What Is The Best Portfolio Size For Value Investing?

July 21, 2015

by John Alberg and Michael Seckler

PDF | Page 2

Einhorn Tells Investors: Tesla Is Gaming S&P 500 Index Committee

david einhorn, reading, valuewalk, internet, investment research, Greenlight Capital, hedge funds, Greenlight Masters, famous hedge fund owners, big value investors, websites, books, reading financials, investment analysis, shortselling, investment conferences, shorting, short biasThe Federal Reserve has poured unprecedented levels of stimulus into the U.S. economy to deal with the pandemic, and most experts agree that inflation is just around the corner. David Einhorn has positioned his Greenlight Capital to benefit from inflation when it arrives. Q2 2020 hedge fund letters, conferences and more SORRY! This content is Read More

Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.

Many traditional value investors have concentrated portfolios of less than 50 names. Many index funds that tilt toward value factors have portfolios that consist of hundreds of names. This begs the question: What is the best portfolio size for value investing?


We performed a simulation to examine portfolios of various sizes where holdings were selected on the basis of their earnings yields. Earnings yield here is a company’s trailing annual earnings before interest and taxes, divided by its enterprise value. In order to show a distribution of simulated outcomes, each portfolio size reflects 100 simulations using a random sample as described in the appendix. In the charts, EW Universe represents a simulation of the performance of an equal-weight allocation to each stock in the entire universe used in the simulations.

Here are some of our findings:

  1. While not shown on these charts, the simulated equal-weight return of approximately 12% is higher than what the market-cap weighted S&P 500 actually delivered across the 40 years of this study. We are intrigued by this result, particularly given the mass of investors who hold market-cap weighted indices. Here is a paper that looks at the performance of equal-weight portfolios.
  2. Across the 40-year simulation period, any effort to select cheap companies and eliminate the most expensive companies would have raised the expected return above what would have been achieved by equally weighting (or market-cap weighting) assets across the entire market.
  3. If using earnings yield alone, you would have wanted to own at least 15% to 20% of this universe (or 200 to 300 names) to maximize your expected investment performance. The implication is that you would need to have some more powerful perspective about a company than its inexpensiveness (as measured by earnings yield) in order to intelligently hold a more concentrated portfolio.
  4. And, these results beg an important question: What is it about the most inexpensive companies that pulls down the median return of the most concentrated portfolios?

At Euclidean, we manage a portfolio with approximately 50 holdings. To do this with confidence, we invested our energy in understanding what gives companies their value. Our goal is to avoid companies that are deservedly cheap and to invest in companies where their market price underappreciates their intrinsic value.

PDF | Page 2

Remember, if you have a question or comment, send it to [email protected]