Considerations in explaining investor behaviour. Occasional paper no.2 Author Michael Wydeveld [1999] NZSecCom 3 (1 September 1999)

The writer is an economist with the New Zealand Securities Commission. The views presented herein are the writer's own and do not necessarily reflect those of the Securities Commission or its Members. The writer would like to acknowledge Henk Berkman, Bryce Wilkinson, Grant Hannis and Jeff Huther for their useful comments on various drafts of the paper.

	Home \| Databases \| WorldLII \| Search \| Feedback New Zealand Securities Commission

The study of investor behaviour has attracted researchers with a variety of backgrounds.1 Economists, sociologists and psychologists have all attempted to explain investor behaviour in various ways. Economists' enquiries into investor behaviour have focused largely on the "rationality" or "irrationality" of investor decision-making processes. Explanations such as "greed" and "fear" are reduced to rational terms. Sociologists explain investor behaviour by focusing on investors' social environments. They suggest investors may be trying to preserve or enhance their stature within a group or society in general. Psychologists explain investor behaviour by focusing on individual characteristics. Investors may be labelled variously as "sophisticated", "gullible", "proud", etc. There is a lot of overlap between the disciplines.

Set against this background the paper has two purposes. First, it provides a framework within which to consider the various explanations of investor behaviour. Secondly, the paper provides background for the targeting of educational and regulatory resources, where the aim is to promote a reasonable and informed approach by investors to their investment behaviour.

It seems obvious that individuals will be able to make their most effective decisions when acting on complete information, so that the exact implications for their actions are known.3 But in complex markets conditions are uncertain. Not only are the implications of actions uncertain, i.e., there are numerous potential outcomes (returns), but the probabilities (risks) associated with the various outcomes are also often difficult to measure. In addition, the actions of others may make it difficult for individuals to make objective decisions.

Presented below is a framework for explaining investor behaviour. It introduces the concepts of "expected returns", "risk preferences", and "personal confidence", and presents the individual factors influencing investor behaviour under these headings. The framework aims to provide order and transparency in considering the explanations of investor behaviour.

It is the contention of this paper that the concepts noted above will underlie each investment decision to varying degrees. However, it does not attempt to explore which are the dominant factors in explaining investor behaviour. It seems likely that, dependent on the circumstances, the different factors will be more or less dominant at various times. Likewise, it is recognised that, while a sole factor may be dominant at one point in time, at other times there will be a number of pressing influences on investor behaviour.

Absent from the framework are direct discussions of the budget constraint facing investors and investors' time preferences. Obviously, as individuals become wealthier, they could potentially invest greater sums, and more frequently. It is assumed that this behaviour holds, but is tempered by the other factors highlighted herein.4 Likewise, it is assumed that investors are willing to roll over (or stick with) an investment, until such time as they come to spend their money, or one of the other factors highlighted in this framework changes.5

The returns an investor expects to receive will be an important determinant of their behaviour. Other things being equal, investors will be motivated to choose that investment with the highest expected returns. "Greed", in this sense, is seen as quite rational.

A best estimate of the potential of the investment and its risks has to be made in determining expected returns. Mathematically speaking expected returns equal the sum of all possible returns adjusted for the level of probability. It should be noted that "possible returns" include potential losses. If appropriate, expected returns can be discounted (to account for the time value of money) to come up with their present values. This is likely to be important for investments of differing maturities.

However, calculating expected returns can be a difficult task. First, the certainty regarding an investment's potential can vary significantly. The potential returns for a bond, on the one hand, will usually be relatively easy to calculate. They are generally equal to the principal invested plus the yield. On the other hand, the potential of a share in a company may be boundless, which may make its potential returns difficult to quantify. The potential future cash flows of the company provide a theoretical starting point for such quantification.

Second, the risks associated with an investment can vary significantly. Some risks can be reasonably quantified, while others are variable and can themselves be affected by a variety of factors, including the changing circumstances of an investment and moral hazards.6 Risks also emerge in various contexts including the characteristics of the product, the product provider (e.g., credit risks, counter-party risks) and the market in general.

Finally, how investors process the information required to determine an investment's expected return can be problematic. Investors become susceptible to poor judgement as the uncertainty of decision-making increases. A number of studies suggest that investors often find patterns in what is statistically random data.7 Other studies have suggested investors disregard information (and risks) that does not support their view, while placing too much weighting on information that does support it.8 Individuals may also invest without complete information. Anecdotal evidence indicates that some investors make up their mind about investing in an initial public offering even before the offer documents (investment statement and prospectus) are available.9

Investors can estimate expected returns from a variety of sources. In New Zealand an investment statement and prospectus serve as the primary offer document for securities. These documents should provide the basis from which a reasonable initial estimate of expected returns can be made. Ongoing disclosures in the form of annual reports provide the basis upon which a recalculation of expected returns can be made. The issue of calculating expected returns is not addressed further in this paper. The variety of possible investment products makes this onerous.

Numerous gambling studies have examined investors' reactions to changes in expected returns under a limited number of possible returns and fixed risks.10 The simplest of these studies examine gambler behaviour in a fixed risk gamble with a certain number of known outcomes (e.g. a gamble where the potential for a $1 win or lose is .5 in either case). In these studies gamblers distinguish possible returns, even if they are uncertain which one will eventuate. These studies conclude that any change in the fixed risks (or returns) will lead individuals to display the same substitution effect as do price changes under conditions of certainty.

In more general terms it can be said that the substitution effect is caused by a change in the relative expected returns of investments. For example, changing interest rates affects the returns to bonds directly, and through a substitution effect the returns to equities are also affected. Predictably equity markets will rise as interest rates fall (and the returns on bonds fall) or visa versa.

Investors are not solely motivated by the maximisation of expected returns in dollar terms. There may well be intrinsic rewards associated with an investment. For example, an investment may provide an investor with the opportunity to fulfil a long held desire to be involved in the making of a movie, or to own a piece of a Pacific island, etc. In the interest of brevity, the intrinsic value of the investments themselves is not dealt with again in this paper. In any case the intrinsic value of investments can be dealt with in a broader concept of expected returns.

Whether individuals accept a particular investment and how much they are willing to pay for that investment will be influenced by their risk preferences. The paper proceeds by considering those factors that determine investors' risk preferences and how these preferences may affect their investment decisions.

Investors who would receive great utility from an increase in wealth will tend to be risk seeking.11 In this regard they may even choose an investment with expected returns that are less than their initial outlay, in the hope that the investment will actually exceed expectations.

At a low level of wealth an investor is more likely to receive much utility from an increase in wealth. Youth is seen as a period of low wealth, middle age as a period of high (and rising) wealth, and retirement as a period of high (but decreasing) wealth. These observations suggest that younger people are probably the risk seekers of an economy, while an ageing population is likely to promote an economy of slower and more stable wealth accumulation.12

Psychological studies suggest that the utility investors attach to possible changes in wealth is more innate. Studies have shown investors became more distressed by prospective losses than they were made happy by equivalent gains, regardless of their level of wealth.13 These studies suggest that investors have an inherent bias toward risk aversion.

Losses that can be quickly recovered through new income may temper the significance an investor places on a possible decrease in wealth. The converse of this logic may explain why investors who have retired from paid work often appear to be strongly risk averse.

Theory suggests that, other things being equal, given a choice of two investments with equivalent expected returns, an investor will chose that investment with the lower degree of risk (lower probability of an adverse outcome). In this regard two investments with similar expected returns, but different degrees of risk, can demand different prices.

The possible investment returns as distinct from the expected returns, can affect individual preferences towards degrees of risk. Studies have shown that investors faced with two options, accepting a certain gain, or accepting a gamble with a marginally better than equivalent expected return, act in a risk averse manner. In this regard investors appear to prefer to "take the money and run". But investors faced with a sure loss, or the chance to recover their money, while risking greater losses, are seen to act in a more risk-seeking manner.14 This suggests investors do not prefer to "cut their losses".

Thaler et al (1997) argue that losses are often given more importance than the possibility of their occurring would suggest. They explain a failure by investors to accept equivalent bets (with similar expected returns and degrees of risk) when considered separately as "myopic loss aversion".15 In more general market observations, Statman (1987) argues portfolios have less funds devoted to them when too much attention is given to the specific investments (and the risks) that make up the portfolio. He believes that, as a result of this, investors are often not as diversified as they should be.

At the other extreme there may be investors who are compulsive risk seekers. They appear to be motivated to pursue investments of high possible returns, but low expected returns, accepting almost any degree of risk. "Greed", in this sense, is not seen as rational. The infrequency of the conditions that would realise the high possible returns leads them to behave compulsively. There are numerous support groups for gambling addicts. It would be generally accepted that these forms of behaviour when apparent do not maximise an investor's utility in the long term.

It is difficult to make certain statements on the desirability of risk preferences with so many conflicting arguments. Siegel & Thaler (1997) investigate the equity premium puzzle. In their introduction they calculate that if US$1,000 had been invested in US Treasury bills in 1925 the cumulated returns would be US$12,720 at the end of 1995. If the money had been invested in a portfolio of equities the accumulated returns would be US$842,000. They question whether such a difference in returns is really justified as a compensation for the higher risk (volatility) of equities.16

Theory suggests that risks can be generally reduced by diversification, because the returns of some investments are inversely related to those of other investments for certain risks. Yet evidence suggests that investors are not highly diversified. In New Zealand, for example, there is significant bias towards real estate and bank term deposits in the weightings of investors' portfolios. In many countries, despite an increasingly global economy, most investors still overwhelming hold assets in their home economy. Transaction and monitoring costs may suggest domestic investing is cheaper, but technology and specialist financial intermediaries are increasingly eroding these costs. It seems likely that other factors, such as risk aversion, must also explain why home investments persistently dominate portfolios.17

For institutional investors it is appropriate for their risk preferences to reflect those of their beneficiaries. The consumption of pension funds, for example, will occur for most beneficiaries many years down the track, when they retire. Risk preferences for these funds should reflect those of its members or outcomes could fall short of expectations and utility maximisation. The implications for institutional investors are that they need to be well informed of their beneficiaries' preferences and to have balanced incentives (those that align their own interests with those of their beneficiaries).18

Investors risk preferences may change over time, as the factors noted above may become more or less significant. For example, a gambling addict may be educated on the implications of their behaviour. Investors may also become overly optimistic regarding future prospects following a successful investment experience, or conversely, if they have lost money, they may become more myopic in their aversion to future potential loses.19

In addition, time preferences can affect or change investors' risk preferences. Studies have shown that uncertainty about expected returns for a diversified equity portfolio over the long run is much smaller than in the short run.20 This suggests that investing for the long term is a rational response by investors to the tendency for the expected returns to regress to a long-term average (whatever that may be). This reasoning has filtered through to the general consciousness of investors. In this regard some investors in times of a share market down turn can be heard to say that they believe "the market will bounce back" and that they are "in for the long haul". These investors may discount the significance of short-term volatility, thus appearing to be more risk seeking in the short term.

Some investors may feel confidence in their own skills and judgement in investing and may desire to distinguish themselves from others or the market (i.e., beating the market). Other investors may lack the confidence and either follow the group or rationalise some course of action that would still generate group approval.

In markets with a diversity of participants, such as the share market, the prices of securities are likely to reflect the expected returns and risk preferences of individual investors. But this does not mean that an investor's decision will always be made objectively.

Herd behaviour has frequently been highlighted in financial markets and suggests investors can be influenced by the actions of others. Investors may trade on noise (caused by the actions of others) as if it were information. Statman (1987) argues this induces investors to trade more frequently than would otherwise be the case.

In general market observations there is evidence of the anchoring effect of past prices.21 Anchoring may explain why stock prices, market indices, and financial ratios become sticky around certain levels. Similarly, overseas prices have been found at times to act as an anchor for domestic ones.22 Confidence indexes also exhibit a form of anchoring effect. Statman (1987) found evidence that market performance led adviser sentiment in explaining the way confidence indexes adjusted.

In certain circumstances news may weaken the current anchor, which may explain why some studies have found that markets over-react or under-react on particular occasions (e.g., significant price changes around profit or dividend announcements).23 In this regard the volatility of initial public offerings may be partly explained by a lack of an anchor.24

Pride has often been highlighted as an explanation of investor behaviour. Studies have found people are generally highly confident in their intuitive judgement, which leaves them susceptible to illusions and poor judgement. There is evidence that investors often think their choices are based on superior information (or superior information processing methods) without even knowing what information is available to the counter-party to their trade.25

Studies suggest investors are more likely to hold securities they have touched or personally selected.26 The frequency with which investors trade can also be explained by pride factors. In this regard the investor takes pride from being seen as playing the share market.

The possible fear of regret is a factor driving some investors' behaviour. This factor is most likely to dominate where investors are not confident of their information, or ability to process it, and where the possible detriment to their pride of a poor decision is considered too much to bear. This helps explain why studies have found that some investors faced with choosing between a popular or unpopular security may choose a popular security, because it would be easier to explain losses if everyone else bought the same security.27 Similarly, studies have shown that investors may use investment advisers as scapegoats thereby reducing their responsibility for poor investment decisions.28 Institutional investors may also display this behaviour when they seek to preserve their reputations.

Studies have shown that investors faced with the decision to sell an investment are affected by whether the security was bought for more or less than the current price. Goetzmann & Peles (1993), observing that money flows in more rapidly to mutual funds that have performed extremely well than flows out from mutual funds that have performed extremely poorly, suggest that investors, in losing funds, are unwilling to confront the evidence that they made a bad investment by selling their investments. In some instances investors may even be prompted to take risky bets to try and break even, and so save face. Reporting of losses often appears to be the source of much embarrassment.

There is no doubt that many investors get pleasure from the very act of investing. The fact that people enjoy a "flutter" is well established.

There is an increasing policy push internationally to encourage individuals to save for their retirement. This may be raising the intrinsic value of the act of investing.29 This push appears to be prompted by the fiscal burden a growing aged class is expected to impose on an economy. In this regard the value of investing is related more to its defensive benefits (i.e., allowing an individual to live well in retirement) than the prospects of increasing wealth per se.30

The considerations of investor behaviour raise broad educational and policy issues. The section on "expected returns" suggests that investors would be well advised to reconsider whether their expectations about an investment are realistic before investing. The expected returns of an investment can be uncertain. It is important to research an investment's potential and read the documents accompanying an offer or supplied annually regarding an investment. These messages should be sent consistently and regularly to investors.

There are ongoing policy questions of how much, and of what quality, initial and on going disclosure should be required from issuers. The appropriate authority will have to recognise the diversity of investors in their particular policy formulations.

The section on "investor preferences" suggests that investors should be comfortable with, and behave in a reasonable manner in accepting, the risks they bear regarding an investment. Investors have various risk preferences. Investors can display extremes of behaviour, which manifest at the one extreme as an overwhelming fear of risks, or at the other as a risk-loving (gambling) mentality. Neither extreme is beneficial in the long run. These messages should be sent consistently and regularly to investors.

There are policy questions on the degree to which investors should be educated regarding the desirability of their risk preferences.31 These questions involve careful consideration of whether investors will learn from their mistakes, and whether the appropriate authority can make a positive and effective contribution. For example, the experiences of a group of investors may provide useful educational material for others. There may be a role for an authority in facilitating such an educational experience.

The section on "personal confidence" suggests investors should be aware that they can be influenced by the actions of others and their own pride. Investors have varying abilities and confidence in investing. Those lacking confidence may choose to seek professional advice. In seeking professional advice, investors should go to someone of demonstrable competence and be wary of approaches by supposed experts. For those seeking to exercise their own judgement, they should be careful that they are not responding to misleading signals. There are intangible benefits to investing. However, these benefits should be measured against the potential costs should the investment turn sour. These messages should be sent consistently and regularly to investors.

More investors are likely to be attracted to markets they are less familiar with, given that the intrinsic value of investing may be rising. The nature of the role of advisers and other financial intermediaries in guiding investors as they enter more complex markets may justify analysis in the context of any examination of market influences.

Bernstein, P.,	Against the Gods: The Remarkable Story of Risk, John Wiley & Sons Inc., 1996.
DeSerpa, A.C.,	Microeconomic Theory: Issues and Applications, Allyn and Bacon Inc., London, 2nd edition, 1988.
Gneezy, U., & Potters, J.,	"An Experiment on Risk Taking and Evaluation Periods", Quarterly Journal of Economics, May 1997.
Goetzmann W. & Peles N.,	Cognitive Dissonance and Mutual Fund Investors, Yale School of Management, 1993.
Kahneman, D, & Riepe, M.,	"Aspects of Investor Psychology", Journal of Portfolio Management, Summer 1998.
Lee, C., et al	"Investor Sentiment and the Closed-End Fund Puzzle", Journal of Finance, March 1991.
Nicholson, W.,	Microeconomic Theory: Basic Principles and Extensions, Dryden Press, London, 4th edition, 1989.
OECD	"International Financial Market Implications of Ageing Populations", Financial Market Trends, No. 71, 1998.
Ritter, J.,	"The Long-Run Performance of Initial Public Offerings", Journal of Finance, March 1991.
Sappington, D.,	"Incentives in Principal-Agent Relationships", Journal of Economic Perspectives, Vol 5, no 2, Spring 1991.
Shafir, E., et al	"Money Illusion", Quarterly Journal of Economics, May 1997.
Shiller, R.,	"Human Behaviour and the Efficiency of the Financial System", Prepared for the Handbook of Macroeconomics, September 1997.
Siegel, J., & Thaler, R.,	"The Equity Premium Puzzle", Journal of Economic Perspectives, Vol. 11, No. 1, Winter 1997.
Statman, M.,	"Investor Psychology and Market Inefficiency", Equity Markets and Valuation Methods, 1987.
Swallow, S., & Fox, M.,	"Investor psychology in New Zealand", New Zealand Journal of Psychology, June 1996.
Thaler, R., et al	"The Effect of Myopia and Loss Aversion on Risk Taking: An Experimental Test", Quarterly Journal of Economics, May 1997.
Wydeveld, M.,	Survey of Investors' Opinions of Offer Documents, Report to the New Zealand Securities Commission, Wellington, January 1999.

"Behaviour" for the purposes of this paper is taken to mean the investment choices individuals make and the frequency with which they make them.

As consumption occurs over time, there will be changes in both the relative prices of products and the budget constraint facing the individual in their ongoing choices. Changes in relative prices lead individuals to substitute products. As individuals become wealthier or poorer they can afford more or less products. Their tastes may also change, maybe developing a taste for the finer or more exotic things as their wealth rises. All these changes inevitably give rise to changes in the relative mix of products purchased. Individuals are seen as responding immediately to these changes.

However, it is possible to argue that investors are in fact being rational in their decision making. If they expect other investors to act the same way as they do, then even small pieces of new information can have a significant impact on the way their investments are valued in the short term, even if it doesn't affect performance in the long term. Personal confidence will determine the degree to which investors are influenced by the reactions of others and whether this represents good or poor judgement.

Another assumption economists often make is that of non-satiation, i.e., that individuals will always prefer more to less. At first glance this assumption appears to be incompatible with the former. However, this will not be the case as long as the marginal utility of wealth does not become negative.

The above example also highlights the issue of "regression towards the mean". In multiple bets with the same odds the mean is less likely to vary from the expected return of each independent bet. However, because individuals may not have the wealth to commit to multiple plays, Samuelson argued that each bet must be considered independently to determine whether behaviour is consistent with utility maximisation. He concluded that his colleague's actions were less than rational.