As Our Man’s mentioned, he’s not the biggest fan of the endowment model and one of the reasons stems from the following question; are they really perpetual and permanent sources of capital?
Given that an endowment is there to support an institution (be it a foundation or a university) the level of risk that an endowment takes should be related to how the institution uses the annual. For example, Universities that get a significant percentage of their operating revenue from endowment disbursals (e.g. Princeton at c45%, Yale at c45%, Harvard at 35%, etc) should have different investment and risk objectives for their endowments than those where the disbursals represent a minimal amount of operating revenue. Larger endowments (in terms of disbursals as % of operating revenue) should be more risk averse than smaller ones, unless the institution is prepared to make significant cuts in its operating budget (e.g. by firing professors, doctors, etc or substantially reduce grants) every time there’s a noticeable fall in the endowments value. As such, can an endowment really be considered permanent when it provides a significant part of the operating income of the attached institution?
Illiquidity and being Short an Option
If we go back and look at the Endowment Model and how Harvard and Yale’s endowments have changed over time - we can see a reduction in fixed income and listed equities. Based on Mebane Faber’s work; at Yale between 1985 and 2008, Fixed Income (10.3% to 4%), Cash (10.1% to -3.9%) and Equity (67.9% to 25.3%) were all sources of cash. They were replaced by investments in Private Equity, Real Assets and Absolute Return strategies. This represents both a move towards more illiquid equity-orientated strategies and a subtle increase in leverage (as a result of investing in strategies, and outside managers, that employ leverage).
Unquestionably, one of Harvard and Yale’s great successes was realizing in the late 80’s and early 90’s that there was substantial illiquidity premium to be reaped by those investors who had a longer-time horizon. However, with the publicizing of the Yale model, and the move by others to copy it post-2000, does that illiquidity premium still exist?
Harvard and Yale’s macro call on the use of leverage has also proven to be successful – as investments that rely on leverage have benefited over the last 20years from the increased availability and low cost of credit. The use of credit and leverage has helped drive Equity and Real Estate markets to well above historical valuations, benefiting the users and their investors. However, given what we’ve seen since late-07 and the subsequent tightening of credit and leverage, is allocating capital to these areas the best strategy going forwards?
My final concern is that many illiquid strategies require advance commitments for funding. Endowments have generally been willing to over-commit vs. their target allocations with the intent of using cash from the successful exit of existing investments to fund future investments. For example, per Yale’s Annual Financial Report, they have $7.5bn+ of unfunded commitments to Private Equity, or almost 50% of the endowment’s total value. This makes complete sense if the standard assumption is that prices are constantly rising, but it also creates a short-option position for the endowment. How?
Well, a down market results in the investor failing to receive their expected cash back (because returns are bad) and/or it takes longer for the cash to come back (e.g. the PE firm can’t IPO the business due to ‘market conditions’ so has to wait to do so). This results in new investments having to be funded out of cash or by selling liquid investments, which creates the short option position. I believe that endowments should count this short option position against illiquid investments and as result undersize rather than oversize them. Especially given that a fall in public equity markets already results in an increased allocation to illiquid markets before the above-mentioned cycle begins. This impact was clear in Jun-09, when both Yale and CalPers (amongst others) increased their target allocation to private equity to reflect the reality of their already increased exposure.
Trusting Efficient Market Hypothesis(“EMH”) and Modern Portfolio Theory (“MPT”) – Why?
One of the arguments for having these alternative assets in the portfolio is that they are less/ un-correlated with equities and bonds. However, this begs the interesting question… what time period do you use for correlations? If you truly believe that an endowment is permanent then you should use long-term ones (e.g. 10 or 30-year correlations). However, if you do that then the endowment and the institution associated with it have to accept that there will be a big down year (and hence budgets will have to be cut, people let go, etc) when the assets correlate. Can endowments really do that and are they willing to live with the consequences Siegel’s paradox?
In addition to this, there are the myriad of well-known problems with the EMH and MPT, the most important of which being the normality assumption. The conceptual conclusion of MPT is that asset-specific risk (non-systematic risk) is minor compared to asset-class risk (systematic risk) but this is dependent upon the assumption that investment risk is normally distributed. However, research has continually found that this normality assumption does not hold true for equities and bonds, let alone other (leveraged and illiquid) assets, resulting in a far larger left tail. Or more simply put, actual losses are regularly far higher than the expected losses.
Why have static allocation targets?
The faith in EMH and MPT leads to a broader question - why have static allocation targets? Shouldn’t the targets be dependent on the risk (i.e. shouldn’t price and ‘value’ matter)? While endowments may well have a long-term time horizon, this presumably does not obligate them to ignore current or medium-term information. One would think, as the attractiveness of asset classes changed both compared to their respective fundamentals and to other asset classes, the allocations would change. However, this seems not to be the case with the target allocations changing slowly (normally annually) and by very small amounts. The largest single reason was mentioned early in Part A, and is related to EMH/MPT, a focus on return targets!
A large part of finance, in particular in equity & equity-orientated world, asks us to believe that you can calculate an expected return for something. Couple with This is EMH/MPT telling us that the volatility is an appropriate measure of risk. It has resulted in a general belief that expected returns (i.e. target returns) are defined and inputs with that the level of risk that we take to achieve them merely being an output. As those of you who stumble to this blog will know, my opinion is the opposite of that; I find it easier to understand the level of risk that I’m taking, and allocate my capital to risk-related buckets, and view the return as an output of those decisions.
How else can one explain why endowments hate bonds (especially sovereign) and cash so much, yet adore equity-orientated strategies. Especially when you consider the propensity of sovereign bonds (particularly the US) to perform strongly at the time when an investor needs them the most; when equities suffer. As for cash, I’ve talked before on the blog about the positive optionality that cash offers; especially when times are uncertain, possessing it makes future decisions easier and allows you to be greedy when everyone else is fearful. In fact, for an endowment that’s heavily invested in illiquid strategies and barely any bonds, I would think holding a large amount of cash would be a prerequisite given the short optionality position that they face.
Given that endowments choose not to hold cash (or even more bizarrely hold negative amounts of cash by directly employing leverage) I can only posit that they must believe in the quaint notion that the value of assets must always be upward-trending and cannot possibly fall.
Until such time as Endowments recognize risk at the core of their portfolio allocation decisions, they are likely to be ‘shocked’ by years like 2008, and have to deal with the consequences of Siegel’s paradox. As my regular readers will know, my personal view is that that we have had a 25year+ era of increasing use of credit and leverage that has hidden many of the flaws in the markets and created numerous investing myths (starting with “markets are self-correcting”). Thus, while the Endowment Model represents a truly excellent way to have invested over the last 25-years, I believe that its lack of consideration for risk and its inflexibility means that it will likely fail to meet the needs of non-profit institutions going forward, something that I fear will be proved spectacularly over the coming decade.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment