Ten years ago, finance professors Martijn Cremers and Antti Petajisto introduced Active Share as a tool to help investors decide if they are getting value for their investment management fees. Simply stated, Active Share measures the difference in the holdings of a portfolio, both in terms of the names and the weights, compared to the portfolio benchmark index. A bottom-up stock picker's portfolio, for example, would have a very high Active Share approaching 100 per cent, whereas an index fund which is designed to mimic the index would report an Active Share of 0 per cent.
Their study concluded that funds with a low Active Share and high management fees, what we now call closet index funds, should be avoided at all costs. The best performing funds were those with the highest Active Score and they significantly outperformed the benchmark. Within this group, concentrated stock pickers did best, followed by diversified stock pickers. Factor bets, such as timing the market by adjusting cash positions or by rotating into different industry sectors did not add value. The research also implied that information about Active Share might have predictive value for the future performance of a fund because the statistic was extremely stable over time.
The concept enjoyed a fair amount of publicity at the time, including two articles in Report on Business, but it did not gain traction in Canada. Perhaps because of the computation involved or possibly because fund managers did the calculation, didn't like the answers and promptly buried the evidence.
To be fair, there was some push-back from others in the academic community who pointed out that high Active Share correlated positively with small-cap funds, so what was being measured was a small-cap phenomenon, not a broader issue. In fact, an article in the Financial Analysts Journal last year (March/April, 2016) stated categorically, "We found no statistically significant evidence that high and low active-share funds have returns that are different from each other."
Faced with this smackdown, Prof. Cremers has returned to the fray with an updated and refined study published in the current issue of the Financial Analysts Journal.
His first update is a simplified calculation of the Active Share percentage. Instead of reviewing every stock in the benchmark index for its presence or absence in the portfolio, the process is reversed. The shorter list of portfolio holdings is now compared to the index and only those positions which overlap are deducted from a 100-per-cent Active Share score. The end result is the same under both methodologies, but the newer version requires much less computation. This may lead to wider industry adoption.
He also points out that a high Active Score can be achieved just by holding stocks which do not feature in the benchmark index, but this tells us nothing about the skill of the portfolio manager in selecting those stocks. A successful portfolio manager requires skill to identify good investment opportunities, judgment and conviction to choose among these opportunities and the ability to execute on those convictions persistently over time. All of a sudden, the manager selection process has become more complicated than simply identifying portfolios with a high Active Share. Now, we have to figure out which managers have the skill, conviction and opportunity to execute on those attributes! The latest study is based on research covering 3,100 U.S. retail equity mutual funds for the period 1990-2015, so the findings are heavily influenced by the bubble and collapse in technology stocks in early 2000. It was during this period that the high Active Share portfolios delivered their best relative performance. For the full time period, the group outperformance was only 0.71 per cent a year and not statistically significant. Low Active Share funds continued to underperform, however, by 1.4 per cent a year and this was statistically significant.
A decade ago, Active Share was a statistic with the promise of predictive value. Now, that claim has been scaled back, but there are still some messages of value to the individual investor in mutual funds from this updated study. Whether or not you needed a PhD in finance to discover them is another matter.
First, closet index funds that charge active management fees are a poor bargain for investors. Prof. Cremers surmises that these funds continue to exist because retail investors have no easy way of identifying them in the marketplace. There is clearly a business opportunity in Canada for a service which provides this statistic for the universe of mutual funds.
Second, if you can replicate the non-Active Share part of your portfolio with an index fund management fee of 20 basis points, then the active portion of the fund must really outperform to justify the active management fee levied on the entire portfolio. Prof. Cremers calls this the Active Fee. You can do the calculation by weighting the passive part of your fund by an index fund management fee and figuring out how much the remainder has to earn in order to cover the overall fund fee.
For example, if the overall management fee on your "actively managed" mutual fund is 2 per cent (200 basis points), but the Active Share is only 50 per cent, then you can mimic the other 50 per cent with an index fund at 20 basis points. Which translates into 10 basis points on the overall portfolio. This means that the Active Share part of your portfolio has to do all the heavy lifting to carry the remaining 1.9 per cent of the management fee. With only half of the portfolio to work with, the fund manager now has to achieve an excess return of 3.8 per cent (1.9 times two) on the stock-pickers portion for an investor to come out ahead of the index. No wonder funds with a low Active Share have a tough time beating the market!
Third, although high Active Share portfolios as a group did not reliably outperform, there was a subgroup within the category which showed persistency – those funds with low portfolio turnover. High Active Share portfolios with rapid turnover and low Active Share portfolios with low turnover did not outperform, so that specific combination is key. Prof. Cremers's hypothesis is that most valuation anomalies are quickly arbitraged away so manager skill is ephemeral except in situations of long term undervaluation. These opportunities are risky for the manager because a value stock may get cheaper before it is recognized and impatient investors may head for the exits. As a result, other investors may see the bargain but are unable to act on it, so the payoff unfolds slowly and rewards the patient investor.
In reading these academic articles, I often feel that the data are being tortured until they confess to crimes they never committed, so it is probable that some of the claims will be retracted or modified in the future. Having said that, the messages of enduring value that I extract are: avoid closet index funds, ensure that when I pay active management fees I actually own a high Active Share portfolio and within that group, to prefer managers with low portfolio turnover.
Robert Tattersall, CFA, is co-founder of the Saxon family of mutual funds and the retired chief investment officer of Mackenzie Investments.
Robert Tattersall