Anomaly Sensitivity To Business Conditions
In a recent edition of this journal, Erickson, Fredman, and Stickels documented the greater importance of the price/earnings ratio relative to the price/sales ratio. They concluded that the price/sales ratio should be used only with great caution and only by experienced analysts. The caution with which investment analysts should use the price/earnings ratio, firm size, and Timeliness rankings criteria, in light of concurrent business conditions, is addressed in this article.
A variable may have explanatory power only in the presence of certain business conditions. Small firms may do extremely well in an expanding economy but show lackluster performance during business contractions. Alternatively, Value Line may find the few start performers easier to identify in a recession than the best of the advancing crowd during periods of economic growth. If the market's condition has no impact on anomaly explanatory power, then the significance of chosen anomaly characteristics would remain unchanged.
Specifically, this paper reports results of an empirical analysis of the relationship between firm size, price/earnings ratios, and Value Line Timeliness rankings under alternative business conditions. Firms continually followed by Value Line over the 1975 through 1984 period form the sample set. This set was chosen because it consists of corporations for which information was widely disseminated. Therefore, it would tend not to be sensitive to the firm neglect/informational deficiency hypotheses put forth by Arbel and Strebel and Arbel.
The next section gives a brief overview of the investments literature related to the anomalies under consideration. A more detailed listing of anomaly research is available in Table 1 of Jacobs and Levy. The third section will provide the research methodology. Results are then presented and a conclusion is derived.
OVERVIEW OF THE RELEVANT ANOMALY LITERATURE
The Firm Size Effect
Among the most commonly studied anomalies have been the firm size, price/earnings ratio, and Value Line's Timeliness rankings. A continuous line of research dating back to Banz suggests that small firms, on average, earn higher risk-adjusted returns than do larger firms. The size term was found to have roughly the same statistical significance in explaining returns as did beta. The primary payoff came from holding the smallest twenty percent of New York Stock Exchange stocks. Furthermore, over the 1936-1975 period, the average annual return differential from buying very small firms versus large firms was 19.8 percent.
Reinganum[26, 27] demonstrated that differences in trading frequency, tax-loss selling, or differences in the bid-ask spread between large and small firms did not fully explain the January effect. Price/earnings ratios, dividend yield, standard deviation of return, and coskewness were unable to explain the negative relationship between firm size and subsequent returns.
The Price/Earnings Ratio Effect
According to Elton and Gruber, Wall Street folklore advocated buying firms with low price/earnings ratios long before Basu rigorously tested the P/E anomaly. Exaggerated optimism was believed to be associated with high P/Es and exaggerated pessimism supposedly led to low P/E figures. Subsequent price adjustments, as investor perception becomes more moderate, supposedly produces an inverse relationship between P/E and security returns.
Abnormal returns of the two lowest P/E quintile portfolios were positive in at least fourteen of eighteen months following portfolio formation in Basu's original sample. Basu's sensitivity analysis indicated that the abnormally good performance of low P/E stocks was insensitive to the choice of either the Sharpe-Lintner version or Black's zero-beta version of the Capital Asset Pricing Model. After controlling for differences in firm size, Basu provided evidence that the common stock of low P/E ratio firms earn, on average, higher risk-adjusted returns than the common stock of high earnings capitalization firms. …