Active share - a measurement of the degree to which an active portfolio manager's portfolio varies from its benchmark index - has become one of the most popular metrics in the investment industry. "Active share" is the percentage of a fund's holdings that differs from the fund's benchmark index's holdings.

Two pioneering studies published in 2009 and 2013 found that stock-picking managers with the highest active share outperformed their benchmark indices even after fees and expenses. These outperformers included concentrated stock-pickers who focused on relatively few stocks as well as diversified stock-pickers - portfolio managers with high active share, but broadly diversified portfolios. "Closet indexers," whose portfolios had low active share, underperformed their benchmarks.

Both studies concluded that active share is significantly predictive of fund performance.

However, other studies raise questions about these conclusions. A 2012 study by Pennsylvania-based Vanguard Group Inc. compared the returns of a sample of 903 U.S. equity funds classified into three categories - concentrated stock-pickers, diversified stock-pickers and closet indexers - over two time periods: Jan. 1, 2001, to Dec. 31, 2005; and Jan. 1, 2006, to Dec. 31, 2011.

Although concentrated stock-pickers outperformed their benchmarks by 2.96% a year during the first period studied, they underperformed by 0.77% during the second period. Diversified stock-pickers eked out a paltry 0.11% outperformance per year during the first period, but lagged their benchmarks by 0.42% a year in the second. Closet indexers were the poorest performers in both periods: they underperformed their benchmarks by 0.67% and 1.22%, respectively.

Although selecting a high active share portfolio manager doesn't necessarily lead to outperformance, paying active management fees for index-like returns is a losing strategy.

A more recent study, detailed in an article entitled "Deactivating Active Share" and published in the Financial Analyst's Journal in 2016, uses the same data as the 2009 and 2013 studies, but found there's no statistically significant evidence that funds with high active share have returns that are different from funds with low active share. This recent study concluded that active share is not a reliable predictor of performance.

Morningstar Canada studied the relationship between active share and performance in actively managed Canadian equity funds from Jan. 1, 2001, to Dec. 31, 2015, and concluded that "active share proved a weak and inconsistent predictor of future returns."

In fact, that study found that active share explained only about 10% of the variability in excess returns among portfolio managers before fees and expenses. Style exposure to value, size and momentum - not stock-picking - were more likely the drivers of relative performance.

Worse yet, funds with high active share tend to charge higher fees, which numerous studies have found to be the biggest determinant of long-term underperformance.

Conceptually, as index and closet index portfolio managers' returns before fees and expenses approximate the overall market's return, you would expect that funds with high active share will approximate the market's returns (before costs). In effect, portfolio managers making winning bets would be expected to be offset by portfolio managers making losing bets.

Active share is not a panacea for selecting outperforming active portfolio managers. But, combined with a management expense analysis, it's a useful tool for avoiding pricey, closet indexers.

Michael Nairne is president of Tacita Capital Inc. of Toronto, a private family office and investment-counselling firm.

© 2017 Investment Executive. All rights reserved.