Mind your behaviour

Hedge fund managers are becoming interested in behavioural theory, anincreasingly influential branch of economics that describes the role of intuitivebias in investment decision-making. According to one finance professor,behaviouralism provides valuable lessons for hedge fund investors, and offersthe best account for the demise of LTCM.

Behavioural finance applies the discoveries of psychology to investment decision-making. It says investors are not rational – they have many irrational biases that influence intuitive decision-making. It has become an increasingly popular field, influencing both economic science and investment practice.

In 2002, psychologist Danny Kahneman was awarded the Nobel Prize for economics for integrating “insights from psychological research into economic science, especially concerning decision-making under uncertainty”. A recent investment decision-making conference led by Kahneman last month at Oxford University was attended by a variety of hedge fund investors, including statistical arbitrage and long/short equity managers.

The presence of hedge fund investors was no surprise to Hersh Shefrin. Shefrin is professor of finance at California’s Santa Clara University. “The behavioural psychology on which behavioural finance relies applies to us all, to investors in particular and more particularly to hedge fund investors,” he explains. Shefrin says this is because hedge fund managers are vulnerable to overconfidence more than most professionals. In his book Beyond Greed and Fear: Understanding Behavioral Finance and the Psychology of Investing, he highlights overconfidence in his behavioural analysis of the demise of Long-Term Capital Management (LTCM).

A lesson in overconfidence

For Shefrin, LTCM’s most important error was in the formulation of its original model. Ignoring the ‘sentiment risk’ arising from behavioural bias, the model assumed mis-pricing opportunities were small and quickly exploitable by smart trading. In fact the market sentiment responsible for mis-pricing increased the inaccuracies before decreasing them.

“The assumption that when you go in to exploit an anomaly market prices will quickly revert towards fundamentals so you can make your profit fast – that may not always hold,” explains Shefrin. “In fact you might get wiped out because it goes in the other direction.” And this is what happened. The liquidity constraints arising from redemptions forced LTCM to unwind positions, selling at prices that were not only inefficient, but also at market lows.

Hubris was a second indication of LTCM’s overconfidence. Reporting the LTCM story in his book When Genius Failed, Roger Lowenstein explains how managers ignored the warning signs. A paper published in 1997 in the Journal of Finance warned that liquidity constraints could force hedge funds such as LTCM into exactly this type of capital crunch. It was read by LTCM partner Robert Merton, who ignored the warning.

A third expression of LTCM’s overconfidence was a trap familiar to many hedge fund managers – strategy drift. LTCM’s large gains from 1994–96 were made via leveraged bets following the option pricing theory for which Myron Scholes and Robert Merton – both LTCM partners – had received the 1997 Nobel Prize. But by 1997 other investors had crowded into options markets, reducing the profit opportunities there. Instead of cutting back on the number of trades and staying in the market they knew, LTCM drifted into new markets such as foreign currency, where this expertise was lacking.

Overconfidence may help to explain the failure of LTCM. But how widely can the lesson be applied? Behavioural finance says that overconfidence affects all investment decision-making because it is a psychological fact.

The roots of overconfidence

Behavioural theory traces overconfidence to two underlying biases. People overestimate the likelihood of desirable outcomes and they exaggerate their ability to beat the market.

Consider the first case. Psychological research shows that we all believe situations are more likely to turn out well for us than for others. Underpinning this rose-tinted view of the future is that we tend to be too confident in our estimates. People asked to estimate with 100% confidence, say, the value of their favourite index a week from now, will be wrong 10–15% of the time. Left to our own devices we set our confidence intervals too wide. In investment, this implies that intuitive financial forecasts will be too bold.

This danger is compounded by the fact that people tend to exaggerate their ability. For example, 80% of people reckon they are above average at driving – a statistical anomaly. And overestimating your ability makes you a worse investor. Paul Willman of Oxford’s Said Business School measured ‘illusion of control’ – a measure of overconfidence – among equities traders at four leading investment banks in London in 2000 and 2003. Those with the greatest illusion of control performed worse: they contributed least to the desks’ profits, were rated as weakest by their managers and were bottom of the remuneration pile.

So overconfidence, if unchecked, can weaken investment decision-making. And behavioural finance has further warnings for investors concerning risk.

Sensing risk

Consider when people avoid risks and when they seek them. In general, psychologists have found that people are risk-averse. Offered a bet on the toss of a coin most people need to win at least twice what they will lose to agree to the bet. Taking a loss is more aversive than taking a gain is pleasurable. “This introduces a non-financial consideration (regret avoidance) into the trade,” explains Kahneman, “and implies that investors will be too timid.”

And there is evidence of regret avoidance in investment. “People will pay to avoid regret in the case of dollar-cost averaging,” explains Kahneman, even though putting on a position slowly exposes you to moves in the market.

Importantly, this risk aversion turns to risk seeking when there is a chance to redeem a loss. Experimentally, subjects will choose odds of 90% to lose £200 over a definite loss of £180. Even though you are more likely to end up poorer, the possibility of avoiding regret makes the risk attractive.

But it’s not all bad news. Behavioural finance says you can combat these biases by changing the way you think about the risk. In the case of loss aversion (the toss of a coin), if subjects are asked to think about the gamble in terms of their entire net worth they will take the bet. But if they are offered a series of bets they will settle for more probabilistic odds. In the investment context, expanding frames of reference means seeing losses over the total portfolio and longer period of time – approaching the trade this way will combat the bias.

Richard Thaler, considered by many to be the father of behavioural finance, lists four key cognitive biases that affect investors. The first is representativeness: “people evaluate the probability of an uncertain future event by the degree to which it is similar to recently observed events”. This leads to people over-reacting to new information – a recent profit warning, for example.

Second, recent information tends to influence investors disproportionately. Especially in the case of rare events “people tend to overestimate the probability of such an event occurring in the future if they have recently observed such an event”. Third, overconfidence causes people to underreact to new information. Finally, investors anchor quantitative estimates to previous values of an item such as that proffered by a prominent analyst.

Thaler attributes certain types of mis-pricing to these irrational heuristics. He claims overreaction caused by representativeness and salience heuristics artificially deflates stock prices, providing opportunities for value and contrarian strategies. By contrast, Thaler says under-reaction is probably the primary source of the alpha associated with earnings surprises and short-term momentum strategies. This happens when analysts who anchor themselves to their most recent estimate under-react to subsequent market news.

Thaler has put these ideas into practice. He is one half of California-based Fuller & Thaler Asset Management, which manages total assets of $2.4 billion, as well as advising JP Morgan Fleming on its popular Behavioural Growth and Behavioural Value Funds. Fuller & Thaler’s top-performing fund is the $700 million Small-Cap Value fund. This targets value opportunities created by anchoring. The fund has returned an impressive 19% net of fees since its inception in 1996.

Value of behaviouralism

Notwithstanding Thaler’s returns, critics of behaviouralism claim that biases per se do not provide predictions about the market. Josef Lakonishok is professor of finance at the University of Illinois and one third of LSV Asset Management, a Chicago-based value equity manager with $25 billion under management. “Behavioural ideas are very interesting, but they don’t give you very specific predictions,” he says.

Lakonishok adds that many of the biases championed by behavioural theory have been recognised by investors for years. The risk of market sentiment, to which Shefrin attributes LTCM’s demise, was noted by John Maynard Keynes in 1931 when he wrote that “the market can remain irrational longer than you can remain solvent”. Stuart Mitchell, manager of JO Hambro’s long/short equity fund, and present at the Oxford behavioural conference, agrees: “Much of behavioural finance has been part of the folklore wisdom of great traders like Warren Buffet. They just don’t describe it in an academic way.”

How can appreciating intuitive biases help hedge fund investors? The key for Shefrin is to focus on your own trading behaviour, rather than trying to second-guess the biases of others through behavioural arbitrage. “Understanding behavioural biases allows you to make predictions only if you understand how an investor builds his market prediction model.” Since other investors will keep this information private, you are best to start with your biases.

The most effective way to attack them, advises Nobel laureate Kahneman, is to keep a log. “By recording details of your decision-making you can identify biases and correct them.” Two very successful hedge fund investors apparently do just that. “Warren Buffet and Charlie Munger go through an avoidance list of 20 biases on every major trade,” says Sherman Roberts, academic director of the Oxford Programme.

Other hedge fund managers have applied the lessons of behavioural finance to their own investing. Alex Pritchard moved from Morgan Stanley to run the long/short small cap fund at JO Hambro Investment Management. “The biggest single lesson I have taken from Kahneman’s work is that if I have a hunch, I often bet against it,” he says.

For Hambro’s Mitchell, the key point is the role our personalities play in our investment decisions. “Most hedge fund traders don’t consider the influence of their personality in this way, but unless you do it is very hard to make those crucial rapid switches between long and short positions that are crucial to success.” This may be the key message of behavioural finance: intuitive biases are part of our psychological make-up; by familiarising ourselves with them we can harness them to support rather than threaten our decision-making.

Risk