Behavioral Issues in Risk Management

Behavioral issues play an important role in risk management. Studies have shown that many New York taxi drivers, set themselves a daily income target. Once they reach their target, they close shop for the day. This tendency to work less on a busy day when easy money is there to be made, defies rational logic. Indeed, such anomalies drive the point home, that while taking decisions with financial implications, logic often takes the backseat.
The behaviours of people are strongly guided by perceptions. Two components of risk influence people’s perceptions – the fear factor and the control factor. When we are very much afraid of the outcome or feel less in control, we perceive the risk to be more. On the other hand, when we are not afraid of the outcome or feel more in control, we perceive the risk to be less.
Scholars, Daniel Kahneman (the 2002 Nobel Prize Winner) and the late Amos Tversky, pioneers of behavioral finance make an important point about how people perceive gains and losses. When looking at a potential gain, people tend to be risk averse and when they look at a potential loss, they are more risk loving. They gain less utility from winning $1000 than what they would forgo if they lose $1000. This asymmetry is especially relevant in the case of a financial loss or gain but can also apply to other situations.
How people perceive gains and losses also depends on the frame of reference. For example, managers who have incurred a major loss may be quite happy if the loss is less than what they had expected. Similarly, the choice of a strategy may depend on the way the possible outcomes are presented.
Cognitive bias in decision making is also an important point to be considered. People tend to give greater weight to information which is more easily available or recalled. The tendency to focus more attention on a particular fact or event, just because it is more visible or fresh in our minds is called availability heuristic. According to Werner De Bondt and Richard Thaler, a significant proportion of market volatility is explained by overreaction to recent news.

People often hold beliefs which are plainly at odds with the evidence, usually because they have been held and cherished for a long time. This is referred to as cognitive dissonance or in more common parlance, denial. Many people also tend to be influenced by outsiders’ suggestions. This may happen even when it is clearly known that the person making the suggestion is not necessarily well informed. Evidence indicates that people also tend to take bigger gambles to maintain the status quo.

People often have an exaggerated notion of their ability to control events. Consequently, they do not pay adequate attention to extreme possibilities. When people think they are in control of circumstances, when they are actually not, they underestimate the risks involved. The tendency on the part of people to think they have a greater influence on events than is actually the case is called magical thinking. Conditions that encourage illusion of control include stress, too much focus on results (without a periodic reflection of what is going on) and a series of positive outcomes.

Barberis, Huang and Santos point out another behavioral factor, the house money effect. When gamblers are ahead, they are more willing to take risks. Similarly, investors who have recently earned high returns will be less risk averse. Over confidence leads to reckless risk taking. This certainly seems to have happened in the case of Barings and Long Term Capital Management.

Another behavioural issue which has an adverse impact on risk management is misinterpretation of past events. Once something happens, people tend to think that they could easily have predicted it. This is called hindsight bias. When something happens and people condition themselves into believing they predicted it, when they actually did not, it is called memory bias.

The tendency to believe that past patterns will repeat themselves in the future is another pitfall in risk management. People are adept at finding patterns even when they do not exist. This phenomenon of treating events as representative of some class or pattern is called representativeness heuristic.

Thaler points out the role of mental accounting which refers to the way individuals and households keep track of financial transactions. People tend to evaluate risks separately than in an integrated fashion. If these risks were evaluated with a broader perspective, investors would be less risk averse. Bernatzl and Thaler have used this concept to explain why equity shares command such a high premium over bonds in the capital markets. Investors tend to focus more on the short-term volatility of shares than their long-term returns. Consequently, they demand a premium as compensation. Instead, if they concentrated on the long term returns offered by shares, they would not perceive them to be much riskier than comparable bonds. In the case of Metallgesellshaft, the German oil refiner, though the long term position was hedged, the top management became pretty much concerned about short term losses. Which is why, they decided to unwind their futures positions even though they were working fine on a long term basis.

Cichdti and Dubin (1994) studied customers who were prepared to pay 45 cents per month as insurance against having to incur a telephone wiring repair cost of $55 with only a .005 profitability. The expected loss in the event of a repair was only (.005) (55) or approximately 28 cents per month. Millions of customers in the US have been known to buy similar protection. If utility-maximising customers had rational expectations about the probability of needing repair, it is unlikely that they would buy the protection.

There are various other behavioral anomalies, a brief mention of some of which is in order here. Contamination effects allow irrelevant but proximate information to influence a decision. The affect heuristic allows preconceived value judgments to interfere with our assessment of costs and benefits. Over confidence in calibration leads us to underestimate the confidence intervals within which our estimates will be robust. Bystander apathy makes us abdicate individual responsibility when in a crowd. The problem of induction makes us generalize on the basis of insufficient information.

Risk management must take into account all these behavioral issues. Ultimately, risks are identified, measured and controlled by people. So human psychology cannot be separated from risk management. It is important to note that “normal” rather than “rational” behaviours are at work while taking risk.

One way to resolve the problem of individual biases is to ask employees to operate in teams. The advantage of a collective approach to beliefs about risk and the frame of reference is that individual biases can be minimised and team members can exercise a restraining influence on each other. Goldman Sachs developed the tradition of partners coming together to evaluate major risks and approve important decisions. This has no doubt contributed to the bank’s strong risk culture.