The Limits Of The Yale Model

The Economist wonders if the so-called Yale model, an aggressive use of conventional and alternative asset classes, will shine as brightly in the years ahead as it has over the past quarter century. The rationale for thinking positively comes from the capable investment hands of David Swensen, who’s managed Yale’s endowment since 1985, delivering stellar results. His strategy, explained in his 2000 book Pioneering Portfolio Management, has been hailed by many as the only way to fly for institutional investors. Individuals, too, can also learn a thing or two from Swensen, argue his supporters.


Success never lacks for a crowd, of course. And in this case, there’s economic logic driving the idea. Owning a broad array of assets with varying degrees of correlation is the basis for building optimal portfolios, a.k.a., funds that maximize return with minimal risk. Harry Markowitz formally introduced the idea more than 50 years ago, and it’s as relevant today as it was when he first wrote about it in his famous 1952 paper. But the problem has always been forecasting the three critical variables: return, correlation and volatility.
Estimating performance is especially tricky. In fact, relatively few money managers are able to deliver much more than middling performance over time. That’s true for efforts within a given asset class, as numerous studies on the power of indexing remind. It’s not widely known that passive asset allocation does quite well too.
For example, consider the track records of more than 1,000 mutual funds with at least 10 years of operating history and a mandate for some degree of multi-asset class investing, as reported by Morningstar Principia software. As you’d expect, there are some spectacular winners—and losers. But most of the results fall into the middle. That’s not surprising, since earning excess returns is a zero-sum game. The winners are financed by the losers.
That profile also lays the foundation for expecting an unmanaged, market-value weighted asset allocation to capture the middle ground with a fair amount of consistency. The numbers tell us as much. Imagine a benchmark that holds all the major asset classes, and weights them based on their respective market values. Set it and forget it. Let’s call it the Global Market Index (GMI). How’s it done over the last decade? Pretty good, as the performance chart below shows. Mindlessly rebalancing the mix every December 31 to the previous year’s mix raises return a bit more.

Compared with the 1,000-plus actively managed mutual funds over the past 10 years, GMI’s performance ranks at roughly the 80th percentile (blue line in chart above). In other words, it beat 80% of its competitors. That’s actually better than expected, although the past 10 years have been unusually tough for navigating the world’s markets. Between financial crises and wars, it’s been a ripe era for black swans.
Indeed, the fundamental challenge in managing asset allocation is deciding how to adjust the weights in real time. Suffice to say, this challenge isn’t getting any easier, and for a familiar reason: the future’s uncertain. No one knows, for instance, how the war in Libya will turn out, or what its effect on oil prices will be. It’s also unclear how the Japanese crisis will end. What is known is that the outcomes of these events, and countless others, will collectively cast a long shadow on risk and return over the long haul. It’s just not clear how, or when. No wonder that a passive strategy of navigating treacherous and uncertain waters has a reliable history of minting average performance. It remains the same even after adjusting for risk.
This isn’t an argument for owning a passive asset allocation, but it is a reminder that beating the market writ large is still hard–even after we add alternative asset classes. Alpha still sums to zero, no matter how broad your asset allocation horizons. Even so, everyone needs to customize their portfolio strategy to match the specifics of their risk tolerance, expectations, time horizon and skill set. But it’s also true that very few of us are David Swensen.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>