosity and suspicion, but they are increasingly coming around, because some of its findings are too compelling to ignore.
Prof Kahneman does not however consider himself an economist. 'Absolutely not,' he says. 'I study judgement and decision-making. I never really made a transition into the field of economics. What happened is that some economists became interested in our work. I learnt some economics from my friends over the years, but these were friends who were interested in what I was doing.'
It is evident from Prof Kahneman's deeply introspective autobiography that his interest in the workings of the human mind goes back to his childhood. At the age of seven, in German-occupied France, he was already convinced, as his mother had told him, that 'people were endlessly complicated and interesting'.
About this fundamental truth, he was to discover more and more, in a lifetime of study of the human psyche. One of his key findings was that people suffer from various cognitive illusions, which affect their decisions and their behaviour. He has documented scores of these and inspired other researchers to find even more.
One of the most common problems, he points out, is overconfidence. 'It is pervasive,' he says. 'It is very common for people to have more confidence in their judgement than they should.
He gives an example: 'There is a great book by Philip Tetlock on political judgement. He asked people who were classified as experts - people in the CIA, for example - to make medium and long term judgements about political and strategic developments. And then he waited 15 years to see what actually happened. And the answer was really quite striking. Basically, he found that these experts - and you know, these are people who are all over the press and so on - they don't know more than the average reader of the New York Times when it comes to predicting the future.
'They have expertise. But their expertise is really in making judgements about the past or the present or, at best, the immediate future. They have no expertise in predicting the long term future.
'The same is true of financial analysts who make predictions about stocks. They have some expertise. They know how to analyse balance sheets. They understand the principles of finance. That gives them the idea that they can guess how well a stock will do.
'But that is nonsense - they can't guess how well a stock will do. But they still feel they can, because they have expertise in some aspects of the job. So this is one of the things I am most curious about - this illegitimate transfer of confidence from areas where there is genuine expertise to areas where there isn't.'
Prof Kahneman points out that overconfidence is also the main problem with individual investors. 'They trade too much,' he points out. 'They think they know things and they don't. For me, one of the most important findings in behavioural finance is a finding by a former student of mine, Terry Odean, who is at the University of California at Berkeley. He looked at the trading records of some 10,000 investors - he had access through a brokerage. So he knew every transaction that those individuals had engaged in. And he could look, in particular, at cases where someone sold one stock and bought another stock within a short period of time.
'That person is telling you he thinks the second stock will yield more profit than the first stock in the future. And now you can wait a year and see how well the two stocks did. The result was published in the Journal of Finance. On average, it turned out that the stocks that people sell do better than the stocks they buy by 3.2 per cent. So if you add transaction costs, the cost of having an idea for an individual investor is almost 4 per cent. That is enormous.'
Investors often compound their erroneous buying decisions by making another common mistake in managing their portfolios. 'The findings show that, by a ratio of 2 to 1 or more, people sell their winners and not their losers,' says Prof Kahneman. 'That's a mistake. It is a way to lose money, actually, and for a couple of reasons. One is that in the short term at least, winners tend to do better than losers (so it's better to keep the winners). The second - at least in the US - is taxes. There is a great tax advantage in the US of selling losers and keeping winners. But that is not what people do, and it's a major cause of losses.'
I ask him how well institutions invest compared with individuals.
'Institutions do less well than they expect, but they do better than individuals,' he says. 'It's not that they know the future better. But they are less likely to trade on 'noise'. A lot of trading is news-driven. And individuals are much more likely to trade on news than institutions. So institutions take advantage of this.'
Another common form of irrationality vis-a-vis money that individuals exhibit, is what Prof Kahneman calls 'loss aversion'.
He explains: 'If I offer you a deal: I toss a coin and if it's heads, you will win $200, but if it's tails you will lose $100. Will you take the bet?
'The majority of people say no. They don't consider it attractive. They weigh pleasure against pain, and for most people, the pain of losing $100 outweighs the pleasure of winning $200. Losses are given a weight of more than 2:1 compared with gains. This is loss aversion.
'But then, suppose you are offered not one, but ten, of the same gambles. Would you accept? Most people would now say yes, because the probability they will lose overall is smaller than in a single gamble.
However, that single gamble that was first offered would not have been the last gamble of their life, so why do people turn it down? If they were rational, they would think of it as one of many gambles to come.
This kind of irrationality is what he calls 'narrow framing'. 'We consider decisions one at a time, we focus narrowly on just what we see at that moment,' he explains. 'We also consider how much we stand to gain and how much we stand to lose from that one decision - in other words, we focus on changes in wealth, not on total wealth.'
Given the cognitive impairments of individuals, does it make more sense to take investment decisions in a group setting?
'Not necessarily,' according to Prof Kahneman. 'Groups are better than individuals in some situations, but not in all.
'For example, if you have, in a group, people who are all inclined to make the same mistake, then the group will make that mistake and will make it with greater confidence than any individual, because people feel supported by the fact that others agree with them. So, you need some structure to prevent errors.'
And what are the most common problems when it comes to group, and corporate, decision-making and what structures can be put in place to prevent them?
Legitimising criticism and dissent is the greatest challenge, he says. 'The big problem is that once the organisation begins to make up its mind, everybody falls in line. And then the information that goes to the top gets biased.
'This is clearly something that happened with the Iraq war. What the decision makers wanted was known, and the intelligence community basically gave them the information they wanted - and of course, it turned out a disaster.'
Preventing this from happening is difficult, he adds, but there are ways.
'I learnt a great idea from a friend of mine, Gary Klein. He recommends what he calls a 'pre-mortem'. It's a very clever idea. You get a group of people who have made a plan, it's not completely final, but they have a broad plan. And then you bring them together for a special session, 45 minutes is usually enough. You tell them, take a sheet of paper and now imagine the following: a year has passed, we have implemented the plan and it is a disaster.
'Now, write down on that sheet of paper what happened. Why did the plan turn out a disaster?
'It's a brilliant idea. Because you have a group of people who are now encouraged to think of difficulties and problems with the plan. And that solves the problem of legitimising dissent very elegantly, and it's easy to implement.
But Prof Kahneman is not optimistic about organisations becoming introspective. In particular, they are reluctant to go back and look at their past mistakes.
'It doesn't happen because it's against the interests of the decision-makers,' he points out. 'Nobody likes to expose themselves. So any systematic attempt to collect the way decisions were made - which is really essential for any organisation to improve - is generally sabotaged within the organisation.
'They just don't want to know, and they're not interested in the past. They also greatly exaggerate the extent to which each problem is unique. They often think each problem is a new morning and so history is not relevant.'
To some extent, incentive structures within companies can be changed to help resolve this problem. But only partially. 'When you talk to CEOs, they will not want their own decisions to be followed too closely. But you could get them interested in having their subordinates' decisions improved,' he points out.
These days, Prof Kahneman is working on what he calls 'well-being' or on measuring happiness, roughly speaking.
'One of the activities of the Gallup organisation, which brought me to Singapore, is to conduct a world poll. I believe more than 250,000 people are polled across 140 countries. One of the questions they were asked is the question of the 'ladder of life'. That is, imagine a ladder with ten rungs, and the top rung is the best life possible, and the bottom rung is the worst possible life. And the question is, where are you on the ladder?
There are enormous differences across countries. The world champions in happiness are the Danes. They average 8 out of 10. And people in some very poor African nations average just above 3. Singapore scored 6.2.
'The correlation between happiness and GDP per capita is extremely high. Across countries, the correlation is 0.84.'
But there is also a positive correlation between stress and GDP. He explains: stress is higher in high GDP countries than in low GDP countries. 'But there is more depression, anger, sadness and physical pain among the poor - though not worry, which is very interesting; the poor do not worry as much as the rich.'
On how public policy can influence well-being, he says: 'If you're looking at peoples' emotional state, then clearly mental illness is a very big problem. A lot of suffering in society is because of mental illness. There should be a large effort to alleviate that.
'Incorrect use of time is another problem. People waste a lot of time. They act as if time were not a constraint, but it is a finite resource and people don't use it well.
'Then, it's very clear that people are happier when they are in social contact. Loneliness is a source of misery. And simply the amount of time people spend in contact with others is a powerful predictor of their emotional state. To some extent there are policies that can influence this - even architectural forms matter; you can shape the environment in which people live so that there will be more contact among them.
'So we are learning a lot about well-being,' he says. 'And that's enough to keep me going for the rest of my life.'
John Hull has a confession to make.
As a professor of finance at the University of Toronto's Rotman School of Management, he has won international acclaim for designing and valuing complex financial tools such as options and other derivatives.
But when it comes to managing his own money, Prof Hull has little use for such exotic instruments. His investment portfolio comprises mainly index funds. And while he keeps reminding his students about the importance of hedging risk, his own liabilities are heavily concentrated in Canadian dollars.
"In retrospect, I certainly haven't behaved optimally," the 62-year-old, UK-born professor says with a wry smile.
Seen from a different angle however, Prof Hull's financial strategy is entirely consistent with the message he hammers home as a teacher, author, consultant and expert witness in derivative-related lawsuits: that is, keep things as simple as possible.
"There's a danger, with all the people with PhDs in physics and maths who have moved into this area, that some of the models become too complicated", Prof Hull says. "There's a tendency for people with that sort of background to just want a really difficult problem to solve. And that's not necessarily what's needed."
His avoidance of this tendency helps explain the respect his work has garnered.
Izzy Nelken, president of Super Computer Consulting in Chicago, who has known Prof Hull for 16 years and co-authored several publications with him, notes that while "some folks like to couch it all in huge mystery, John doesn't do that. He always brings the latest developments in a simple, easy-to-understand manner."
Armed with a Cambridge maths degree, Prof Hull's first job was in operational research at a UK shoe manufacturer.
An interest in applying mathematics to finance led him to do a PhD at Cranfield University.
He had his first taste of Canada as a visiting lecturer at the University of British Columbia in Vancouver in 1979. He then moved to York University in Toronto two years later so he could work in a leading financial centre. He has been at Rotman since 1988.
Quantitative analysis and analysts have made deep in-roads in trading rooms and financial research departments since two University of Chicago economists, Fischer Black and Myron Scholes, devised a mathematical model for pricing options and corporate liabilities in the early 1970s.
However, the recent turmoil in financial markets has jolted faith in the so-called "quants".
Prof Hull agrees that "there's some ground for concern" that traders and analysts have relied too heavily on mathematical models in their decision-making.
"We need a much more common-sense approach to risk management and must not let quants and traders run free-rein for short-term profits," he says.
The problem, in Prof Hull's view, has been an overdependence on models that are based chiefly on recent market trends.
Over the past three years, for instance, "we were looking at a period when volatilities were very low", he says, "so values at risk were lower".
"To some extent, that model led to a false confidence on the part of the banks. Somebody should have been saying: 'Let's look at the big picture, what could go wrong? How well will we come out if it does go wrong?'
"In most institutions I don't think anybody was doing that. They were just relying on: 'We're making a lot of money, the value-at-risk model says we're okay'."
But heavy losses since the onset of the US subprime mortgage crisis have prompted a good deal of soul-searching among quants, and those who employ them. "I don't think there is a substitute for sound managerial judgment," Prof Hull says.
"In a few companies, however, rather than senior managers letting the traders run loose on this, they sat back and thought about the environment out there; about what could go wrong and how badly they would suffer if it did."
As for his own approach, he says that "my claim to fame is not being an out-and-out quant.
"It's more looking at a situation, coming up with the simplest model that captures the essence of it, and then writing it up in such a way that people will easily be able to understand it."
His seminal book Options, Futures and Other Derivatives , now in its seventh edition, is widely regarded as the bible of the subject. According to Mr Nelken, "he's had a substantial impact on a whole generation of financial mathematicians".
Prof Hull tries to bring equations to life by including actual figures in them. His books are dotted with "Business Snapshots" that show how quantitative theories and models are applied in real-life situations.
His books also demonstrate his appreciation of the shortcomings, as well as the advantages, of exotic financial instruments.
For example, the chapter on asset-backed securities in the latest edition observes that these financial products have been created from exotic assets such as royalties from the future sales of a piece of music.
"Dealers have been very creative - perhaps too creative - in their use of this type of structure," he concludes.