Close

Investment Characteristics of Stock Market Winners

by Marc Reinganum

Investment Characteristics Of Stock Market Winners Splash image

For many years, academic research supported the idea that investors cannot consistently outperform a simple strategy of buying and holding a diversified portfolio of common stocks without taking on greater risk. This implies that technical and fundamental research based on publicly available information would, at best, improve investment performance only marginally, and throwing darts to select stocks would most likely be just about as effective.

Serious chinks in this investment view have appeared recently, as several studies have found stock market anomalies—investment characteristics that consistently produce returns above those that would be expected based on the investment's level of risk. For instance, studies have indicated that stocks with low price-earnings ratios outperform those with high price-earnings ratios, and stocks with small market capitalizations outperform large-capitalization companies. Other anomalies characterizing peculiar patterns in the timing of stock returns also emerged, ranging from a month-of-the-year effect to a week-of-the-month effect to a day-of-the-week effect and even down to an hour-of-the-day effect. These anomalies point to a similar conclusion: Investors may be able to beat stock performance benchmarks using publicly available information.

Stock market winners may also yield some insights into anomalies. For this study, we singled out stocks with exceptionally high returns to see whether these firms share any common attributes. If history does repeat itself, these common attributes may suggest some profitable investment strategies.

The Study

The companies chosen for the study were based on companies in "The Greatest Stock Market Winners: 1970-1983," published by William O'Neil & Co. To be considered a great winner, a company typically had to at least double in value within a calendar year, although there were a few exceptions to this guideline, and not all companies that doubled in value were selected. Historical fundamental and technical information on these firms was taken from the Datagraph books (also published by William O'Neil & Co. and sold primarily to institutional investors).

Returns were measured based on the price appreciation of the winners between their buy and sell dates. On average, the 222 winners increased in value by 349%. While this average was buoyed by a few winners with astronomical increases, more than half the firms increased in value by at least 237%.

The variables we examined were divided into one of five categories. The first, "smart money," includes the behaviors of professionally managed funds and corporate insiders. The second contains valuation measures such as price-to-book-value and price-earnings ratios. The third grouping includes the technical indicator of relative strength. The fourth consists of accounting earnings and profitability measures. The final group contains miscellaneous variables that did not fit into the other four groups.

Smart Money Variables

The smart money variables reveal the stock holdings of professionally managed investment funds and corporate insiders. Even if they are not clairvoyants, money managers and corporate insiders are probably well-informed. We broke professionally managed funds down into four groups—investment advisers, banks, mutual funds and insurance companies. For each of these groups, Datagraph reports the number of institutions holding a particular issue as well as the aggregate holdings of these institutions as percentages of the outstanding common stock.

Table 1 summarizes the results. We can draw several general observations from the professionally managed funds as a whole. If hindsight were foresight, one would like to know of impending significant increases in the sponsorship of stock held by banks, mutual funds and investment advisers. Between the buy and sell dates, these three groups of professionally managed funds increased their average ownership stakes in the 222 winners by 25%, 60% and 107%, respectively. At least at the conclusion of the rapid price advance, these funds were where the action was. Prior to the buy quarter, the ownership claims of these managed funds tended to rise only slightly. Thus professional money managers may participate in, but do not prophesy, extraordinary price appreciation.

Corporate insiders form another group that may be privy to inform

Smart Money Variables Average Median
Number of Investment Advisers Owning Shares 9.3 0
Percent of Outstanding Stock Held by Investment Advisers 7.2 0
Number of Insiders Buying Stock 0.37 0
Number of Insiders Selling Stock 1.38 1
     
Valuation Measures    
Price/Book Value 0.95 0.6
Price/Earnings 13.6 10
Share Price ($) 27.69 24.07
Stock Market Capitalization ($ millions) 484.3 120.1
Stock Beta 1.14 1.14
     
Technical Indicators    
Relative Strength Rankings (99 = Highest, 1 = Lowest) 90.2 93
     
Earnings and Profitability Measures    
Pretax Profit Margins (%) 12.7 11.2
Changes in Quarterly Earnings (%) 45.9 7.4
Changes in Quarterly Sales (%) 9.5 7.3
Five-Year Earnings Growth Rates (%) 23 17
     
Miscellaneous Variables    
Common Shares Outstanding (in thousands) 13,885 5,740
Ratio of Buy Date Price to Maximum Price During 2 Previous Years 0.899 0.922

 

ation about a company's prospects. Will tracking their transactions lead to profitable trading? Several prior studies have suggested that it may. However, the data among the winning stocks does not indicate any great changes in the pattern of insider trading.

For most companies, no corporate insiders bought stock either prior to the large price advance or after it. Selling transactions seem equally uninformative. One might expect insider selling to subside prior to the major price advance. In fact, insider selling of these 222 companies actually increased slightly before the advance, rising from an average of 0.84 insider sales per company to 1.38.

While the smart money variables may reflect the actions of well-informed investors, the evidence suggests that they do not predict major price advances.

Valuation Measures

Five different valuation variables were examined: price-to-book-value ratio, price-earnings ratio, stock price level, stock market capitalization, and beta. Table 1 provides a brief summary of the findings.

Price-to-book-value ratio compares the market value of the stock to its book value. A ratio less than 1.0 indicates that the market value of a company is less than its book value, suggesting that the stock is underpriced.

Among the 222 winners, 164 were selling for less than book value in the quarter in which the buy occurred. The median price-to-book-value ratio was 0.60, while the average ratio was 0.95. (The median is the exact mid-point, where half of all ratios fall above and half fall below; it is used because it is less influenced by extreme values.) While a price-to-book-value ratio of less than one may not be a perfect indicator of a stock market winner, it does seem to be a common characteristic. This suggests an investment strategy that isolates firms selling below book value.

Start Your FREE 30-day AAII Trial Membership Now!
Get full access to AAII.com, including our market-beating Model Stock Portfolio, currently outperforming the S&P 500 by 4-to-1. Plus 60 stock screens based on the winning strategies of legendary investors like Warren Buffett and Peter Lynch, unbiased investor education with our award-winning AAII Journal, sought-after Investing Guides and more – completely FREE for 30 days.
sign-up

The distribution of price-earnings ratios provided quite different results. In the buy quarter, the average price-earnings ratio equaled 13.6, while the median was 10. In general, the price-earnings ratios for the set of winners do not tend to be very small. In fact, only one out of every 10 of these firms had price-earnings ratios of less than 5 in the buy quarter. This indicates that very low price-earnings ratios are not a necessary ingredient of a successful investment strategy.

The results also indicated that the winners are not characterized by either low stock prices or small stock market capitalizations (number of shares times price per share). The median share price on the buy date was $24.07, while the average was $27.69. The median market capitalization was $120.1 million—a figure that falls in the upper half of New York Stock Exchange and American Stock Exchange-listed stocks. Only one of the 222 winners had a market capitalization less than $10 million, and only 12 had capitalizations less than $20 million. This suggests that small size, whether measured by share price or stock market capitalization, is not a necessary component of a successful investment strategy.

The betas of the stock market winners were examined to see if their extraordinary rates of return might be compensation for riskiness. Beta measures the variation of a stock relative to the stock market. The stock market has a beta of 1.00; a beta greater than 1.00 implies that when the stock market changes, the stock will change to a greater degree and is therefore riskier. The average and median beta of these firms was 1.14, and fewer than 5% had betas greater than 2.00. While the firms as a group were slightly riskier than the market as a whole, the additional stock market risk cannot account for the extraordinary returns of these winners.

Technical Indicators

The technical indicator measured among the stock winners was relative strength (see Table 1). The relative strength of a stock is the average of quarterly price changes during the previous year, but giving more weight (40%) to the most recent quarter and less weight (20% each) to the three other quarters; these average changes are then ranked among all stocks, ranging from 1 (lowest) to 99 (highest). Among the winners, the median rank in the buy quarter was 93; fully 212 of the 222 firms possessed relative-strength rankings of greater than 70. In addition, the relative-strength rankings for 170 of the 222 winners increased between the quarter prior to the buy and the quarter during which the stock was purchased. These findings have two implications for investment strategy: First, investors should seek out firms with high relative-strength rankings; and second, investors should try to identify firms that exhibit a positive change in their relative-strength ranking from the prior quarter.

Earnings and Profitability Measures

Several measures of earnings and profitability among the stock market winners were examined: pretax profit margins, changes in quarterly earnings, changes in quarterly sales, and five-year quarterly earnings growth rates. Table 1 provides a summary of the results.

The average pretax profit margin in the buy quarter was 12.7%. In the quarter prior to the buy, the profit margin was slightly smaller, while by the sell quarter, the average profit margin had increased to 14.5%. In fact, the nearly 2% increase in the pretax profit margins may have contributed to the significant price appreciation of these firms. Fully 216 of the 222 winners had positive pretax margins in the buy quarter and 215 had positive pretax margins in the quarter prior to the purchase. This evidence clearly indicates that a high, positive pretax profit margin should be one of the selection screens in an investment strategy.

On average, quarterly earnings in the buy quarter rose nearly 45.9% from the previous quarter; these were not seasonally adjusted and they represent changes in the raw accounting earnings. Interestingly, quarterly earnings in the quarter prior to the buy increased an average of 60.8%, while in the quarter prior to that they increased an average of 50.4%—in other words, there was an acceleration in quarterly earnings. Thus, another investment rule suggested by the winners is to seek out firms with a positive change in quarterly earnings—earnings acceleration.

The pattern of changes in quarterly sales closely parallels that of changes in quarterly earnings. During the two quarters prior to the buy, quarterly sales were positive and increasing. During the buy quarter, quarterly sales on average increased 9.5%.

A longer-term picture of earnings growth was gleaned from the five-year quarterly earnings growth rates. These rates were determined using five years of quarterly earnings data that were annualized. In the buy quarter, the earnings growth rates averaged 23.0% while in the quarter prior to the buy, they averaged 21.6%. By the sell quarter, the average earnings growth rate increased dramatically, to 38.2%. This is most likely due to the method of calculating the growth rates, which discards the low rates from the prior years and replaces them with the high earnings growth rates experienced during the period of dramatic stock price appreciation. During the quarters prior to the quarter just before the buy, the average five-year rates remained stable. However, during the quarter just before the buy, over 85% of the firms exhibited positive five-year earnings growth rates. This suggests that investors should select companies that have positive five-year quarterly earnings growth rates.

Cumulative Excess    
Holding-Period Returns* Average Median
First Quarter following purchase 3.04% 1.40%
Second Quarter following purchase 8.19 4.4
Third Quarter following purchase 12.65 7.3
One Year following purchase 16.67 9.7
Fifth Quarter following purchase 20.84 12.7
Sixth Quarter following purchase 26.1 15.6
Seventh Quarter following purchase 31.13 18.5
Two Years following purchase 37.14 22.3

 

Miscellaneous Variables

Two variables that didn’t fit into the other categories were also examined: the number of common shares outstanding and the ratio of the price on the buy date to the maximum price during the two previous years.

The average stock market winner had 13.8 million outstanding shares during the buy quarter; the median was a much lower 5.7 million. During the sell quarter, the average and median number of outstanding shares nearly doubled, most likely indicating that many of the firms split their shares of stock during the time of rapid share price increases. However, the most meaningful statistic for investors is that nearly 90% of the firms had fewer than 20 million shares of stock outstanding. Investors may want to select companies with fewer than 20 million outstanding shares of stock.

The ratio of the price on the buy date to the maximum price during the two previous years provides a measure of whether these firms had fallen out of favor among investors. This ratio measures the extent to which the extraordinary success of these 222 winners might have been captured by a contrarian investment strategy of selecting stocks that have suffered substantial price declines. However, the results indicate that a contrarian rule would not have led to the selection of these stocks. On the buy date, more than half the winners were selling within 8% of their previous two-year highs, and only one was selling at a price of less than half its previous two-year high. More than 80% of the firms were selling within 15% of their previous two-year highs. Thus, an investment strategy that selects stocks selling within 15% of their two-year highs would capture a common characteristic of these winners.

A Test of Possible Screens

Given the number of variables examined, there are myriad potential investment strategies that could be developed. To get an idea of the usefulness of some of these screens, we tested one strategy that employed four of the variables. The four chosen were those variables that produced the highest median returns when each of the variables were used as screens for the 222 winners. The four screens were:

  • A price-to-book-value ratio of less than 1.0;
  • Accelerating quarterly earnings;
  • A relative-strength ranking of the stock in the current quarter greater than the ranking in the previous quarter;
  • Fewer than 20 million common shares outstanding.

These screens were applied against a universe of 2,057 stocks listed on the New York Stock Exchange and the American Stock Exchange over the 1970 through 1983 period; the 222 winners were excluded from this list. Stocks were “purchased” 63 days after a buy signal, which ensured that accounting information assumed known had actually been released. The stock was held for two years, and cumulative holding period returns for each stock were determined and compared against the cumulative returns for the S&P 500 over the same time period. The difference between the stock return and the S&P 500 return is the “excess” return generated by the screen.

Table 2 provides the performance results for this strategy, which did quite well relative to the S&P 500. After one year, the selected firms on average provided holding-period returns that were 16.67% above those of the S&P 500; after two years they provided on average holding-period returns that were 37.14% above those of the S&P 500.

Other tests implied that increasing the number of investment rules suggested by our investigation of stock market winners would improve performance. These results should not be construed to mean that these four investment screens are the best four filters. They do, however, illustrate that the lessons learned from an examination of the biggest winners may be profitably applied to a broader universe of firms. In addition, it seems unlikely that any one of the investment rules will yield better performance than all jointly.

Investors may also want to take note of the absence of certain characteristics from the trading strategies mentioned. These strategies do not exploit characteristics that prior research has revealed to be associated with superior performance. The strategies are not tilted in favor of stocks with very small market capitalizations. Nor are firms with low share prices singled out, or those with low price-earnings ratios. The strategies are not contrarian in the sense that companies with substantial previous price decliines are selected. Despite the absence of these characteristics, the trading strategies produce excess returns that are economically significant. This suggests that there may be more than one way to skin the performance cat!

Marc Reinganum is the Phillips professor of finance in the College of Business Administration at the University of Iowa. This article was adapted from an article that appeared in the March-April 1988 issue of the Financial Analysts Journal.


Discussion

No comments have been added yet. Add your thoughts to the discussion!

You need to log in as a registered AAII user before commenting.
Create an account

Log In