The Motley Fool Discussion Boards

Previous Page

Fool Community Help Desk / Fool News & Commentary


Subject:  Re: TMF Jeff Short Articles Date:  8/28/2003  1:01 PM
Author:  howardroark Number:  9147 of 10419

You criticize me for looking at only the last 8 months (a rather forgettable 8 months, even with the rise) and then you look at mid-99 (near an all-time peak) and use that conspicuous data (and logic you just criticized) to support your own argument. (And your argument is what? That stocks fell in 2000, and when they did, shorts did well. Naturally.)

The argument is that you wrote an article which datamined a meaningless result from a miniscule, biased sample and consequently drew a spurious conclusion. My counterexample was intentionally as weak as yours -- my point was to show how easy it is to support any conclusion you wish, including the polar opposite conclusion, using an insignificant patch of data infected with selection bias. I also included by far the most robust research ever done on the performance of heavily shorted stocks which was neither insignificant nor datamined, and its conclusion was directly contrary to that drawn in your article. I'm disappointed that after presenting better evidence and pointing out the fallacy of relying on such a biased sample you've responded without reflecting at all on your conclusion.

This doesn't change the fact that most money managers lose to the market every year and over history -- short funds included. That was my argument. If you want to take on that argument, you have quite the task ahead of you. You'll need to spin almost as good as the industry does.

I really don't know how to respond to this politely. By spin, I assume you mean the liberal manipulation of facts to support a preconceived conclusion? I point out that you have done just that in your article, and your response is that the conclusion is independently accurate regardless... so there? I am taking on the argument that there's no reason to massage bad data to support a preconceived notion, regardless of its merits on other grounds.

But since you brought it up, the active management question is not nearly as straightfoward as The Motley Fool makes it out to be, and the oft-cited 80% underperformance statistic is not even close to the best evidence on the issue, and it doesn't consider some important issues about skewed distributions (80% of stocks in the S&P500 underperform the S&P500!) and survivorship bias (the number is probably a good deal lower than 80% over time). Russ Wermers' research eclipses the scope of the old sound bites. From 1977-1994, the averaged managed fund performed exactly the same as the Vanguard 500 index fund, after costs and fees. A value weighting of active funds underperformed by only 0.6% a year, which is statistically insignificant to zero, i.e. identical performance. Even worse, the dreaded high turnover funds performed twice as well as the indexes over the period, which in fact is statistically significant outperformance. Wermers has also found the stock picks of actively managed funds outperform the indexes by 1.3% a year, which means that stock picks of everyone else in the world including individuals and newsletters underperform the indexes by the same amount. The Wermers paper has been long discussed here but rarely cited in articles:

Of course, even better statistical evidence is complicated, and subject to the next set of better data. In this case, you have to deal with how best to account for risk and how to estimate the additional frictional costs of taxation in taxable accounts, among other things. You also have a world that isn't necessarily stationary. The overall performance of active funds is an open question, and all things considered I think TMF used to give very good advice that most investors are better off passively indexing their holdings, especially in taxable accounts. Its rhetoric, on the other hand, and failure to give adequate consideration to countervailing evidence while grasping every opportunity to decry institutional money management shirks intellectual honesty and borders on what you would call spin. Although TMF's version of anti-active management spin is preferrable to some of the disgusting practices of active managers who tout survivorship biased outperformers, it's not necessary. If you strongly believed that actively managed mutual funds were a pox on investors, I'd think you'd want to ensure your supporting evidence was meaningful, as opposed to cherrypicked. This is especially true now that TMF is in the business of selling stockpicks itself. Because unlike mutual funds, the evidence seems pretty clear that newsletters as a whole don't show stock-picking ability:

Copyright 1996-2019 trademark and the "Fool" logo is a trademark of The Motley Fool, Inc. Contact Us