According to the most recent NACUBO-Commonfund study, the total size of assets controlled by university endowments was about $522 billion in 2016, with an average size of about $641 million. Harvard University has the largest endowment, with $34.5 billion. Endowments control a large pool of capital, and there are significant differences in their sizes. For example, the top 5% account for about 60% of the total assets. As expected, the largest endowments, having the most resources and benefiting from economies of scale, have had the best performance since 2000.
However, the differences are not as significant as would be expected and they have been shrinking in recent years.
Exhibit 1 shows annual returns on all endowments, largest endowments, median endowments and the global 60/40. It appears that in recent years the largest endowments have lost the edge they displayed in early 2000 relative to other endowments and the 60/40 portfolio.
Also, the 60/40 portfolio has performed better than most endowments in recent years (Exhibit 2).
The index portfolio of Exhibit 2 applies the reported asset allocations of endowments to a set of indexes representing those asset classes held by endowments. For instance, the Russell 3000 is used to represent U.S. equity while CISDM-Morningstar and Cambridge Associates indexes were used to represent hedge funds, private equity and venture capital investments. Since 2008, asset allocation and security selection decisions of most endowments have not made any positive contribution to their performance.
Since NACUBO- Commonfund provides information about the performance of different asset classes held by endowments, we can see how the playing field has become more level in recent years. That is, not only have the largest endowments lost their edge, but also collectively they have lost whatever edge they have had relative to passive indexes (see Exhibit 3).
While in 2011 the largest endowments had the top-ranked performance in almost every category (the sole exception being fixed income), the rankings for 2016 appear to be almost random. They had the best performance in only three categories and were tied for best performance in two other. Equally important, the indexes representing the asset classes held by endowments appear to outperform the securities selected by them. Of course, reported returns of endowments are net of fees while returns on some of the indexes are gross of fees. However, endowments' reported returns are not net of all costs. NACUBO reports that in addition to the fees paid to external managers, endowments spend around 60 basis points internally to manage their assets.
Why has there been such a lack of alpha on the part of endowments, and why has the performance deteriorated in recent years?
It is hard to imagine why endowments should underperform indexes representing liquid, publicly traded securities. Here, for instance, the U.S. equity market is represented by the Russell 3000 while non-U.S. equity is represented by an 80/20 portfolio of MSCI's EAFE and Emerging Market indexes. Liquid exchange-traded funds that represent these and a few other classes (e.g., indexes of publicly traded securities were used to represent fixed income, private real estate, and energy and natural resources) are readily available. Endowments could easily have matched the performance of these indexes.
Their hedge fund, private equity and venture capital investments have not performed well in recent years. Given the large allocations that endowments make to these asset classes, it is crucial for them to find best of breed and not to allocate to these asset classes only because they have a target weight to meet.
Also, fees spent on managing assets that are unlikely to be sources of alpha have reduced performance. Finally, staff turnover at some of the largest endowments has not helped either.
What should they do to improve performance? First, increasing liquidity and reducing costs can go a long way toward improving returns. Increased liquidity will allow endowments to take advantage of temporary market dislocations and be buyers when others are sellers. It also would prevent the forced liquidations seen during the 2008-2009 financial crisis.
Second, they should move most of their allocations of traditional assets classes into liquid ETFs.
Third, they should use a bottom-up approach to determine optimal allocations to alternatives. The largest endowments allocate around 50% of their assets to alternatives, and there is remarkably little variation among them. It is hard to believe that all large endowments can identify about the same number of highly skilled managers in this space such that they end up with the same allocation. Herd mentality appears to be the driving force here. Using a bottom-up approach, the endowments will invest in these highly heterogenous asset classes only if they are certain that they have identified a reliable source of alpha.
Finally, the so-called endowment model, which advocated aggressive allocations to illiquid assets to harvest the illiquidity premium is no longer valid. The supply of illiquidity premium is finite and not every illiquid asset is a source of illiquidity premium.
At the extreme, if an endowment values liquidity highly and/or cannot find any reliable source of alpha, a highly diversified portfolio of ETFs will do fine. In fact, a portfolio of 18 ETFs using allocation weights provided by NACUBO has performed remarkably well since 2000.
Hossein Kazemi is senior adviser to the Chartered Alternative Investment Analyst Association and professor of finance at the Isenberg School of Management at the University of Massachusetts, Amherst. This article represents the views of the author. It was submitted and edited under Pensions & Investments guidelines, but is not a product of P&I's editorial team.