Beyond Greed and Fear

Home > Other > Beyond Greed and Fear > Page 16
Beyond Greed and Fear Page 16

by Hersh Shefrin


  Some Theories

  The return pattern in post-earnings-announcement drift is part of a more general phenomenon involving momentum in the intermediate term and overreaction in the long term. There have been several theories put forward seeking to explain why this pattern comes about.

  In the first theory, Nicholas Barberis, Andrei Shleifer, and Robert Vishny (1998) hypothesize that analysts and investors have difficulty interpreting information about earnings. They suggest that analysts shift back and forth between two different mind-sets. In their first mind-set, analysts think that the rate of change is temporary. This means that they expect that earnings growth will revert. This is their mean-reverting mind-set. In their second mind-set, analysts believethat earnings are in a growth-spurt phase and that earnings growth will soar. This is their continuation mind-set. According to this theory, analysts are prone to have a mean-reverting mind-set. Therefore, when a permanent positive change in the earnings picture takes place for some company, analysts fail to recognize it. The first positive surprise will tend to be followed by a few more. However, after a succession of such changes, analysts will rethink their position and shift to a continuation mind-set. In doing so, however, they extrapolate past growth rates—they bet on trends. Therefore, they overreact. They wind up with a Goldilocks-like problem: Their first reaction is too cold, and their second reaction is too hot. According to this argument, analysts and investors are just too coarsely calibrated to deal effectively with the middle case.

  The second theory, by Kent Daniel, David Hirshleifer, and Avanidhar Subrahmanyam (1998), is a little different. Theirs is not an underreaction-based explanation. Rather, they argue that analysts and investors suffer from a combination of overconfidence and self-attribution bias. Self-attribution bias occurs when people attribute successful outcomes to their own skill but blame unsuccessful outcomes on bad luck.

  According to the theory, the combination of biases leads investors to underreact to information obtained from public sources and overreact to either information or analysis they arrive at on their own. To see how this applies to post-announcement-earnings drift, consider what happens when a company experiences a streak of good news. Suppose that an analyst has managed to uncover the first bit of good news as private information.8 Having made a good call, the analyst’s confidence grows and he becomes more overconfident. Therefore, he recommends a larger position in the stock to the investors he advises. As chance would have it, the streak of good news leads investors to earn a handsome abnormal profit. Note that the higher stake leads the price to jump as well, thereby giving rise to a momentum pattern. Emboldened further, the analyst recommends a larger share. And the momentum continues—that is, until the arrival of bad news. At this time the analyst’s overconfidence disappears, and the price drops dramatically, giving rise to an overreaction pattern.

  Harrison Hong, Terence Lim, and Jeremy Stein (1999) have proposed a third theory. In their theory, no single investor is in possession of all relevant information. Instead, the information is distributed across different investors. Moreover, cognitive limitations prevent investors from using market prices to infer what others know. Therefore, dispersed information diffuses slowly. The result is momentum. The authors hypothesize that wider analyst coverage and a larger investor base will help speed diffusion.

  To test their theory, Hong, Lim, and Stein examine how the strength of the momentum effect varies with the number of analysts following a firm and with firm size. They find that the momentum effect is smaller for larger firms that are more closely followed.9 They also find that analyst coverage is more pronounced for stocks that are past losers than for stocks that are past winners.

  Although these theories are all motivated by behavioral concepts, they differ from one another. Daniel, Hirshleifer, and Subrahmanyam posit that momentum stems from overreaction, whereas the others propose that it stems from underreaction. A key aspect of this difference is that the behavioral traits hypothesized in these papers have not been documented in psychological studies. Rather, the authors have used behavioral bits and pieces to string together their own behaviorally motivated explanations. The explanations are interesting, but I see a flashing yellow light here. In the past, when economists have developed their own psychology, the result has been both bad psychology and bad economics.

  Summary

  Scholars have produced ample evidence that a trading strategy based on post-earnings-announcement drift has consistently generated positive abnormal returns. What causes intermediate-term momentum but long-term overreaction—is it random chance or heuristic-driven bias? This question has been prominent in the debate on market efficiency. The answer is heuristic-driven bias.

  Both analysts and investors react to earnings announcements the way a poor heating system reacts to a sharp drop in winter temperature. At first, the reaction is too slow—the interior temperature stays low for too long and people freeze. Eventually the interior heats up, but instead of turning off at the desired temperature, the heating system overshoots and people boil.

  A combination of overconfidence, together with anchoring-and-adjustment leads investors and analysts to adapt insufficiently to the arrival of new information. The result is conservatism. Permanent changes in circumstances get mistaken for temporary ones, at least up to a point. Salience is key here. If the earlier history is especially salient, the information about recent change will be underweighted. If the recent information about change becomes more salient, it will be overweighted.

  Part III Individual Investors

  Chapter 9 “Get-Evenitis”: Riding Losers Too Long

  People appear to be predisposed to “get-evenitis.” Are they?

  Get-evenitis refers to the difficulty people experience in making peace with their losses. I first discussed get-evenitis in chapter 3, when I introduced the concept of loss aversion. Meir Statman and I (Shefrin and Statman 1985) coined the term disposition effect, as shorthand for the predisposition toward get-evenitis. In this chapter, I describe the extent to which the disposition effect permeates the investment landscape.

  This chapter discusses the following:

  • instances of loss aversion in equity markets, mutual fund management, and real estate

  • the documentation of loss aversion in the general financial literature

  Case Study 1: Cy Lewis

  Get-evenitis afflicts both sophisticated and unsophisticated investors. In this first case, I present a sophisticated investor.

  Alan “Ace” Greenberg, currently Bear Stearns Company’s chairman, believes that “the definition of a good trader is a guy who takes losses.”1 This view almost prevented him from rising to the head of Bear Stearns. His predecessor, Salim “Cy” Lewis, was known for his resistance to take a loss, holding on to every stock he bought.2 The two clashed. At one point in the 1960s, Mr. Greenberg was able to convince Mr. Lewis to permit him to take losses, but only by threatening to resign. “Before then, he wouldn’t let me sell,” Mr. Greenberg recalls. “From then on, he let me sell anything I wanted to.”

  Case Study 2: Investors in Steadman Mutual Funds

  Next, consider investors who are not quite as sophisticated as Cy Lewis. Some of the most insightful and entertaining examples of loss aversion have involved holders of Steadman mutual funds. According to quarterly ranking data from Lipper Analytical Services, between June 30, 1994, and April 15, 1997, at least one of the four Steadman funds has had the worst ten-year returns.

  Through the 1950s, 1960s, and early 1970s, Melvin Klahr, a math instructor at Broward Community College in Pembroke Pines, Florida, invested about $1,000 in Steadman American Industry and a predecessor fund. In June 1997 his position was worth about $434. His last Steadman purchase was at the end of 1974. Had he placed $1,000 in the average capital-appreciation fund at that time, his position would have appreciated to $29,000 in June 1997. But Klahr stubbornly continues to hold onto his Steadman shares. Why has he not been able to bring himself to sell them?
>
  Mr. Klahr says: “‘Cause I’m so stupid. … Every time I think about selling it, I think, oh, I think it’s going up a bit more.” To be sure, the Steadman funds do have their moments. In January 1997, Steadman American Industry and Steadman Associated were up 13.6 percent and 14.7 percent, respectively. This placed them in the top four of diversified stock funds for the month.

  In spite of this kind of occasional performance, Klahr does not see his resistance to selling his Steadman position as rational. “Maybe I don’t need a financial planner so much as I need a psychiatrist,” he says.

  Loss Aversion: The General Phenomenon

  Although psychiatric intervention might be a bit of an extreme, the aversion to selling at a loss definitely has very strong psychological roots. In chapter 3, I indicated that the phenomenon of get-evenitis is central to prospect theory, the framework developed by Daniel Kahneman and Amos Tversky (1979). Investors who behave in accordance with prospect theory do not mark their assets to market, at least internally. Rather they keep track of their trades in terms of gains or losses relative to the price they originally paid.

  Think back to Cy Lewis and Melvin Klahr. Both exhibit loss aversion. Mentally, neither marks to market. Both are eager to get even before they will close out a position. Both suffer from get-evenitis.

  Case Study 3: Charles Steadman’s Strategy

  Get-evenitis leads people to take chances in order to avoid taking a loss. I have already discussed Steadman shareholder Melvin Klahr. But let’s also look at Charles Steadman, the late manager of the Steadman funds. He too experienced losses, and his path to those losses is intriguing.

  William Steadman, Charles’s older brother, started Steadman funds in the 1950s. According to the Wall Street Journal, when William died in the 1960s, Charles took over and acquired the right to manage several other mutual funds. Some did well, and assets in all of the Steadman funds together reached $160 million.

  A few years ago, the performance of the funds deteriorated, but not abysmally. But then Steadman became ensnared in a legal issue with state securities regulators and the Securities and Exchange Commission. Although the SEC lost a lawsuit against Steadman, it has for the most part blocked the majority of Steadman funds from selling new shares to the public. New inflows did not offset redemptions; thus, as a proportion of money managed, expenses soared. In a January 1997 regulatory filing, two of the funds reported expenses of 25 percent of assets a year, compared to an average of 1 to 2 percent for most stock funds.

  How did Mr. Steadman react? To try and earn more than his huge expenses, Mr. Steadman used leveraged, risky investments. On Dec. 31, 1996, Steadman Associates had 16.7 percent of its portfolio invested in Intel Corp. warrants. These warrants gave an investor the right to buy Intel stock at a fixed price. Because it is a leveraged instrument—like a call option—the return to a warrant tends to be quite volatile. If Intel were to close below the exercise price when the warrant expired, then the warrant would expire worthless.

  In an April 1997 article, Robert McGough of the Wall Street Journal reported that Mr. Steadman came into 1997 with high hopes. In January, Charles Steadman wrote to shareholders that the “performance in 1997 should be reasonably good, and the fund will continue to participate in stocks of high-quality companies. … [U]pward movement of stock prices is strongly correlated to the extensive transformation of the national economy from technological discovery and accompanying increased output that we have experienced in recent years, a phenomenon which I disclosed to you a year ago.”3

  Case Study 4: Investing to Fund College Education

  Real-world events have a lot of texture, more so than the simple questions involving prizes and probabilities that I discussed in chapter 3. This next case offers a considerable amount of detail to help you understand the context, range of emotions, and unpredictability associated with real-world decisions.

  Reality looks much more obvious in hindsight than in foresight. People who experience hindsight bias misapply current hindsight to past foresight. They perceive events that occurred to have been more predictable before the fact than was actually the case. As we study case 4, which involved a real estate investment, I will ask you a series of questions to mitigate the effect of hindsight bias. Some of the questions pertain directly to hindsight bias.

  Bill and his wife are in their early thirties and have just had their first child. A good friend of theirs named James has been actively investing in real estate for the last several years. James has displayed a knack for finding undeveloped property that looks like a mess but has great potential. After purchasing a property, James’s formula has been to clean it up, divide it into parcels, and resell it at a substantial profit (without adding any structures to the land). In the last three years, James has been able to resell most of his properties for between one and two times what he paid for them. Since his initial purchases were made with borrowed funds, a common practice in real estate investment, James has been earning an extremely high rate of return.

  Several years ago James had encountered some personal problems, and Bill was very helpful and supportive during that time. Since then, James has always tried to repay Bill for his kindness in whatever way he could. Last year James recommended that Bill join James by investing in a small rural tract, which Bill did. This year James had been able to sell the tract for 75 percent more than their initial investment. Bill and his wife bumped into James one evening, and James was quite enthusiastic about another deal, predicting: “You’ll want to go in with us on this.”

  With the birth of their child, Bill and his wife have begun to think about setting aside more money for the future, particularly in respect to funding their child’s college education. They know that even when the returns on an investment look attractive, those returns can turn out to be quite modest once inflation and taxes are taken into account. They are wondering whether the deal mentioned by James might be a suitable investment for funding their child’s future college education.

  Later when they meet with James, he tells them the details. A development company that has just declared bankruptcy purchased Clear Lake Development and is eager to sell. James believes that the property has great potential for retirees. It lies in a rural area, on the shore of a lovely lake.

  The price tag for Clear Lake would be $205,000, and James suggests that they go in jointly, as equal partners. That is, James would invest $102,500, and so would Bill and his wife. James plans to follow his usual formula. After subdivision, he would sell all the lots within a year or so, for a total of $459,000. Bill and his wife currently have $17,500 in savings, and James assures them that he can arrange for them to borrow the remaining $85,000 at an attractive interest rate. James says that he is a great believer in leverage because it can offset inflation and taxes that really eat into initial returns.

  1. On a scale of 1 to 10, how would you rank Clear Lake as an investment whose purpose is to fund college education in fifteen years? Here 10 means extremely suitable and 1 means entirely unsuitable.

  During the first few months, Bill has been making payments on his $85,000 loan. In addition, he is no longer earning a return on the $17,500 of his own money that he invested. When he sees James a few months later, Bill asks him how Clear Lake is doing. James replies that a small glitch has come up.

  Apparently, the property had never been surveyed properly, and he has now commissioned a survey. However, no lots can be sold until the survey is complete, and the survey is taking more time than he anticipated. Moreover, the original $85,000 loan has come due and must be renewed. Renewal is not a problem, but the bank only wants to renew the loan for $75,000. Consequently, Bill will have to come up with the additional $10,000 himself. James is clearly embarrassed, but he also remains upbeat about the ultimate success of the project. Fortunately, Bill and his wife have been able to save exactly $10,000 in recent months, and so have the necessary funds.

  2. If You were in Bill’s situation at this point, how would you reac
t emotionally? Would you be worried? or anxious? Or would you be patient, believing that most investments incur some glitches here and there and that you are better off having minor problems than major ones? How might you feel, if you found yourself in this situation: worried, anxious, or patient?

  3. If you were Bill or his wife, would you begin to experience any feelings of regret? Specifically, would you feel that anyone should be blamed?

  4. If you answered yes, whom would you blame most: yourself, James, or the situation?

  5. Think about how the scenario has unfolded so far. At the outset, how obvious does it seem to you that it would have gone this way? That is, were any telltale indications present? How would you answer this question on a scale of 1 to 10, where 10 is very obvious, and 1 is impossible to predict?

  Time passes. In fact, a full year has passed since Bill made his initial investment in Clear Lake. In a conversation, James apologizes that it has taken a year for things to get moving, but says that they now are on the move. He realizes that this has been a period of negative cash flow for Bill, but James says he can work things out so that Bill and his wife won’t have to make any more payments on their loan. He, James, will handle the financing details. Moreover, to speed up sales, he will put up a model home on one of the lots. However, he thinks he can take care of the associated costs without asking Bill to contribute additional funds.

  The next time Bill sees James, James tells Bill that he has good news and bad news. The good news is that the model home has been built and sold. The bad news is that the sale has not stimulated additional interest in Clear Lake. James says that he feels just terrible that he brought Bill into this venture, and is concerned that it will turn into a cash drain for Bill before too long. Therefore, he proposes taking over Bill’s interest in Clear Lake, including all further interest payments, and asks Bill if he would like to sign his interest over to James. If Bill does so, he basically pulls out of this investment. In the event of a further cash drain, Bill would avoid the extra loss. But if the investment turns profitable, Bill loses out on the chance to lower his loss, recover his investment, or make a positive return.

 

‹ Prev