Fama: Active Management a Bad Bet

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
gkaplan
Posts: 7034
Joined: Sat Mar 03, 2007 7:34 pm
Location: Portland, Oregon

Fama: Active Management a Bad Bet

Post by gkaplan »

Active fund management is a zero-sum game in which luck and skill are virtually indistinguishable, says Nobel Prize-winning economist Eugene Fama.

Fama, a University of Chicago finance professor and a co-winner of the Nobel Prize in economics last year for his efficient markets hypothesis, spoke to an audience of financial advisors and institutional investors Thursday at the Morningstar ETF Conference in Chicago....
http://news.morningstar.com/articlenet/ ... ?id=665712
Gordon
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Fama: Active Management a Bad Bet

Post by pascalwager »

As for his own portfolio, Fama said he primarily owns equity index funds and some Treasury Inflation Protected Securities (TIPS), adding that he has no interest in real estate or bonds.

"I'm a tenured professor," he said. "The university issued me a bond."
So Fama considers his very dependable salary and presumed future pension as his personal bond allocation.
VT 60% / VFSUX 20% / TIPS 20%
User avatar
Kevin M
Posts: 15789
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Fama: Active Management a Bad Bet

Post by Kevin M »

Interesting (to me) quotes:
He spoke about the idea of isolating various risk factors within a portfolio--sometimes called strategic beta or smart beta strategies--and said that, while real, such "dimensions of risk" are still not well understood. He said such factor-based investing works, "But why? All we know is that they seem to be there."
He also said that combining factors can be counter-productive, adding that "once you get beyond two dimensions of return, the third adds very little." He offered value and momentum as examples of factors that can work against one another within a portfolio.
He said he doesn't recommend any specific allocation strategy, but said "you pick your risk exposures and then you diversify the hell out of it. That's the best you can do."
Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Robert T
Posts: 2803
Joined: Tue Feb 27, 2007 8:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Fama: Active Management a Bad Bet

Post by Robert T »

He also said that combining factors can be counter-productive, adding that "once you get beyond two dimensions of return, the third adds very little." He offered value and momentum as examples of factors that can work against one another within a portfolio.
I generally agree that if you add too many factors you may simply end up with a market portfolio. In this repect, my preference is to focus on 3 (market, size, value), but try to get 'pure exposure' (zero alpha) to these factors. I agree that some factors can work against each other - thats why I prefer mid-cap value to large cap value. If targeting a positive size load on a portfolio, including a large cap value fund (with a negative size load) takes you a step further away from, not closer to, the positive size load target. As a result you need to add a larger share to small caps to meet size load targets which likely comes at higher cost. While I can understand his view on momentum and value, the challange is most value funds have negative momentum loads which are typically larger, the larger the value load. In a three factor model, this shows up as negative alpha. In this sense, the returns you get are not the same as those suggested by the value load, given the negative alpha. In my view, adding a seperate momentum allocation can offset the negative momentum load on value funds, but would come at a cost of a reduction in the combined portfolio value load (as does the DFA momentum screens), so would need to increase the allocation to value to get back to the value load target. While there is an additional cost to this, in my view, its likely less than the cost of the expected negative alpha. In addition, if the low correlation between momentum and value persists there would be ositive rebalancing return (a.k.a. Markowitz - and as suggested by backtests, obviously no guarantee).

Robert
.
Post Reply