By Teh Hooi Ling.
13 July 2002
Business Times Singapore
Researchers have found investors are not rational and because of this, many US asset management firms invest based on behavioural finance theories
TAKE a crack at the problems below. Assume the following information: 1) The probability that a woman has breast cancer is one per cent. 2) If she has breast cancer, the probability that the radiologist will correctly diagnose it is 80 per cent. 3) If she has a benign lesion (no breast cancer), the probability that the radiologist will incorrectly diagnose it as cancer is 10 per cent.
The question: What is the probability that a woman with a positive mammogram actually has breast cancer?
Ninety-five of the 100 physicians presented with this problem (Eddy 1982) estimated the probability to be about 75 per cent.
Does that sound about right?
Not even close. The correct probability, as given by Bayes' rule (a theorem in probability), is 7.5 per cent.
Try another
Q: What is the probability that a 20 to 30-year-old Singaporean male who does not engage in risky sexual behaviour is in fact infected with HIV if he gets a positive result on an Aids test?
The background information is as follows: 1) The prevalence rate is 0.01 per cent. 2) The sensitivity of the test is 99.8 per cent. 3) The false positive rate for the test is 0.01 per cent.
Clearly, this is a tremendously accurate test, identifying almost everyone who has the virus while hardly ever incorrectly identifying an HIV-negative man as positive.
So if the test says a man has the virus, what is the probability that he really does?
Or asked another way, what is the probability of an uninfected man being tested positive for HIV?
The answer: 50 per cent!
Here's how you work out the probability. Imagine 10,000 men in the low-risk group taking the test. One of them is infected (the prevalence rate is 0.01 per cent). He will almost certainly test positive (the sensitivity of the test is 99.8 per cent).
Of the remaining 9,999 uninfected men taking the test, one will also test positive (the false positive rate for the test is 0.01 per cent). So two men test positive; one actually has the virus, the other doesn't.
Therefore, if a man from the low-risk group tests positive, the probability that he has the virus is about 50 per cent.
We can come away with a few lessons from the above problems. First, probability can be a difficult concept to grasp.
Often, our intuition about the likelihood of an event occurring may be flawed.
I used the above two problems to illustrate the cognitive errors that we all make. Such errors or biases can lead to irrational decisions, including investment decisions.
A substantial amount of work has been done in the field of behavioural finance.
Many researchers are of the opinion that such biases are consistent enough to be exploited for profit.
Indeed, US asset management companies such as LSV Asset Management, Fuller & Thaler, David Dreman, Ken Fisher and Martingale invest based on behavioural finance theories.
The investment philosophy of LSV - which manages US$8.5 billion worth of assets - is to exploit the judgmental biases and behavioural weaknesses that influence the decisions of many investors.
These include: the tendency to extrapolate the past too far into the future, wrongly equating a good company with a good investment irrespective of price, ignoring statistical evidence and developing a 'mindset' about a company.
Investors are irrational
Standard finance and economic theories are based on the notion that investors are risk-averse, rational and consider all available information in the decision-making process.
But there are persistent anomalies in the market such as the excess returns from value stocks which called into questions these assumptions.
And many researchers have uncovered that investors are not rational. In other words, they are swayed by their emotions, alternating between greed and fear, they are loss-averse, they lack self-control, they become over-confident and they are prone to regret, among other things.
For example, some believe that the outperformance of value investing results from investors' irrational over-confidence in exciting growth companies and from the fact that investors generate pleasure and pride from owning growth stocks.
Some noted academics in the field include Daniel Kahneman (Princeton), Meir Statman (Santa Clara), Richard Thaler (University of Chicago), Robert J Shiller (Yale), and Amos Tversky. Tversky died in 1996 and is frequently cited as the forefather of the field.
Common examples of irrational behaviour (many interrelated) that researchers have documented include the following.
Loss aversion
Tversky and Kahneman found that people are much more distressed by prospective losses than they are happy by equivalent gains.
They also found that faced with sure gain, most investors are risk-averse, but faced with sure loss, investors become risk-takers. For example, given a choice of: a) a sure gain of $240 or b) a 25 per cent chance of gaining $1,000, most people would opt for (a).
However, given a choice of: c) sure loss of $750 or d) an 80 per cent chance of $1,000 loss, most would go for a gamble and opt for (d).
In the first instance, choice (b) has a higher expected value, while in the second, choice (d) has higher expected loss.
You would have made irrational choices in both instances had you chosen (a) and (d).
And once a stock purchased falls in value, many people tend to put off realising the losses.
Their reasoning is that, as long as I don't sell the stock, I have not actually lost the money. Consequently, people are predisposed to hold their losers too long, and correspondingly sell their winners too early.
Related to this is the finding of Prof Statman that people tend to feel sorrow and grief after having made an error in judgment. Thus they avoid selling stocks that have gone down in order to avoid the pain and regret of having made a bad investment.
This also extends to corporate executives who, reluctant to terminate losing projects, continue to throw good money after bad.
Following the crowd
Some researchers theorise that investors follow the crowd and conventional wisdom to avoid feeling regret if their decisions prove to be incorrect. Many investors find it easier to console themselves on a loss from a popular stock since everyone else owned it and thought highly of it.
Buying a stock with a poor image is harder to rationalise if it goes down. Additionally, many believe that money managers favour well-known companies because they are less likely to be fired if they underperform.
This could explain why stocks with low price-to-book ratios or down-trodden stocks have consistently outperformed the 'glamour stocks', and by a wide margin at that.
Linear extrapolation
People typically give too much weight to recent experience and extrapolate recent trends that are at odds with long-run averages and statistical odds. They tend to become more optimistic when the market goes up and more pessimistic when the market goes down.
As an example, Prof Shiller found that at the peak of the Japanese market, 14 per cent of Japanese investors expected a crash, but after it did crash, 32 per cent expected a crash.
Many believe that when high percentages of participants become overly optimistic or pessimistic about the future, it is a signal that the opposite scenario will occur.
Over-confidence
People are also generally over-confident about their own abilities, and investors and analysts are particularly over-confident in areas where they have some knowledge.
However, greater confidence frequently doesn't correlate with greater success.
For instance, studies show that men consistently over-estimate their own abilities in many areas including athletic skills, leadership and ability to get along with others. Money managers, advisers, and investors are consistently over-confident in their ability to outperform the market. However, most fail to do so.
Analysts get the impression that they are in a good position to forecast the earnings of a company after having visited it.
A recent study by Chan, Karceski and Lakonishok, however, concluded that 'over long horizons, however, there is little forecastability in earnings, and analysts' estimates tend to be overly optimistic'.
All in all, they said, evidence suggests that the odds of an investor successfully uncovering the next stellar growth stock are about the same as correctly calling coin tosses.
On the whole, the absence of predictability in growth fits in with the economic intuition that competitive pressures ultimately work to correct excessively high or excessively low profitability growth, they concluded. More than just monetary gains Many researchers theorise that the tendency to gamble and assume unnecessary risks is a basic human trait. Entertainment and ego appear to be some of the motivations for people's tendency to speculate.
This would explain why many are willing to act on market gossip, which often turns out to be just noise.
People also tend to remember their successes, but not their failures, thereby unjustifiably increasing their confidence.
Conclusion
Arnold S Wood, president and CEO of Martingale Asset Management, noted that one risk in the investment world which is often overlooked is behavioural risk.
He contends that flaws in behaviour in the investment arena are prevalent, dependable and exploitable.
'Investors may recognise such flaws, but cognitive dissonance often prevents overcoming them. Once minds are set, disconfirming evidence is resisted with gusto,' he wrote.
'As the 1950s cartoon raccoon Pogo summarised: 'We have met the enemy, and he is us."
