Back in mid-March, I came across a link to an interview with David Swensen, chief investment officer for Yale University’s endowment and author of the 2005 book Unconventional Success: A Fundamental Approach to Personal Investment. Swensen has a phenomenal record running the Yale endowment and pioneered the use of hedge funds, private equity and a wide array of other alternative investments among institutional investors.
But his book offers a very different approach for individuals, who don’t have access to most of the investment managers used by Yale. Instead, Swensen recommended that individuals allocate their assets across a simple and fixed mix of six index funds (30% in U.S. stocks, 20% in real estate, 15% in U.S. Treasury bonds, 15% in U.S. Treasury Inflation Protected Securities, 15% in foreign, developed stocks and 5% emerging market stocks). Beyond occasional rebalancing, Swensen’s formula was fixed, unchanging and perfect for all. It seemed like the magic formula for investing success, according to Swensen.
One particular out-to-lunch passage struck me from the March interview, when Swensen asserted that someone following his magic formula “probably did reasonably well” through the recent credit crunch and market turmoil. That wasn’t true. The magic formula lost fully one-third of its value over the prior year, nearly as bad as the loss of the S&P 500 by itself and certainly not what people following Swensen’s advice (or reading his quote in the interview) would have expected. And so I posted my critique here.
The post has received a number of comments, most quite critical of my analysis. I posted a reply in the comments field today but I thought it was worth expanding my reply into a full-blown blog post to further the discussion.
[UPDATE: I asked Swensen if he would comment. He sent a reply that he did not want published. Needless to say, he doesn’t agree with me.]
First, I think it’s important to acknowledge that there's a huge (and misleading) disconnect between how Swensen ran the Yale endowment to produce such amazing results and the magic formula he recommended for ordinary investors. Yale uses active managers, has nothing like 30% of its assets in US stocks (or any such fixed asset allocations at all) and relies on a wide array of alternative assets, many of which aren't easily accessed by individuals. Plenty of investors turned on by Yale's results looked to Swensen for advice but the advice they received has no connection to the endowment's superior results. Those commenters who cite Swensen's Yale track record as evidence that his magic formula is a good idea have fallen into the same trap. “Swensen Fan,” for example, quotes Swensen mixing Yale’s results and the book’s formula.
Until he wrote a book for individual investors, Swensen regularly admitted that individual investors could not follow Yale's investing program. I suspect he got tired of giving that non-answer and, to paraphrase Woody Allen in the great flick "Manhattan," let's face it, he wanted to sell some books here. And there’s really not much offered to support the magic formula’s particular allocations beyond considerations of how those kinds of assets have performed in the past.
Second, commenters like “MSS,” “Stu” and “Finn” say one year is too short a time frame to evaluate how the magic formula has done. “Ignorance Arbitrage” says it’s singularly idiotic. But the point isn’t to assess Swensen’s formula over one-year periods. Swensen himself had misstated the performance of the recent formula (without even getting into how long it would take to earn back a 32% loss). It’s critical to assess how an investment strategy performs in all kinds of markets. Swensen’s seemingly conservative strategy worked fine under “normal” market conditions but failed miserably in the bear market. We’ve also just experienced a 30-year or so trend of declining long-term interest rates and inflation in the U.S. that gave a nice tailwind to certain assets. An asset mix constructed based on returns and volatility during that period might not work so well if the next 30 years see a very different interest rate backdrop.
[UPDATE 6/18/09: Running the results of the six-fund magic formula over longer periods doesn't help Swensen's case much. If you invested $10,000 in the magic formula at the beginning of 2005, when the book came out, how much would you have 4-1/2 years later on May 31, 2009 with annual rebalancing? $10,275. That's worse than if you'd kept the whole sum in T-bills or a money market fund and virtually identical to the $10,268 you'd have investing a BENCHMARK EXAMPLE of 70% U.S. Stocks and 30% U.S. Treasuries.]
Third, commenters Lawrence Weinman and “Finn” criticize asset allocations I supposedly recommended in the post. I didn’t actually recommend any strategy, offering the performance of a simpler mix for context. We do write about that topic frequently in Businessweek and on the blog, though. Check out this post about Mebane Faber or this one looking at age-appropriate portfolio designs, for a few examples. And I recently wrote a magazine article looking at the pro’s and con’s of asset allocation funds. There are also many great blogs that cover the subject like CXO Advisory, Bespoke Investment Group and Interlake Capital.
Finally, commenter "Finn" make reference to modern portfolio theory, the efficient frontier and the benefits of diversification. These are excellent theories but they're constructed on top of numerous simplifying assumptions about how the world works. I'd suggest taking a look at some of the more recent research about how MPT has failed of late. Particularly, see what Nassim Nicholas Taleb has said. Here's a quote from a recent interview with Taleb (free registration required):
The field of statistics is based on something called the law of large numbers: as you increase your sample size, no single observation is going to hurt you. Sometimes that works. But the rules are based on classes of distribution that don’t always hold in our world.The comments are open for further debate!
All statistics come from games. But our world doesn’t resemble games. We don’t have dice that can deliver. Instead of dice with one through six, the real world can have one through five—and then a trillion. The real world can do that. In the 1920s, the German mark went from three marks to a dollar to three trillion to a dollar in no time.
That’s why portfolio theory simply doesn’t work. It uses metrics like variance to describe risk, while most real risk comes from a single observation, so variance is a volatility that doesn’t really describe the risk. It’s very foolish to use variance.