A private foundation is an independent legal entity that provides a vehicle for charitable giving. Private foundations play an increasingly important role in modern society through their support of public charities and their own philanthropic programs. In tax year 2016, private foundations had roughly $900 billion in total assets and distributed nearly $65 billion to support charitable objectives. While the number of private foundations has steadily increased over the past 20 years, just over 1,000 foundations make up 63 percent of total assets and 50 percent of charitable dollars.
Most private foundations receive minimal contributions after their founding, but instead rely heavily on their investment portfolios to support their charitable spending. For example, the Ford Foundation was founded in 1936 by Henry Ford’s son with an initial gift of $25,000 and an additional gift of $250,000 from the family in the mid-1940s. Since then, the foundation has received no additional contributions, but has grown its endowment to more than $16 billion through prudent investing. Since 1976, private foundations have been required to pay out a minimum of five percent of their average fair market value of net investment assets each year. Despite their systematic importance to the United States charitable sector and unique operating structure, surprisingly little is known about private foundations’ investment performance, asset allocation decisions, investment fees, and spending behavior.
Private foundations are relatively unconstrained long-term investors with the desire to spend the investment earnings from their endowment in support of their charitable causes. Private foundations differ in their liability structure from pension funds, as private foundations are constrained only in the short-term to give five percent annually, whereas pension plan beneficiaries represent liabilities that must be met over time. Moreover, private foundations often receive their original wealth from successful families or individuals in the form of stock (e.g., Bill and Melinda Gates Foundation: Microsoft stock, Kellogg Foundation: Kellogg Stock, and Robert Wood Johnson Foundation: Johnson and Johnson stock), which makes them less diversified than other institutional investors, such as university endowments. Private foundations rely heavily on their endowment investment income to meet the five percent required spending rule each year and to maintain the real value of their endowment, because they seldom engage in fund-raising activities or receive government support. In contrast, most colleges and universities in the United States rely on a mix of government grants and contracts, tuition and fees, investment return, and private gifts and grants to support their operating budgets. Finally, private foundations are important as they contribute to the efficient allocation of philanthropic capital between donors and charitable entities and provide a credible signal to donors of the potential for charities to achieve their missions. These unique features speak to the importance of a comprehensive study of private foundations’ asset allocation, investment performance, and payout policies.
There is little research to date on the investment performance of private foundations, mostly due to the lack of high-quality data. To fill this void, we use forms 990-PF filed with the Internal Revenue Services (IRS) between 1991 and 2016. We collect data on dividends and interests from securities, net capital gains (or losses) from the sale of assets, contributions, distributions, expenses, and a breakdown of investment assets and liabilities for the universe of private foundations in the United States. Because we cannot observe unrealized capital gains (or losses) we estimate (and validate) total returns using changes in net asset values (NAV) unrelated to charitable inflows and outflows.
The average (median) foundation has $37 million ($5.5 million) in total assets, while total assets are about $3.5 billion on an asset-weighted basis. Distributions to charitable sectors represent, on average, six percent of assets and about three-fourths of total expenses. We also document that private foundations have, on average, strong investment performance. The large heterogeneity we observe in investment performance likely stems from a combination of asset allocation decisions and the ability to select high-performing managers. We document a shift over time towards riskier assets – in particular, alternative investments for large private foundations – and a steady decline in fixed-income investments, such as government bonds.
Asset Allocation and Risk Taking
Asset allocation is one of the most important decisions an investor makes, as it is a key determinant of both investment return performance and risk exposure. We document that larger foundations are associated with higher allocations to risky assets, such as public equity and alternatives. Interestingly, as private foundations age, their share of capital allocated to public equity decreases. This is consistent with private foundations receiving initial endowments in the form of common stock from wealthy families and later seeking diversification across asset classes. Increased investment sophistication also explains trends in asset allocation: foundations with larger teams of highly paid individuals are associated with larger allocations to alternatives.
We explore one channel for the increased risk-taking behavior of private foundations over time. As most foundations seek to live in perpetuity, the task of spending five percent of their corpus each year while maintaining their real principal value becomes increasingly challenging, especially for those foundations relying on a constant investment income stream to support their charitable goals. We show that private foundations are more likely to “reach for yield” when conservative asset allocation policies are not sufficient to cover distributions without eroding their principal. This reach for yield behavior results in foundations shifting their asset allocation away from safer investments (government and corporate bonds) to riskier investments (equity and alternatives).
Our data also allow us to study the investment performance of the universe of private foundations. First, we attribute a large share of return variability to asset allocation in domestic and international equity, fixed income, hedge funds, and private equity. However, larger foundations seem to carry out more active investment programs, as their returns cannot be fully explained by these benchmark indices. On a risk-adjusted basis, foundations with more than $500 million in total assets generate alphas ranging from 150 to 230 basis points per year. On the other hand, smaller foundations do not generate alphas, on average. These risk-adjusted performance results suggest that larger private foundations are sophisticated investors whose superior investment performance enables them to have a greater impact towards their philanthropic missions.
Prior studies find that nonprofits underperform in their investment performance using data from 2009 to 2018 (Yermack & Dahiya, 2021; Lo, Matveyev, & Zeume, 2021). Consistent with their results, we also find that private foundations underperform during this more recent period. However, we document variation over time in alphas and stress the importance of analyzing longer time periods to capture this variation and provide more meaningful conclusions. Second, we show that investment performance exhibits some persistence over time, which further provides evidence of the investment skill of private foundations. This performance persistence is likely a feature of each foundation’s strategic asset allocation and ability to consistently select better investments (e.g., Binfarè, Brown, Harris, & Lundblad, 2022). Third, we show that concentrated foundations (e.g., foundations that have a large proportion of assets invested in a single stock) outperform, but the much higher idiosyncratic risk of these concentrated foundations completely offsets the higher expected returns.
Performance and Capital Preservation
“In a foundation, I have a mandate of 5 percent payout. So, I have to have at least 70 percent of equity risk in this portfolio to be able to achieve, on a long-term basis, the objective which is to grow or maintain the real spending power of the institution’’– Ana Marshall, Chief Investment Officer for the William and Flora Hewlett Foundation.
Although some foundations perform well, the minimum five percent rule places private foundations under great strain to achieve a net investment return of five percent to sustain nominal principal or five percent plus inflation to sustain their real principal.
To make broader recommendations for maximizing the real value of private foundations’ giving moving forward, we conduct a simulation study to examine how private foundations’ real principal values are expected to change over the next 25- and 100-year periods under varying investment strategies. Our results suggest that many foundations are in danger of losing their real purchasing power of assets without increases in risk-taking and broader exposure to alternatives to overcome the low-yield investment environment. When first passed into law in 1976, the five percent minimum spending rule was created to inhibit private foundations from solely hoarding wealth and to provide a sustainable benchmark that private foundations could meet philanthropic needs and still maintain their real value of invested principal. For many private foundations, the ability to meet the five percent rule and sustain operations in perpetuity is out of reach as all investment strategies, except the aggressive growth with alternatives portfolio, result in a greater than 50 percent chance of a foundation experiencing a decline in real principal over longer time horizons. While it seems uncharitable to set a lower mandated giving rate, the optimal spending rate to maximize future charitable distributions must enable private foundations to continue to grow their endowment base rather than deplete it.
The inflexibility of the five percent spending rule is another inefficient legislative feature as the optimal spending rate for a private foundation likely results from a complex set of interactions between its strategic asset allocation, return (asset) volatility, mission, and time-horizon. The mission and goal of a nonprofit necessarily capture its rate of time preference for future expected consumption (e.g., the real spending on charitable goals). For example, private foundations seeking to eradicate poverty, hunger, or clean water crises would be rational to spend their current invested principal more aggressively in fighting these needs due to the high-value creation of these projects. On the other hand, private foundations seeking to support inter-generational causes such as art and higher education should seek to solely maximize the present value of their distributions by selecting a spending rate near the optimal spending rate that can be supported over long periods. It is important to note that many foundations already give in excess of the five percent mandate suggesting this reduced benchmark would not necessarily reduce charitable giving in the short-term, but instead provide greater flexibility to private foundations to select a spending rate based on the urgency of the mission they support and time horizon they seek to operate.
The expected life of a foundation is another important dimension to consider. Foundations that decide to live in perpetuity are willing to smooth out their spending over time to be sustainable. On the other hand, other foundations might decide to deplete their capital over a pre-determined time frame (e.g., The Bill & Melinda Gates Foundation will spend all its assets within 50 years of them both dying).
Our simulation results suggest that over long horizons (i.e., 100 years), the optimal spending rate is almost always strictly less than 5 percent. However, our simulations mask large variations in spending rates depending on the rate of time preference. For a discount rate which mirrors the discount rate of the average university endowment, the optimal spending rate for a portfolio with alternative assets is about 4.10 percent which resembles the typical spending rate for colleges and universities. In summary, given the results from our simulation, we believe that lowering the mandated spending rate and providing greater flexibility to foundations based on their missions and time horizons would result in greater efficiency and productivity of the United States charitable sector moving forward.
Matteo Binfarè is an Assistant Professor of Finance at the Robert J. Trulaske Sr. College of Business at the University of Missouri. He earned his Ph.D. in Finance from the University of North Carolina at Chapel Hill in 2020.
Kyle Zimmerschied is a doctoral student in Finance at the Robert J. Trulaske Sr. College of Business at the University of Missouri.
The views expressed in this post are those of the authors and do not represent the views of the Global Financial Markets Center or Duke Law.