INTRODUCTION

If capital markets were efficient, no advantages could be gained by issuers attempting to time their stock offerings since their share price would always reflect all available information. Despite the general acceptance of the efficiency paradigm in its various strengths (from weak to strong), investment bankers and issuers alike persist in spending great efforts and money in finding the right "window of opportunity" in the market, where temporary "mispricings" would optimize the proceeds of their planned financial operations. To reconcile these two positions, it is necessary to investigate the efficiency of the market in some critical, limit conditions, such as those prevailing when attempting to price stocks which have historically experienced extremely high growth rates.

 

Standard valuation theories assume that the price of a share is simply the discounted value of all cash flows accruing to that stock in the future. Under an efficient market hypothesis, the resulting prices should at all times reflect rational expectations about the future, so that the realized returns on such stock holdings should on average be commensurate with their risk level. However, empirical tests in the literature have highlighted many instances in which the pricing of stocks appears to deviate from that which would be expected under efficient markets.

A first line of research indicated the existence of abnormal returns for small firms, which, in subsequent investigations, were further shown to occur primarily in early January (what became known as the "January effect", although it should more adequately be referred to as the "Small Firm January Effect") [Banz (1981); Keim (1983); Reinganum (1983); Blume and Stambaugh (1983); Ritter and Chopra (1989)]. Other studies have identified additional anomalies including a "day-of-the-week effect": statistically significant differences in returns are observed on different days of the week, and in particular largely negative returns during the period from the close of the market on Friday to its close on Monday [French (1980); Keim and Stambaugh (1984)]; the "earnings report effect": abnormal returns seem to occur around quarterly earnings announcements which differ markedly from prior analysts’ expectations [Rendleman, Jones, and Latane (1982)]; and an "overreaction/reversal effect": stocks having experienced high returns during one period tend to underperform in subsequent periods, and vice-versa for the underperformers [DeBondt and Thaler (1985); Clayman (1987); Chopra, Lakonishok, and Ritter (1992); Haugen (1995)].

 

Supergrowth firms, which for the purpose of this study will be defined as the publicly traded firms having experienced the highest annual sales growth over the prior 5 year period, are especially likely to stretch the efficiency concept because: (1) future growth rates are difficult to estimate and very volatile; and (2) most of these firms belong to high beta cohorts where deviations from CAPM pricing have already been outlined in the empirical literature. Moreover, behavioral studies of decision-making under uncertainty indicate a tendency for estimates of future outcomes to be biased by observations of prior outcomes. Anchoring, for example, has been shown to occur where an initial best guess is adjusted upward or downward to predict a future outcome: the resulting estimated ranges tend to be too narrow and biased in the direction of the initial estimate. Issues of small sample size and the representativeness of prior observed experience can lead to further biases in estimates of future outcomes, as detailed by Kahneman, Slovic and Tversky (1982).

 

The estimation of future growth rates of supergrowth firms can be hypothesized to be unduly affected by investors’ observations of recent growth rates, and therefore lead to expectations of stock prices which are biased upward from their intrinsic values. Indeed, a statistically significant overreaction of this type has been observed by Chopra, Lakonishok and Ritter (1992). The over-reaction effect was observed to persist even after adjusting for firm size and risk and was shown to be more pronounced for smaller firms. The noted tendency for individuals to weight recent data more heavily in the making of judgments about the future might be expected to create further deviations from the conformance of observed returns to those predicted by the CAPM.

 

Even though stock prices at all times should reflect the present value of all future cash flows, it is clear that estimating the cash flows for supergrowth firms is no easy task. Moreover, many of these stocks may not be as closely followed by institutional investors as those from larger, more established companies. This lack of scrutiny may in many instances lead to a diminution of the quality of information available about the supergrowth stocks as a group and to a further lack of precision in the market’s pricing of these assets [Arbel (1985)]. In addition, these stocks tend to exhibit high systematic risk, a factor which has been associated in previous papers with deviations from standard asset pricing models (high beta stocks tend in general to perform worse than expected under the risk/return model).

 

The objective of the study is then, using simple investment portfolio strategies, to determine the extent to which abnormal returns could be earned by selectively investing in these supergrowth stock based solely on public information. More specifically, all trades are based on the Inc. magazine annual ranking of the 100 fastest growing public companies in America. This ranking is likely to identify a group of companies particularly difficult to price for the market because of their extreme historical rates of growth (which are unlikely to persist) and their high level of overall risk (most firms belong to emerging, high technology fields).

 

The results tend to support the general hypothesis: opportunities for statistically significant abnormal returns seem to exist, even after correcting for risk and the possibility of a survivorship bias in the sample. An equally weighted investment in all firms listed in the Inc.100 rankings for the period 1979 through 1990 inclusive would have generated cumulative S&P500 adjusted excess returns of approximately 23% in the 36 months following the ranking (with a t-test equal to 3.166, thus enabling us to reject the hypothesis that abnormal returns are equal to zero with a degree of confidence greater than 1%). The results are generally not consistent with market efficiency for these supergrowth firms, opening up the door to the possibility of strategic games, such as issue timing, by issuers and investors alike. These results also shed new light on the contradiction observed under the traditional efficiency paradigm between the theoretical absence of benefits to timing activities and the observed reliance of many issuers and investment bankers on such factors. If the market is indeed less efficient in "limit" conditions (high growth, high risk, etc.), then mispricings may occur and timing may be a valuable investment.

 

The study is structured as follows. Section 2 lays down the research hypotheses and their conceptual justifications. Section 3 outlines the database created for the purpose of this research program. It is followed by an extensive presentation of the methodologies used to measure abnormal returns in the supergrowth portfolios. Section 5 provides some descriptive statistics of the sample, and introduces the extensive analyses of the results conducted in section 6. Conclus-ions and discussions ensue.

 

RESEARCH HYPOTHESES

The fundamental objective of the paper is to investigate the market’s ability to properly price a group of firms characterized by very high historical growth rates in sales. As outlined in the introduction, these firms are likely to stretch the market efficiency concept to the limit because of the intrinsic volatility of such firms and the difficulty of forecasting future growth and risk. Testing for pricing errors in supergrowth firms thus becomes a proxy for the larger question of market efficiency in critical (limit) conditions. Within that broad statement of objectives, a series of more specific hypotheses are being analyzed.

 

Hypothesis 1: Supergrowth firms have higher-than-normal exposure to market risk, i.e. they exhibit statistically higher levels of systematic risk than the market as a whole.

 

Hypothesis 1 serves important conceptual and methodological purposes. Conceptually, the evolution of market risk exposure for high growth firms is still largely uncharted territory, as outlined by Cotter (1992) and Loughran and Ritter (1994). The common understanding is that firms going public through an IPO, being relatively young and having products with untested futures, have "high risk", as mentioned by Loughran, Ritter and Rydqvist (1994). A significant problem with this approach, highlighted in Leleux (1993), is that total risk is often mistaken for market-related risk (also referred to as systematic risk, or beta). In other words, these firms seem to exhibit large variances in their returns but the sensitivity of these returns to variations in market returns may not be "high". Still presented differently, most of the risk inherent in these young, high-growth firms may be of a non-systematic (idiosyncratic) nature and thus easily diversifiable in a portfolio. These risks may include production problems, marketing and logistics constraints, etc., most of which are unconnected to the performance of the market. No extra returns should then be expected to carry unsystematic risk, as underlined in the finance literature.

 

The nature of the risk has important methodological implications for the choice of the returns adjustment procedure. If the systematic risk is close to 1.0 on average, a simple market adjustment is acceptable. On the other hand, if beta is statistically different from 1.0, an additional risk adjustment is required, using some form of risk/return model, usually the CAPM.

 

Hypothesis 2: Abnormal returns can be observed for supergrowth firms before, upon, or after the Inc.100 rankings publishing date.

 

Hypothesis 2 is essentially testing the market efficiency concept. In an efficient capital market, stock prices correctly reflect all publicly available information, so that changes in stock prices around information announcements (such as the publication of the Inc.100 rankings) provide an unbiased assessment of the economic effect of the event on the target/acquiror company's shareholders [Schwert (1984); Brown and Warner (1980,1985)]. Furthermore, companies only earn a "normal" rate of return over the long-term, where normal is defined with respect to their respective risk category.

 

Hypothesis 3: Cumulative abnormal returns post-ranking can be related to growth-related variables, both historical and forward-looking.

 

Hypothesis 3 extends the results obtained while testing hypothesis 2 if significant abnormal returns are highlighted there. It is an attempt to determine the factors driving abnormal returns over the long-run, factors which could include prior growth rates, growth in employees, growth in net income, market-to-book ratio, price/earning multiple, etc.

Top of Page
Previous Page | Next Page


1997 Babson College All Rights Reserved
Last Updated 4/12/97 by Cheryl Ann Lopez & Dennis Valencia

To sign-up for the Center for Entrepreneurial Studies' publication lists,
please register with the
Entrepreneurship WebTeam.