Introduction
Much ink has been spilt over the investment portfolios of higher education endowments, particularly those in the United States. These analyses typically rely on university endowment disclosures and data released to NACUBO (National Association of College and University Business Officers). The following report investigates non-profit endowments and foundations which are not involved in higher education. We utilize a unique dataset, U.S. Internal Revenue Service Form 990 and Form 990-PF filings, and their associated Schedules, to investigate the size and performance of non-higher education endowments.
The asset owners that are the focus of this report, namely larger non-profit endowments with regular investment activity, typically have long-term investment horizons, a wide range of investible securities and strategies, and predictable liquidity needs. Given the wide range of investment options, non-profit fiduciaries can benefit from benchmarking their portfolios and performance to others operating under similar constraints.
Download a copy of the report here
U.S. Non-Profit Endowment Dataset
Our dataset covers endowments which manage more than $100 million in assets and have electronically filed Form 990 or Form 990-PF from 2016 through the 2024 tax year. The tax filing dates covered in this report are dictated by the IRS’s disclosure schedule, a handful of 2025 filings available at time of writing. Our final sample includes over 2,500 foundations and endowments managing an estimated $1.76 trillion through 2024, after a number of modifications to the full dataset as elucidated below. The fuller dataset is available to clients of Nasdaq eVestment’s Peer Benchmarking solution.
The primary goal of this piece is to investigate endowments with regular and meaningful investment activity. The first modification we make to the dataset is to set an asset threshold, specifically at $100 million. However, we do keep the full historical records of any entity that crosses the $100 million threshold at any point in the 2016 to 2024 tax periods. Secondly, university endowments are removed from the sample set as we seek to investigate non-higher education endowment performance – this is a sizeable marginal reduction with over 1,000 entities removed.
The last systematic modification we make to the dataset involves the estimation of endowment investment performance. The precise methodology is discussed in the following sections, but for the purposes of triangulating our desired sample set, we remove endowments with irregular performance. There exist instances in which contributions or distributions, as well as endowment liquidations, produce performance estimates which do not reflect investment returns – these are removed from our sample. Further reductions in the sample have relatively minor impacts including duplicative entries and cases in which end-of-year assets do not balance out against beginning-of-year assets and reported income and expenses.
U.S. Non-Profit Endowment Assets
Like other institutional investor types, non-profit endowment assets are concentrated at the right tail. Those with more than $5 billion in assets account for 3% of entities, but 45% of total assets (roughly $800 billion) through the most recent period, as shown in Figure 1. Zooming out, endowments with more than $1 billion in assets account for 13% of entities and 71% of total assets.
At the left tail, the $100 million to $250 million range is the most populous and represents half of the final cohort, but only 11% of total assets. Lastly, as mentioned previously, our sample excludes endowments with less than $100 million in assets unless they did report managing $100 million or more at any point between 2016 and 2024 – even with this restriction, 4% of endowments fit in the sub-$100 million category at most recent reading. We note that our full sample is substantially, but not fully captured in the chart below as we incorporate endowments that have since liquidated or have delayed 2024 filing.
Figure 1: U.S. Non-Profit Endowment Asset Distribution
Note: Assets based on 2024 filing year.
At this juncture, we must address the fact that endowments have varying filing dates. Figure 1 and the statistics cited above commingle asset data based on each endowment’s most recent filing. The asset data is reproduced with cohorts grouped by filing date in the appendix of the full report. We find that asset distributions maintain similar characteristics to those outlined above across all filing periods.
U.S. Non-Profit Endowment Portfolio Returns
Non-profit endowments and foundations typically do not disclose investment performance to the general public. However, we can use the same tax filings to estimate investment returns. We follow the existing literature1,2 to produce whole-portfolio performance estimates using the following formula.
Figure 2 shows net investment returns grouped by calendar year. Non-profit endowments exhibit considerable dispersion in investment performance. Top quartile performance in the 2024 tax year ranged from 13.6% to 27.6%, while the bottom quartile ranged from -2.1% to 8.1%. The average 2024 tax year return was 11.2%, above the median of 7.0% – from 2016 to 2023, the arithmetic average of the average annual return was 6.0%. Filing date periodicity, i.e.: when endowments’ specific tax years end, does have an impact on the dispersion shown here, but we find dispersion to be high even after controlling for this factor. We cover periodicity more fully in the following sections and include a breakdown of performance by filing date in the appendix of the full report.
Figure 2: U.S. Non-Profit Endowment Performance Distribution
Note: Investment performance is grouped by calendar year of filing irrespective of filing date. The top and bottom quartiles exclude the top and bottom 1% of returns, respectively.
Endowment size did not have a significant effect on investment performance across our sample. Below we show performance distributions for endowments with assets above and below $1 billion at the start of the fiscal year. We see that the average and median returns of the two cohorts are closely linked. The annualized return for the average large endowment was 6.4% versus 6.3% for the average smaller endowment.
However, we do find some dissimilarities. Firstly, large endowments tend to exhibit a narrower range of outcomes, albeit still quite large in relation to other institutional investors. This is likely reflective of large endowments’ exhibiting more homogeneity in their asset allocations, and less portfolio volatility, in relation to smaller endowments. Secondly, we find that smaller endowments tend to outperform large endowments when public equity markets outperform and vice versa. For example, large endowments posted an average return of -7.6% in tax year 2022 versus smaller endowment’s average of -10.0%; the MSCI ACWI index posted a total return of -18.0% during calendar year 2022. Similarly, the average smaller endowment outperformed the average large endowment by 1.4% in tax year 2023 (MSCI ACWI 22.8%) and 0.7% in tax year 2024 (MSCI ACWI 18.0%). This points strongly to the smaller cohort taking more equity risk and comports with the academic literature (Lo, Matveyev, and Zeume (2025)).
While Form 990 and 990-PF do offer some insight into asset allocations, it is fairly limited in scope. For instance, “publicly-traded securities” comprised 37.4% of assets for smaller endowments in 2024 versus 31.2% for large endowments. However, publicly-traded securities do not encompass these investors’ full exposure to public markets as a large portion would be subsumed under managed funds, partnerships, etc. A fuller accounting of asset allocations will be left for future explorations into endowment and foundation investing.
Figure 3: Non-Profit Endowment Performance Distribution by AUM Category
Note: Investment performance is grouped by calendar year of filing irrespective of filing date. The average tax year return is distinguished by color, yellow indicates the higher return and black the lower return. The top and bottom quartiles exclude the top and bottom 1% of returns, respectively.
Lastly, we look at non-profit endowment returns against two model portfolios, shown in Figure 4. The first is a U.S.-biased 60/40 portfolio and the second is a stylized “institutional” portfolio with more non-U.S. exposure, a broader spectrum of liquid fixed income, and alternative investments. Due to filing date periodicity, we disaggregate our sample into three groups to enable more apt comparisons to our model portfolios. The groups include those that file within 7 days of June-30, September-30, and December-31, e.g.: January-6 filers would be included in the December-31 cohort, and together comprise over 90% of our sample. The remaining filers (March-31 and non-quarter end filers) are excluded from this analysis as they were limited in number and because our private markets benchmark (Nasdaq Private Capital Global™ Index) and hedge fund benchmark are calculated quarterly.
Across all three filing groups, the average endowment’s correlation with the model portfolios ranges from 0.95 to 0.99 with the “institutional” model portfolio showing a slightly higher correlation against all cohorts. The average endowment also underperforms against both of our model portfolios in almost all periods. The magnitude of this underperformance suggests behavioral and structural effects, and cannot be accounted for by investment management fees alone. Endowment and foundation investment officers may find the exercise useful in benchmarking their own portfolios.
Figure 4: Rolling Annual Returns vs. Model Portfolios
Note: Endowment performance grouped by filing date, e.g. the 2016 return for the June Filers group covers the period from July 2015 through June 2016. Model portfolios rebalance annually by filing group at the filing date. Index returns are measured in USD on a total return basis. The Nasdaq Private Capital Global™ Index and hedge fund returns are NAV weighted.
Conclusion
University endowments have undergone multiple evolutions with the “Yale model” having an outsized impact over the past two decades and the “Total Portfolio Approach” now forcing CIOs to rethink investment processes and governance. Non-higher education endowments have similarly evolved, but thought leadership on the unique challenges for these institutions has been scant. Differences in scale, distribution decisions, regulations, governance and so forth produce significant downstream effects on portfolio construction. Fiduciaries of non-profit endowments, and their various stakeholders, can benefit from benchmarking their portfolios against other institutional investors – particularly those operating under similar constraints. We look forward to publishing more on this underserved segment.
Endnotes
- Lo, Matveyev, and Zeume, 2025, “The Risk, Reward, and Asset Allocation of Nonprofit Endowment Funds.”
- Dahiya and Yermack, 2021, “Investment Returns and Distribution Policies of Non-Profit Endowment Funds.”
