Skip to main content

The Globe and Mail

Can a factor-based strategy emulate styles of Buffett and Soros?

Warren Buffett of Berkshire Hathaway

Nati Harnik/The Associated Press

John Reese is CEO of Validea.com and Validea Capital, and portfolio manager for the National Bank Consensus funds. Globe Investor has a distribution agreement with Validea.ca, a premium Canadian stock screen service. Try it.

A new body of research shows being an extraordinary investor such as Warren Buffett or George Soros may in fact be possible for other investors if they use a few tried and true yardsticks to evaluate stocks.

The report, titled Alternative Thinking: Superstar Investors, was produced by AQR Capital Management, the program-driven investment firm. The researchers compared returns for superstar investors to portfolios it constructed with a small set of buy and sell signals that tracked the investing styles of Mr. Buffett, Mr. Soros and others. It managed to get results that were pretty close to the real thing. The conclusion of AQR's research is investors don't have to pick exactly the same stocks as the masters to generate benchmark-beating returns.

Story continues below advertisement

Factor and style investing are gaining in prominence with the explosion of interest in exchange-traded funds and quantitative strategies that threatened to replace active managers. The ETFs are an easy way for investors to track specific styles such as value, momentum or low volatility, or factors such as balance-sheet stability.

Mr. Buffett, who is best known for picking undervalued stocks of companies he views as long-term industry leaders and largely sticking with that strategy over years and decades, may not rely on a concrete set of factors, per se, but the stocks he buys fit a pattern. First off, he looks for companies that have a long-term competitive advantage in their industries, such as a strong brand or a dominant market share, with earnings that are predictable. Berkshire Hathaway has returned 17.6 per cent a year since 1977, twice as much as the broad Standard & Poor's 500 over that time.

AQR developed a portfolio that tracked Mr. Buffett's style broken down by factors such as value, quality (stocks of companies with the strongest balance sheets), low volatility (smallest price swings) and the performance of the market as a whole. This hypothetical portfolio run over 40 years would get within four percentage points a year [the graph in the AQR report says 3.6 if you want to be more precise] of Mr. Buffett's own results over that period, the research said.

But how does one define value? One way to look at it is observing that companies with high book-to-market ratios, or value stocks, outperform those with low ratios, or growth stocks. Traditionally these ratios are calculated using 18-month-old price and book data that don't take into account the most recent price movements. AQR developed a formula that uses more recent pricing data to more accurately forecast the ratio.

Another approach is to look at data over a long period of time and to stack fundamentals criteria on top of each other. To construct our Buffett-inspired portfolio at Validea, we look for companies that have 10 years of consistent earnings, 10 years of higher than average return on equity as well as the ability to pay off debt, increasing share buybacks and other measurements. Integrating multiple factors and selecting only the best stocks that meet those investment criteria, such as those mentioned above, may be a more real-world way that someone such as Mr. Buffett may actually go about picking stocks. In the U.S., my 10-stock Buffett-based portfolio has returned 141.1 per cent since 2003, beating the market by more than 28 per cent. In Canada, it has done even better. The Buffett model is the No. 1 ranked performer of all my guru-based models in Canada since we launched them in August, 2010, with a return of 137.5 per cent versus 29.5 per cent for the S&P/TSX composite index.

AQR also looked at Mr. Soros, whose Quantum Fund generated 20 per cent annually from 1985 to 2004. AQR ran a portfolio based on his style, breaking it down by value (for currencies), momentum (for currencies), trend (for stocks and for asset classes), and also the broad market and got pretty close to the same returns just by tracking similar types of investments.

Interestingly enough, AQR also looked at the great Fidelity fund manager, Peter Lynch, and the results weren't as strong as their results for Mr. Buffett and Mr. Soros. Mr. Lynch, who produced an annual return of 20.8 per cent from May, 1977, to May, 1990 (compared to 7.1 per cent for U.S. equities), seemed to have a tilt toward small caps and momentum, according to the research. In the report, the authors point out that Mr. Lynch "had various checklists for various categories of companies." Compared to the portfolio AQR tried to simulate, Mr. Lynch's still generated outperformance of 8 per cent each year.

Story continues below advertisement

Factor-based investing is evolving rapidly, and the AQR research, as well as that of others, is providing long term data to support the view that certain types of factors, and the integration of factors, can drive long term performance. One takeaway for investors is that success isn't luck or chance but the reward for investing long-term using fundamental factors that have a proven track record of delivering returns. That isn't to take away from the great success of Mr. Buffett or Mr. Soros or even Mr. Lynch.

"These great investors figured it out first, had the ability to stick to their philosophies, and rightly deserve their reputations," the firm said in a recent research note. Easier said than done for most investors.

Report an error
About the Author

John Reese is CEO of Validea.com and Validea Capital, the manager of an actively managed ETF. Globe Investor has a distribution agreement with Validea.ca, a premium Canadian stock screen service. More

Comments

The Globe invites you to share your views. Please stay on topic and be respectful to everyone. For more information on our commenting policies and how our community-based moderation works, please read our Community Guidelines and our Terms and Conditions.

We’ve made some technical updates to our commenting software. If you are experiencing any issues posting comments, simply log out and log back in.

Discussion loading… ✨