To borrow from a Vanguard study1, "active share" is defined as the percentage of a portfolio that differs from a benchmark index. It provides a simple measure of the degree of active management in a portfolio at a given point in time. An active share of 100% would indicate a completely different set of securities than the benchmark, while an active share of 0% would indicate a complete replication of the benchmark. The study that introduced the concept of active share2 showed a link between high active share and outperformance of a benchmark (i.e., funds with a high active share had a higher frequency of outperforming). This finding is not surprising, because active funds with a low active share are essentially closet index funds that charge the high fees associated with active management, so they have virtually no chance of delivering outperformance. Furthermore, a portfolio with a high active share may simply not have the proper benchmark (or blend of benchmarks) assigned to it.
The authors of the Vanguard study took a different and, in my opinion, a more useful approach. They asked the question of whether active share has predictive power for determining future outperformance. Rather than focusing on active share as of the end of the period and seeing if there was a connection to funds that had shown outperformance, they calculated active share at the beginning of the period and then evaluated the subsequent returns. They accomplished this by dividing up the 903 funds into ten deciles based on active share level. The top decile of active-share funds had an average active share of almost 98%, calling into question the validity of the Morningstar-assigned benchmark. In contrast to the finding of the original active share study, the top decile (along with the nine other deciles) delivered negative returns relative to the benchmark during the six-year performance period ending on 12/31/2011. Statistically speaking, there was no significant performance difference among any of the deciles, indicating that active share has no predictive value. Furthermore, there was no significant difference between funds that achieved active share by taking highly concentrated positions within the benchmark or by venturing into positions outside the benchmark (style drift).To put it in mathematical terms, the authors of the Vanguard study demonstrated that active share is a necessary but by no means sufficient condition for outperformance.
The promoters of active share tout it as a valuable component of the toolkit used for the purpose of manager-picking. From its inception, IFA has counseled investors to refrain from playing the mug's game of manager-picking. While we definitely agree with the conclusion that investors should avoid closet-index funds, we take the position that most investors are best served by lower-cost index funds that are based on well-designed indexes that are not susceptible to gaming by opportunistic traders. An example of an index that does not meet this criterion is the Russell 2000 small cap index. According to a study published in the Financial Analysts Journal3, the loss to an investor in a Russell 2000 index fund is about 1.3% per year and can be as high as 1.8%. IFA reminds investors that the single best predictor of a fund's future performance (relative to a properly defined benchmark) is its expense ratio4, and the authors of the Vanguard study found that the funds with the highest active share also had the highest expense ratios.
1Schlanger, Todd, Christopher B. Philips, and Karin Peterson LaBarge, 2012. The Search for Outperformance: Evaluating ‘Active Share'. Valley Forge, Pa.: The Vanguard Group.
2Cremers, K.J. Martijn, and Antii Petajisto, 2009. How Active Is Your Fund Manager? A New Measure that Predicts Performance. Review of Financial Studies 22(9): 3329-65.
3Chen, Honghui, Gregory Noronha, and Vijay Singal, 2006. Index Changes and Losses to Index Fund Investors. Financial Analysts Journal 62(4) : 31-47.
4Philips, Christopher B., 2012. The Case for Indexing. Valley Forge, Pa.: The Vanguard Group.