All of these explanations beg the question of why hot IPOs trade to what Hambrecht & Quist describes as “irrational prices in the aftermarket.”20 It suggests the following distribution for what investors are willing to pay for a hot issue. During the roadshow, H&Q finds that 200 to 300 accounts are willing to pay an amount that is within the range specified in the initial prospectus. If the issue looks attractive, so that the upper end of the range increases, then at the order stage there will be 100 to 200 accounts placing orders. When the offer price is set, at yet a higher amount, there will be 50 to 100 accounts receiving an allocation. In the aftermarket, H&Q suggests that between 10 and 20 accounts will be willing to pay a “reasonable premium.” However, when an issue is “hot,” it indicates that there will be a few large holders, between 1 and 5 accounts, willing to pay almost “any price.” In the view of Hambrecht & Quist, it is this last group that causes the IPO to trade to an irrational price.
Apparently, the SEC has also had an interest in initial underpricing. In 1999, it was investigating several securities firms for “spinning”—distributing underpriced IPO shares to favored clients in hopes of winning future business. One of those firms was Hambrecht & Quist. Interestingly, William Hambrecht, founder and former chief executive of Hambrecht & Quist, claims that the practice of “spinning” motivated him to start a firm that would change things. W. R. Hambrecht + Co. uses the Internet to enable individual investors to participate in IPOs before stock trades on a public exchange.21 Time will tell if initial underpricing disappears.
What about long-run underperformance? Before asking what causes underperformance, we should ask whether underperformance is a general phenomenon. This is a complicated issue. Looking at anecdotal evidence, such as the Boston Chicken and Netscape IPOs, can be very misleading. The way to establish whether or not there is a special underperformance effect for IPOs is to compare the long-term performance of IPOs with similar firms that have not issued new shares in the time frame under investigation. Figure 17-4, from Tim Loughran and Jay Ritter (1995), suggests that IPOs underperform their comparison group of nonissuers by 7 percent per year in the five years after issuing, where the comparison group is determined by market capitalization (size).
A study by Alon Brav and Paul Gompers (1997) suggests that the underperformance effect is considerably more complicated. They find that underperformance only seems to occur in small firms that are not backed by venture capitalists. They also employ a different matching procedure, where IPO firms are matched not just by market capitalization but also by market-to-book. In chapter 7, I discussed the fact that historically, small stocks with low market-to-book ratios have earned positive abnormal returns. Brav and Gompers argue that the underperformance does not stem from the issue’s being an IPO. Rather the low returns stem from the fact that IPOs tend to be high market-to-book stocks. This is because the original backers and investment bankers prefer to take firms public after some demonstration of success, which usually means a high market-to-book ratio.
Either way, the behavioral implications are much the same. As I argued in chapter 7, high market-to-book stocks tend to earn inferior returns because investors overreact to the positive events that led to the run-up in stock price. They overweight the recent good news and wind up being disappointed, on average, over the long run.
Robert Shiller (1989) has suggested that the market for IPOs is subject to fads, and that investment bankers behave like impresarios who organize rock concerts. To make the concert an “event,” the impresario
Figure 17-4 Average Annual Returns for the Five Years after the Offering Date
Average annual returns for the five years after the offering date for 5,821 IPOs in the United States from 1970–1993, and for nonissuing firms that are bought and sold on the same dates as the IPOs. Nonissuing firms are matched on market capitalization, have been listed on the CRSP tapes for at least five years, and have not issued equity in a general cash offer during the prior five years. The returns (dividends plus capital gains) exclude the first-day returns. Returns for IPOs from 1992–1993 are measured through Dec. 31, 1996. Source: Loughran and Ritter (1995), as updated. underprices admission charges. On one hand, the strategy does work to create an event, and for IPOs, it induces investors to overvalue the offering initially. However, over time, the market will correct its originally optimistic opinion. Therefore, long-run underperformance will be strongest among stocks with the best initial performance.
Similar reasoning applies to hot-issue markets and windows of opportunity. The owners of firms going public clearly hope to obtain the best price for their shares that they can. Therefore, it will be to their benefit to time the issue for when sentiment is positive, meaning investors are especially optimistic. The evidence for hot markets is actually strong. Historically, the IPO market has moved in cycles, both for average initial returns and for the volume of IPOs. Figure 17-5 below depicts the return cycle, and figure 17-6 depicts the number of deals cycle.
Figure 17-5 Average initial returns (i.e., on first day) by month for SEC-registered IPOs in the United States during 1977–1998.
Source: Ibbotson, Sindelar, and Ritter (1994), as updated.
Figure 17-6 The number of IPOs by month in the United States during 1977–1998, excluding closed-end fund IPOs.
Source: Ibbotson, Sindelar, and Ritter (1994), as updated.
The hot-issue market appears to have become part of the financial landscape. For example, in late January 1998, a Wall Street Journal headline read: “Tough Road Is Predicted for IPOs.” A tough road means a lukewarm market. The article states: “By the end of last week, underwriters had brought a paltry five new companies to the market. … The least appealing deals to investors right now are small deals without lengthy histories.”22 In a similar vein, at the end of 1997 Hambrecht & Quist predicted that the 1998 IPO market would feature fewer but larger deals.
What about seasoned equity offerings, an issue that comes up again in the next chapter? Their story is similar to that of IPOs. Loughran and Ritter (1995) describe the situation for the average case. In the year leading up to the offering, the stock of the average issuing firm has risen dramatically, 72 percent. The profit margin stands at 5.2 percent. Moreover sales have been growing and the firm’s managers have been investing heavily in acquiring assets. As we might expect, after the new issue, the managers continue to invest.
Unfortunately, the successful performance that led up to the offering does not continue. Four years later, profit margins have declined from their peak (at issue) to 2.5 percent. Recall that the average IPO underperforms by 7 percent in the five years after issue. For a seasoned equity offering the number is about the same—8 percent.23
Why? Are the same factors at work? Consider the situation from the perspective of the firm’s management. When will they want to bring out a new issue? When they think that their firm’s stock price is undervalued? Hardly—they are more likely to issue when they think they face a window of opportunity and the stock is overvalued. Indeed, if the stock actually is overvalued, then it should come as no surprise that its subsequent returns will be abnormally low.
What about investors? Do they recognize that newly issued shares tend to be overvalued? Or do they instead suffer from excessive optimism? Well, firms that issue new shares tend to have high market-to-book ratios. And as we discussed previously, investors bet on trends. For the most part, they expect continuation, and they overweight the recent past when making long-term projections. As a general matter, high market-to-book stocks and the stocks of firms that have been growing rapidly underperform in the long term. Seasoned equity offerings are part of this group. Interestingly, seasoned equity offerings actually do worse than comparable growth firms do. It is unclear why this occurs. However, a possible explanation centers on the marketing effort by the investment bank bringing out the issue. Are investors and managers unduly optimistic about the future prospects of the issuing firm? Loughran and Ritter suggest so. As I mentioned
in chapter 10, psychological studies document that people are predisposed to be excessively optimistic. When psychologist Neil Weinstein (1980) first studied optimism, he asked people to indicate how likely they were to live past 80, relative to other people of the same age and gender. He also asked them how likely they were to be fired from a job, again, relative to their age and gender. What he found was that most people think that they are more likely to live beyond 80, and less likely to be fired from a job. The subjects in his study were undergraduate students. I have replicated Weinstein’s study with MBA students ranging in age from 24 to 45, and find the same pattern he did, though not as strong. It seems that people are predisposed to be optimistic, at least until experience leads them to think differently.
Summary
This chapter presented three IPO phenomena: initial underpricing, long-term underperformance, and hot-issue markets. All three are indicative of inefficient markets, largely stemming from heuristic-driven bias. Which behavioral elements are involved? Loughran and Ritter suggest that both investors and managers are unduly optimistic about the future prospects of the issuing firm. It may be that similarity, regret, betting on trends, and frame dependence also play key roles in explaining the three phenomena.
Chapter 18 Optimism in Analysts’ Earnings Predictions and Stock Recommendations
Investors are slow to learn that security analysts do not always mean what they say.
Corporate executives, security analysts, and investors play two games, a recommendation game and an earnings prediction game. In both games, analysts follow a company, solicit information from its executives, predict its earnings, and/or issue buy or sell recommendations on its stock. Investors pay attention to the pronouncements of executives and analysts, so these pronouncements affect stock prices. But do the actions of investors lead prices to correctly reflect fundamental values?
Edward Keon Jr. edited the I/B/E/S Innovator, a newsletter published by I/B/E/S International, Inc. I/B/E/S collects and disseminates the earnings forecasts and stock recommendations of financial analysts.1 On August 1, 1997, Keon appeared as a guest on the public television program Wall $treet Week with Louis Rukeyser. During that appearance, he made a number of important points about the biases and errors in analysts’ recommendations and earnings predictions.
But analysts are not alone in being prone to bias and error. In fact, all three players—executives, analysts, and investors—in the two games exhibit heuristic-driven bias. Moreover, some players appear to exhibit frame dependence as well, with reference-point effects in the earnings game. So, what is the effect of heuristic-driven bias and frame dependence? As usual, market inefficiency.
The two games also have a “wink, wink, nod, nod” character. What investors hear is not always what analysts mean. This is partly because analysts do not always mean what they say. They frequently say “hold” but mean “sell,” or say “buy” when they mean “hold.” But investors often miss the “wink” and misinterpret the “nod.”
This chapter discusses the following:
• some short cases that illustrate that the optimism displayed in some analysts’ recommendations need to be taken with what Keon calls a “grain of salt”
• general evidence about the recommendations of lead analysts, and the fact that investors underreact to these recommendations
• excessive optimism in the earnings game
• the role of reference points in earnings announcements
The Recommendation Game: Case Studies About Grains of Salt
In order to understand the recommendation game, we need to identify the clients who make use of analysts’ services. Retail stockbrokers certainly rely on recommendations to induce trading by individual investors. But the buy-side institutions are also a very important recipient. Not only do they pay attention to analysts’ pronouncements, they reward the analysts they value by placing their trades with those analysts’ firms. Because of the commissions involved, analysts’ activities lie at the center of the action.
But analysts’ firms also engage in investment banking. That led Edward Keon, in his Wall Street Week with Louis Rukeyser appearance, to state: “[W]hen there’s an investment banking relationship, the analysts are almost always more optimistic than their fellow analysts. So those forecasts of investment bankers who underwrite a company’s stock should be taken with a grain of salt.”
Here’s an example that succinctly describes why such forecasts should be viewed warily. The October 27, 1997, issue of Fortune contains the following discussion about underwriters and analyst recommendations:
Consider the case of Atmel, a small semiconductor company that went public in 1991. Although Alex. Brown was the lead underwriter, bankers from other firms soon began paying regular visits to Atmel’s CFO, Chris Chellam, hoping to land the company’s next deal. “We were looking for more coverage from Wall Street, and we told that to all the bankers we talked to,” says Chellam. “If they weren’t willing to pick up coverage, they were never going to get our banking business.” (Nocera 1997)
Did Chellam mean unqualified coverage in exchange for Atmel’s banking business, or did he mean buy recommendations? In 1994, Atmel did an equity offering. Prudential was one of the underwriters. Earlier one of its analysts had initiated coverage of the company with a “buy” recommendation. The analyst acknowledged: “We got the banking business because of the research.” CFO Chellum put it more succinctly when he said: “We liked his reports.”
Keon’s warning becomes more valid. Let’s return to the year 1991 and underwriter Alex. Brown. On November 1 of that year, Alex. Brown took public a development-stage pharmaceutical company named Alteon.2 Alteon’s prospectus stated that it was “engaged in the discovery and development of novel therapeutic and diagnostic products for the complications of diabetes and aging. … Numerous studies have demonstrated that the company’s lead compound, aminoguanidine, inhibits AGE formation, cross-linking and the progression of the major complications of diabetes in animals.” Alteon’s potential market at the time was known to be very large.
Here is how the Wall Street Journal reported the first day of trading in Alteon shares:
If ever there was proof that biotechnology has taken Wall Street by storm, it came Friday with the eye-popping, 92% gain in first-day trading of Alteon Inc., a Northvale, N.J., developer of drugs to treat complications of diabetes and aging.
Though it has just 30-odd employees and no sales in sight, Alteon saw its total market value balloon to more than $316 million by the end of the feverish trading day. The embryonic company’s initial offering of three million shares, priced at $15, rose a riveting 13 to wind up at 28¾, leaving investors to wonder whether the feeding frenzy signaled madness or genius among biotechnology investors.3
How did Alteon’s stock do after its rip-roaring first day? Over the ensuing seven weeks Alteon’s share price fell to a low of $17.75 on November 25, and then recovered, closing at $24 on Friday, December 20. However, this was still $4.75 below its first-day closing price.
The first twenty-five calendar days after an IPO is a “quiet period.” Release #5180 from the SEC precludes the firm and its investment bankers from issuing forecasts and opinions relating to revenues, income, or earnings per share during this period. At the conclusion of the twenty-five days, analysts who work for the underwriter are permitted to issue a stock recommendation.
Recall the statements made by Alex. Brown’s client Atmel about what they expect in return for their banking business. Those expectations might involve “booster shots,” the issuance of positive recommendations intended to boost the stock price of a new IPO, if the price has been retreating. Indeed, on Monday, December 23, after the quiet period, an Alex. Brown analyst issued a “buy” recommendation on Alteon. The stock immediately rose by $4, closing at $27.75 that day. Apparently investors did not view this recommendation as a booster shot.
Was Alex. Brown’s “buy” recommendation intended as a booster shot, or did
they actually know something that the market did not? After all, the underwriter would supposedly be especially well informed about Alteon. We cannot say with certainty whether the buy recommendation was information based or intended to boost the price. However, figure 18-1, which depicts the behavior of Alteon’s stock price over its first two years, suggests that the recommendation
Figure 18-1 Share Price of Alteon, November 1, 1991–October 29, 1993
How the price of Alteon stock behaved during the two years following its IPO. Did the positive recommendation issued by its underwriter Alex. Brown on December 23, 1991, contain positive information about its long-term prospects? Or was it a “booster shot,” designed to inflate the price of Alteon shares? caused the positive price movement for several weeks, and then the “truth played out.”
The downward trajectory in figure 18-1 occurred as biotechnology stocks generally fell out of favor in early 1992. But in addition, the Food and Drug Administration had requested more efficacy data on Alteon’s main compound aminoguanidine, thereby delaying clinical trials for more than a year.4
Is the case of Alteon typical or unusual? An article by Roni Michaely and Kent Womack (1999) demonstrates that the situation with Alteon is in fact typical. Michaely and Womack study “buy” recommendations in the first year after IPOs for the period 1990–1991. They obtained their data from First Call Corporation. First Call collects daily commentary of security analysts, portfolio strategists, and economists at major U.S. and international brokerage firms. It then sells this information to professional investors through an online PC-based system, providing their subscribers with almost instantaneous access to information generated by brokerage firms.
Beyond Greed and Fear Page 32