Investment Commentary – May 2013

Active Share: An Investor’s Crystal Ball?

When the California Public Employee Retirement System (CalPERS), the second largest pension fund in the US, announced in late March that the fund was contemplating abandoning active investment management – i.e. strategies aimed at outperforming the market – some considered it a very telling development in the decades-long active versus passive debate.

In our March commentary we touched on the fact that the merits of active management have long been questioned. Proponents of passive index strategies cite numerous academic studies and investor experiences that show consistently beating the market is quite difficult. It is interesting to note that although the majority of Canadian investment managers have been up to the task recently, their US and international counterparts have not fared so well. The latest S&P Dow Jones Indices Versus Active (SPIVA) report found that more than 63% of US large-cap equity managers underperformed the S&P 500 last year, while greater than 56% of international funds also fell short of their index. The apparent ineptitude of active management in these regions is further highlighted when some longer-term performance records are studied.

An optimist can examine these numbers and conclude that, on average, roughly 30% of active managers in any given market can deliver consistent, long-term outperformance. However, pessimists will note that the task of finding these managers in advance is exceedingly difficult. One issue is the fact that outperformance can be wiped out once investment management fees are considered. Furthermore, even if an active fund has achieved a track record of outpacing the benchmark after fees, it has been shown that chasing yesterday’s winners tends to be a losing strategy.

With necessity often breeding innovation, researchers at Yale University devised a relatively new metric called “active share” to help identify managers that are poised to continue outperforming. This metric is the percentage difference between a fund and the corresponding index. An active share of 0% indicates that the portfolio is identical to the index, having the same allocation to the same stocks in the benchmark. Higher active share indicates that the portfolio is different from the index, either in terms of the weights allocated, the stocks held, or a combination of both.

The graph above shows the results of their 2010 study. Drawing on a database of over 2500 US Equity mutual funds from 1990 through 2009, each fund’s active share was calculated and sorted from highest to lowest. They found that managers with the highest active share consistently outperformed their corresponding index, on average by 1.26% per year after fees. Furthermore, they found that “closet indexers” – funds that charge active management fees, but construct relatively index-like portfolios – underperformed both gross and net of expenses. More importantly, the study found active share tends to be a reliable indicator. Funds with high active share in one period had high active share in subsequent periods and continued to achieve better performance.

It would be a stretch to conclude that active share measurements provide us with a crystal ball of sorts, which can perfectly forecast fund performance. However, it is a useful quantitative tool that can be combined with careful qualitative judgments to tip the scales in our favour. While the task of selecting top managers is always a challenge, T.E. Investment Counsel firmly believes in the merits of active investment strategies. With fundamentals increasingly driving returns, funds with an active approach to stock selection are likely to benefit.

Print Friendly, PDF & Email

These articles are for general informational purposes only. Please obtain professional advice before taking any action based on this information. No endorsement or approval of any third parties or their advice, information, products or services should be implied by any references to third parties contained in any article. Trademarks cited in these articles are the respective properties of their owners.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *

three × 2 =