Skip to main content

Common wisdom has it that sunny summer afternoons are best enjoyed on the beach, picnicking, or otherwise frolicking in the great outdoors. Such activities work for balanced souls, but I think there’s nothing better than spending the day exploring financial databases and testing investment strategies.

I recently indulged in my passion by focusing on back-testing concentrated value strategies using the S&P 500 index of large U.S. stocks. The idea being to sort the 500 stock index by a value ratio, select an equal-dollar amount of the 20 stocks with the lowest ratios, hold them for a year and then repeat the process to see how the portfolio fared over the long term.

For instance, a portfolio of 20 stocks with the lowest price-to-earnings ratios (P/E), rebalanced each year, generated compound annual returns of 14.7 per cent over the 25 years through to the end of 2017. It beat the index, which climbed at a 9.7-per-cent annual rate over the same period.

(All of the returns figures herein are quoted in U.S. dollar terms and include dividend reinvestment. They do not include fund fees, commissions, taxes or other trading frictions.)

Eight different value ratios were put to the test and the results presented in the accompanying table. More specifically, 20-stock portfolios were formed based on low price-to-earnings, price-to-book-value (P/B), price-to-tangible-book-value (P/TB), price-to-sales (P/S), price-to-cash-flow (P/CF), price-to-free-cash-flow (P/FCF), enterprise-value-to-earnings-before-interest-and-taxes (EV/EBIT), and enterprise-value-to-earnings-before-interest-taxes-deprecation-and-amortization (EV/EBITDA) ratios. In addition, a 20-stock high-dividend-yield portfolio was also investigated.

The value portfolios won in each and every case over the full 25-year period. The high-yield portfolio was the weakest with an average annual return of 10.6 per cent, which isn’t too surprising because dividend yield has historically been a relatively weak value factor in the United States. The low-P/B portfolio also fared relatively poorly with returns of only 11.1 per cent annually, which is unfortunate because the ratio is used by many value-oriented index funds and exchange-traded funds.

On the other hand, some of the other value portfolios fared quite well including the aforementioned low-P/E portfolio. The best result was posted by the low-EV/EBIT portfolio with average annual gains of 16.5 per cent. It’s a remarkable performance because the period included one of the biggest stock market calamities of the past century.

The cheery 25-year results might whet your appetite for value stocks. But before you dive in, it’s instructive to consider the most recent 10-year period spanning from the end of 2007 to the end of 2017. The table shows the grim 10-year results.

While the S&P 500 gained an average of 8.5 per cent annually over the past 10 years, four of the nine value portfolios trailed the index and, in some cases, by a large margin. For instance, the low-P/CF portfolio gained just 3.8 per cent annually, the high-yield portfolio gained 4 per cent annually, while the classic low-P/B portfolio trailed the market with annual gains of 5.8 per cent.

The miserable results came with a good deal of extra effort. Just imagine how discouraging it would have been to trail do-nothing buy-and-hold index investors despite being active in the market year after year for a decade.

Mind you, better results were obtained from the other value portfolios. For instance, the low-P/E portfolio gained a respectable 11 per cent annually over the past decade while the low-P/S portfolio fared surprisingly well with average gains of 15.6 per cent annually.

The results also provide a sense that the success of any particular value ratio varies over time. It’s a big reason why many quantitatively oriented investors use multiple value factors when looking for promising stocks. For instance, the average low-ratio portfolio outperformed the index by 4.4 percentage points annually over 25 years and by 0.8 of a percentage point annually over the past 10 years.

While it was a hard decade for many value investors, such periods happen from time to time. Hopefully the next 25 years will be similar to the past 25 years rather than the past 10. If they aren’t, it might have been better to spend a few more days at the beach.

Norman Rothery, PhD, CFA, is the founder of

Concentrated U.S. Value Portfolio Performance

10 years25 years
PortfolioAnnual Return (%)vs Index (pp)*Annual Return (%)vs Index (pp)*
S&P 5008.59.7
Low P/E11.02.514.75.0
Low P/B5.8-2.711.11.4
Low P/TB10.
Low P/S15.
Low P/CF3.8-4.712.52.8
Low P/FCF8.2-0.312.73.0
Low EV/EBIT10.41.916.56.8
Low EV/EBITDA9.30.815.35.6
High Div. Yield4.0-4.510.60.9

*Percentage points; Source: Bloomberg

Report an error

Editorial code of conduct