In April 2003, I released a whitepaper entitled “Active vs. Passive: The Debate Continues” that highlighted exactly what has recently been happening in the saga of exchanges about skill versus luck in active management on Advisor Perspectives.

My original 2003 paper highlighted that there is a lot of marketing going on by BOTH “passive pundits” and “active advocates” instead of objective research. That paper demonstrated how it isn’t objective science if the so called “research” uses essentially the same data (as the Wermers and Howard pieces on Advisor Perspectives, 2008 did) yet comes to opposite conclusions. True objectivity and science are not that forgiving. Marketing, on the other hand, is explicitly designed to spin such data into a story that sounds compelling, or at least plausible, free of the bounds of scientific objectivity.

At the highest conceptual level, Wermers’ work is stating something simple, rational and mathematically provable. Essentially, all Wermers is saying is that just because a fund out performed (or under performed) for some period of time, that doesn’t mean it was necessarily skill (or the lack thereof), it may have just been luck. He didn’t say skill doesn’t exist. He didn’t say he could tell which funds had skill. His FDR (False Discovery Rate) is in essence a statistical attempt to estimate how much luck exists in the universe of funds, back out that likely statistical luck from the universe, and thus estimate how much skill remains across that universe. He even acknowledges that there were data problems because of the weakness of fund objective consistency over the entire time period.

Wermers is saying that if 10,000 monkeys were picking stocks, that some monkeys would out perform and some would under perform. In the case of monkeys it would be all luck, but in the case of money managers only some of them would be luck. Wermers is saying that it is erroneous to assume 100% of all managers that outperform are skilled, and he is making a statistical attempt to identify how much luck is present in a universe of funds and separating that out to identify how much skill remains in the universe.

Enter Howard, who responds to Wermers work but apparently thinks that all monkeys that outperform are “skilled” and completely discounts the notion that a lucky monkey could exist. All monkeys that out perform are skilled, according to Howard, at least according to his “research” methodology. There does not exist a lucky, unskilled manager in the Howard research. Surprisingly, in response to the Advisor Perspectives article about Wermers, even Ron Surz (whose PODS universes simulates all potential monkeys) chimed in as well with the standard industry bromides that are not measurable or provable (but are marketable) with “People, Process, and Philosophy.”  (A.k.a. the “Three Ps” of investment manager research.)
 
In looking at the Wermers work, as a skeptical scientist, I’m wondering if there is enough statistical significance to necessarily draw all of the conclusions he did, especially since the classification criteria is fairly arbitrary and would introduce a significant amount of error that he - to his credit - objectively acknowledges as a potential weakness (i.e., classification as Aggressive Growth Funds, Growth, Growth & Income, etc.).

But, in looking at Howard’s response to Wermers, I do not find the same level of objectivity; I see marketing spin. Howard’s first exhibit in his Wermers response showed a trend line of average negative alpha of 2%, “trending” upward to almost positive 2%. I wonder what Howard was saying back in the 80’s and early 90’s when his average alpha was negative? He must have been an indexer back then and now has become more enlightened with the upward trending slope that shows all managers combined can beat the entire market combined…hmmm…doesn’t the market have to equal itself? (To be fair, Howard used average and not dollar weighted measurements. So, in theory, the market doesn’t have to equal itself with his method.)

In reality, Howard’s “alpha slope” is a slippery slope indeed. He is comparing all domestic equity funds (with very minor limitations) to the S&P500. Forgetting that there is a large universe of small cap, mid cap, value and growth funds (many of which have index benchmarks that outperformed the S&P500 over the last five years) in his domestic equity universe, he compounds the sleight of hand by attributing to all funds that outperformed the wrong benchmark a “growing alpha skill.”

Display article as PDF for printing.

Would you like to send this article to a friend?