Perhaps the most-referenced source in the “active vs. passive” debate is the ongoing series of Standard and Poor’s SPIVA scorecards. The SPIVA studies (updated twice each year) consistently show that most actively managed mutual funds fail to outperform their benchmark index. In other words, if you randomly picked a fund, you would have a less than 50% chance of outperforming an index that tracks the performance of the asset class in question.
While that’s interesting information (and it’s nice to have a consistently up-to-date source for figures), it’s worth noting that the question the studies answer has some limitations with regard to its real-life applicability.
First, the studies compare the performance of real-life actively managed funds to the performance of an index, yet an index is not something we can invest in. We can only invest in index funds, which (unlike the indexes themselves) have expenses
Second, most investors aren’t picking funds randomly. They’re doing better than that. For example, John Reckenthaler (Morningstar’s Vice President of Research) found last year that in most fund categories, the asset-weighted performance is better than the equal-weighted performance (meaning that investors are putting the majority of their assets into funds that perform better than the middle of the pack).
This shouldn’t be terribly surprising. Given the evidence that the poorest-performing group of funds tends to continue to perform poorly, all investors have to do is avoid choosing funds with terrible past performance in order to have a decent likelihood of a better-than-random selection.
Third, looking at the “active vs. passive” question on a fund-by-fund basis is probably less applicable than looking at things on a whole-portfolio basis.
A New Look at the Active vs. Passive Question
In a recent study, Rick Ferri (of Portfolio Solutions) and Alex Benke (of Betterment) looked at the likelihood of a portfolio of actively managed funds outperforming a portfolio of index funds. In other words, they took two big steps closer to reality by comparing actively managed mutual funds to actual, investable index funds and by looking at how an active fund portfolio is likely to compare to an index fund portfolio.
The specific figures vary depending on the period and asset allocation tested, but the general conclusion is that a portfolio of actively managed funds has a much less than 50% chance of outperforming an index fund portfolio. And that chance gets smaller as you look at longer periods of time. In addition, in scenarios in which the portfolio of active funds outperforms the portfolio of index funds, it tends to outperform by a smaller amount than the amount by which it underperforms in scenarios in which it loses.
Perhaps the most interesting part of the study was that Ferri and Benke also checked to see how the results change when you exclude actively managed funds with above-average costs from the active fund portfolios. (I think this is a useful test given that investors do tend to put the majority of their money into funds on the low-cost half of the spectrum.) The answer: Excluding high-cost funds meaningfully improves the results for the active portfolios, but the degree of improvement is still well below what would be necessary for an active fund portfolio to be a good bet.
I’d be interested to see what happens if the very same idea is carried further. For example, what if the active portfolios were constructed such that they only included funds in the bottom 25% in terms of expenses? What about the bottom 10%? Is there a point at which the cost difference is small enough for the actively manage funds to, on average, outperform (or roughly tie) the index fund portfolios?