20 February 2007
 The weekly investment newsletter “Outside the Box” picks up where a previous edition left off - with James Montier’s assertion that “CAPM is CRAP“. This week’s edition contains an interview with Peter Bernstein and James Montier conducted by Kathryn Welling. As newsletter editor John Mauldin correctly observes, the article is “double the length of normal”. Mauldin bills the interview as “Bernstein taking on (Montier’s) criticism of CAPM”. But unfortunately for those pining for a grudge-match (like us), there seems to be a lot of agreement between Montier and Bernstein.
As you may recall, Montier argues that CAPM is too theoretical to be of any use. He reiterates his position to Welling:
“CAPM is still the only thing that is taught in business schools around the world–perhaps with arbitrage pricing theory, but the central formula is the same there. So students still come out with their MBAs, having been drilled in CAPM, and practitioners still use these models. Yet I can’t help but think that the behavioral critiques that have been leveled at a lot of your “capital ideas” do actually invalidate the use of CAPM, for instance, in all sorts of ways.
“…whilst Peter is emphasizing the benefits that CAPM has provided in terms of alpha and beta separation, I suspect those benefits are probably clearer on paper than they are in practice, put it that way. But even if we accept that there are benefits from alpha and beta separation, it should also be recognized that CAPM has delivered some very undesirable side effects.”
In response, Bernstein acknowledges these shortcomings, but holds that CAPM is a good pace to start.
“CAPM and the rest provided a place to begin. Just as clinical psychology would not be what it is today if there hadn’t been Freud.”
“Nobody claims that the theories work. In particular, the Capital Asset Pricing Model has been proven, over and over again, even as far back as by Fischer Black, to not “work.” (But) we know things about how markets work and the centrality of the risk/reward trade off and diversification and so forth that just weren’t part of the investment process before these ideas were set forth.”
Although the conversation tends to ramble somewhat from then on, it contains several references that readers of AllAboutAlpha might find particularly interesting.
Montier on Hedge Fund Replication
“The whole idea that you can create alpha by re-weighting beta seems to me to just demonstrate that the nature of alpha and beta can be blurred. Another example, if you like, is hedge fund replication, the fact that some of the investment banks have launched clone funds of hedge funds. Again, the supposed alpha engines are now being duplicated with six-factor effectively beta-style models. So I doubt that alpha and beta are as distinct as CAPM says.”
Montier on Alternative Beta
While Montier does not discuss “alternative beta” per se, he alludes to it in an attack on hedge fund herding.
“…the correlation amongst hedge funds themselves has soared. It’s .7 or .8 on a range of strategies supposedly as diverse as convertible arbitrage and emerging market equities. Obviously, these things should have zero correlation, but they don’t, because everybody is doing the same thing.”
Bernstein on Fundamental Indexation
“I think they are distinct but there’s no law that says if somebody is generating an alpha and other people begin to copy it, it doesn’t turn into beta. There’s no question about it. Alpha is very ephemeral and transient in a market that is, if you’ll pardon the expression, as efficient as we have. Everybody is looking for opportunities like crazy and you’ve got computers transmitting these great volumes of information, so it’s very likely that alphas will turn into betas. And if enough people…copy fundamental indexing, then the character of fundamental indexing is going to change…small-cap stocks are not necessarily small companies. They’re very often the depressed stocks of larger companies. This was the genesis of Rob’s idea that the market capitalization and the size of the company aren’t necessarily identical and they surely aren’t.”
Bernstein on Portable Alpha
“…there is a profound difference between your asset allocation decision and deciding which particular items to include in each asset class. It’s from that idea that we’ve now developed a lot of strategies called “portable alpha,” in which the whole search for performance that beats the market is entirely separate from the people who are doing the asset allocations. This is a big step forward in management and I think it’s again something that will tend to make markets more efficient–because it exposes the opportunities in a way that we never thought about them before.”
Montier & Bernstein on Benchmarking
Montier: “Take Wall Street’s obsession with benchmarking. That undoubtedly comes from CAPM, as well as the constraints themselves. The ideas of benchmarking and of relative performance; this obsession with following a capitalization-weighted index falls out of CAPM. Also the whole idea of tracking error and of professional fund managers worrying more about tracking error than they worry about delivering positive returns. So whilst CAPM may have generated benefits, it has also generated some pretty severe problems for the financial system.”
Bernstein: “The source of the tracking error problem is not the portfolio manager, but the client….This is a point I didn’t make in the new book but I probably should have: That clients are the villains of the piece all-too-often and, as Kate suggested, consultants. They are the villains to a much greater extent than the managers, who would love to be freer.”
Bernstein on Time Horizon & Risk Tolerance
“I’m a consultant to a very big family trust–it’s in the billions of dollars–that was set up so that it could not distribute the principal for a very long time to the family. Actually, it’s based on the life on an individual who has turned out to live to 102 years old, so far. So everybody is sitting around waiting for this poor lady to meet her maker.
“This trust was set up in the early 1970s. They put the money 100% into common stocks and said: We don’t care what happens between here and there, because liquidation is a long way off and we want to be sure we can beat inflation and have an income stream that rises. They’ve been religious about this and perfectly happy and they see market declines as opportunities because they can take lower capital gains if they want to make a change to the portfolio. But that sort of investor is a very rare bird. In particular, the steadfastness with which these people have held to their policy is extremely rare. And it is beginning to weaken now because the lady is 102.”
Both men believe that price volatility is a myopic measure of true risk. So Bernstein’s story of the 102 year-old woman makes us wonder if the modern concept of risk (price volatility) is a direct result of increasing market and information liquidity. Bernstein has been around a long time and he remembers a time - pre-Markowitz - when risk was not a critical measure. He tells Welling:
“When I think back to my early days in the business, in 1951, we used the word “risk” only casually. But now it is central to every sophisticated investment decision.”
We wonder if the pre-Markowitz investor may have been more akin to Bernstein’s “very big family trust” due to the natural liquidity constraints of markets at the time. Securities were priced less frequently and information flowed like molasses by snail mail and telephone. We guess that price volatility probably seemed lower and that serial autocorrelation might have existed in return streams. In fact, it may have been a lot like today’s world of private equity…
Read Full Article
 Email this post to a friend
 Print This Post
|
March 9th, 2007 at 2:59 pm
[…] I saw that debate between Peter Bernstein and James Montier. I think Bernstein is spot-on when he says CAPM shapes the way we think about markets and about managing money. I’d argue that the whole idea of benchmarking comes from that sort of thinking. But we now need to move on. I’d argue that there is something beyond these ideas from the 1960’s. I attempt to address this by writing a book - but it contains just one point of view. It’s certainly not going to replace financial textbooks. But we need to move beyond orthodox finance which argues that markets are perfectly efficient and random. That is so untrue it hurts. […]
March 14th, 2007 at 8:45 pm
[…] University of Toronto’s Felix Chee on efficient markets: “Markets are efficient, just not effective. Markets are virtually frictionless – that is, they price securities according to all available information very quickly. But markets are not good at determining if this information is fact or fiction.” (Ed: This “ineffectiveness” is reminiscent of Bernstein’s “macro-inefficiency” argument). […]
March 19th, 2007 at 8:01 am
[…] Even if markets for many individual securities are efficient, inefficiencies can also exist on a macro scale. In a recent interview, Peter Bernstein characterized markets as ”macro-inefficient”: “The market is hard to beat. Nobody says it isn’t. But the markets are macro-inefficient and this means that risk and return for the market as a whole can go haywire.” […]
March 24th, 2008 at 9:09 pm
[…] Vanguard puts a decidedly behavioral spin on the search for alpha that is reminiscent of a debate between Peter Bernstein and James Montier last year (see related posting). “As long as investors make behavioral and informational errors, there will be opportunities to outperform the market. Many investing decisions are emotionally driven. On any given day, investors may underreact or overreact to news, trading securities at prices unequal to their real values.” […]