<!-- Swiftype Variables -->

Portfolio Management

Are concentrated portfolios more volatile than a diversified index?

Concentration, quality key factors in portfolio construction

The perception that concentrated investment managers are riskier or more volatile than a diversified index is slowly changing. Despite academic research that dates back several decades, concentrated managers continue to encounter the instinctive belief from some investors that holding only a few securities in a portfolio will lead to more volatile performance outcomes. While more volatile performance returns certainly could be the case for any given concentrated portfolio, it is far less certain that it should be the case.

It is just as likely, if not more likely, that concentrated portfolios could deliver superior risk-adjusted returns compared to a diversified index. And this is without taking into account the types of holdings for a particular investment manager. For equity investment managers, emphasizing factors such as quality could help further reduce risk profiles without necessarily reducing performance potential. In fact, the combination of concentration and quality in equity portfolios might be one of the more optimal investment frameworks available to investors today.

Setting the stage

To test this theory, financial information and analytics provider FactSet Research Systems Inc. generated 10,000 unique, highly concentrated portfolios with stocks that would have been available from 1989 to 2014. The parameters for the experiment included:

20-stock portfolio: Maintain a truly concentrated portfolio of no more than 20 stocks at any given time.

Five-year regeneration: The portfolios were completely turned over and randomly regenerated every five years until 2014.

Equal weightings: For simplicity purposes, holdings were equally weighted in the sample portfolios.

Substantial market caps only: U.S. stocks were screened for a minimum market cap of $1 billion to account for the fact that market caps were generally lower in 1989 compared with today. Also, to factor in generally rising stock prices over the time period studied, the minimum market cap was gradually raised to $3 billion, allowing for added screening flexibility. The market cap increase was done ratably across the entire 25-year period. Any stock with an unadjusted share price of less than $1 was excluded.

The results of this experiment are shown in the following chart:

The key observation here is how the median portfolio performed relative to the S&P 500, a proxy for the performance of large-cap U.S. equities. The median randomly generated portfolio of 20 stocks produced an annualized return of 12.2% with a standard deviation of 19.6%, compared to the S&P 500's results of 10.4% and 18.3%, respectively.

Some simple math reveals the median concentrated stock portfolio was a better (i.e., more efficient) investment than the S&P 500, even when accounting for volatility. In fact, a stunning 87% of the simulated portfolios outperformed the S&P 500 and roughly 70% of the portfolios had a higher return/volatility ratio than the S&P 500 during the period studied. Even the most volatile portfolios (those in the upper right quadrant) still outperformed the S&P 500, in some instances quite significantly, despite some rather extreme volatility.

The case for concentration

As a result of the initial parameters — specifically the restrictions around market cap and country — the experiment produced a universe of stocks with a high degree of overlap with the S&P 500. However, despite the fact the S&P 500 has extremely low turnover and is cap-weighted, these randomly concentrated portfolios were still, on average, a better bet.

Intuitively, this makes sense. The big winners in the simulated concentrated portfolios would have benefited from much larger position sizes before rebalancing, and therefore would have had a more significant impact on returns relative to the weaker performing stocks, which would have become smaller position sizes due to underperformance. The fact that the annual returns of the median concentrated portfolio were only marginally more volatile than the market's returns echoes well-known research from 1968 by John L. Evans and Stephen H. Archer (“Diversification and the reduction of dispersion: an empirical analysis”) and is strong evidence against the generally held notion that concentrated portfolios are much more volatile than their diversified counterparts.

Factoring in quality

To take this experiment a step further, the same relevant data for the MSCI USA Quality index were also included, which, in its creation methodology, utilizes investment criteria such as high return on equity and low debt/equity. These guidelines are intended to narrow the universe of potential stocks to only the highest confidence investment ideas, with a focus on businesses with a competitive advantage and sustainable growth and profitability. In other words, it allows managers to fish from a pond that is already stocked with the highest-quality fish.

As shown in the exhibit, the annualized return and volatility metrics of the MSCI USA Quality index are nearly identical to the median portfolio, but still notably better than the S&P 500 on a risk-adjusted basis.

Conclusion

The data make clear that concentrated equity portfolios, even when randomly constructed, can produce very competitive returns relative to the market, without a significant increase in volatility. And despite the common argument against quality — that quality stocks tend to trade at premium valuations relative to their lower quality brethren — the experiment shows this need not be an impediment to long-term investment success. With markets more correlated than ever, investors should give appropriate consideration to concentrated equity portfolios that are composed of high-quality businesses.

Stephen Atkins is a research analyst with Polen Capital Management LLC, Boca Raton, Fla. This content represents the views of the author. It was submitted and edited under P&I guidelines, but is not a product of P&I's editorial team.