Dispersion and the ability of fund managers to outperform
Yesterday, I discussed how investors who focus on correlation rather than dispersion underestimate the benefits of international diversification. But if we go one level deeper and look at the dispersion and correlation of stock returns within a given universe, dispersion really becomes important.
Since 2010, the drive towards passive investment products has been relentless and actively managed funds as well as hedge funds have had little to show for their efforts. Not only did the average actively managed fund underperform its benchmark after fees, but even funds with a long record of beating their benchmarks had performance issues.
One reason why performance for actively managed funds and equity-related hedge funds may have been so poor was the low dispersion amongst stocks. Our chart shows the dispersion amongst the stocks in the S&P 500 in comparison to the average correlation between these stocks. If return dispersion between stocks is low, no matter how good a stock picker you are and no matter how uncorrelated stock returns are, you will have a hard time picking stocks that will create sufficient outperformance over the index that can exceed the fees you charge.
A study by Anna von Reibnitz from the Australian National University showed that the performance difference between the least active and most active fund managers is negligible when return dispersion is low or average. She looked at the performance of US equity funds and split the market environment in five quintiles of dispersion. The three quintiles with the lowest dispersion show no significant performance difference between highly active funds and closet indexers, and the actively managed equity funds on average underperformed the S&P 500. This is exactly the environment that prevailed from 2010 to 2018, with the exception of two brief spikes in dispersion in 2011 and late 2015.
Once return dispersion enters the second highest quintile, returns of actively managed funds start to pick up, but only for the most actively managed funds. The 20% most active and most concentrated funds manage to outperform the index by about 0.5% p.a. Once markets become highly dispersed and return dispersion is in the highest quintile, the most active funds start to outperform the index by 4.8% p.a. to 9.2% p.a. while the close indexers continue to underperform the index. Overall, the return difference between the 20% most active and the 20% least active funds increases to 9.9% p.a. when return dispersion is high.
Similar results were reported by David Smith for equity hedge funds. Depending on the adjustments for style, size etc. used to calculate hedge fund alpha, equity hedge funds created an average annual alpha over the benchmark of 2% to 4% p.a. when return dispersion was high compared to an alpha below 0.7% p.a. when dispersion was low. Particularly large return differences between regimes of high and low return dispersion were found for hedge funds focusing on large cap stocks.
What this means is also that investors should judge active manager performance not only versus a benchmark but also based on the market environment and the opportunity set it creates. Low correlation between stocks in this case is of little help in creating benefits for investors. Instead, investors want to assess fund performance given the prevailing dispersion of stock returns. And in this regard, neither actively managed funds nor hedge funds had much opportunity to differentiate themselves in the last eight years or so.
But with the recent market turmoil that situation could change. In the fourth quarter of 2018, return dispersion increased at the same time as correlation between stocks increased. We are not quite in the top quintile of return dispersion but close to it, and in this environment the most active and focused funds should start to differentiate themselves again. It will be one of the interesting developments to watch in 2019, to see whether this environment of high return dispersion can prevail for longer and if fund mangers can take advantage of it.
Correlation vs. dispersion amongst S&P 500 stocks
Source: Bloomberg, Fidante Capital.