Odds of Success
It is a critical question that is asked far too infrequently in the investment business and few attempt to answer it. Unfortunately, it's also not a question that is easily answered in most cases. Statistical analysis of the probability of success resulting from skill are subject to numerous assumptions, some of which tend to vary over time. Additionally, the average tenure of equity portfolio managers is typically six years or less, which is usually less than the time required to determine whether skill or luck is responsible for outperformance or underperformance.
The research cited below strongly implies that it is very difficult to identify skilled managers. I believe there are skilled managers out there, but I have some unique experiences that allow me to make that statement and it's very complicated to attempt to identify them and invest with them in advance. Not only have I worked directly with some talented managers that I believe have skill, I've had the opportunity to analyze their trades (for them) over extended periods, so I've personally seen cases that lead me to believe there are clear examples of skill in stock picking. However, I believe they are a small minority of the number of managers that think they have skill (many of which have substantial assets under management).
There are many "buts" in trying to profit from skilled managers. First, it's very difficult to identify them in advance, especially without proprietary information. Second, once they become successful they tend to attract significant assets, thus making it harder to take advantage of their skill (in fact plenty of successful managers close their funds to additional investment). Third, it's not unusual for teams to split up, or for staff members with knowledge of the skill to break off, which can ultimately lead to more money chasing the same anomalies. Fourth, unsuccessful active management efforts tend to get terminated, while successful efforts tend to survive, creating a survivorship bias issue that often complicates the analysis.
Despite all the complications, in recent years there have been some very serious and professional studies addressing the question of whether skill can be differentiated from luck in the investment business. The bottom line is that it is very complicated and difficult to identify skilled managers and generate profits from investing with them. Below is a collection of studies and research on the topic.
Links and References
- Do Past Winners Repeat?
Forthcoming in Journal of Finance is
Luck Versus Skill in the Cross Section of Mutual Fund Returns by Eugene Fama
Ken French. There is also an extensive Discussion of the article (11/30/2009) on the DFA website. See also Top Mutual Funds: Luck or Skill? New Study Questions 'Active' Managers from WSJ 12/3/2009
- Untangling Skill & Luck by Michael J. Mauboussin (7/15/2010) is a very interesting report discussing how luck and skill interact and determine results in investing, sports, and other games (for instance see "The Skill-Luck Continuum" on page 3 of this 42 page report). Differentiating Skill and Luck in Financial Markets with Streaks by Andrew Mauboussin and Samuel Arbesman (7/23/2010) discusses investing and sports streaks. See also Separating Luck From Skill by John W. Rogers in Forbes (11/30/2009) - "investing is not like playing chess, where a highly skilled player will beat a novice almost every time, nor is it like playing a slot machine, where there is no skill involved. Rather, it is like poker, where a good player can lose if he gets dealt a bad hand."
- Yale’s Endowment Returns: Manager Skill or Risk Exposure? from Peter Mladina and Jeffery Coyle (Waterline) in Journal of Wealth Management Summer 2010.
See also Is David Swensen Lucky or Good? By Larry Swedroe (6/30/2010).
- False Discoveries in Mutual Fund Performance: Measuring Luck in Estimating Alphas from Laurent Barras, O. Scaillet, and Russ R. Wermers
Journal of Finance Feb 2010 (earlier Version) See also
The Prescient Are Few from Mark Hulbert in the NYTimes (7/13/2008). The study used the “False Discovery Rate,” which is used in other fields like computational biology and astronomy. They found declining pass rates for managers in recent decades. For the full sample through 2006 they found virtually no managers with genuine stock-picking ability. The authors suspect investment costs are a cause for the lack of success.
- Luck, Skill, and Investment Performance by Bradford Cornell in The Journal of Portfolio Management Winter 2009 (Earlier Version). See also
Are Outperforming Active Managers Lucky or Skillful? By Larry Swedroe (5/10/2010)
- Active investing: another look at luck vs. skill by Steve Merrell (12/9/2009) and The Inconvenient Truth about Active Investing
- According to Jeremy Siegel (in Stocks for the Long Run), in order to be able to say with 95% confidence that a fund manager’s performance is due to skill rather than luck, he’d need to outperform the market by an average of 4% per year for roughly 15 years. (If the fund was only outperforming by 3% annually, it would take more than 20 years.)
- According to Barr Rosenberg, it would take 70 years of observations to show conclusively that even 200 basis points of incremental annual return resulted from superior investment management skill rather than chance? Source: Investment Policy: How to Win the Loser's Game by Charles Ellis.
Let’s start with the Investment Industry component. It is so obvious in this business that it’s a zero sum game. We collectively add nothing but costs. We produce no widgets; we merely shuffle the existing value of all stocks and all bonds in a cosmic poker game. At the end of each year, the investment community is behind the markets in total by about 1% costs and individuals by 2%. And the costs have steadily grown. As our industry’s assets grew tenfold from 1989 to 2007, despite huge economics of scale, the fees per dollar also grew. There was no fee competition, contrary to theory. Why? Clients can’t easily distinguish talent from luck or risk taking.
Jeremy Grantham in GMO's 2000-2009 Review.
© COPYRIGHT 2010 ALL RIGHTS
RESERVED SUMGAMES.COM DISCLAIMER