The Human Element… in Individual and Institutional Investing Lawrence Speidell Laffer Associates email@example.com I thank Meir Statman, Glen Klimek Professor, Department of Finance, Leavey School of Business, Santa Clara University and Arnold S. Wood, President and CEO, Martingale Asset Management for all their help in preparing this paper. “The real trouble with this world of ours is not that it is an unreasonable world, nor even that it is a reasonable one. The commonest kind of trouble is that it is nearly reasonable, but not quite. Life is not an illogicality; yet it is a trap for logicians. It looks just a little more mathematical and regular than it is: its exactitude is obvious, but its inexactitude is hidden; its wildness lies in wait.” – G. K. Chesterton The problem with finance is that markets are almost rational, but not quite. As Walt Kelley’s cartoon character Pogo said as he surveyed a swamp littered with trash, “We have met the enemy and he is us.” The human element provides an emotional bias to economic decisions that is reflected in our behavior as individual investors and as institutional investors or members of investment committees. Understanding our emotional instincts can help us to make better, more profitable investment decisions. Looking at some individual stories may help to illustrate the magnitude of the human element in investing. Statman (2001) reports that a PBS Frontline program in 1997 told the story of a young couple, Sharon and Russ, who watch CNBC most of the time. They have almost all their savings in the market and trade frequently. They concentrate their portfolio in a few stocks. Their objective is to make some aggressive money very quickly, so they can build their dream home and retire early. They put a large position their money into Aylin Pharmaceuticals. The program was aired in 1997 and as the chart shows, the stock went from 15 to 2…we don’t know when they sold it…. At the other end of the investor spectrum is Larry Ellison, Chairman of Oracle and one of the world’s richest men. He is an avid yachtsman, but his quest for the perfect yacht began in 1996, when a girlfriend slighted his eighty foot sailboat, Sayonara. “That’s a yacht!” she exclaimed, pointing to a larger white luxury craft floating nearby. “She made me feel terribly inadequate,” Ellison said, “As a result, I went on a search to reclaim my adequacy.” One of his rivals, Jim Clark, founder of Netscape and Healtheon, expressed his financial goals differently. He said,” I just want to have more money than Larry Ellison. I don’t know why.” What do investors want? They want everything: the highest returns, no risk, no taxes and no fees. Let’s dig a little deeper into the behavior of individual investors, before turning to institutional investors and investment committees. 1) The Human Element in Individual Investors: Loss Aversion Over-confidence Regret Lottery & Insurance Portfolios Individuals have emotions that influence their response to losses, their confidence and their feelings of regret over actions taken…or not taken. One result is that individuals buy both insurance and lottery tickets, seeking both conservatism and risk in their investments. To examine risk seeking and risk avoidance, consider the following gamble, posed by Kahneman & Tversky (1979): Gamble A: First you receive $1000. Next you are asked to choose between a sure gain of $500 and a 50:50 chance to either win $1000 or nothing. Which do you choose? Now consider a different gamble: Gamble B: First you receive $2000. Next you are asked to choose between a sure loss of $500 and a 50:50 chance to either lose nothing or lose $1000. Which do you choose this time? Studies have found that 84% of people choose the “sure thing” in Gamble A, whereas most people (69%) chose the risky option in Gamble B. In fact, the outcome of both the first and the second gamble are the same when the initial amounts received are considered. The difference in behavior reflects two human characteristics: 1) People hate loses more than they like gains. They would rather gamble to avoid a loss than to make a profit. 2) People do not frame portfolios as a whole. They consider the amount they have as history and irrelevant in making a risky choice. Loss Aversion This risk seeking and risk avoiding behavior leads to the concept of the utility function, a curve that converts financial payoffs into emotional payoffs. Professor Paul Samuelson once offered a colleague a wager on a coin toss: heads you win $200, tails you lose $100. The colleague’s response was “no” for one toss, but “yes” for a hundred tosses. This is completely logical if we construct a utility function where losses count for 2.5 times as much as gains. In this case, the formula for a wager on one toss is as follows: 1 toss Utility = 50%*1*200 + 50%*2.5*(-100) = -25 On the other hand, the formula for two tosses results in positive utility: 25%*win both+50%*split+25%*lose both = .25*400 + .5*100 + .25*2.5*(-100) = +25 Empirical results from Tversky and Kahnemen in 1992 produced the utility function chart shown below. As shown, the utility of gains falls to less than half the dollar value, whereas losses have utility roughly equivalent to their dollar value. In fact, at higher levels of wealth increases, the utility per dollar falls off even more ($400 million does not have twice the utility of $200 million), while the negative utility of large dollar losses becomes even more painful. Utility Function Loss Aversion Factor = 2.1x @ $500, 2.5x @ $2000 1000 810 500 450 250 Utility 0 -500 Source: Tversky & Kahneman ( 1992) -1000 -1000 -500 0 500 1000 1500 2000 2500 Losses…… Change in Wealth ($) ……Gains Over-confidence While we may be very sensitive to losses, we are often also overoptimistic about gains. Why else would millions play the lottery? O. Svenson in a 1981 survey, found that 90% of Swedish drivers believed that their driving was less risky and more skillful than other drivers. Garrison Keillor ends each episode of Prairie Home Companion by saying Lake Wobegon is where “all the women are strong, the men are good looking and all the children are above average”. In truth, most of us believe that our children are…. “What is the Expected Return of Stocks?” (High expected returns follow high realized returns) 15% 39.8% 40% Expected 13.2% S&P Return 6.3% during the Return on Market preceding 12 Sept. Sept. months 1999 2001 -24.4% Source: Gallup Similar optimism applies to our investment portfolios. Gallup surveys done in September 1999, when the market was rising sharply, and in September 2001, after six months of steep declines, showed interesting results. The expected return on the market, for example, is strongly related to its recent return, with investors expecting a 13.2% return after a 39.8% gain in the prior twelve months, but only a 6.3% return after a 24.4% year-over-year drop. “What is the Expected Return of YOUR Stocks?” (I will do better than the market) 14.9% Own portfolio 7.9% 13.2% Stock 6.3% Market Sept. Sept. 1999 2001 Source: Gallup In both surveys, however, investors were expecting superior returns from their own portfolios than those of the market as a whole. “Is the market overvalued?” An overvalued market offers higher expected returns…. 50 49 15% Overvalued 13.2% 27 Expected Return on Market 6.3% Sept. Sept. Source: Gallup 1999 2001 Interestingly, although people expected higher returns from the market in September 1999 than they did a year later, more people felt the market was overvalued then too…. “Is this a good time to invest?” Yes …but the market is overvalued Good Time to Invest 72% 53% 49% 27% Overvalued Sept. Sept. 1999 2001 Source: Gallup Yet while more people felt the market was overvalued in September 1999, more people also felt that it was a good time to invest…. Regret In addition to our skewed loss aversion and over-confidence, our investment behavior is also influenced by our feelings of regret, based on “20-20 hindsight.” Part of this is due to the fact that the market responds to social behavior rather than to physical laws. As the weather turns colder, our purchase of an overcoat will not cause warmer weather, but investors who sell in a falling market help clear the market so it can rise again. Emotions can play havoc with investment success, and there is a tendency for investors’ attitudes to be heavily influenced by recent price movements. Investors tend to anchor their feelings about a stock to the price at which they purchased it. As Kahneman and Riepe (1998) point out, if a stock drops from $160 to $150, an investor whose cost is $200 will feel more upset than one whose cost is $100. Also the investor with the higher cost will be less likely to sell and thus recognize his loss, than the investor who is taking a gain. Many investors regret having held NASDAQ stocks at higher levels in 2001 when in hindsight they were “obvious” sales. Sadly, investors are often influenced by recent past performance in making investment decisions. The technology stock mutual funds above all had strong gains in 1999, but most investors bought those funds late, suffered the losses of 2000 and 2001 and had losses overall, while several of the funds had net long term gains. A study by the DALBAR Group estimated that from 1984 to 2000, investors in Diversified Stock Mutual Funds had a compound return of only 5.3%, while the funds themselves had returns of 13.3% and the S&P 500 returned 16.2%. The shortfall of the funds versus the index was due to trading costs and fees, but the shortfall of investors was due to ill-timed buy and sell decisions. Investors have habitually bought and sold late, chasing trends after the opportunities have passed. Statman (2001) points out that investors are simultaneously loss averse and over- confident about opportunities for gain (and over-confident about their trading skills). This is represented in the image on the right below: In finance theory, investors are rational and seek optimized portfolios which minimize risk and maximize expected returns, while in practice investors keep different portfolios in separate mental accounts, holding both insurance and lottery portfolios. Some money is kept safe for a “rainy day” and some is to “play” with. 2) The Human Element in Institutional Investors: Over-confidence – Pattern Recognition Growth & Value Traps Turning to institutional investors, there are some additional biases, added to the ones each individual brings to the job. We will explore over-confidence as it effects pattern recognition and traps particular to the major investment styles of growth and value. Institutional investors practice security analysis in disciplined ways that go beyond those of most individual investors. They study financial statements, analyze price movements and trading patterns and talk with company managements, competitors and suppliers. All these efforts, however, are designed to identify the characteristics that are predictive of positive stock price performance. This is pattern recognition, based on many of the same evolutionary instincts that led our ancestors to learn the danger of lions and lightning strikes. Sometimes patterns are easy to identify. Other times they are hidden. Kidder Peabody, a brokerage firm no longer in business, highlighted the importance of finding hidden patterns with this ad in the 1980s. a) Over-confidence: Pattern Recognition Our eyes can fool us: The bottom line may look longer…. a) Over-confidence: Pattern Recognition But both lines are the same length. We may see two similar faces upside down…. But right side up, these images of Margaret Thatcher are clearly different! Statman (2001) reports on a simple test comparing the pattern recognition skills of mice and men. Groups of humans and rats were presented two alternatives. One choice, B, pays a reward one time out of five, while the other choice, S, pays a reward four times out of five. After several trials both groups improve their performance, but as the chart below illustrates, rats catch on much faster. If S stands for stocks and B for bonds, rats would have 100% of their portfolios in stocks…. Touching Button S Always Rats 100% 1/5 chance Rats B of winning 80% Men 60% 40% S 4/5 chance 20% of winning 0% Never 1 2 3 4 5 Time A consequence of deficiencies in pattern recognition is that overconfidence in our stock-picking skills can lead to disappointing investment results. Barron’s weekly financial paper holds an Annual Roundtable of Wall Street Super-stars, men and women who are famous for their superior investment skill. Desai and Jain (1995) studied the results of the stocks recommended in these interviews over 24 years. Out of 1751 recommendations, 1599 were buys and 152 sells. The buy stocks had outperformed the market by 1.9% prior to publication of the roundtable issues, but had a zero relative return after publication. There was some information content in the sells, which are rare on Wall Street. They did drop -8.1% after publication, compared with - 1.2% before. Traps of Active Management 1.2 Love Conviction Concern Interest 0 0 Curiousity Panic Hate -1.2 If we take the trend out of stock prices over time, what remains is like a sine curve. When a stock has risen faster than the market, investors’ interest grows, reaching greatest conviction and even love just as the fundamentals start to deteriorate. On the downside, concern grows. By the time general opinion has passed from panic to hate, there are probably no sellers left. Traps of Active Management 1.2 Buy Late Buy Expensive Stocks 0 0 Buy Cheap Stocks Buy Early -1.2 Growth and Value managers each have biases of their own that make them prone to repeat certain mistakes. Growth managers may wait too long to identify attractive stocks and thus buy expensive stocks late, while Value managers may be too eager to buy cheap stocks early. Growth Manager Traps –Over-confidence in Forecasts –Over-confidence in New Eras –Buying too slowly (ideas decay) –Seeking Comfort –Falling in Love Growth managers tend to buy late. They look backwards for confirmation that a company is on a growth track and then buy slowly while they build conviction. As their knowledge accumulates, their conviction level rises. They may become over-confident in their forecasting skills and believe that there is a New Era starting in which historical valuations have become irrelevant. At the peak of a growth stock’s relative performance, growth managers may be true experts in what the company is and does; however, there are few new converts to the crowd to propel the stocks higher. Other investors believe the rosy fundamentals are fully reflected in the price. The stocks are “priced to perfection” and vulnerable to any minor disappointment. Value manager traps –Over-confidence in Accounting Data –Over-confidence in Mean Reversion –Over-confidence in Back-testing –Failure to use Timely Information –Arrogance of Contrary Ideas Value managers tend to buy early. They believe in mean reversion or “regression to the mean,” which says that all stocks have a central tendency (the horizontal line around which a sine curve fluctuates). Thus, they become interested when a stock dips below its trend of price or of some valuation measure, such as Price to Book, Price to Earnings, Price to Discounted Cash Flow, etc. Many of these valuation measures are based on accounting data and are vulnerable to errors, if, for example, depreciation has not kept pace with obsolescence. Furthermore, delays in the timeliness of reporting can make accounting information less valuable. Also, companies and markets change, and there is no guarantee that valuation levels or models which have worked in the past or back-tested well will work in the future. Finally, some value investors pride themselves on opinions that are contrary to the “crowd.” While contrary opinions may be correct at turning points, the crowd is often right. For an investor, there is little difference between being early and being wrong. An old saying is: “Even a stopped clock is right twice a day.” Another problem for institutional investors is greed, leading to failure to recognize diminishing returns to scale. While management fees rise with assets under management, the cost of trading rises inexorably with trade size. As firms get larger, their size can overwhelm their potential for excess returns. Pensions&Investments 3) The Human Element in Investment Committees: The Beauty Contest Agent Risks Regret Over-confidence in Random Events Finally, we examine some of the emotional pitfalls of investment committees. These include the human elements that each person brings to the table as an individual. And they also include the investment biases that many individuals on committees have as institutional investors themselves. When brought together on a committee, some of these elements can be magnified, making investment results secondary to satisfying other needs of committee members. We will examine some of these pitfalls by looking at the beauty contest, agent risks, regret and over-confidence in random events. The Beauty Contest As Wood (2001) observes, Miss Iowa, above, illustrates the possibility that investment committees may look at criteria other than investment credentials in selecting and retaining managers. Beauty, dress and appearance can influence investment manager selection. Dressing for success as an advisor to retail investors often requires stockbrokers to spend money on their wardrobe: jewelry for women, tailored suits, cufflinks and fancy ties for men. In the institutional market, however, the tone is more subdued. Before important meetings, investment managers dress conservatively, sometimes in funereal clothes, to convey seriousness and austerity. Sometimes other external issues can intrude on the decisions of investment committees. Politics within the committee, for example, may pit management representatives against union members, with one side or the other seeking to control certain decisions. Social issues, excluding tobacco, alcohol or defense stocks, can limit the investment universe and burden managers with extra research and reporting tasks that may subtract from investment returns. Proxy voting decisions of managers may limit their acceptability to managements. Batterymarch was once fired by a client for voting against its proxy measures to protect against a hostile takeover. Finally, the choreographed setting of a “finals” presentation, where managers rush through descriptions of philosophy, process and performance in 20 minutes identifies a good presenter, not a good stock-picker… Agent Risks A significant challenge in institutional investing is the separation of the actual beneficiaries of portfolios from the investment decision makers by a labyrinth of agents who may have differing objectives. For a corporate pension plan, the planning horizon of the beneficiaries extends into their retirement, on the order of 30 years. The ultimate guarantor of their benefits is the company itself, which has a planning horizon extending indefinitely, assuming the company succeeds in its primary objective of staying in business. Administering the pension plan, however, are several levels of agents whose time horizons are shorter. In many corporations, the pension officer is on a career path within the Treasury Department and hopes to move on in three to five years. The same time frame may represent the tenure of members of the investment oversight board, who are further hampered by the challenge of getting a large group to agree on strategies which might be upsetting to the most conservative members. Consultants can provide an important stabilizing and educational influence, but sometimes they are only present to offer a convenient target for blame when results are disappointing. Finally, the money managers are reviewed monthly, quarterly and annually under investment guidelines that often require out-performance relative to benchmarks over three to five years. The result of this hierarchy of agents is that the long term planning horizon is subordinated to short term evaluation horizons. Corporate Pension Plan Horizon = Forever XYZ Company CFO Investment Board Horizon = 3-5 Years Pension Staff (Onlookers?) Consultants Horizon = Forever Pension Fund (Education…Blame?) Horizon = 3-5 Years Money Managers Horizon = 30 Years Beneficiaries 8 The situation at endowment and foundation funds is further complicated because the interests of future beneficiaries are not directly represented. Instead, the pressures of annual program budgets are added to the concerns of the investment staff and board, further leading them to a shorter horizon. Endowment or Foundation Horizon = Forever ABC Foundation CFO Investment Board Horizon = 3-5 Years Investment Staff (Onlookers?) Consultants Horizon = Forever Endowment Fund (Education…Blame?) ??? Horizon = 3-5 Years Money Managers Future Beneficiaries Horizon = Forever Horizon = 1 Year Programs 9 The impact of agents plays a part in the “equity premium puzzle.” Benartzi and Thayler (1995) point out that since 1926 stocks have had a 7% real return, compared with 1% for bonds. $1 invested in stocks would have compounded to roughly $2000 today, whereas $1 invested in bonds would have grown to only $35. Why is the equity premium so high? Since it is so high, why does anyone hold bonds? An answer may lie in investors’ aversion to losses over short periods. If we look at stocks over one-month periods, the risk/rewards trade-off is not very attractive: stocks have been up 62% of all months since 1926, but they were down 38% of the time. Furthermore, the down months had declines 97% as large as the gains. Based on studies of loss aversion and investors’ utility functions, this is not an attractive payoff for most investors. On the other hand, if we look at stocks over a five-year evaluation period, stocks gain 90% of the time, and when they lose, the losses are only 63% of the average gain. Benartzi and Thayler drew 100,000 simulations of returns for stocks, bonds and T-bills from the CRSP tapes and examined the relative utility of holding different assets mixed over different evaluation periods. They found that investors’ utility would be maximized by holding fewer stocks at shorter evaluation periods and more stocks at longer periods because the higher volatility of stocks is smoothed over a longer time frame. A 50:50 asset mix between stocks and bonds is consistent with an evaluation period of one year. If investors used a longer period, they would choose to hold more stocks. Another way of looking at this result is to say that investors with a one-year horizon require that stocks have a risk premium of 6.5%. The authors go on to say: “Someone with a twenty-year horizon would be indifferent between stocks and bonds if the equity premium were only 1.4% and the remaining 5.1% differential is potential rents payable to those who are able to resist the temptation to count their money so often. In a sense, 5.1% is the price of excessive vigilance.” In addition to skewing asset allocation, loss aversion causes investment committees to forget the “Prudent Man Rule” of portfolio diversification and focus on the losses. In the portfolio from Wood (2001), below, the one stock held at a loss is likely to be the focus of discussion at a client review meeting. Pension officers frequently receive worried calls from investment board members asking whether the portfolio holds stocks like Enron or K-Mart or in countries like Argentina. In fact, broad diversification through an index fund guarantees that those stocks will be held. Furthermore, seeking opportunities in risky assets like small cap stocks and emerging markets guarantees that there will be periodic embarrassments. Loss Aversion vs the “Prudent Man Rule” (Framing Portfolios as a Whole) Gain (or a) Cost Market loss) Undisclosed Perceptions, Inc. 200,000 300,000 100,000 Crispy Cream Diet Centers 200,000 310,000 110,000 Harley Safety Equipment 200,000 280,000 80,000 Boston Red Sox B Shares 200,000 120,000 (80,000) New Orleans Health Foods 200,000 260,000 60,000 TOTAL 1,000,000 1,270,000 270,000 10 Once, in the early 1980s a money manager, Batterymarch, bought a group of stocks that were considered candidates for bankruptcy. They believed that all stocks in the group could not go bankrupt, but all were priced as though they already had. Although a few stocks held did go bankrupt, the remainder provided handsome returns. On the other hand, there were many anxious phone calls from clients and some refused to participate, considering the strategy too risky. Regret As with individual investors, regret plays a role in institutional investing. The chart below shows the market environment at the end of March 2001, when the rolling 12- month return of growth stocks and growth managers in the Callan Style Group had achieved a record out-performance versus the value style of over 20%. Investment committees responded by firing value managers and moving the money to growth managers. Callan Associates, March 2000 Rolling 4 Quarter Relative Returns Relative To Russell 1000 Index for 15 Years Ended March 31, 2000 15.0 12.0 Russell 1000 Growth 9.0 CAI Lg Cap:Growth Style 6.0 Relative Returns 3.0 0.0 (3.0) (6.0) (9.0) (12.0) Russell 1000 Value CAI Lg Cap:Value Style (15.0) 19 5 19 6 19 7 19 8 19 9 19 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 2099 00 8 8 8 8 8 9 9 9 9 9 9 9 9 9 19 Callan Associates, Sept 2001 Rolling 4 Quarter Relative Returns Relative To Russell 1000 Index for 15 Years Ended September 30, 2001 40.0 30.0 CAI Lg Cap:Value Style Russell 1000 Value 20.0 Relative Returns 10.0 0.0 (10.0) CAI Lg Cap:Growth Style (20.0) Russell 1000 Growth (30.0) (40.0) 19 86 19 7 19 8 19 9 19 0 19 1 19 2 19 3 19 4 19 5 19 6 19 7 19 8 20 9 20 0 01 8 8 8 9 9 9 9 9 9 9 9 9 9 0 19 In less than 18 months, the story changed dramatically. The chart above shows the market environment at the end of September 2002, when the rolling 12 month return of value stocks and value managers in the Callan Style Group had achieved a record out- performance versus the growth style of over 50%. Investment committees responded this time by firing growth managers and moving the money to value managers. Overconfidence Overconfidence has been discussed in the context of individual and institutional investors. For investment committees, it has particular problems relating to the misinterpretation of events that seem meaningful but are in fact random. In the world of basketball, a study was done of the persistence of baskets made by players. Fans generally believe that hits follow hits and misses follow misses. An interview of 100 knowledgeable fans found that a hypothetical player who makes 50% of his hits would make 61% of shots after just having made a hit, but only 46% of shots after a miss. Naturally, they believe that players should pass the ball to the one who has a “hot hand.” In fact, evidence from the Philadelphia 76ers (the only team to keep such records) shows that players who have just missed shots are more likely to make their next shot and that this probability increases with the number of missed shots: Made last shots(s) Missed last shot(s) Last Shot(s) 1 2 3 1 2 3 % Hits 51% 50% 46% 54% 53% 56% Another example of the deceptive power of random behavior is in coin tosses. Strings of four, five or six heads or tails in a row would seem unusual, but these patterns have probabilities of 50%, 25% and 20% in 20 flips of a fair coin. Turning to the world of investing, investors believe that newsletters that have performed well in the past year are a good source of recommendations. Mark Hulbert, of the Hulbert Digest, studied this and found that picking the top newsletter from the prior year and following its advice for a year would have shrunk a $100,000 portfolio to $70,752 from 1991 to 2000, while a similar investment in the S&P would have grown to $497,371. An example of the misuse of random events is in the advertising of investment products, which are often selected with hindsight on the basis of favorable results. In an advertisement in Kiplinger’s Personal Finance, July 2000, a fund family highlighted two funds with returns of 37% and 33% in the latest year, while six other funds in their family with similar objectives only had returns ranging from 1.3% to 17.5%. This is similar to the pattern in ads for lotteries and casinos, 70% of which feature winners. Frank Russell, 1998 Random events can also propel money managers into the limelight in the manager selection process. In the chart above, out of 41 managers in the top quartile of Frank Russell Universes for four years ended 1998, only eight had been in the top quartile in the prior four-year period and 14 had actually been in the bottom quartile. Furthermore, of the 41 top quartile managers in the earlier period ended 1994, only 8 remained in the top quartile in the subsequent four-year period. Callan Associates, 2001 A similar study for the Callan Database for the period ended 2001 shows similar results. Of 127 top quartile managers over four years ended 2001, 64 were in the bottom half of their universes in the prior four years. Of 129 top quartile managers for the four years ended 1997, 66 were in the bottom half of their universes in the following four years. Picking “hot dot” managers is no guarantee of superior results in the future. The difficulty of identifying manager skill over short periods is further illustrated by Wood (2001) using the hypothetical results of dart throwers and managers who have either skill, an Information ratio = 0.1, or “unskill”, and Information Ratio = -0.1. After one year, 46 of unskilled managers will outperform, 43 will still outperform after 3 years, 41 after 5 years! Meanwhile, in the skilled managers group, 54 will out-perform over one year, 57 over two and 59 over three, still 41 will have under-performed after five years. A final illustration of the challenge of separating skill from noise is to construct a simulation of a manager with a constant 3% alpha and a 6% tracking error. For each year, we generate an excess return relative to the benchmark by scaling a series of normally distributed random numbers to correspond to the 6% tracking error around the 3% alpha. The results for one hundred years are shown below. With the annual excess returns in blue bars and the trendline tracking error plotted as well. The trendline is not flat because this one hundred year sample is not long enough to reveal that the alpha is in fact constant. More troubling is the series of red bars, which show the draw-down or shortfall of the manager’s results relative to the last peak relative to the benchmark. In years 29 through 36, for example, the manager is “underwater” relative to the benchmark. In year 33, this draw-down is a seemingly insurmountable 14% before a string of positive years pulls the portfolio ahead. Under most circumstances, the manager would be terminated by the client sometime during this negative performance period. Years 2001-2100 20 15 Annual Difference vs Benchmark 10 5 0 -5 -10 Draw Down vs Benchmark -15 -20 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 1 4 7 We continued our simulation of manager results and found that the combination of a 3% alpha with a 6% tracking error is not sufficient to prevent the manager from being fired many times over. Performance under-performs the benchmark three years in a row twenty times, four years in a row eight times and five years in a row three times. And the portfolio is underwater relative to the benchmark for three years 66 times, for ten years 10 times and once for 23 years. It is doubtful that any money management firm could survive such a long dry spell. Yet these are the consequences of the normal behavior of a portfolio with 6% tracking error. Clearly separating skill from noise is a challenge for even the most patient client, but the current focus on three to five year performance records and quarterly evaluations is placing excessive emphasis on short- term random events that have little information content. Conclusion The market is a fierce warrior Draw them in with the prospect of gain. Take them by confusion. Use anger to throw them into disarray.” Sun Tzu - “Art of War”, Statman (2001) The human element in investing presents a challenge to investors and investment committees. Many of our most basic instincts do not serve us well when we attempt to make effective long-term investment decisions. There are some lessons we can take from a review of human behavior in finance. Loss Aversion Frame Portfolios as a Whole: We hate losses more than we like gains. By framing portfolios as a whole, we can view total portfolio risk rather than overemphasizing small segments. Follow the Prudent Man Rule: Portfolio diversification, as a “prudent man” would do, shifts the focus from individual stocks that may decline to the process of investing the overall portfolio. This is more productive than “second guessing” a manager’s mistakes, and it allows the portfolio to capture returns in market segments that are known to be risky. Stretch Evaluation Horizon toward Planning Horizon: Long-term funds should not be invested for short term comfort. The highest return assets will be volatile and their higher risk premiums can only be realized by holding them through difficult periods. More frequent reviews only heighten the anxiety of fiduciaries and agents, making it more difficult to maintain a commitment to a long term asset allocation plan. Over-confidence Understand the Risks in Pattern Recognition and the Probabilities of Random Events: We can be easily mislead by events that seem to fit a pattern but that are in fact random. Both stock-picking techniques and manager selection criteria need to resist easy solutions and focus on deep understanding of the fundamentals. Regret Avoid Crowds: Following fad and fashion is both tempting and comfortable, but it is rarely rewarding. Avoid Beauty Contests but Seek Inner Beauty: Appearances can be deceiving, both at the company and the money manager level. If managers look like they were sent in from Hollywood by Acme Casting…perhaps they were! There is no substitute for intimate knowledge to penetrate beyond superficial appearance. References: Benartzi, Shlomo, and Richard Taylor, Myopic Loss Aversion and the Equity Premium Puzzle, Quarterly Journal of Economics February 1995 DALBAR, Leuthold Group, February 2002 Desai, H., and P.C. Jain, Journal of Finance, Sept 1995, Gilovich, Thomas, How We Know What Isn’t So, The Free Press, 1991 Hulbert, Mark, Hulbert Digest, New York Times, Sunday, Jan 21, 2001 Kahneman, D., & M. Riepe, Aspects of Investor Psychology, Journal of Portfolio Management, Spring 1998 Kahneman, D. And Amos Tversky, “Prospect Theory: An Analysis of Decision Making under Risk”, Econometrica, 263-291, 1979 Speidell, Lawrence, Deborah Miller & James Ullman, Portfolio Optimization: A Primer, Financial Analysts Journal, January/February 1989 Statman, Meir, & Kenneth Fisher, Cognitive Biases in Market Forecasts, The Journal of Portfolio Management, Fall 2000 Statman, Investor Psychology (Silencing Investment Noise), October 2001 Statman, Lottery Traders, Santa Clara University, July 2001 Statman, Meir, The Psychology of Risk and Taxes, Investment Counseling for Private Clients, Association for Investment Management and Research, 2001 Svenson, O., Acta Psychologica 1981 “The Technology Peaks of 1999 Become the Valleys of 2001” The New York Times, Oct. 7, 2001 Wood, Arnold, Behavioral Risks: Anecdotes and Disturbing Evidence, The Journal of Investing, Spring 1997 Wood, Arnold S. Beauty Contests, hot Hands and manager Selection, Institute of Psychology & Markets Conference, October 11, 2001 Disclosure: The Standard & Poor’s 500 Stock Index (S&P 500) is an unmanaged index generally representative of the U.S. Stock Market, without regard to company size. The Morgan Stanley Capital International (“MSCI”) Europe, Australasia, Far East Index ("EAFE") is an unmanaged index of over 900 companies, and is a generally accepted benchmark for major overseas markets. Index weightings represent the relative capitalizations of the major overseas markets included in the index on a U.S. dollar adjusted basis. The Lehman Brothers Government/Credit Bond Index is an unmanaged, market-weighted index generally representative of intermediate and long-term government and investment grade corporate debt securities having maturities of greater than one year. The Russell 1000 Index consists of the 1,000 largest securities in the Russell 3000 Index, which represents approximately 90% of the total market capitalization of the Russell 3000 Index. It is a large-cap, market-oriented index and is highly correlated with the S&P 500 Index. Russell 1000 Growth Index measures the performance of those Russell 1000 companies with higher price-to-book ratios and higher forecasted growth values. Russell 1000 Value Index measures the performance of those Russell 1000 companies with lower price-to-book ratios and lower forecasted growth values. Unless otherwise noted, equity index returns reflect the reinvestment of all income dividends and capital gains, if any, and bond index returns include all payments to bondholders, if any. Index return calculations do not reflect fees, brokerage commissions or other expenses of investing. Investors may not make direct investments into any index. Past performance is no indication of future performance.
Pages to are hidden for
"The Human Element…"Please download to view full document