Search This Blog

Wednesday, April 22, 2009

Harvard Yale And Individual Investing

Two new books: How Harvard and Yale Beat the Market by Matthew Tuttle and and The Ivy Portfolio: How to Invest Like the Top Endowments and Avoid Bear Markets - Mebane T. Faber The Have appeared purporting to show how individuals can use the investing techniques of the Yale and Harvard endowments in their individual portfolios. This is despite the fact that books by David Swensen of Yale and Mohammed el Arian formerly head of Harvard Management do not recommend individuals attempt that approach. I gave both books a "latte read" (a short review over a coffee at my local bookstore) and I was not particularly impressed. The Tuttle book stresses the efficacy of using active managers despite their high fees, the Faber book (more on that book in a later post) is somewhat better although it's system to "avoid bear markets' amounts to market timing based on a moving average technical trading model.

The Tuttle book is reviewed in the WSJ today. The reviewer correctly points out that individual investors can incorporate part of the Yale/Harvard approach by adding asset classes such as real estate and commodities. Tuttle argues in his book that individuals can make use of hedge funds through "funds of funds" and other vehicles. In the book he argues that access to skilled managers is worthwhile despite the relatively high fees ( a view diametrically opposed to that of Swensen in his book Unconventional Success which is directed to individual investors.)

The reviewer, Dave Kansas notes in his review

The best part of Mr. Tuttle's book is his critique of mutual-fund investing. He notes that mutual funds are organized primarily around "style," meaning that a fund focuses on certain investing themes, like large-cap growth or small-cap value. Mr. Tuttle argues that style investing boxes in money managers. If large-cap growth isn't doing well, the manager is still "trapped" into investing in that arena or building up cash. The goal: just beat the large-cap growth benchmark. So if the large-cap index is down 25% and a large-cap fund is down 21%, it is doing "better" than the benchmark and the goal has been met. But that hardly helps the fund's investors.

Mr. Tuttle argues that investors should focus on skills rather than style. In other words, an investment fund should have a mandate to make money, not to stick to a certain benchmark. (This strategy is similar to the one followed by most hedge funds.) Some mutual funds now favor a skills-based approach, and Mr. Tuttle argues that more such funds will become available as style-based investing comes in for more scrutiny and criticism

I find fault with the above argument on several counts. If one accepts that "styles" correspond to asset classes that are not highly correlated, then it is logical to argue that consistently holding at least some allocation to all of these asset classes should over the opportunity to increase returns without increasing risk. This is a fundamental tenet of modern portfolio theory and although much of the theory is certainly under severe criticism of late this premise still holds as a useful foundation for building a portfolio.

Certainly the alternative put forward by Tuttle has even less intellectual consistency. To argue that "investors should focus on skills rather than style" and that skills based investors that can invest in any asset class will produce outstanding returns, is to put it mildly quite a leap of faith for the following reasons:

1. If there is overwhelming evidence that active managers do not display skill in beating their benchmark when they are constrained by style (see the other post for today) it would be reasonable to assume that investing skill is a rare commodity. Why would it be reasonable to assume that there are "skilled" investors that would consistently produce outstanding returns simply because they are allowed to invest across asset classes.

2. Since there is virtually no consistency of outstanding performance among active managers within asset classes why would one assume that managers of outstanding skill among the "go anywhere" asset managers could be identified and that once chosen their "skill" would continue.

3. "Skilled Managers" are not an asset class. Therefore relying on skilled managers can increase the risk on a portfolio considerably. If one chooses a single "skilled manager" one has traded the transparency and diversification of a multi asset strategy using index instruments for an "all in" strategy betting on the skill of the manager, it could be highly concentrated on a single asset class or even a very limited number of securities. Adding additional "skilled managers" doesn’t necessarily reduce the risk of the portfolio since they all could be following similar strategies and have high exposure to the same asset classes. Furthermore, unlike the transparent indexed portfolio there would be no way to reduce the risk of the portfolio other than liquidating part of the accounts invested with the "skilled managers" and investing cash or near cash instruments.

Many if not most of the absolute manager concentrate on a particular style although they are different than “traditional “ asset classes. In contrast to Mr. Kansas’ argument, few absolute return asset managers really “go anywhere” and can enter into any strategy that they think can make money. Thus hedge fund managers tend to specialize in areas such as merger arbitrage, convertible bond arbitrage, long short stock trades and others. Not surprisingly the returns within these styles tend to correlate highly.

Choosing absolute return managers may not solve the problem of “style boxes” one is simply swapping the style boxes of “traditional” asset classes for the style boxes of the hedge fund world. The widely used hedge fund indices break out performance by style and the hedge “funds of funds” available to retail investors usually strive to provide a mix of styles.

The problem of having mangers that are “locked in” to a style” is not solved by simply adding absolute return managers to a portfolio. For example choosing an absolute return manager that concentrated on convertible bonds would have locked one into a dismal performing asset class last year. Since the range of hedge fund strategies is so broad, a truly diversified portfolio of “skills based manger” would likely require successfully choosing managers that have outstanding skills exercising a whole range of strategies . One thing that is certain when one adds absolute return managers instead of index instruments to one’s portfolio: fees will be higher. And because of the short term trading and leverage used by many of these managers taxes and volatility are likely to be higher as well.

In sum Tuttle's "strategy" consists of giving money to a "skilled manager" (screened one assumes by an advisor like Tuttle)and hoping for the best from the "genius manager".
I am a big believer that investors can be easily (in the words of Naseem Taleb) be fooled by randomness and that what seems to be investing skill is in fact often luck or at best no indicator of future success. While all of us are rethinking some of our approaches to investing in light of recent developments, throwing out the idea of exposure to a range of asset classes (styles) and instead relying on finding "skilled" managers is in my view simply a recipe for increasing risk that is unlikely to meet much success.

No comments: