Daniel Kahneman, a pioneer in the field of behavioral economics, passed away yesterday at the age of 90, according to published reports.
In late April, 2004, Financial Advisor contributor Harold Evensky interviewed Kahneman, his client at the time, about the emerging field of behavioral finance and how advisors can learn from it. It says something about how young this discipline is when Kahneman, a psychologist and not an economist, was awarded a Nobel Prize for economics in 2002. In this far-ranging interview, Kahneman identifies several common mistakes individual investors are prone to make.
Evensky: A good place to start off is certainly the whole field of financial advice, where behavioral finance is a hot subject. But is it behavioral economics or behavioral finance? You know, what is the field-what's the proper description?
Kahneman: Well, the broader field is behavioral economics. Behavioral finance is what finance people do when they sort of give up the standard assumptions.
Evensky: What are the standard assumptions that we in the finance field should be giving up?
Kahneman: A basic assumption is that in standard economics and in financial analysis, people are assumed to be rational. What is meant by that is not that they know everything, but that they have a consistent system of beliefs, that they act on their beliefs and that they have consistent preferences.
Now, when you abandon those assumptions of rationality or similar assumptions that people make within the context of finance, and you accept assumptions that are psychologically realistic, then you are doing behavioral economics, and if you are doing finance you are doing behavioral finance.
Evensky: What are some of the examples of the biggest behavioral errors that the traditional retail investor makes?
Kahneman: The biggest of them all is probably what is called narrow framing-looking at investment problems in isolation from the whole portfolio. This is probably the most common and fundamental source of mistakes.
Evensky: Almost every question I have, with our clients, and for that matter probably with ourselves every day, is what can we learn from the field of behavioral economics to help us reframe or help our clients reframe their approach?
Kahneman: People are probably more rational when they take a very broad view and derive their specific decisions from a broader view. The idea is to begin with strategic decisions-like how safe you want to be, what the basic allocation is going to be. These general rules should govern decisions. As an individual investor, you shouldn't be spending too much effort in trying to guess about a particular stock or about the trends of particular economic indicators. That, I think, is one of the main lessons of behavioral economics. When people put too much emphasis on being right in very specific choices, they end up doing the wrong thing.
One of the major findings from academic studies of the behavior of individual investors is the disposition effect: When people are selling stocks from their portfolio, they tend to sell winners and hang on to their losers. It's very natural why people do that. If you sell a stock that is currently worth more than you paid for it, then you can pat yourself on the shoulder for a successful investment. When you cut your losses on the stock, in contrast, you have to accept punishment right now for having made a choice that didn't turn out well. It is not surprising that people prefer rewarding themselves rather than punishing themselves. But recent research by Terry Odean has shown that this is really very costly, for several reasons. On average, you will do much better if you sell losers and hang on to your winners. This is a pretty good example of normal investor behavior leading to generally bad outcomes.
Evensky: What, if anything, can we do to help our clients make better decisions?
Kahneman: Well, in this case, telling people that how much they paid for a stock is really not relevant anymore. It is necessary for people to understand this principle, but it is not easy. So what you can do depends on people's willingness to take advice. But the problem that I see is that people very frequently ask for advice about a specific investment-should I sell this, should I buy that? In principle, advisors should resist such requests to give advice about specific investments without knowing the whole picture about that individual's portfolio, the amount of risks that they are carrying. People should be educated to ask for advice about the big picture, and always to consider particular decisions in the context of their overall situation and objectives.
Evensky: A somewhat related problem that all of us see with clients is an unwillingness to sell. People often have portfolios that are significantly undiversified. They have just a few positions, typically in the company they work for, and getting them out of it is almost impossible. They own a stock which could collapse, they don't know anything about it, but there was a big loss and they are just unwilling to part with it. All these are caused by the same kinds of errors. Are there any different techniques to try and deal with those different kinds of problems?
Kahneman: What you just described as the unwillingness to sell dogs, now that's what we call the disposition effect and it's something that we understand really well. There are many reasons not to want to sell a stock that has lost a lot of money. So you would be punishing yourself right now by selling it, and in addition it seems cheap to hang on to a loser. But perhaps the key point is that for an untrained individual to make decisions about specific stocks is already a mistake. Some of your sensible clients-I hope I am one of them-want you to make these decisions, and don't really want to know what you are doing: "I think you make such choices better than I could." Merely being dispassionate about the outcome gives you, the advisors, a big advantage in getting things right. It is clear from research on individual investors that the great majority of them are out of their league. They are bound to lose money, on average, because they are hyperactive, they churn their accounts, and they don't even know how well they are doing. Many people think they are successful when in fact they are not. Somehow they manage to delude themselves.
Evensky: Let's go back to the issue that we often see: People working for large corporations and they have huge positions in their company stock. Those stocks they may perceive as winners, but we can't get them to sell those either. Is it the same or is it a separate phenomenon?
Kahneman: This is a sad thing to say, but maybe there are by now enough examples to have some educational impact. People tend to be influenced by examples more than by general rules, and so Enron and a few other salient cases illustrate how incredibly stupid this behavior can be. Persuasion by such vivid examples is the best way to go, but you are still going to encounter resistance. People like to buy stocks that they are familiar with; this is the home bias. People who live in New Jersey like to buy New Jersey-based companies, and somehow they feel they know more about these companies than they know about a company in Hong Kong, which often they don't. These are very strong biases.
Evensky: Byron Wien of Morgan Stanley has said that he can't find much difference between the behavior of individuals and institutional investors. Does the same problem apply to professionals as well as individuals?
Kahneman: Terrance Odean at the University of California is studying these issues. There are good reasons to expect that the considerations that affect institutions and individuals are different. The people who act as agents on behalf of institutions have reputations to maintain. There is evidence of a lot of window dressing before reporting periods and a lot of cleaning up. But the professionals have a broader picture of what is going on. They probably have much better information.
Odean has been doing studies on individual investors. He gets records from brokerage houses, so that he knows every transaction that an investor engaged in over a period of years. This data enables him to identify when an investor has an idea. If you see that an investor sold one stock and bought another within a short period of time, then you know that this individual had an idea about the value of these stocks. What Odean did a number of years ago was to follow those stocks in order to find out what happened to the stocks the individuals bought and to the stocks they sold, one year later. The remarkable result is that on average, the stock that people sell does better than the stock they buy by 3.5%. If you add transaction costs, then the cost to an individual investor of having an idea is roughly 4%.
Now, clearly there is somebody on the other side of the transaction, and probably the other people on the other side of the transaction are institutions. So institutions, although they are not very good at picking stocks themselves, can do better than individuals.
Evensky: A quick question about professionals. Most studies show that 80% or 90% of money managers think that they are above the average. Are they living in the same kind of bubble, only they're viewing the world a lot less narrowly and not doing the narrow framing that got a lot of individual investors in big trouble?
Kahneman: They are in a business in which every participant thinks he or she is better than average, and able to beat the market even if no one else can. The people who self-select into that occupation are bound to exaggerate their skill. This is almost a statistical regularity. It has to be that way. If they didn't think they were above average, they would be doing something else where they think they would have a competitive advantage.
So I don't think that overconfidence is fixable, it is always going to be there. But it's also true that many financial professionals have a lot of confidence in their ability to do things that they simply cannot do. They are confident about individual decisions, and about particular opinions and general forecasts about what the market will do, even when there is no foundation for that confidence.
Evensky: I take from that that behavioral finance theorists would at least be reasonably warm supporters of index investing and passive investing?
Kahneman: That's true. You know, most of us are very skeptical about people who claim that they can do miracles. We are all very aware of how easy it is for people to exaggerate their own skills and how biased people are in evaluating their own past performance. And there are some exceptions, although I [question] even some of the exceptions. Of course there are people, such as Warren Buffett, who collect a lot of information about particular firms and they buy the firms, not the stocks. Clearly there is skill in doing that.
But many of us are very skeptical about stock picking. Even if there are a few people who can do it, most of those who claim they can do it certainly cannot.
Evensky: Well, one of the most common issues of discussion among practitioners in the financial planning profession is this whole issue of risk. We think in terms of risk statically, standard deviations. Is there any research on what risk means to investors?
Kahneman: Well, it certainly doesn't mean standard deviation. People mainly think of risk in terms of downside risk. They are concerned about the maximum they can lose. So that's what risk means. In contrast, the professional view defines risk in terms of variance, and doesn't discriminate gains from losses. There is a great deal miscommunication and misunderstanding because of these very different views of risk. Beta does not do it for most people, who are more concerned with the possibility of loss.
Evensky: The other related issue that I am sure all practitioners ran into in '98 and '99 is that if a portfolio is reasonably diversified but other people seem to be getting rich with no risks, then investors are feeling unhappy. There is a risk of not being with the pack, or the risk of underperformance.
Kahneman: This is why individual investors typically will get in too late. By the time they convince themselves that everybody else is getting rich and they are the only ones not getting rich, it is probably late in the day. When those investors get in, it may be time for other people to get out. There is a psychological phenomenon that we are aware of which is that people see patterns in pure noise, where in fact there are no patterns at all. So people are prone to think that the world has changed, and they do not appreciate that the next turn is near.
During the bubble, there was some talk that the rules of the economy had been suspended, that we were in a new world where things can go straight up forever. Quite a few people almost believed that. We are very vulnerable to seeing random fluctuations as indications of permanent change, a new pattern.
Evensky: There is obviously a lot of research and a lot of understanding about these kinds of mistakes that the investors make. The question advisors all have is, what can we do about it? Are there any specific kinds of tools, techniques-other than being aware of it and trying to warn people, but often they don't pay attention. They would much rather chase the hot story.
Kahneman: The advisor is in a very difficult position, in that you want to do the best for your client but you also want to have clients. So it is a difficult world, but that's the place we live in. The best advice may well be, "Do little and don't think too much," and "Leave it to me and don't check your results too frequently because that will cause you to make mistakes," but this advice can sound very self-serving. For good advice to be acceptable, we are dependent on the public getting educated. This is happening, but it's a slow process.
Still, there is more money under passive management. People are aware of the difference between long term and short term, and the idea of staying the course is more popular than it was. So there are broad changes at the level of education of the public. What an individual advisor can do is to lay down a line that is reasonable. Then they have to say that, of course, you [the client] are a free agent and if you want me to do something else, I'm an advisor, I am not your father. Proposing a reasonable line basically means for individual investors to become relatively [passive], not do much, not check too often and not make too many decisions.
Evensky: Dick Thaler [of the University Of Chicago] had suggested that an optimal period for reviewing your performance was 13 months.
Kahneman: Actually what Dick was saying is that for people who have invested for the long run, the optimal period is several years. If you are highly diversified, then you may do better by just not looking at the intervening results, because responding to fluctuations will lead you to be too active. What Dick found is that you can explain the behaviors of the market as a whole by assuming that many people-I think that applies to institutions more than individuals-have the horizon of about one year. That is, they think of the outcome of current investments in terms of gains and losses where the horizon is about one year.
Now, losses loom larger than gains in people's minds, and that causes them by and large to be risk averse. And risk aversion causes the price of equities to be relatively too low. Thaler wasn't saying it's optimal. He was only saying that the relative prices of assets suggest that one year is the prevalent horizon, and it is much too short.
Evensky: I think all of us wish we could have clients who had a one-year horizon instead of one-minute horizon. Let's go a little more into the relative feeling about losses versus gains.
Kahneman: People view losses and gains very differently. For an example, consider this question: "We will toss a coin. If it shows tails you lose $1,000, and if it shows heads you win $X. What would X have to be for this gamble to be attractive for you?" Many people will ask for more than $2,000 to compensate them for the risk. This is really crazy if you start to think about it. If you get 1.5 to 1, this is a better investment than most and the returns are immediate. So what we measure here is pure loss aversion: People just hate the idea of losing that amount, and that emotion governs their decision. And this decision is foolish at many levels. You, the decision maker, should not be thinking of the outcomes as gains and losses, you should be thinking of your possible states of wealth after the game is played. You should not be thinking of a gamble in isolation, you should be thinking of all the opportunities that will come your way in the future. In these various ways, the advice is for people to frame the decision problem broadly. Broad framing is the best way to control loss aversion, which is one of the main sources of irrationality in individual decision making.
Evensky: It seems quite extraordinary that you won the Nobel Prize in economics and you are a psychologist, not an economist. Do you think this shows classical economists are beginning to recognize or take an interest in behavioral issues, or is there still a significant divide in beliefs?
Kahneman: The prize that I got is really a testimonial to the achievements of behavioral economists, in particular Richard Thaler, who is the leading genius in that field. As I see it, I got the prize because of their success. Now, behavioral economics is still a minority movement within economics. But it seems to be a growing minority, and there are many signs that it's becoming increasingly influential. Two of the last four John Bates Clark medals [for achievement by an economist under the age of 40] have gone to behavioral economists. So it is clearly spreading, but it still is a minority movement.
Evensky: If a professional wanted to pursue this, where should we as a nonacademic practitioner be looking in the retail market?
Kahneman: Quite a few good books on the subject have been published over the last few years. I know of at least three that have been authored or co-authored by respected academics, which were intended for the general public. The names of Shefrin and Statman, Bazerman and Gilovich come to mind, and there are probably some I don't know about. Many other books, some very good, have been written as behavioral finance has become popular, by journalists and experts who are not academics. But I am biased in favor of my tribes, so those are the books that I am aware of.
Evensky: What we are talking about is the individual versus retail investors, and the question I had-because you are talking about some of the mistakes that individuals make, not selling their losers-is that one reason why you think that institutions do better?
Kahneman: The systematic study of how institutions make decisions is just beginning and not much is known, at least known to me. Terry Odean and Brad Barber would be the source, because they have been working on it. But, of course, you might expect that institutions would be better at handling some of these problems. One conclusion that I remember was that they do a better job than the public in distinguishing news that is worth acting on from noise. And, of course, they have more information. So I think they are better at dealing with risks, because they are more likely to adopt a broad frame and view life as a repeated game. They are professional and there is an advantage of being professional. I don't mean to suggest that professionals can guess what individual stocks will do, but they will be able to avoid certain mistakes.
Evensky: Thanks a lot, Danny.