Are truly active managers more likely to beat the market than those who stay closer to their benchmark index? In my latest podcast, I discuss this idea with Christopher Davis, a strategist and researcher at Morningstar Canada.
The idea of active share was introduced in a 2009 paper called How Active is Your Fund Manager? A New Measure That Predicts Performance, by Martijn Cremers and Antti Petajisto. For example, a large-cap US equity fund that holds only half the stocks in the S&P 500 would have an active share of 50%. An index fund, by definition, has an active share of zero.
The researchers presented evidence that the most active funds “significantly outperform their benchmarks, both before and after expenses, and they exhibit strong performance persistence.” A year later, Petajisto published a follow-up paper called Active Share and Mutual Fund Performance, which included more support for the idea that the most active stock pickers can be expected to add value.
Both papers take aim at so-called closet index funds, which charge the higher fees one expects for active management yet differ little from their benchmark. Closet indexing is a recipe for underperformance: it might be worth paying more for truly active management, but it’s never worth shelling out 2% to track a broad-market index when an ETF will do it for a few basis points. As Christopher and I discuss in our interview, closet indexing is rampant in Canada, in large part because our small, poorly diversified market doesn’t give stock-picking managers many options.
Davis’s research looked at data from 2001 through 2015 to see if Canadian equity funds with high active share went on to produce higher returns in subsequent years. During the first period he examined (from 2001–2010), funds with the highest active share significantly underperformed their benchmark, while in the subsequent sample (2006–15), the most active funds did outperform—at least before fees. As might be expected, the most active funds were also the most expensive.
Davis concluded that active share does not seem to have any predictive power. Indeed, Morningstar’s research has consistently found that the best predictor of a fund’s future performance is cost.
If you’re interested in other criticisms of active share, see the following:
Dissecting The Active Share Myth, Larry Swedroe: “Although much has been made by the active management community regarding active share, it appears to be much ado about nothing.”
The Search for Outperformance: Evaluating Active Share, Vanguard: “Vanguard’s study found no relationship between high levels of active share and subsequent fund outperformance for the period studied.”
Deactivating Active Share, AQR Capital Management: “Our conclusions do not support an emphasis on active share as a tool for selecting managers or as an appropriate guideline for institutional portfolios.”
How to invest in the rear-view mirror
In the Bad Investment Advice segment of the podcast, I pick some low-hanging fruit, courtesy of a recent Q&A column by Gordon Pape. Reading the whole article may require a subscription to The Globe and Mail, but here’s the gist: a reader asked why he should care about mutual fund MERs. After all, isn’t an expensive fund worth the cost if it outperforms? You betcha, answered Mr. Pape:
“I’m glad some people understand what I’ve been saying for years. The only number that should matter when judging mutual funds is the net return to you. If a mutual fund with a 2.5 per cent management expense ratio returns 10 per cent a year after expenses and an ETF with a 0.5 per cent MER gains just 8 per cent, which would you choose?”
This is a classic example of outcome bias, the same logical error people use when they call a low-probability bet in poker and then get lucky when their miracle draw arrives. Unfortunately, investors need to choose their funds before they know what their returns will be. And the rational choice is to choose lower-fee funds, since we know cost is reliable predictor of future performance, while past returns have no predictive power.
Mr. Pape has a huge audience and a mutual fund newsletter that charges $79.95 annually, so I feel justified in pointing out that this is dreadful advice for investors. Unless, of course, you have a crystal ball that allows you to predict future returns, or a time machine that allows you to buy them.
Are markets really overvalued?
Finally, in the Ask the Spud segment, I address a question from a blog reader who is nervous about investing a large lump sum during a period of “historically high valuations” for stocks.
Are valuations really historically high? While US stocks are expensive according to traditional measures, the same isn’t true for most other markets. The German firm StarCapital publishes a summary of current valuations in all major stock markets, focusing on the cyclically adjusted price-to-earnings (CAPE) ratio. According to their most recent data (as of December 30, 3016) the US and Japan are the only major markets that appear “expensive” according to their CAPE ratios.
OK, you say, but indexes like the S&P/TSX Composite and the S&P 500 are near their all-time highs. Shouldn’t we be nervous about that? The implication seems to be that an all-time high signals an imminent downturn—maybe even a bubble.
But stock indexes don’t move up and down according to a fixed reference point: they’re designed to grow with the stock market, so we should expect them to touch all-time highs pretty frequently. And they do, as this blog by financial planner Jason Lina points out. “The S&P 500 Index has closed at an all-time high on 6.9% of all days,” he writes. That’s about once every couple of weeks, on average. “More tellingly,” Lina says, “the index has been within 5% of its all-time high on 46% of all trading days since 1950.” So if you’re concerned about investing when the market is near an all-time high, in turns out that’s true about half the time.
In the end, I advise the reader to consider moving into the market gradually if that would make him less anxious. The evidence is clear that dollar-cost averaging with a lump sum does not increase expected returns, but that’s not the point here. Investing gradually—according to a fixed schedule, such as 25% of the lump sum every three or four months—is far better than procrastinating or trying to time the market.