Investors have no chance of adding alpha by pursuing an “endowment” model. New research shows that even the most sophisticated institutions do worse when they increase exposure to alternative asset classes, and that investors would be better served with a passive, 60/40 allocation.
At the end of fiscal 2019, the National Association of College and University Business Officers (NACUBO) compendium of annual endowment data showed that the 774 reporting endowments held more than $600 billion of assets. The average asset size was about $800 million, though the median asset size was only about $100 million.
The pressure on budgets, along with high equity valuations and historically low bond yields, led many endowments to try to improve returns by increasing their exposures to alternative investments such as private equity, private real estate and hedge funds. Unfortunately, the research from the 2013 study “Do (Some) University Endowments Earn Alpha?”, the 2018 study “Investment Returns and Distribution Policies of Non-Profit Endowment Funds,” the 2020 study “Institutional Investment Strategy and Manager Choice: A Critique,” and the 2020 study, “Endowment Performance,” show that factor models explain virtually all of the variation in performance of endowments, and that despite taking on more risks in the form of often opaque and illiquid investments (such as hedge funds, venture capital and private equity), there has been no evidence that the average endowment is able to deliver alpha relative to public stock/bond benchmarks. In fact, alternative asset classes have failed to deliver diversification benefits and have had an adverse effect on endowment performance.
Dennis Hammond contributes to the literature with his 2020 study “A Better Approach to Systematic Outperformance? 58 Years of Endowment Performance,” published in the August 2020 issue of The Journal of Investing. His study covers the 58-fiscal-year period ending June 2019. The data is from the NACUBO compendium of annual endowment data, which extends back to July 1961. His objectives were to determine whether endowments met various return goals over time and whether returns on average could be systematically improved. Because a disproportionate ownership of endowment assets (75%) is held by a relatively small percentage (14%) of large endowments, Hammond used the equal-weighted rather than the dollar-weighted mean of endowment returns to define the average return. (Dollar weighting would provide insight into the comparative performance of the average endowment dollar, but not the average endowment.)
By convention, endowment performance data are reported to NACUBO net of external expenses but gross of internal expenses. Thus, net returns are overstated. Following is a summary of his findings:
- Over the 58-year period data are available, the average endowment significantly underperformed its annual return need, its long-term return objective, and the traditional 60/40 passive benchmark.
- Over the 58-year period, the average endowment return had compounded 2.1% below the long-term return objective. On an annual basis, over the shorter, most recent 46-year period, endowments underperformed the long-term return objective by 0.8% annually.
- Over the fiscal years 2009-18, the average endowment failed to earn its long-term return objective in even a single year.
- Over the 58-year period, the long-term return objective was reduced from 10% to 7.1%.
- Relative to the average endowment, the traditional 60% U.S. stocks/40% U.S. bonds mix provided superior absolute and risk-adjusted returns over the 58-year period by 1% (similar to the difference between the costs of active versus passive strategies).
- The average endowment return underperformed the 60/40 benchmark in each of the last 58-, 50-, 40-, 30- and 10-year periods. The only period in which the average return outperformed the 60/40 mix was the trailing 20-year period, in which it outperformed (on a gross basis) by 50 basis points (bps) annualized.
- The largest cohort of endowments outperformed both the average and the smallest cohort. This could be the result of access to superior managers, the ability to negotiate lower fees, or a combination of both.
- The large cohort outperformed the 60/40 mix over the last 50- (0.1%), 40- (1.8%), 30- (1.1%) and 20-year (2.7%) periods. However, its outperformance of the 50-year period was just 10 bps, and that doesn’t consider internal expenses (the large cohorts tend to have higher internal expenses, including their own CIOs). In addition, over the most recent 10-year period, it underperformed by 1.5% per year.
These findings led Hammond to conclude that rather than shifting allocations to riskier (and much more expensive) alternative investments, the average endowment, and especially the smaller endowments, would have been better served by simply adopting passive, low-cost strategies.
Summary
Hammond found that over the 58-year period data are available, the average endowment significantly underperformed its annual return need, its long-term return objective, and the traditional 60/40 passive benchmark. And while the largest cohort of endowments basically managed to match the return of the benchmark over the full period, it underperformed by 1.5% over the most recent 10-year period. The result could be reflective of the increasing difficulty managers have experienced due to the “Incredible Shrinking Alpha.”
Hammond concluded: “The fact that the average endowment failed to meet both internal goals and the traditional external benchmark in so many extended periods, both recently and long-term, is cause for serious concern. Perhaps even more compelling is the finding that the average endowment could have significantly improved its performance by simply replicating a passive benchmark. Trustees of the majority of endowments would be well served to consider the implications of these findings in managing their endowment and funding their institution’s prime directives.”
Larry Swedroe is the chief research officer for Buckingham Strategic Wealth and Buckingham Strategic Partners.
Read more articles by Larry Swedroe
