|
© Copyright 2000 Rogers Media. The following article first appeared in the December 2000 edition of
BENEFITS CANADA magazine.
Behavioural Finance 101
Investors don't always act rationally. That's true of defined benefit plan trustees and defined
contribution plan members. Behavioural finance helps us understand our many faces.
By John Gilfoyle
When it comes to managing pension funds, behaviour is often a problem. Outbursts of joy when the markets
soar and temper tantrums when they head south aren't the issue at hand though. The behavourial issues
confronting plan sponsors are rooted in the reality that, as human beings, we are not completely rational
when making investment decisions, as numerous studies have shown.
An area of research called behavioural finance examines the behaviour of investors. Whether this analysis
has an impact on how we behave in the future remains to be seen, but it's helpful to be aware of the
theories. Pension committees or boards of trustees will inevitably be subject to behavioural influences
when they decide on an investment structure or to hire a manager, and defined contribution plan members can
also fall prey to behavioural errors.
The theories of financial economics are based on the assumption that agents make decisions based on
rational expectations. But behavioural finance challenges this notion by proposing that people are, more
often than not, irrational.
Here are some of the typical human behavioural tendencies that have been observed. A greater understanding
of them may help avoid their recurrence.
OVER-CONFIDENCE
Probably the most potent of human behavioural tendencies, over-confidence, may be associated with survival.
In investment management, or manager selection, it manifests itself in the belief that we can pick winners.
The theory states that experts tend to believe strongly in their abilities, and as a result they are most
often surprised by unfolding events.
With pension committees, over-confidence can be seen in the area of manager selection as committees often
state their confidence in a fund manager's ability to perform in the future as well as they have in the
past, adding 300 basis points a year to the index return. Unrealistic expectations and the possibility of
disappointment result.
COGNITIVE DISSONANCE
This condition is often closely related to over-confidence. An individual with an illusion of control may
only subscribe to evidence that confirms their view of the world, rejecting that which contradicts this
outlook. This may manifest itself as the mental anguish that results from being presented with evidence
that assumptions or beliefs are wrong. To counteract this, we hang on to our biases and often deliberately
ignore new information or develop convoluted arguments to refute the new while justifying the old.
An example of this state is the reluctance to admit a selection error and holding on to a bad investment or
poorly performing investment manager too long. The relationship with over-confidence is illustrated by the
hindsight bias exhibited when irrational markets or unpredictable events are blamed for a bad decision, but
not when decisions turn out positive. This tendency results in persistent over-confidence in
decision-making capabilities.
LOSS AVERSION
It's a human tendency to avoid tangible loss. Often the desire results in choosing not to participate in
potential gain. One obvious example took place in the market crash of October 1987, when investors were too
focused on further loss to take advantage of buying opportunities that emerged.
The natural desire to gain something while protecting against loss might also explain why retail investors
are prepared to pay quite high fees for life insurance segregated fund products. The guarantee of no loss
after 10 years seems to have some appeal even though there have been few 10-year periods in the past where
market returns have been negative.
Investors that are prone to loss aversion may shy away from assets with short-term volatility even though
long-term performance is superior.
FRAMING
This is the importance of context in decision-making. The way a question is framed can significantly affect
the answer given. Similarly, how information is presented impacts the response.
In a famous study of pension plan members in the U.S. conducted nearly a decade ago, the allocation between
stocks and bonds was impacted by how past returns were presented. If 30 one-year returns were exhibited,
the proportion in stocks was much less than if the average 30-year numbers were given.
The reference point for comparisons is also relevant. Consider, for example, a pension committee presented
with its U.S. equity manager's results over the last four years. A comparison with the Standard &
Poor's 500 index might show significant underperformance, while a comparison to a peer group might
illustrate above-average results. How the committee views the manager may be very different as a result of
the comparison.
MENTAL ACCOUNTING
Closely associated with framing and loss aversion, this state refers to how we think about specific
situations. For example, investors tend not to consider their total wealth when making a decision. Jumping
through hoops to maximize foreign content inside a registered retirement savings plan while maintaining
significant non-registered so-called Canadian investments illustrates the point. Holding low risk
investments inside a pension plan for retirement and higher risk assets outside of it demonstrates a
similar mindset.
Again, in fund manager selection a committee may have concerns with a manager whose risk level is
considered too high. Take the example of a manager with a volatility of 16 and a median manager with a
volatility of 12. Considered as part of an overall portfolio (which, after all, is what is most important
to a pension plan), the asset class may represent only 30% to 40% of the total (at most).
Add to that the ability to smooth asset values and amortize deficits in the actuarial valuation process and
you may come up with volatility that is much more acceptable, with limited impact on expected contribution
volatility.
OVER-REACTION
If there is an air crash, passenger traffic tends to drop with the focus on recent events. In investment
management, this behaviour may manifest itself in an over-reaction to a company profit warning. In manager
selection or de-selection, it may result in too great an emphasis being placed on recent performance--bad
or good.
In manager selection, committees have been known to focus too much on short-term performance. Assume that
long-term performance has been very good but the last six months have been poor. There will be a hesitancy
to make a decision in favour of that manager, which is an over-reaction.
HERDING AND REGRET
This is the safety in numbers syndrome. The power and influence of a peer group is an obvious example of
this mentality in investment management. There is also less of a fear of regret when everyone else is
getting on the bandwagon. This results in consensus decision-making or decision by committee.
The herding instinct has probably had an impact on forecasting. The quality of analysts' predictions of
future company earnings has deteriorated since the advent of services like I/B/E/S International Inc.
(formerly known as Institutional Brokers Estimate System) in the U.S.
In manager selection, herding results in a brand premium effect where well-known names are chosen even if,
overall, less well-known companies fit the investment criteria better. There is a much higher comfort level
associated with hiring a manager with $10 billion under management compared with one with a few hundred
million dollars.
Since regret is the feeling of sorrow experienced after making a decision that turns out to be wrong, one
of the ways to avoid it is not to make a decision. Paralysis of the process can result.
OVER-SIMPLIFICATION
Many decision-making situations we find ourselves in are complex, but we try to simplify matters, whenever
possible. We take shortcuts based on previous experiences, seeing patterns where there may be none. This is
reinforced by cognitive dissonance as we maintain these shortcuts for as long as we can.
One example of a shortcut is the one-number heuristic. For example, when an investor is given a choice of
six similar investment options he puts one-sixth of his money in each. This can be extended to committees
making manager selection decisions. When they cannot make a choice among managers, the inclination is to
split the money among several parties.
Some of these traits, in combination, can lead to continuing bullish forecasts for specific stocks or
markets; over-confidence in an ability to correctly predict the markets; cognitive dissonance in not
reacting properly to contradictory information, but trying to explain it away; as well as regret aversion
or avoiding a bearish outlook when everyone else is bullish, especially considering history has taught us
that markets go up more than they go down.
Loss aversion results in a skewed distribution of payoffs. Research has shown that people place twice as
much significance on a loss as they do on a gain. The normal measure of risk used by investors is standard
deviation, which assumes a symmetric, normal distribution.
If investors are irrational, then markets will be inefficient. Over-reaction will result in too great a
mark down for negative earnings surprises. Studies have shown that such stocks tend to outperform over the
long term.
Stocks with positive earnings surprises outperform over the short term, but they also tend to underperform
over longer periods. Efficient market theory suggests that information is available to all participants and
it will be quickly reflected in market prices.
However, this does not reflect the reality of over-reaction. Behavioural scientists believe that it takes
some time for information to be properly analyzed. It's worth noting that some quantitative active managers
refer to this theory to justify their systematic approach to finding and profiting from the anomalies
created by market inefficiency.
THE CURE
Plan members, fiduciaries and fund managers are human. We all bring biases and a tendency to adopt a
certain mindset to investing, and must look for ways to minimize any negative effects this has on
decision-making processes in the hiring, firing and organization of investment managers. Here are a few
remedies:
-
Guard against over-confident optimism. All organizations need optimism to state and then achieve goals,
but it must be tempered with reasonable expectations of our ability.
-
Recognize group dynamics and play to the strengths of the various participants while preventing the
group from being overly influenced by one strong individual. Sometimes, the quieter, perhaps less
experienced, members ask the best questions.
-
Use individual doubters to compensate for the natural optimism that group decisions generate.
-
Avoid over-confidence and cognitive dissonance by using unbiased information and analysis. If possible,
obtain an independent second opinion.
-
Use limited information to arrive at a decision. Remember, it's quality not quantity that counts.
-
Beware of leading questions--the ones that are framed to elicit a particular response.
-
Avoid primacy bias (the first choice offered is usually more popular than the others) by rotating
potential questions.
John Gilfoyle is senior investment consultant at Watson Wyatt Canada in Toronto.
john_gilfoyle@watsonwyatt.com.
|
News
News
