- Research suggests that less than one percent of professional active manages are “skilled” at what they do.
- Given this, the odds of picking a solid active manager is not just slightly worse than a coin toss; in fact, the chances are closer to slim to none.
- Therefore, a passive investing strategy may be more sensible over the long run for most investors.
Bringing It Home
We just discussed the long-term impact of an active investing strategy versus a passive investing strategy as it relates to opportunity cost in Part 6 of this series. What we came to understand is that the outcomes from passively investing are more likely to mitigate opportunity cost (the cost of not hitting our benchmark’s returns), while the outcomes from investing in active mutual funds are more uncertain.
Further, we highlighted that actively managed mutual funds do not beat their benchmarks on average when studied over long periods. Indeed, as we discussed in Part 3 and Part 4 of this series, active investment products charge much higher fees than passive products, but nevertheless, the performance of active funds do not necessarily get better as their fees go up, and ultimately active products are more likely to underperform their benchmarks by a considerable amount as their fees increase.
But for more insight into this, let’s dig into just how poorly active funds tend to perform by digging into some published research from academia. The results are scary to say the least. In their research paper titled, “False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas”, published in the February 2010 edition of the Journal of Finance, authors Barras, Scaillet, and Wermers tackle an important question: Just how much value do active managers bring to the table? Turns out, not much.
First the results: this compelling research suggests that less than one percent of professional active manages are “skilled” at what they do.
As a retail investor, the odds of you picking a solid active manager is not just slightly worse than a coin toss; in fact, the chances are closer to slim to none.
Details Are Important
Now, let’s dig into the research. For the study, the authors use, “the monthly returns of 2,076 actively managed U.S. open-end, domestic-equity mutual funds that exist at any time between 1975 and 2006. (inclusive)” This universe of mutual funds is large and encompassing, which instills greater confidence in the results due to the large sample size that was used, while also reducing the impact of survivorship bias and selection bias, which might skew results.
Survivorship bias is the notion that if you only include funds that are alive today, your analysis will be biased given that dead funds might paint a completely different picture. Selection bias is the notion that if you only run your study on a biased subset of the population, your results may be also biased given that this subpopulation my not be representative of the entire population. The authors mitigate both of these concerns by using a broad universe over a 30+ year period that includes funds that survived as well as those that did not.
We investigated excess returns (the measure of a fund’s return relative to its benchmark, a.k.a., alpha) of large-cap mutual funds in Part 4 of this series. But the authors of this paper take it a step further; they factor in the impact of luck into their analysis in order to better measure skill. The notion is that some funds may have just been lucky with respect to their ability to provide positive alpha.
After all, whenever you go to the casino with a group of people, someone usually comes out a winner, but this doesn’t necessarily mean that this individual has an actual edge on the casino; it could be that this person was simply lucky. As such, you wouldn’t invest your hard-earned money based upon pure luck alone, would you?
A Simple, Hypothetical Example
It’s important to understand how the authors define luck and skill in their research. Here’s an hypothetical example of this notion. Let’s say you have a bag of 100 colored balls (10 green, 75 yellow, 15 red); what are the chances of picking a green ball on the first try? Right, 10% (10 out of 100). If you run this experiment many times (let’s say with a hundred individuals), how many green balls would get picked on average? Assuming an individual always picks from the same 100 balls, an individual would expect to pick a green ball about 10% of the time, or about 10 out of the 100 individuals should pick a green ball. Similarly, there should be around 75 yellow balls picked, and around 15 red balls picked across the 100 individual ball-pickers.
Now in this hypothetical scenario, let’s label our green-ball-pickers as our “outperforming active managers”, i.e., those with positive alphas (α > 0); and let’s label our yellow-ball-pickers as our “non-performing active managers”, i.e., those with zero alphas (α = 0); and finally let’s label our red-ball-pickers as our “underperforming active managers”, i.e., those with negative alphas (α < 0).
At a high-level, what we are trying to say is that our 10 expected positive alpha pickers (α > 0) outperformed not because they had any special skills, but rather, they were just the lucky ones. After all, 10% of individuals should “outperform” just by pure luck alone according to our random, hypothetical ball-picking scenario. And indeed, the same is true with investing; i.e., some portion of active funds should outperform just by random chance alone (basically luck).
But now let’s say we end up with 15 “outperforming active managers” rather than the 10 green-ball-pickers that we would normally expect. What we might conclude is that 5 of these 15 are truly skilled at picking green balls. We don’t know which 5 out of the 15 are truly skilled, but we have a one-in-three chance of picking a truly skilled picker at random. Of course, these 5 truly skilled ball pickers could have gotten lucky as well, but let’s give them the benefit of the doubt on this. So in this hypothetical scenario, we’re saying that 5% of the ball pickers are “outperforming active managers” that are truly skilled (true α > 0) while 10% were “outperforming active managers” that just got lucky (α > 0).
By the same line of thinking, if we were to end up with more red-ball pickers than we would expect (e.g., if we were to end up with 20 red-ball-pickers when we should only end up with only 15), then we might consider 5 of these red-ball-pickers to be truly unskilled (true α < 0), while the other 15 were just unlucky (α < 0).
Back to the paper. The research finds that of the universe of funds studied, “75.4% are zero-alpha funds—funds that have managers with some stock-picking ability, but that extract all of the rents generated by these abilities through fees.” Based upon our hypothetical analogy, these funds are good at picking yellow balls. So essentially, according to the paper, most active funds fall into this category, as they do not provide any value to their investors once you factor in their fees.
But it gets worse. The authors further conclude that, “24.0% of the funds are unskilled (true α < 0), while only 0.6% are skilled (true α > 0)—the latter being statistically indistinguishable from zero.”
Accordingly, almost a quarter of funds ended up being not just unlucky, but as being truly unskilled (true α < 0). These funds represent the truly unskilled red-ball-pickers from our hypothetical scenario above. Also astonishing, practically zero funds ended up as being truly skilled (true α > 0). These funds represent the truly skilled green-ball-pickers from our hypothetical scenario above.
Further, the research suggests that over time, the prevalence of truly skilled active managers has all but disappeared. The authors find “a significant proportion of skilled (positive alpha) funds prior to 1996, but almost none by 2006.” Specifically, they, “observe that the proportion of skilled funds decreases from 14.4% in early 1990 to 0.6% in late 2006, while the proportion of unskilled funds increases from 9.2% to 24.0%.”
The authors observed 14.4% “outperforming active managers” that are truly skilled in 1990, but only 0.6% “outperforming active managers” that are truly skilled in 2006. That’s not encouraging at all, is it?
Basically, according to the authors, there used to be a good proportion of truly skilled green-ball-pickers prior to 1996, but by 2006 the truly skilled green-ball-pickers had all but vanished. This means that by 2006, if you were to find an active manager that is outperforming, it is more likely that the manager was just lucky rather than truly skillful. This also implies that the chance of a fund that can outperform its benchmark beyond luck alone is slim to none. Accordingly, mutual fund investors may have had some chance of finding a good actively managed fund in the 1990s, but that’s not the case anymore according to this research.
The Big Picture
It’s hard to say exactly why all the truly skilled funds have disappeared, but the authors do provide some insight into this, “Either the growth of the fund industry has coincided with greater levels of stock market efficiency, making stock-picking a more difficult and costly endeavor, or the large number of new managers simply have inadequate skills.” Essentially, it could be that competition and an increased supply of funds has driven alphas down to zero, just as the Efficient Market Hypothesis would suggest. Further, “Although increased competition may have decreased the average level of alpha, it is also possible that funds do not achieve superior performance in the long run because flows compete away any alpha surplus.” This further implies that the mutual fund space is simply too crowded at this point.
Also critical to understand, the frequency of truly unskilled funds has dramatically increased over time. As noted above, we’ve gone from 9.2% to 24.0% of funds being truly unskilled by 2006. According to the authors, “This increase [in unskilled funds from 1996 onwards] is likely to be due to rising expenses charged by funds with weak stock selection abilities, or the introduction of new funds with high expense ratios and marginal stock-picking skills.” Not very encouraging either. Understandably, the authors of the paper are astonished as I am with all this:
Still, it is puzzling why investors seem to increasingly tolerate the existence of a large minority of funds that produce negative alphas, when an increasing array of passively managed funds have become available (such as ETFs). (Barras 2010)
Case in point, when was the last time you critically interrogated the performance of your actively managed fund portfolio? Are your active funds really living up to their hype? Or have they not really been providing you with substantive value for quite some time? Furthermore, given all we’ve reviewed so far, why would you invest in an active fund ever again? Why, indeed.
At this point, we’ve made a strong case on the merits of passive funds over active ones. But also, consider this, if professionals have such a tough time beating the market, does it really make sense for non-professionals to put their valuable free time into this endeavor? In Part 2 of my series, The World’s Most Expensive Hobby, we’ll shift gears and look into how average investors fare in their quest to beat The Street. As you might suspect by now, these results are not pretty either.