Which U.S. Stocks Will Outperform?

Reward PrizeWhile low P/E stocks, also known as value stocks, tend, on average over the long run, to outperform, my research shows that about 39% of all low P/E stocks have a negative return for the 12 months following their selection. In this article, I want to determine the U.S. value stocks that are most likely to outperform and those that are most likely to underperform over the next 12 months.

Canadian stocks will be examined in a separate article at another day.

I used data from COMPUSTAT and examined positive P/E stocks, excluding AMEX companies, high business risk companies, such as Software & Services, Semiconductors & Semiconductor Equipment, Transportation, Automobiles & Components, Real Estate/Construction Materials and Pharmaceuticals, Biotechnology & Life Sciences Capital Goods, and companies that had reported extraordinary items the year before. These stocks were excluded as I found that over the past 40 years they tended to have lower returns than the rest of U.S. stocks. I also excluded stocks that traded for less than $1.

At the end of April 2013, I sorted all NYSE and NASDAQ stocks by trailing P/E ratios from low to high and formed quartiles. Value stocks, those of interest in this article, were those that fell in the lowest quartile.

These stocks were then sorted by the following historical financial metrics (using the most recently reported financials): market cap,  annual trading volume to shares outstanding, assets to revenues, total debt to equity, cash to assets and year-over-year EBIT annual growth rate, one financial metric at a time. A SCORE was given to each firm-specific metric based on its rank.  All individual SCOREs per firm were then added to determine an overall SCORE for each firm.  Finally, firms were sorted by SCORE and seven portfolios of firms were formed with SCOREs from low (SCORE 0) to high (SCORE 6). The way the indicator was constructed implied that the lower the SCORE the better it was.

The tables below show the U.S. (non AMEX, non-high business risk and non-reporting extraordinary items) value stocks that received SCOREs 0 and 1 and those that received SCOREs 5 and 6 as at April 30, 2013. In other words, they identify the value stocks that will most likely outperform (i.e., SCOREs 0 & 1) from May 1, 2013 to April 30, 2014 and those that will most likely underperform (i.e., SCOREs 5 & 6).  It is interesting to note that there are more higher SCORE (possibly underperforming) stocks than lower SCORE (possibly outperforming) stocks, which may imply something about the stock market valuations in the US.

As a note, when I back tested this portfolio formation strategy covering 5,509 unique companies for the period May 1, 1969 to April 30, 2011, I found that value firms with the lowest SCORE had a mean annual return of 54.38%, while the highest SCORE value firms had a mean annual return of 13.32%. For comparison, the mean annual return of all value stocks in my sample was 22.36%. Clearly, the lowest SCORE stocks outperformed while the highest SCORE stocks underperformed.

GA 1

GA 2

GA 3

GA 4