Investment Temperament
by Meir Statman Glenn Klimek Professor of Finance Santa Clara University Leavey School of Business Santa Clara, CA 95053
[email protected] and
Vincent Wood, CFA President Advisor Team 340 Brannan St, #402 San Francisco, CA 94107
[email protected]
June 2004
We thank Allison Beezer, Ramie Fernandez, Kenneth Fisher and Michael Kay for their help. Meir Statman acknowledges support from The Dean Witter Foundation.
Investment Temperament Abstract Financial advisors ask ‘Can we have a better risk questionnaire so we might assess better the risk tolerance of our investors?’ Risk questionnaires are rooted in mean-variance portfolio theory, with its focus on risk and expected returns. We argue that advisors need to combine the insights of mean-variance portfolio theory with those of behavioral portfolio theory, and combine a focus on risk and expected returns with a focus on goals, hopes and fears. The goals, hopes and fears of investors are related to their temperaments and we offer a temperament-based questionnaire that would help advisors focus on their investors’ goals, hopes and fears, and guide them to wealth and well-being. Temperament affects all life choices. While some people choose to be physicians, others choose to be teachers, engineers or entrepreneurs. Temperament is also likely to affect the choice of investment goals and the importance the people assign to each. People who cherish downside protection are likely to gravitate towards professions that offer secure income even if they offer little chance for riches, while those who cherish upside potential gravitate towards professions that offer better chances for riches even if they do not offer secure income. We examine the relationship between temperaments and attitudes toward risk and the relationship between temperaments and attitudes beyond risk, such as preferences for domestic or international stocks and preferences for stocks of socially responsible companies or stocks of conventional companies.
Investment Temperament Financial advisors ask ‘Can we have a better risk questionnaire so we might assess better the risk tolerance of our investors?’ Risk questionnaires are rooted in mean-variance portfolio theory, with its focus on risk and expected returns. We argue that advisors need to combine the insights of mean-variance portfolio theory with those of behavioral portfolio theory and combine a focus on risk and expected returns with a focus on goals, hopes and fears. The goals, hopes and fears of investors are related to their temperaments and we offer a temperament-based questionnaire that would help advisors guide investors to wealth and well-being. Mean-variance and behavior portfolio theories Investors in mean-variance portfolio theory care only about the risk and expected returns of overall portfolios. They measure risk by the standard deviation of portfolio returns, and choose portfolios that balance optimally risk and expected returns. The role of advisors in the mean-variance framework is to help investors identify, through “risk questionnaires,” their attitudes toward risk and expected returns and identify the asset allocation of their optimal portfolios on the efficient frontier. Investors in behavioral portfolio theory (Shefrin and Statman (2000)) care about risk and expected returns but measure risk in units of loss relative to benchmarks rather than in units of standard deviation. Moreover, behavioral investors do not consider portfolios as a whole. Instead, behavioral investors focus on aspirations or goals, arranged in mental accounts, such as retirement, college education, or a cruise around the world. The role of advisors in the behavioral framework is to promote wealth and well-being by helping investors identify their goals, modify them, and create portfolios that would satisfy as many goals, as best as possible.
1
Investors have many different goals. Investors are also different in the importance they assign to each goal. For example, some investors assign great importance to downside protection goals, such as secure retirement income (“I don’t want to be poor”) while other investors assign great importance to upside potential goals, such as the ability to travel whenever and wherever they wish, (“I want to be rich”). In his book, Love and Money, Opdyke (2004) lists his personal goals and those of his wife, some weighty and some less so. •See the running of bulls in Pamplona and the Indianapolis 500 car race •Build a retirement home of our dreams. •Have no mortgage. •Go back to school, maybe a chef. •Help our son financially, if he needs it. •Help any grandchildren we have pay for college. •Own a 1966 Mosport green Corvette (that’s the year I was born and my alltime favorite car). •Run a small business, maybe a restaurant. •Have the money to visit our son frequently, wherever he lives. •Travel as a volunteer to help children in developing nations. •Don’t have to worry about money. (p. 200-201) Many goals are easy to express but not all. A mother might find it difficult to say that she is concerned about the stability of her daughter’s marriage and wants to set aside funds to help her if she is divorced. A man is embarrassed to admit that an old slight by a wealthy brother-inlaw still burns in him years later and he wants to get even. Financial advisors who establish trusting relationships with their investors are more likely to elicit their true goals. We offer measures of temperament as instruments that would help financial advisors focus discussions with investors about their goals and guide them toward shaping and achieving them. Some financial advisors see themselves as promoters of wealth alone and are reluctant to assume their role as promoters of well being. Such advisors often say that they do not want to intrude into their clients’ personal lives or introduce psychology into finance. But psychology
2
has always been a part of finance and will continue to be. Risk-aversion and risk-seeking are central concepts in finance and advisors who consider them are already within the realm of psychology. Risk questionnaires are a common tool in the hands of advisors. Temperament surveys are one more tool for the toolbox. We believe that much of the resistance to switching from promotion of wealth alone to promotion of both wealth and well being comes because advisors lack tools for the promotion of well being. Physicians face the same problem and medical schools are taking steps to solve it. Today’s medical students examine patients in the presence of instructors and receive feedback on their ability to pace the interview, educate without jargon, and encourage questions – all ingredient of good bedside manner. Financial advisors need similar training and similar tools. Temperaments Some people are shy while others are outgoing, some have good self-control while others are impulsive. Temperaments are measures that provide important insights into preferences and behavior. Temperaments combine “nature” with “nurture.” We know that some people are shy by nature and others are outgoing since even babies exhibit shy or outgoing behavior. But we also know that shy babies do not always grow into shy adults. Nurture can help shy children counter some of their shy nature. Moreover, temperaments are likely to change as people age. For example, Zuckerman (1994) found that “sensation-seeking” behavior peaks in adolescence and diminishes with age. Temperament affects all life choices. While some people choose to be physicians, others choose to be teachers, engineers or entrepreneurs. Temperament is also likely to affect the choice of investment goals and the importance that people assign to each. People who cherish downside protection are likely to gravitate towards professions that offer secure income even if
3
they offer little chance for riches, while those who cherish upside potential gravitate towards professions that offer better chances for riches even if they do not offer secure income. We use the Keirsey Temperament Sorter® as our measure of temperament. The Keirsey Temperament Sorter® (KTS®) is based on Keirsey’s (1978) Temperament Theory and it has been used by psychologists, counselors and human resources professionals for over 25 years. The Keirsey sorter classifies people into four basic temperament groups and 16 temperament variants. The four basic temperament groups are Guardians, Artisans, Rationals, and Idealists and AdvisorTeam.com (the official provider of KTS®) describes them. Guardians Guardians are the cornerstone of society, given to serving and preserving our most important social institutions. Guardians have natural talents in managing and they use these talents to keep things running smoothly in families, communities, schools, churches, hospitals, and businesses. Guardians are cautious, loyal and disciplined. They follow the rules and cooperate with others. Artisans Artisans have a natural ability to excel in any of the arts, the fine arts, the performing arts, the athletic, military, political, mechanical, and industrial arts, as well as the “art of the deal” in business. Artisans want to be where the action is; they are impulsive, competitive, and believe that the next throw of the dice will be the lucky one. Above all, Artisans resist being tied or confined or obligated; they would rather not wait, or save, or live for tomorrow. Idealists Idealists are passionately concerned with personal growth and development. Idealists strive to discover who they are and how they can become their best possible self. And they want
4
to help others make the journey. Idealists are naturally drawn to working with people, and whether in education or counseling, in social services or personnel work, in journalism or the ministry, they are gifted at helping others find their way in life, often inspiring them to grow as individuals and to fulfill their potentials. Rationals Rationals are problem solving people. They might tackle problems in organic systems such as plants and animals, or in mechanical systems such as railroads and computers, or in social systems such as families and companies and governments. Rationals are rigorously logical and fiercely independent in their thinking. They are skeptical of all ideas, even their own. Often they are seen as cold and distant, but this is really the absorbed concentration they give to whatever problem they’re working on. Temperaments are associated with the choice of college major and profession. Fairhurst and Fairhurst (1995) reported that while Idealists make up 12% of the general U.S. population, they make up 76% of college students who major in counselor education. Similarly, Rationals make up 12% of U.S. population but 57% of science majors, and Guardians make up 38% of the U.S. population but 43% of finance and commerce majors. (See Table 1) The Keirsey sorter identifies temperaments from questions such as: At work it is more natural for you to: ____ Point out mistakes ____ Try to please others Which seems the greater fault: ____ To be too compassionate ____ To be too discompassionate
5
At a party do you: ____ Interact with many, even strangers. ____ Interact with few friends. AdvisorTeam offers the Keirsey Temperament sorter on the web at www.AdvisorTeam.com. On average, 3,500 to 4,000 people use the sorter each day and their median age is 31. We asked people who used the Keirsey sorter to answer additional questions and studied the relationship between their answers and their temperaments. The answers support the descriptions of temperament. For example, tradition is a word associated with Guardians. Forty percent of Guardians agreed with the statement “I respect tradition,” almost double the 21% of Rationals and higher than the 27% of Artisans and Idealists. Artisans prefer spontaneity over discipline and their preference is reflected in the statement “I like self-discipline.” Only 22% of Artisans strongly agreed with this statement, less than half the 51% of Guardians and lower than the 33% of Rationals and 30% of the Idealists. Caring and compassion are evident among Idealists. Sixty percent of Idealists agreed with the statement “I feel compassion for the needy,” almost triple the 23% rate of Rationals and higher than the 38% rate of Artisans and 42% of Guardians. “Financial advisors deserve our trust” is a statement that links temperament with financial advisors. Only 25% of Rationals agreed strongly or somewhat strongly with this statement, lower than the 35% of Idealists, the 42% Artisans and the 45% of Guardians. The skeptical attitude of Rationals towards financial advisors is part of their general skepticism towards people and social institutions. Only 34% of Rationals agreed strongly or somewhat strongly with the statement “I trust people are social institutions,” compared with 52% of Guardians and 56% of Artisans and Idealists. (See Table 2)
6
Risk attitudes Financial advisors often think of risk questionnaires that are offered by mutual fund companies and others as instruments within mean-variance portfolio theory but, in truth, risk questionnaires come closer to behavioral portfolio theory (Fisher and Statman (1997)). Consider, for example, the investment pyramid that the Putnam mutual fund company (2004) prescribes to its mutual fund investors, as presented in Figure 1. Assets are arranged by goals, layer by layer, as described in behavioral portfolio theory, and the portfolio is not seen as a whole, as prescribed by mean-variance portfolio theory. Income funds in the Putnam Pyramid are “[d]esigned to help provide a regular stream of income,” and their place is at the bottom of the pyramid, while growth funds are “[d]esigned to help build the value of your investment over time,” and their place is at the top of the pyramid. Risk is measured in units of standard deviation in mean-variance portfolio theory but it is measured in units of gain and losses relative to benchmarks in behavioral portfolio theory. The benchmark might be the value of assets today but it might also be the purchase price of assets years ago or the value of assets we aspire to have. Questions in risk questionnaires are typically phrased in units of gains and losses. Consider, for example, the following question from the Vanguard Risk Tolerance questionnaire (2004): How you feel about risk affects how aggressively or conservatively you should invest. The chart below shows how much an investment of $10,000 may fluctuate during a one-year period, depending on how the amount is invested. However, the maximum gain or loss on an investment is impossible to predict. The ranges shown in the chart are hypothetical and are designed solely to gauge an investor’s risk tolerance. Given the potential gain or loss in any one year, in which would you invest? A. Loss of $164, gain of $593.
7
B. Loss of $1,020, gain of $1,921. C. Loss of $3,639, gain of $4,229 This question suffers two interpretation difficulties, one related to variations in time horizons and the other to variations in the meaning of dollar amounts. First, the time horizon in this question is one year. Would the preferences of investors be the same if the horizon were 10 years, or 30 years? Second, the dollar amounts in this question have different meanings to different investors. A loss of $3,639, as in Choice C, is a substantial loss for a woman earning $50,000 a year with $100,000 worth of assets in her portfolios but it is a less substantial loss for a woman earning $300,000 a year with $600,000 worth of assets in her portfolio. Many who would choose a 50-50 chance for a $30 gain or zero over a sure $10, would not choose a 50-50 chance for a $300,000 gain or zero over a sure $100,000. Barsky et al (1996) sought to overcome the time horizon and dollar amount difficulties with a question that focused the attention of respondents on their own incomes and lifetimes. Suppose that you are the only income earner in the family, and you have a good job guaranteed to give you your current (family) income every year for life. You are given the opportunity to take a new and equally good job, with a 50-50 chance that it will double your (family) income and a 50-50 chance that it will cut your (family) income by a third. Would you take a new job? (p. 540)
Barsky et al probed deeper into risk-attitudes with two follow-on questions. They probed those who accepted the 50-50 chance for doubling their income or seeing it decline by a third to see if they are sufficiently risk-tolerant to take a 50-50 chance for doubling their income or seeing it decline by a half. Similarly, they probed those who rejected the 50-50 chance for doubling their income or seeing it decline by a third to find if they would accept the less risky 50-50 chance for doubling their income or seeing it decline by a fifth.
8
Barsky et al noted in their review of the literature that there is no general agreement that risk tolerance, such as reflected in the answers to the question they posed, predicts risk-taking in real life. People who are risk averse in social settings might not be risk averse in financial settings. They quoted Slovic who concluded that “the majority of evidence argues against the existence of risk-taking propensity as a generalized characteristic of individuals.” (p. 550) However, Barsky et al also noted that more recent evidence on a biological basis for behavior suggests that personality traits related to risk aversion, such as “harm avoidance” and “novelty seeking” have some consistency across settings. Barsky et al found that attitudes toward risk vary somewhat with personal characteristics. For example, they found that people in the 55-70 year old group have less risk tolerance than those in the younger than 55 group or in the older than 70 group. They also found that males are more risk tolerant than females, that Asians and Hispanics are more risk tolerant than Whites or Blacks, and that Jews are more risk tolerant than Protestants or Catholics. They also found that behavior varies somewhat with risk attitudes. For example, people who smoke now are more risk-tolerant than those who quit smoking, and those who quit smoking are more risk tolerant than those who have never smoked. However, the magnitudes of the differences are small. For example, 65.1% of females rejected both the one-third and one-fifth versions of the question but only 64% of males did. Similarly, while 66.7% of Blacks rejected both the one-third and onefifth version of the question, only 64.9% of Whites, 59.3% of Hispanics and 57.9% of Asians did. Still, Barsky et al found that risk tolerance has some predictive power in the investment arena. People with higher risk tolerance held higher proportions of stocks and lower proportions of Treasury bills and savings accounts than those with low risk-tolerance. However, the relationship between risk-tolerance and allocation to risky assets is much weaker than predicted
9
by the standard Capital Asset Pricing model (CAPM). In particular, according to the CAPM the elasticity of the proportion of risky assets with respect to risk-tolerance should by one, but Barsky et al estimate it at only 0.17. As they wrote: “consequently, there is inadequate sensitivity of portfolio shares to risk-tolerance compared with the prediction of the standard model.” (p. 561).
Temperaments and risk-attitudes We study the relationship between temperaments and risk-attitudes as measured by the Barsky et al question. Rationals have the highest risk tolerance among all temperaments when they weight a 50-50 chance for doubling their income or seeing it cut by a third; 58% of Rationals would take the chance while only 51% of Artisans, 44% of Idealists and 38% of Guardians would. (See Table 3). Rationals are always at the high end of the risk-tolerance scale, whether the downside involves a cut of one-half, one-third or one-fifth of income, but while differences by temperament are statistically significant for the one-half and one-third cuts, they are not statistically significant for the one-fifth cut. Artisans are least sensitive to the magnitude of the downside, so they require the least compensation for an increase in risk; 39% of Artisans would take the chance if the downside is a loss of one-half and 69% would take it if the downside were a loss of one-fifth. In contrast, Guardians and Idealists are willing to take risk if they are well compensated. Only 27% of Idealists and Guardians would take a chance if the downside is a loss of one-half. But 72% of Idealists and 71% of Guardians would take it if downside is a loss of one-fifth. The pattern where Rationals are most willing to take risk is also evident, but only weakly, when subjects are
10
asked about their preferred allocations to stocks and bonds. Consider a question that begins with a statement of facts and continues with the elicitation of preferences. Facts: Historically, stocks have had higher returns, on average, than bonds but they were more volatile. In years when stocks gained, they tended to gain more than bonds but in years when they lost, they tended to lose more than bonds. Preferences: Investors vary in their preferences. Some investors are very concerned about the possibility of losses with stocks and prefer to keep a lower proportion of stocks in their portfolios. Other investors are less concerned about the possibility of losses with stocks and prefer to keep a higher proportion of stocks in their portfolio. Which statement best describes your preference: 1. 2. 3. 4. 5.
I prefer a much higher proportion of stocks than bonds in my portfolio. I prefer a somewhat higher proportion of stocks than bonds in my portfolio. I prefer equal proportions of stocks and bonds in my portfolio. I prefer a somewhat higher proportion of bonds than stocks in my portfolio. I prefer a much higher proportion of bonds than stocks in my portfolio.
Rationals have the highest preference for stocks; 45% of them preferred a much higher or somewhat higher proportion of stocks than bonds in their portfolios. Only 38% of Artisans and 35% Guardians and Idealists shared that preference. However, echoing Barsky et al’s finding on the weak link between risk attitudes and asset allocation, we find that differences in preferences by temperament are not statistically significant. (See Table 4) There are no layers or mental accounts in mean-variance portfolio theory. Investors in mean-variance portfolio theory consider the portfolios as a whole, not as a collection of mental accounts. Investors who increase the proportion of stocks in their portfolio indicate, in effect, that they are moving from one point on the efficient frontier to another. They are willing to assume greater risk in their portfolio in exchange for higher expected returns. But answers to Question 13 in the Asset Allocation Planner of Fidelity Investments (2004) are consistent with behavioral portfolio theory, not with mean-variance portfolio theory.
11
If you could increase you chances of improving your returns by taking more risk, would you: 1. 2. 3. 4. 5.
Be willing to take a lot more risk with all your money. Be willing to take a lot more risk with some of your money. Be willing to take a little more risk with all your money. Be willing to take a little more risk with some of your money. Be unlikely to take much more risk.
Answers 1 and 3 make sense within the mean-variance framework. In that framework, only the risk of the overall portfolio (i.e. all your money) matters. But answers 2 and 4 make no sense within the mean-variance framework. This is because answers 2 and 4 segment the portfolio into layers where investors are willing to take a lot more risk or a little more risk with some of their money. Mean-variance investors have single attitude toward risk, not a set of attitudes, layer by layer. In contrast, behavioral investors have many attitudes toward risk, layer by layer. So they might be willing to take a lot more risk with some of their money. (See Table 5) The vast majority of respondents is willing to take a lot more risk or a little more risk with some of their money while only a small minority is willing to take a lot more risk or a little more risk with all their money. Note that the effect of taking a lot more risk with some of your money on overall portfolio risk and expected returns is similar to the effect of taking a little more risk with all your money. Yet the proportion of those who chose to take a lot more risk with some of their money was almost ten times higher than the proportion of those who chose to take a little more risk with all their money. For example, while 42% of Guardians were ready to take a lot more risk with some of their money, only 5% were ready to take a little more risk with all their money. We see, once more, that Rationals place the greatest weight on the upside potential layer of their portfolios. Fifty-six percent of Rationals are willing to take a lot more risk with some of their money while 46% of Artisans, 42% of Guardians and 41% of Idealists are willing to do so.
12
Preferences beyond risk Investors in mean-variance portfolio theory care only about risk and expected returns, but investors in behavioral portfolio theory care about other characteristics as well. For example, while mean-variance investors do not care about the distinction between foreign stocks and domestic stocks or the distinction between stocks of socially responsible companies and those of conventional companies, behavioral investors do. (See Table 6) Consider preferences between U.S. stocks and foreign stocks. Facts: Historically, foreign stocks such as those from Germany, Japan or Britain, had returns that were approximately equal, on average, to the returns of U.S. stocks, but the returns of U.S. stocks and foreign stocks were very different in many years. Preferences: Some U.S. investors prefer U.S. stocks over foreign stocks while others prefer foreign stocks over U.S. stocks. Which statement best describes your preference: 1. 2. 3. 4. 5.
I strongly prefer U.S. stocks over foreign stocks. I somewhat prefer U.S. stocks over foreign stocks. I have equal preference for U.S. stocks and foreign stocks. I somewhat prefer foreign stocks over U.S. stocks. I strongly prefer foreign stocks over U.S. stocks.
People of all temperaments display “home bias” but Artisans and Guardians display greater home bias than Rationals and Idealists. Sixty-five percent of Artisans and 62% of Guardians prefer strongly or somewhat strongly U.S. stocks over foreign stocks, while only 49% of Rationals and 48% of Idealists do. Now consider preferences for socially responsible stocks and conventional stocks. Facts: Historically, stock returns of socially responsible companies were approximately equal to the stock returns of conventional companies. Investors vary in their definition of socially responsible. Some investors exclude tobacco and alcohol producing companies while others exclude companies that harm the environment. Preferences: Some investors prefer stocks of socially responsible companies over stocks of conventional companies. Other investors prefer stocks of conventional companies
13
over stocks of socially responsible companies. Which statement best describes your preference: 1. I strongly prefer stocks of socially responsible companies over stocks of conventional companies. 2. I somewhat prefer stocks of socially responsible companies over stocks of conventional companies. 3. I have equal preference of socially responsible companies and stocks of conventional companies. 4. I somewhat prefer stocks of conventional companies over stocks of socially responsible companies. 5. I strongly prefer stocks of conventional companies over stocks of socially responsible companies. People of all temperaments display a preference for stocks of socially responsible companies over stocks of conventional companies. Idealists have the strongest preferences for socially responsible investing. While 48% of Artisans, 50% of Rationals and 54% of Guardians preferred strongly or somewhat strongly stocks of socially responsible companies, 58% of Idealists preferred them. But differences by temperament are not statistically significant. As noted earlier, Idealists express the highest compassion for the needy but they seem to express that compassion outside the realm of investments. The distinction between compassion for the needy and socially responsible investing is supported by the observation that there are only small differences by temperament in agreement with the statement, “Investing has an ethical dimension” (See Table 7). Last, consider a pair of questions, one asking whether subjects believe that success in picking stocks that earn higher than average returns is due mostly to skill or luck and the other asking whether subjects believe that they can pick stocks that would earn higher than average returns. Some people believe that success in picking stocks that earn higher than average returns is mostly due to skill. Other people believe that success in
14
picking stocks that earn higher than average returns is mostly due to luck. What do you believe? Some people believe that they can pick stocks that would earn higher than average returns. Other people believe that they are unable to do so. What is your belief? Most subjects believe that success in picking stocks is due mostly to skill and differences by temperament are not statistically significant; 67% of Artisans, 69% of Guardians, 69% of Rationals and 62% of Idealists share that belief. Most subjects also display overconfidence and differences in overconfidence by temperament are statistically significant. The proportion of those who think that they can pick winning stocks would be close to zero if people calibrated their abilities properly. Instead, it is greater than 50% in all temperament classifications. For example, while 69% of Guardians and equal 69% of Rationals believe that picking winning stocks is mostly due to skill, 71% of Rationals believe that they can pick winning stocks, while only 54% of Guardians share that belief. (See Table 8) Application A client has taken the temperament survey at the request of his advisor and found that he is a Guardian. “Does the description of a Guardian fit you?” asks the advisor. The client says that the description fits; he is cautious by nature, plans ahead, manages well and places more emphasis on financial security the on riches. The client is a lawyer who works in a small law firm whose principals are a father and his son. He is unhappy about his setting, feeling dominated and unappreciated by the principals. The client has good management skills and could establish a law firm of his own, but he is reluctant to do so. “I’m not an entrepreneur,” he says. “I don’t want to risk my paycheck.”
15
The advisor takes an inventory of the client’s assets and liabilities, performs a Monte Carlo simulation and finds that the client has a 98% probability of having retirement income equal to 80% of his current income, with assets to spare. “What would you like to do in retirement?” asks the advisor. The client says that he has not thought much about retirement beyond the desire to know if he can provide for it. The advisor asks the client to make a list of ideas about retirement as they occur to him. In a subsequent meeting the client says that he would like to teach a few classes at a local college, expand his volunteer work at a charity, and travel in the U.S. and abroad. He also says that he would like to leave specific amounts of money to each of his three children and donate a specific amount to the charity. The portfolio pyramid of the client has three layers, each representing a goal: comfortable retirement income at the bottom layer, with a 98% probability of reaching it, money for the children in the middle layer with a 70% probability of reaching it, and money for charity with some probability, however small, of reaching it (See Figure 2). The advisor asks further questions about investment preferences. The client says that he prefers to have no more than 20% of his portfolio in international securities and that he has no preference for socially responsible investments over conventional investments. The advisor performs three separate Monte Carlo simulations, one for each layer, and find that the first two goals, retirement income and money for the children can be reached at the specified probabilities and that there is a 7% probability that the third, charity, would be reached as well. The advisor prepares a written plan that specifies amounts in the actual accounts in the portfolio, so much in bond fund A, so much in stock fund B, but also specifies “mental accounts,” so much of bond fund A and so much of stock fund B is for the retirement mental account and so much of them is for the children and charity mental accounts. The client leaves
16
the meeting knowing that his retirement income is as secure as can be in a world where uncertainty cannot be fully eliminated, that he has a high probability of leaving money to his children and that he has hope, however small, of leaving money for charity as well. Conclusion Good physicians do more than promote health, they promote well being. And good financial advisors do more than promote wealth, they promote well being. We offer behavioral portfolio theory as a framework and temperament classifications as a tool for financial advisors. Classification by temperament can help financial advisors and investors in a number of ways. First, people have an intuitive sense of their own temperament and are rarely surprised when the temperament sorter classifies them as Guardians, Artisans, Idealists or Rationals. The congruence between their intuition and classification tells investors that advisors use valid tools of analysis and adds to the trust of investors in their advisors. Second, classification of temperament leads naturally to discussions about goals, aspirations, hopes and fears. Why did you choose to be a physician rather than a teacher? Why did you choose to be an entrepreneur rather than an employee? Exploration of life choices are likely to help investors articulate their goals, aspirations, hopes and fears, and they are likely to help advisors design portfolios with the right balance of upside potential and downside protection. Third, classification of temperament can help advisors communicate with investors in language that resonates with them. Guardians respect the opinions of authority figures such as well-known investors and scholars. They prefer orderly and linear presentations supported by citation of facts. Artisans prefer to learn by doing and appreciate opportunities for active involvement. They prefer presentations that include anecdotes, humor and opportunities to ask
17
questions. Idealists are nurturing and sincere and value highly personalized relationships. They prefer presentations that rely on metaphors and analogies. Rationals are in search for knowledge. They focus on the more theoretical features of investment and prefer presentations in scholarly language with precise definition. Last, financial advisors would benefit by recognizing their own temperaments. We tend to think that our own goals, aspirations, hopes and fears are shared by all and that all prefer our style of communication. Advisors who know their own temperament and that of investors can tailor their communication to fit their investors.1
1
It would be useful to examine the temperaments of money managers as well. For example, are Guardian drawn to the management of fixed income securities? What is the predominant temperament among hedge fund managers? Do the least successful money managers perform poorly because their temperaments are ill-suited to the assets they manage?
18
Reference: Barsky, Robert, F. Thomas Juster, Miles Kimball and Matthew Shapiro (1997). “Preference Parameters and Behavioral Heterogeneity: An Experimental Approach in the Health and Retirement Study,” Quarterly Journal of Economics, May: 537-579. Fairhurst, Alice and Lisa Fairhurst (1995). Effective Teaching. Effective Learning. Palo Alto, CA: Davies-Black Publishing. Fisher, Kenneth and Meir Statman (1997). “Investment Advice From Mutual Fund Companies,” Journal of Portfolio Management, Fall: 9-25. Keirsey, David (1978). Please understand me II: Temperament, character, intelligence. Del Mar, CA: Prometheus Nemesis. Opdyke, Jeff (2004). Love & Money: A life guide for financial success. Hoboken, New Jersey: John Wiley and Sons Shefrin, Hersh and Meir Statman (2000). “Behavioral portfolio theory,” The Journal of Financial and Quantative Analysis. Vol, 35, No. 2, June: 127-151. Zuckerman, Marvin (1994). Behavioral Expression and Biosocial Bases of Sensation Seeking, Cambridge, New York: Cambridge University Press.
19
Figure 1: The portfolio pyramid.
Growth Risk/Reward
Growth and Income Funds
Income Funds
Tax-Free Funds
Source: Putnam (2004).
20
Figure 2: The portfolio pyramid of a client.
Goal: Money for Charity 7% probability of reaching the goal
Goal: Money for Children 70% probability of reaching the goal
Goal: Retirement Income 98% probability of reaching the goal
21
Table 1: Proportions of temperaments in the general U.S. population and among college students and professors.
1 2 3 4 5 6 7
General U.S. population Engineering majors Science majors Counselor education majors Finance and Commerce majors Law Students Professors
Guardians 38% 27% 10% 10% 43% 30% 30%
Source: Fairhurst and Fairhurst (1995), Table B1, p. 304.
Artisans 38% 8% 7% 5% 28% 11% 6%
Idealists 12% 22% 26% 76% 10% 16% 33%
Rationals 12% 43% 57% 8% 18% 43% 31%
Table 2: Proportion of subjects who strongly agree with the statements:
Guardian Artisan Idealist Rational
1
I respect tradition.
I like self-discipline.
I feel compassion for the needy.
Investing has an ethical dimesion
Financial advisors deserve our trust.1
40% 27% 27% 21%
51% 22% 30% 33%
42% 38% 60% 23%
29% 28% 36% 33%
45% 42% 25% 35%
n = 956 Chi-squared = 12.96***
n = 951 Chi-squared = 6.31*
n = 955 Chi-squared = 21.08***
n = 895 Chi-square = 3.23
n = 897 Chi-squared = 7.82*
Combines subjects who strongly agree and somewhat agree. * Statistically significant at the 0.10 level ** Statistically significant at the 0.05 level *** Statistically significant at the 0.01 level
Table 3: Proportion of subjects who accept an offer for a job with a 50-50 chance for a higher income or a lower one. Imagine a situation where you are the only income earner in the family, and you have a good job guaranteed to give you your current (family) income every year for life. You are given the opportunity to take a new and equally good job, with a 50-50 chance it will double your (family) income and a 50-50 chance that it will cut your (family) income X%. Would you take the new job? Temperament Artisans Guardians Rationals Idealists
X=
50% 39% 27% 38% 27%
33% 51% 38% 58% 44%
20% 69% 71% 79% 72%
n = 966 Chi-square = 8.34**
n = 964 Chi-square = 9.95**
n = 960 Chi-square = 0.69
** Statistically significant at the 0.05 level
Table 4: Preferences for stocks and bonds Fact: Historically, stocks have had higher returns, on average, than bonds but they were more volatile. In years when stocks gained, they tended to gain more than bonds but in years when they lost, they tended to lose more than bonds.
Preferences: Investors vary in their preferences. Some investors are very concerned about the possibility of losses with stocks and prefer to keep a lower proportion of stocks in their portfolios. Other investors are less concerned about the possibility of losses with stocks and prefer to keep a higher proportion of stocks in their portfolio. Which statement best describes your preference: More Bonds than More Stocks than Bonds Equal Proportions Stocks Temperament Guardians 35% 33% 32% Artisans 38% 36% 27% Idealists 35% 40% 26% Rationals 45% 32% 23% n = 1634 Chi-square = 3.33 Chi-square is for those who chose more stocks than bonds.
Table 5: Mean-variance or behavioral portfolios? If you could increase your chances of improving your investment returns by taking more risk, would you?
Temperament Guardians Artisans Idealists Rationals
Be willing to take a lot more risk with all your money. 3% 4% 4% 8%
Be willing to take a lot more risk with some of your money. 42% 46% 40% 56%
Be willing to take a little more risk with all of your money. 5% 5% 7% 5%
Be willing to take a little more risk with some of your money. 40% 35% 40% 26%
n = 1984 Chi-square = 7.34* Chi-square is for those who are willing to take a lot more risk with some of their money. * Statistically significant at the 0.10 level
Be unlikely to take much more risk. 10% 10% 9% 5%
Table 6: Preferences for domestic stocks and foreign stocks. Fact: Historically, foreign stocks such as those from Germany, Japan or Britain, had returns that were approximately equal, on average, to the returns of U.S. stocks, but the returns of U.S. stocks and foreign stocks were very different in many years.
Preferences: Some U.S. investors prefer U.S. stocks over foreign stocks while others prefer foreign stocks over U.S. stocks. Which statement best describes your preference:
Temperament Guardians Artisans Idealists Rationals
Prefer Domestic over Foreign Equal Preference 62% 27% 65% 23% 48% 39% 49% 34%
Prefer Foreign over Domestic 12% 12% 14% 17%
n = 1597 Chi-square = 12.84*** Chi-square is for those who prefer domestic stocks over foreign stocks. *** Statistically significant at the 0.01 level
Table 7: Preferences for socially responsible stocks and conventional stocks
Fact: Historically, stock returns of socially responsible companies were approximately equal to the stock returns of conventional companies. Investors vary in their definition of socially responsible. Some investors exclude tobacco and alcohol producing companies while others exclude companies that harm the environment.
Preferences: Some investors prefer stocks of socially responsible companies over stocks of conventional companies. Other investors prefer stocks of conventional companies over stocks of socially responsible companies. Which statement best describes your preference:
Temperament Guardians Artisans Idealists Rationals
Prefer SRI over Conventional 54% 48% 58% 50%
Equal Preference 34% 40% 31% 36%
n = 1584 Chi-square = 3.37 Chi-square is for those who prefer SRI stocks over conventional stocks.
Prefer Conventional over SRI 12% 12% 11% 15%
Table 8: Beliefs about success in picking stocks. Question: Some people believe that success in picking stocks that earn higher than average returns is mostly due to skill. Other people believe that success in picking stocks that earn higher than average returns is mostly due to luck. What do you believe? Question: Some people believe that they can pick stocks that would earn higher than average returns. Other people believe that they are unable to do so. What do you believe?
Temperament Guardians Artisans Idealists Rationals
I believe that success in I believe that I can pick picking stocks that earn stocks that would earn higher than average returns higher than average 1 1 is due mostly due to skill. returns. 69% 54% 67% 63% 62% 56% 69% 71% n = 1884 Chi-square = 3.05
n = 1889 Chi-square = 8.54**
1. Combines those who believe strongly or somewhat. ** Statistically significant at the 0.05 level
