just 2.5 %. The endowment model, then, does not seem to be an asset allocation story but rather a story of superior manager selection.
3.4.2 Effective Investment Manager Research
From a risk-budgeting perspective, many endowment managers prefer to spend the majority of their active risk budget in alternatives. On average, these large investors allocate to eight managers within traditional asset classes and more than 11 managers in alternative asset classes. Endowments have exploited both their networks of successful alumni and their first-mover advantage in allocating to the best-performing managers, many of whom have now closed their funds to new investors. While common wisdom assumes that the sole secret to endowment success is the large allocation to alternative investments, the top endowments enhance performance further by allocating to managers who outperform. In fact, the largest endowments have historically outperformed in nearly every asset class, both traditional and alternative. Those investors seeking to replicate the success of endowments should be cautioned that although this outperformance within asset classes can add up to 2 % per year to performance, it is unlikely to be replicated by an alternatives-heavy allocation if investors lack the talented staff and valuable network of invested managers that many endowments have cultivated.
An examination of university annual reports shows that in addition to an alternatives-heavy asset allocation, which can enhance returns and reduce risk, many universities have superior security selection and manager selection skills. This skill, or high risk tolerance, is seen when examining the return in each asset class portfolio compared to that asset class's benchmark. Harvard's 2014 annual report shows that returns beat asset class benchmarks annually over five years, including 1.9 % in real assets and 3.9 % in absolute return, while falling 0.5 % behind the private equity benchmark. Even in the asset classes considered to be most efficiently priced, such as publicly traded equity and fixed income, Harvard outperformed annually by 0.6 % and 2.9 %, respectively, over the same period. In total, Harvard's entire endowment portfolio beat the return on the asset-weighted benchmark by 1.4 % annually, adding more than $1 billion in excess returns in just five years.
Swensen (2009) demonstrates the importance of manager selection within the alternative investment universe. In liquid, efficient markets, the dispersion of returns across asset managers is relatively small. For example, in U.S. fixed income over the 10-year period ending in 2005, there was a mere 0.5 % difference in returns between managers at the first and third quartiles of returns. Equity markets have return dispersion across managers of between 1.9 % and 4.8 %. Pension plans often invest in passive index funds rather than with active managers in traditional asset classes. However, index funds are generally unavailable in most alternative investment asset classes. Siegel (2008) cites literature that documents the difference in annual returns between managers ranked at the 25th and 75th percentiles. These numbers are relatively small in traditional assets, with 0.5 % in fixed income, 2.7 % in U.S. equity, and 3.9 % in non-U.S. equity.
In contrast, the value added by active managers in alternative investments can be quite substantial. In inefficient markets, managers have a greater opportunity to profit from skill, information, and access to deal flow. Dispersion in alternative investments is much higher, with 7.4 % in hedge funds, 14.2 % in private equity buyouts, and 35.6 % in venture capital. In many cases, especially in private equity, investments are not attractive when investing in the median manager. In order for a private equity investment to outperform public equity on a risk-adjusted basis and adequately compensate for the liquidity risk in these investments, investors need to allocate to managers who deliver returns far above the median manager in each asset class.
3.4.3 First-Mover Advantage
It appears that the largest endowments have significant skill in selecting the top-performing managers within each asset class. Lerner, Schoar, and Wang (2008) explain that this ability to select top managers may be related to the first-mover advantage (i.e., benefits emanating from being an initial participant in a competitive environment): large endowments invested in many alternative asset classes years earlier than pension funds and smaller endowments did, and may therefore have an advantage. For example, Takahashi and Alexander (2002) explain that Yale University made its first investments in natural resources in 1950, leveraged buyouts in 1973, venture capital in 1976, and real estate in 1978. In contrast, Lerner, Schoar, and Wang explain that corporate pensions began investing in venture capital only in the 1980s, while public pension plans did not make their first venture capital investments until the 1990s.
Many of the funds of managers who have earned top-quartile performance in these asset classes have been closed to new investors for many years. Newer investors seeking access to top managers in alternative investment asset classes, especially in venture capital, are destined to underperform when the top managers allow commitments only from those investors who participated in their earlier funds.
Lerner, Schoar, and Wongsunwai (2007) show that endowments earn higher average returns in private equity, likely due to the greater sophistication of their fund-selection process. Endowment funds have higher returns than do other investors when making allocations to first-time private equity fund managers. Once an endowment fund has become a limited partner in a private equity fund, it seems to be more efficient at processing the information provided by each general partner. The follow-on funds that endowments select for future investment outperform funds to which endowments decline to make future commitments.
Mladina and Coyle (2010) identify Yale University's investments in private equity as the driving factor in the endowment's exceptional performance. It can be difficult to emulate Yale's outperformance in private equity and venture capital investments, as its venture capital portfolio has earned average annual returns of 31.4 % since its inception through fiscal year 2007. In fact, this study suggests that without the private equity and venture capital investments, the returns to the Yale endowment would be close to that of the proxy portfolio.
3.4.4 Access to a Network of Talented Alumni
Perhaps the first-mover advantage and the manager-selection skill of top endowments can be attributed to the superior network effect. An institution has a positive network effect when it has built relationships with successful people and businesses that may be difficult for others to emulate. Alumni of the universities in Exhibit 3.1 are noted for being among the most successful U.S. college graduates, in terms of both academics and business. As measured by scores on the SAT® exam,20 Harvard, Yale, Stanford, Princeton, and the Massachusetts Institute of Technology routinely select from the top 1 % to 5 % of students. In 2003, the median SAT score for all college-bound students, including both verbal and mathematics scores, was approximately 1,000. Top universities attract students with average scores exceeding 1,400. Graduates of these schools also tend to have the highest initial and midcareer salaries.
A study by Li, Zhang, and Zhao (2011) correlated manager-specific characteristics to the returns of the hedge funds they managed. In contrast to the median SAT score of all college-bound students of 1,000, Li, Zhang, and Zhao found that the middle 50 % of hedge fund managers attended colleges and universities with average SAT scores between 1,199 and 1,421 (the 79th and 97th percentiles, respectively), demonstrating that the majority of hedge fund managers attended the most competitive colleges and universities. Within the group of studied hedge fund managers, the research showed that those who attended undergraduate colleges with higher average SAT scores have higher returns and lower risk. For example, a 200-point difference in SAT scores, such as that between 1,280 and 1,480, was correlated with higher annual returns of 0.73 %. Not only did managers who attended top universities have higher returns, but they did so at lower risk and earned greater inflows during their tenures as fund managers. The authors suggest that talented managers are attracted to hedge funds due to the incentive fee structure, which rewards performance over asset gathering. In contrast to their studies on hedge fund managers, Li, Zhang, and Zhao found that SAT scores did not seem to affect the asset gathering or excess returns earned by mutual fund managers, as these managers are compensated for gathering assets, not for earning excess returns.
Many alumni of top universities wish to continue an association with their alma mater, the university from which they received