In search of information content: portfolio performance of The 100 Best Stocks to Own in America

Financial Services Review, Summer 2005 by Anderson, Randy I, Loviscek, Anthony L

Table 5 reports the results of the tests. In the case of the first test, none of the portfolios registers a probability value, or p-value, less than 5%. The top five stocks register the best performance, beating the S&P 500 in five of the six editions and showing ap-value of 12.5%. The worst performance is by all 100 stocks, with a p-value of 62.3%. The top 10 and top 20 stocks register p-values of 49.9% and 54.1%, respectively, and the EGP stocks produce a p-value of 23.2%. These results indicate that the individual investor's best bet from one edition to the next-one to be placed cautiously, of course-lies with the top five stocks.

The portfolios fare better in the test of portfolio growth. As indicated above, the geometric mean of the Sharpe ratios of every portfolio exceeds that of the S&P 500, leading to p-value of 2.5%, which is less than 5%. This finding suggests that The 100 Best Stocks to Own In America contains some information content, and therefore is worth reading for stock selections and investment advice on developing an effective stock selection strategy. However, given the results from the first test, we recommend that the individual investor focus on the top five stocks. They have the largest geometric mean, exceeding that of the S&P 500 by 40% (i.e., 0.278% vs. 0.199%), and they outperform the S&P 500 in five of the six editions. None of the remaining portfolios appears to offer convincing, market-beating results, an observation that is consistent with the hypothesis that guides this study.

5. Conclusions

Gene Walden's The 100 Best Stocks to Own in America is an enduring and lucid reference for the active individual investor. Now in its seventh edition, the book has reportedly sold over 300,000 copies through the first six editions, indicating that it might contain information content, or stocks that can beat a broad market index on a risk-adjusted basis. Does it? As a response, we set forth the following hypothesis: portfolios of Walden's stock selections do not consistently outperform a broad market index on a risk-adjusted basis. To test this, using the Wilcoxon signed-ranks statistic, we compare the out-of-sample Sharpe ratios of 30 portfolios constructed from the first six editions of Walden's rankings to the Sharpe ratios of the S&P 500. The 30 portfolios consist of the top five, 10, 20, and all 100 stocks from each edition. They also include portfolios constructed from the Elton-Gruber-Padberg procedure, which is based on Markowitz's mean-variance criteria.

We find some evidence of information content and suggest that the individual investor focus on the top five stocks. The remaining portfolios do not appear to offer risk-adjusted returns that consistently exceed those of the S&P 500. All things considered, we conclude that The 100 Best Stocks to Own in America should be classified among the selected trade publications, investment newsletters, and media reports that have the potential to help the individual investor beat a broad market index on a risk-adjusted basis.