Search This Blog

Tuesday, August 25, 2009

The Fallout From The "Yale Investing Model" Continues This Time at The University of Chicago

I have a particular interest in the investment results now being reported by the major university endowments. I teach a course in investment management for not for profits and one of the major themes in the class is whether the "Yale Endowment Investment Strategy" is appropriate for smaller not for profits. It seems that after the model has been subject to the real world mega stress test of the past year many large endowments have been asking that question as well.

The Yale model outlined most clearly in the book by David Swensen the head of Yales's endowment: Pioneering Portfolio Management allocates large proportions of the portfolios to alternative asset classes:
hedge funds, private equity, and real assets such as timber with low allocations to more conventional asset classes(such as publicly traded bonds and stocks) with a far lower allocation to bonds than such endowments held in the past. Yale posted extremely high returns for over a decade prompting many other unversity endowments large and small to try to "be like Yale".

But as James Stewart of the WSJ wrote (and I blogged on ) the strategy includes a fatal flaw. The instruments generating those high returns are extremely illiquid. And those times when an institution would need access to funds often corresponds to when those illiquid vehicles (like hedge funds) close their gates.

From the WSJ article on the University of Chicago endowment:(my highlights in bold my comments in bold italics)

The WSJ reported on the University of Chicago and the story seems to be repeating itself elsewhere as well as these institutions may rethink their investment approach (my bolds, my comments in bold italics)

More such judgments will be passed beginning later this month, when colleges begin disclosing how their portfolios fared over the fiscal year that ended June 30. Northern Trust Corp. estimates that, over the period, the average U.S. college endowment has a 20% decline.

Harvard University has predicted the asset value of its endowment, the nation's largest, would drop as much as 30%. Yale University and Princeton University have projected declines of about 25% each. The University of Chicago has predicted a 25% decline in its portfolio value for the fiscal year.

Though loath to discuss their money-management strategies, many universities appear to have considered moving to more conservative investments. Harvard tried last year to sell a chunk of its private-equity portfolio but didn't receive an acceptable offer; it has been cashing out some of its hedge-fund investments, say people with knowledge of the situation.

Most college endowments used to favor stocks and bonds. David Swensen, hired in 1985 as Yale chief investment officer, argued that endowments -- long-term investors that were unconcerned about redemptions or short-term market fluctuations -- were ideal for the likes of real estate, leveraged buyouts and distressed debt. Yale beat markets by a wide margin.

But the stock market's nosedive last year showed what some see as flaws in the model. "The endowment model contained a colossal intellectual error in thinking -- that long-term investors don't need short-term liquidity," says Robert Jaeger of BNY Mellon Asset Management, a unit of Bank of New York Mellon, who advises endowments on how to structure their portfolios.
Some endowments maintain that, despite steep losses, they aren't second-guessing themselves. "That would require moving away from equity-oriented investments that have served institutions with long time-horizons well," Mr. Swensen said in an interview earlier this year.

I have great respect for Mr Swensen but the above statement is more than a little disingenuous. The major achilles heel of the Yale model as reveal last year was not related simply because of a significant allocation to "equity oriented investments."And a strategy other than the "Yale Model" could easily include a significant allocation to equity oriented investments. The distinctive aspect of the Yale model was not it's use of equity oriented investments, it was the large use of illiquid and leverage equity oriented investments.

As others have pointed out, while hedge funds and private equity may seem like asset classes that are not driven by the same market forces, in fact both are crucially dependent on access to capital (creating leverage). During a credit crunch they both suffer severely. In addition the level of leverage is often the main reason for the high returns which of course increases the risk. If a hedge fund is leveraged 10 or 15:1 and generates returns far in excess of the S+P 500 is the fund really generating alpha (higher risk adjusted returns) or just leveraging up both the risk and return. In fact Swensen himself has expressed skepticism at the risk adjusted returns of some investment vehicles.

Warren Buffett has written that when the tide goes out we find out who is naked. Could it be that in the market crash of 2008 the Yale model has proven to be not all it seemed to be ? It certainly seems that in evaluating the model one must in fact expect a significantly higher than a traditional portfolio in order to compensate for the higher risk and lower liquidity in these portfolios.

And it is not the case as Swensen seems to assert that abandoning the Yale model means moving out of equity like investments. As an alternative one could create a liquid portfolio of etds that covered the broad range of asset classes. In fact it seems we may have come full circle. In the most recent edition of Pioneering Portfolio Investments and in his book for individuals. Swensen advocates just that type of more simple liquid portfolio.

In addition to the liquidity issue another factor is our old friend behavioral finance. Swensen may well have educated folk at Yale (and he has a strong track record behind him) to weather the volatility of the portfolio with large allocations to illiquid alternative investments. But it may be that many university endowments found, just like many individuals, they are less able to weather large falls in the value of their portfolio than they previously thought

Ironically enough one can see this played out in the recent episode concerning the University of Chicago's endowment. In a little bit of inside baseball the episode must strike those of us in the finance field as a bit ironic. The Univesity of Chicago Finance and Economics department are considered the cornerstones of the efficient markets school and modern portfolio theory which preach strongly (to say the least) against market timing. Over the last few years though, even this bastion of the rational has been breached with scholars of the behavioral finance school like Richard Thaler joining the faculty.

I think this experience will be taught (certainly in my class) as a case study in the problems with the "Yale Model" when subject to the stress test.

It shows that:

As is well known (certainly to practitioners) "the only thing that goes up in a down market is correlation (among all risky assets).

In calm periods investors often overlook the issue of liquidity in their portfolio Yet in periods of turmoil the non traditional vehicles (hedge funds, private equity, direct investment in real estate) become the least liquid if not totally illiquid. Furthermore many of the strategies in these asset classes are dependent on leverage and ready access to credit. When that credit dries up the funds themselves may be obliged to enter into forced liquidations. Furthermore with many of these funds following similar strategies (momentum long/short) they all wind up liquidating at the same time, aggravating losses. What this means is that looked at from a risk/reward basis many of these asset classes were not really outperforming traditional asset classes. Adjusting for the leverage and lack of liquidity the risk adjusted returns were not so outstanding. Only in a crisis is this clear. An excellent academic paper (which I hope to blog on) on these issues is here.

Investor behavior: As many in the field of behavioral finance (and most people in my field) have experienced, people overestimate their risk tolerance when asked in the abstract, hence in my view making most questionnaires on risk tolerance useless. People have a strong loss avoidance and faced with large losses many who stated on a questionnaire that they would ride out such losses wind up selling (sell low buy high). It seems even the investment professionals at the University of Chicago experienced this. And the consequence was they had to sell the more liquid assets leaving the portfolio with even a higher portion of their investments in the illiquid assets.

The episode is described below from the WSJ
( Again my bolds my comments in bold italics.

The crisis of confidence has been playing out at the University of Chicago, in a previously unreported battle that divided the overseers of the nation's 10th-largest university endowment, a committee that included hedge-fund managers Sanford "Sandy" Grossman and Cliff Asness. Amid last fall's market turmoil, the committee argued over whether their portfolio's assets, $6.6 billion in June 2008 but falling fast, were too risky and volatile.

The endowment's managers staged a $600 million share selloff to buy safer instruments, an unusually abrupt no-confidence vote in its portfolio model. In a late October email to committee members, Kathryn Gould, a venture capitalist who heads the college's endowment committee, and Chief Investment Officer Peter Stein, wrote: "We will no doubt look like heroes AND idiots in the next couple of months."....

In 2005, the University of Chicago hired Mr. Stein from Princeton, where he had worked under a protégé of Mr. Swensen. In June 2008, the university's endowment had 77% of assets in "equity-like investments" -- foreign and domestic stocks, private equity and hedge funds -- according to the 2008 annual report.

That September, around the time that Lehman Brothers collapsed, members of the investment committee took a hard look at their mix.

"We had underestimated the value of liquidity and overestimated our degree of diversification," said Andrew Alper, chairman of the university's board of trustees and a committee member. He says the committee hoped to change the portfolio's long-term exposure to risk and volatility, and would have preferred to unload its stakes in private-equity firms.

But with the market in these illiquid assets essentially frozen and hedge-fund redemptions coming slowly, they began to talk about selling stock.

In early October, the Dow tumbled 18% in one week. In an Oct. 28 email viewed by The Wall Street Journal, Ms. Gould and Mr. Stein told committee members they were considering "an outright sale" of $500 million in stocks -- "virtually all the immediately saleable equities."

By early November, according to people familiar with the matter, Ms. Gould had instructed Mr. Stein to sell $200 million of equities.

James Crown, a trustee and a general partner at Henry Crown & Co., a Chicago investment firm, expressed confusion. "Where are we going with the endowment and why?" he wrote to committee members in a Nov. 2 email viewed by the Journal.

A force behind the sales push was Mr. Grossman, a Greenwich, Conn., hedge-fund manager and economist, say those familiar with the situation. On Nov. 6, during a three-hour committee meeting at university's business school, Mr. Grossman sketched out formulas on an easel to explain the portfolio's relationship to market risk. Other times he referred to the 2008 returns at his own hedge fund, QFS Asset Management -- some were ahead double-digits that year -- to make a case for selling, these people say.

They say some of Mr. Grossman's points were challenged by Martin Leibowitz, a managing director at Morgan Stanley, and by top trustee Mr. Alper.

Mr. Grossman says he advocated reducing risk and volatility but doesn't remember whether he pushed to sell stocks.

The night of the committee meeting, Ms. Gould wrote several members that Mr. Stein was preparing to sell $300 million in stocks. A sale of another $100 million followed. Some proceeds went into fixed-income funds.

Mr. Asness, co-founder of Greenwich, Conn., hedge-fund firm AQR Capital Management, expressed dismay at the response to Mr. Grossman's presentation. "Was Sandy that convincing?" he wrote in an email the next day. "I felt a consensus was building not to be so short-term."

University of Chicago officials won't say when they sold their stock, so it is impossible to calculate returns. Mr. Alper says the proceeds weren't invested into other stocks but that the endowment continues to hold equities. (which means to me the portfolio must consist mostly of cash or bonds + illiquid investments...just the asset allocation that got them into trouble in the first place)

Fallout continues. "We cannot time the market to this degree and should not be trying," Mr. Asness wrote in a January email to members of the committee.

Last year, the university changed its policy so only trustees could serve on its investment committee. That led three nontrustee members, including Messrs. Asness and Leibowitz, to step down in June.

Endowment CIO Mr. Stein announced his resignation in January. Mr. Stein said in a statement at the time that he left for family reasons.

No comments: