Trading Decisions of Individual Investors: Evidence of Psychological Biases - Literature review Example

Trading decisions of individual investors: Evidence of psychological biases Behavioural Finance Behavioural finance seeks to investigate the seemingly irrational behaviour of individual investors in the markets that runs contrary to the theory of an efficient market. Individual investors are supposed to trade only when incremental cost is exceeded by the incremental benefit to be realized. This traditional view of the rational investor has often failed to meet the empirical evidence concerning investor actions in the market, such as stock market bubbles. (Ritter, 2003, p. 429). This paper discusses two studies in this field: “Are Investors Reluctant to Realize Their Losses?”, Odean (1998), and “Trading is Hazardous to Your Wealth? The Common Stock Investment Performance of Individual Investors”, Barber and Odean (2000). The Effect of Overconfidence Barber and Odean (2000) studied data of stock market transactions undertaken by 78,000 households, from January 1991 to December 1996. Under the overconfidence model, investors who are overconfident about executing a profitable trade will trade more frequently in the market, and because much of their market action will be based on emotion (overconfidence) rather than deliberate and pragmatic study, their trades will be of lower expected utility. The resulting net return of households with high turnover will be inferior to those less frequently traded accounts. By comparison, the rational expectation framework of Grossman and Stiglitz posit that when investors trade, it is because they perceive that the marginal benefit they will realize is greater than the marginal cost they will incur. Since such investors trade only when such opportunity presents itself, which probably will be as often as not, then the rational investor transacts less frequently, incurring a lower aggregate transaction cost. The study discovered that households that have lower turnover (and thus traded less frequently) had larger accounts that those households that had higher turnover. This may be explained by the fact that investors who trade less frequently are longer-term investors whose objective in entering the market is for capital appreciation rather than the “quick buck”. They will tend to select stocks of “blue chip”, investor, quality, and to maintain that position for years. The Disposition Effect The earlier investigation done by Odean (1998) sought to discover whether or not individual investors tended to maintain a losing position too long and, conversely, close out on their gaining stocks too soon. This has direct bearing on the Prospect Theory by Kahneman and Tversky (1979), originally conceived as a study of behaviour of lottery participants, but as applied now to investments. This theory states that a person would tend towards greater risk aversion as his stock position increases in value; however, the same investor is capable of tolerating greater risk the more his position loses in attrition. Graphically this is seen as an S-curve that shows greater convexity in the "gains" quadrant and greater concavity in the "losses" quadrant. As an example, consider an investor who had two stocks. Based on their purchase prices, one stock was gaining and the other was losing. Assuming the investor has no additional information or motivation to keep either stock but was in need of liquidity, he would be more predisposed to sell the stock that was appreciating rather than that which was falling. This is, notwithstanding the fact that the price of the stock which was disposed would more often than not continue to rise, and that of the losing stock which was maintained would continue to remain in negative territory. The rationale for this behaviour is that the investor is reluctant to realize the losses on a stock he had chosen (maybe reluctant to admit he had made an error in choosing the stock when he did). Despite information to the contrary and no sign of imminent recovery in the charts (and no reason to expect one), the investor chooses to hold on in the hope of one, then this is irrational behaviour and unexplainable by the conventional theory of the rational investor. This particular study was based on observation of 10,000 accounts randomly chosen from all accounts active in 1987 in a nationwide discount brokerage. Among other findings, the study determined that stocks that gained and were sold averaged returns of 3.4 percent higher than the average returns for stocks that lost and were maintained. Comparison of research design of both studies Essentially, the studies pertain to different aspects of individual investors trading habits. The study on Disposition Effect has to do with the gains realized for specific positions, and the holdover of acquisitions to which paper loss is attributed. The nature of this study is different from that of Overconfidence, which takes a cumulative view of the entire equity portfolio of the household-investor (as is housed in the subject brokerage) and his philosophy in managing this account. It does not give attention to particular positions or transactions. In any case, a cursory review of the data gathering, treatment and analysis is presented here. (An indepth comparison is not called for given the disparity in the bases of the research.) As to the subjects studied The study on the Disposition Effect covered 6,380 individual accounts involving 97,483 transactions. A one-to-one correspondence is traced for every sale to its purchase, in order to determine the holding period and the gain on single transaction. Positions that were not sold were computed on the basis of their paper gains and losses. On the other hand, the overconfidence theory did not study individual accounts. Instead, it based its findings on households, stratified according to categories of classification employed by the brokerage (general, affluent, and active trader). A total of 66,467 households were made the subject of study. Thus, while the study on the disposition effect involved fewer investor accounts, it however entailed the tallying of more data incidences since the individual transactions were collated and analyzed. As to the data gathered and analyzed As already mentioned, the data on the Disposition Effect involved the analysis of individual transactions; the study measured the time from purchase to disposal of each position, as well as the difference between purchase price and selling price. The delimitation of the study covered transactions entered into between January 1998 and December 1993. On the other hand, the study on the Overconfidence theory did not need to take into account the buying and selling of particular positions, just the frequency. Rather, data was gathered on the frequency of trades (turnover), the size of the accounts, and transaction costs of accounts which were in existence for the period January 1991 to December 1996. Transactions, rather than taken individually as in the Disposition Study, were in this case assumed to occur at end of month, ignoring intra-month trading activity. As to the analysis methodology and treatment of data Concerning the Disposition Effect, ratios were devised and used by the researcher. The Proportion of Gaines Realized (PGR) ratio was devised to measure the proportion of realized gains to the sum of realized gains and paper gains; and Proportion of losses realized (PLR), to designate realized losses as a proportion of the total loss (sum of realized and paper losses) at any one time. Hypotheses testing was conducted with the use of the t-statistic of inferencing (on whether the proportions, or their differences from each other, vary significantly among groups). The method is applied to groups categorized according to period and trading activity. On the other hand, the study on the Overconfidence Theory made use of four methods to determine portfolio performance adjusted for risk: (1) own-benchmark abnormal return, which involves taking the difference of portfolio returns and the returns the portfolio would have realized had it not been transacted (this yields a 72-month time-series statistically verified for significance using the t-test); (2) mean monthly market-adjusted abnormal return (subtracting return of value-weighted index from individual investors return); (3) the capital asset pricing model to calculate Jensens alpha (the CAPM intercept); (4) the intercept test using French and Famas three-factor model.. Furthermore, the study calculated bid-ask spreads for purchases and sales, which was important for determining a reasonable estimate for transaction costs. Computation was based on the days closing prices as gathered from secondary data. Results were added to commission to gain an estimate of total transaction costs Analysis of the results of studies on overconfidence and disposition effect The preceding discussions deal with two phenomena that support the theory that investors tend to act irrationally in the course of their stock investing, and these irrational decisions are reflected in discernible trading patterns. The first pattern is described by the greater volumes and more frequent trade turnovers attributable to household investors than to pension funds and mutual funds. The second pattern is characterized by the propensity of individual investors to sell their gainers sooner than the stocks in which they are losing. By methods that confirm the researchers’ primary theory and other methods that exclude the alternative possibilities, the researchers have proven that the first pattern is due to a tendency for household investors to act on the basis of overconfidence, while the second pattern was explained by the hope investors nurse for the eventual reversion of losing trends, and their reluctance to admit that they had made the wrong decision in creating a position in that particular stock. The significance of the two theories taken jointly lies in the fact that, with the proof provided by these studies, empirical evidence moves closer to the validation of the tenets of behavioural finance – that investors will take market action on the basis of psychological motivation rather than in response to rational decision-making. Conventional theories have operated on the basis laid down by the efficient market theory, that markets efficiently discount all information there is to know about all assets, and that investors, moving on the basis of this information, inputs its value into the price of the asset. Thus, the rise and fall of prices should, theoretically, be the result of studied valuation of the assets’ intrinsic worth; that is, the result of a rational decision. The two theories are different in a way that complement rather than contradict each other. The first, overconfidence, is a predilection that tends to goad investors towards action – in fact, over-action – in transacting more often for no compelling reason. On the other hand, the disposition theory observes inaction in instances when the investor ought to. The irrational protracted holding of losing stocks and preference for liquidating stock positions that gain is based on the refusal to acknowledge an error in discernment. The first is precipitated by perception of a chance to gain, the second is denial of realizing a loss. More often the first is brought about by the positive experience of the investor in having made quick money in the past; believing this to be because of his investing acumen, and not just the natural consequence of a rising market, he is confident that he will be able to replicate the experience. This encourages him to try again and again, turning over trades more often. Usually, however, the second phenomenon is the result of a sideways moving or consolidating market, where selective stocks rise while a good number languish. In such a market, investors with positions in the lacklustre counters, rather than cut short their losses, insist on denial that the gambit is lost, and rely on hope that the expected gain is but postponed. Hope, denial, greed and fear – these are the seeming motivations that introduce biases in market behaviour that has been investigated by these studies. These concepts are wholly inconsistent with the efficient market theory and the rational expectations framework. An important segment of the financial market is dominated by individual investors motivated by personal perceptions and biases. An understanding of what ultimately moves the individual investor to buy and sell may be the key to understanding the creation of market value in a trading environment. It is for this reason that these two studies are commendable for their credible investigation and documentation of the significance of investor psychology. This acknowledgement will open avenues for a greater understanding of our financial markets. REFERENCES Barber, B M & Odean, R 2000, ‘Trading is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors’ The Journal of Finance, vol 55 no. 2 pp. 773-806 Bodie, Z, Kane, A and Marcus. A J 1996, Investments, 3rd Edition, Richard D. Irwin, Chicago. Brigham, E & Gapenski L 2004, Intermediate Financial Management, Dryden Press, Fort Worth, TX Damodaran, A 1994, Damodaran on Valuation, John Wiley and Sons, Canada DeBondt, W & Thaler R 1985, ‘Does the stock market overreact?’, Journal of Finance, vol. 40, pp. 793-807, as cited in Odean T 1998, ‘Are Investors Reluctant to Realize Their Losses?’ The Journal of Finance, p. 1790p. 1790 DeBondt, W & Thaler R 1987, ‘Further evidence on investor overreaction and stock market seasonality’, Journal of Finance, vol. 42, pp. 557-581, as cited in Odean T 1998, ‘Are Investors Reluctant to Realize Their Losses?’ The Journal of Finance, p. 1790p. 1790 Jegadeesh, N & Titman, S 1993, ‘Returns to buying winners and selling losers: Implications for stock market efficiency, Journal of Finance, vol 48, pp. 65-91, as cited in Odean T 1998, ‘Are Investors Reluctant to Realize Their Losses?’ The Journal of Finance, p. 1790 Odean, T 1998, ‘Are Investors Reluctant to Realize Their Losses?’ The Journal of Finance, vol 53 no. 5, pp. 1775-1798 Reilly, F K and Brown K C 2006, Investment Analysis and Portfolio Management, 8th Edition, Thomson South-Western, Mason, OH Ritter, J R 2003, ‘Behavioral Finance’, Pacific-Basin Finance Journal, vol. 11 no. 4 pp. 429-437 Thaler, R H 1999, ‘The End of Behavioral Finance’, Association for Investment Management and Research, Nov/Dec 1999 Read More