SSIR: Endowment for a Rainy Day
Judging from media accounts, U.S. nonprofits are facing unprecedented, if not catastrophic, financial distress because of endowment losses. Hiring is being frozen, facility maintenance is being deferred, programs are being dropped, performance seasons are being shortened, and construction projects are being cut back or even halted. As the president of Harvard University, Drew Gilpin Faust, put it when defending her decision to sharply reduce expenditures following a 30 percent drop in the value of the school's endowment, "Tinkering around the edges will not be enough."
Harvard isn't the only institution making dramatic cuts in response to a falling endowment. The J. Paul Getty Trust, which runs the J. Paul Getty Museum in Los Angeles, slashed 14 percent of its workforce and delayed exhibitions and acquisitions after its endowment fell from $6.4 billion to $4.2 billion. Yale University cut capital expenditures by $2 billion and staff salaries and benefits by 7.5 percent after its endowment fell from about $23 billion to about $16 billion. And the Shriners Hospitals for Children considered closing six of its twenty-two children's hospitals after its endowment fell from $8.3 billion to $5 billion. The Shriners tabled that motion, but are considering billing insurance and Medicaid for treating children, a profound change from the free patient care that they have traditionally provided.
Of course, no organization likes losing a quarter or more of its wealth. And any organization that does must make financial adjustments. But the actual story behind falling endowments and the resulting financial crisis is much more complex. It turns out that only a small percentage of nonprofits — albeit prominent ones with large endowments — suffered losses of 25 percent or more. Most nonprofits have no endowment at all. And those nonprofits that do have endowments generally have small ones, which, ironically, declined less in value than did the large endowments.
Even more important is how those nonprofits that have endowments responded to the declines in the value of their holdings and other revenue (such as donations and user fees). Most of these nonprofits took the same path as Harvard, Getty, and Yale, choosing to cut expenditures to compensate for reduced income. Though many of these organizations have substantial endowments to fall back on — enough to offset many years of lower income streams — most choose not to draw down the endowment to make up for reduced income.
We believe that most nonprofits are taking the wrong approach when it comes to managing their endowment. Instead of husbanding money for the future, nonprofits should treat at least some portion of their endowment as a rainy day fund, a source of money that is available to make up for those unexpected yet perennial times when income drops or demand for services increases.
The Endowment Boom
The first thing to understand about endowments is that their plummeting value is from all-time record highs. A 30 percent drop actually returns most nonprofit endowments to the level they were at just four years ago, even after controlling for inflation. Consider Brown University: school officials estimate that its endowment will end the 2009 fiscal year at $2 billion, a shade more than the $1.8 billion the school had at the end of the 2005 fiscal year. In 2004 and 2005 nonprofit universities, hospitals, and museums were not crying poverty and were not cutting construction, programs, faculty, or financial aid to students. To the contrary, they were expanding programs and services.
Wealthier nonprofits grew their endowments faster than poorer ones. Rich organizations were able to achieve a higher rate of return because they took on greater risk by investing in products like hedge funds, commodities, and private equity. One of the risks of investing in these types of products is that they are volatile and relatively illiquid, making it difficult to sell the asset when markets are weak, as in the recent financial decline. Of course, it is exactly during poor financial times when a nonprofit might need to sell assets and access its cash to offset declines in donations, fees, and other types of income.
It was not chance that the endowments of the wealthiest nonprofits were growing so fast. The rich can afford to invest in high-risk, high-return products. The poor cannot. And their comparative returns on endowments show the net effect. According to the National Association of College and University Business Officers' NACUBO Endowment Study 2007, between 1998 and 2007 colleges and universities with endowments of less than $25 million posted a 6.7 percent average annual rate of return, while colleges and universities with endowments of more than $1 billion posted an 11.1 percent average annual rate of return. A typical nonprofit college had an $80 million endowment in 2007 and received about a 7.9 percent annual return. The impact of these different rates of return, when compounded over several years, can be dramatic. Over twenty years the difference between a 7.9 percent and an 11.1 percent annual rate of return is a whopping 81 percent higher return for a wealthy college when compared to a typical college.
The vast majority of colleges and universities, however, have little or no endowment to fall back on in a recession. One third of all two-year colleges and 11 percent of all public and nonprofit four-year schools report they have no endowment. (These figures do not include for-profit schools, which have no endowments.) And 85 percent of the two- and four-year schools reporting data have endowments of less than $100 million. These smaller endowments contribute at best only a few million dollars to a school's annual budget — a small portion for most schools. Schools with small or no endowments spend essentially all of their income — from tuition, fees, donations, and other sources — every year. These schools have no savings to sustain activities and expenditures when revenues drop.
The Purpose of an Endowment
In the 1890s, Samuel Gompers, the president of the American Federation of Labor, famously responded to the question, "What does labor want?" by saying, "More." That seems to be the answer to today's question, "What size of endowment does each nonprofit want?" But this answer is unsatisfactory. A nonprofit may want a larger endowment, and it can have it — but at the cost of cutting current spending. It can have a bigger endowment or it can spend more now, but it achieves one by forgoing some of the other.
If the purpose of an endowment — that is, of holding down current expenditures in order to save for the future — is to limit tuition, hospital patient charges, or museum admission fees later, then the issue is one of intergenerational transfers. The holding, let alone the expansion, of endowment is a matter of weighing the relative importance of today's and tomorrow's users. With long-term economic growth of per capita income a virtual certainty, however, it is not clear why the present generations of college students, patients, and museum goers should pay higher charges and fees in order to preserve or expand endowment so that future generations of wealthier people will benefit via lower prices.
The primary goal of a college endowment should be to protect a school's educational and research programs. Similarly, a hospital's goal should be to treat patients and advance medical research. A museum's goal should be to advance cultural education and preserve cultural heritage. Building an endowment is a means to sustain these programs. It should not be the goal of the programs to protect the endowment, cutting them back to sustain or rebuild the endowment. "In conversation with our donors, their motive was to support the university, not to grow the endowment," observed Sandy Wilcox, president of the University of Wisconsin Foundation.
The basic rationale for adding resources to endowment rather than using them to achieve these immediate goals is simple: to save for a rainy day when revenue falls sharply. The principal motive for saving for the future rather than spending now is the same, whether for individuals or organizations, nonprofit or for-profit — uncertainty about the flow of future revenues.
Every nonprofit will, at some time, encounter a situation that is beyond its control and that has an immediate and substantial impact on its operations. It might be a natural disaster, like an earthquake or fire, or it could be a financial collapse, such as the one that the world is currently undergoing. Nonprofits that have held money in reserve for just these sorts of situations — a rainy day fund — will be better able to weather these storms than nonprofits that did not prepare.
Rainy Days Happen
In 2005, Hurricane Katrina struck the Gulf Coast of the United States. Extensive damage forced colleges and universities such as Tulane University, Loyola University New Orleans, and Dillard University to close down temporarily. Tulane, for example, suffered a 10 percent drop in its total revenues between fiscal years 2005 and 2006, including a stunning 45 percent drop in hospital revenues. Xavier University lost more than 12 percent of its revenue, and Loyola University New Orleans was hit with a staggering 26 percent decline in revenue.
The current economic crisis certainly constitutes a rainy day. Most nonprofits, however, have not adequately planned for this event. Many have not created any endowment and so have no reserves to ride out the storm. Some have substantial endowments that would allow them to ride out the current economic crisis with no cuts, if they choose. But instead of drawing down their endowments to cover the rainy days, they have focused on cutting budgets.
The American Conservatory Theater in San Francisco, for example, cut $1.5 million from its $21 million budget, letting go three high-level managers, suspending non-main-stage productions, and shortening performance runs. The Rhode Island School of Design closed its Museum of Art for the month of August, laid off school staff, froze wages, and reduced its contribution to employees' retirement accounts. Princeton University, even after the loss of 24 percent in endowment in the last year, still has a massive endowment of $12.6 billion — ten times its total annual budget. But the university reduced its spending for 2009-2010 by $50 million, choosing to cut programs and staff rather than dip further into the endowment.
Some nonprofits are trying to compensate for financial shortfalls by increasing revenues. In a depressed economy, however, increasing donations, government funding, or user fees is difficult and may undercut the organization's mission. Colleges can increase the number of students, as Columbia University has, to increase revenue. Or they can admit fewer low-income students and more students whose families can afford to pay full tuition. Reed College, for example, was unable to sustain its "need-blind" admissions policy this year and admitted more than a hundred students who could pay their way instead of students, already on the admissions list, who needed financial aid. Other nonprofits are increasing fees. The Philadelphia Museum of Art, for example, raised its admission prices in July 2009 for the second time in two years.
A few nonprofits are considering merging to reduce operating costs. Andover Newton Theological School and Colgate Rochester Crozer Divinity School, for example, are discussing merging, despite being in different states (Massachusetts and New York, respectively).
Other nonprofits are choosing to borrow money. In April 2009, Stanford University, reeling from endowment losses and in need of short-term liquidity, sold more than $1 billion of bonds and is now holding $800 million in low-yield money market funds. Stanford CFO Randy Livingston describes these actions by saying, "We've created a rainy day fund."
It's a Matter of Choice
Endowment is not simply a sum of assets that is determined by outside forces. It is, instead, a fund that nonprofits have a great deal of control over. It is a mistake to assume that a school's endowment is determined by donors' decisions to require that most of their contributions be retained and only a small percentage of their yield be spent. Rather, hospitals, symphonies, colleges, and other nonprofits decide how much of their income from all sources to spend and how much to save. They decide how much revenue to generate via tuition, patient fees, or ticket sales, and how much to cut into that revenue by granting student financial aid, providing charity care, or giving free performances. They decide how much to spend on developing alumni giving or corporate gifts, on lobbying legislators for government grants, and on building luxury skyboxes at their football stadiums.
We do not claim that nonprofits can have whatever endowment they want — only that they have significant control over how much they save for the future and how much they spend now. Contrary to common belief, there is no legal minimum or maximum amount that nonprofits must withdraw (payout rate) from their endowment each year. (Foundations are required to take a 5 percent payout rate.) The commonly used payout rate of 4.5 percent is chosen by the trustees, not by external authorities. Endowment can grow more rapidly if a school gives less financial aid, if it drops unprofitable initiatives such as specialized science programs requiring costly laboratory facilities that bring in little grant revenue from government or private sources, or if it replaces expensive tenure-track faculty with cheaper adjunct and lecturer faculty. Or, endowment can grow more slowly, or even decline, if the school chooses to spend more money on current educational and research programs.
It is also wrong to assume that donations commonly specify that the gift must be added to endowment or that it be spent promptly — in either case dictating whether the contribution must or must not be added to endowment, putting the decision beyond the school's or museum's control. In fact, nonprofits are only rarely confronted by a "take it or leave it" donation offer in which the donor insists that a gift be used for something that the nonprofit does not want to do. Much more commonly, the relationship between donor and donee is collaborative, with the donor wanting to help advance the nonprofit's agenda rather than trying to force the nonprofit into another direction.
Another erroneous assumption is that gifts to endowment are typically restricted. According to the NACUBO Endowment Study 2008, of the 77 colleges and universities reporting more than $1 billion in endowments in 2008, 51 percent of their endowments on average are "true" or, as the IRS calls it, "permanently restricted" endowment, 9 percent of the endowments are "temporarily restricted," and 35 percent on average are "quasi-endowment" or "unrestricted" endowments. (An additional 5 percent of endowment funds are "held in trust by others.") But there is a gigantic difference, of enormous importance, between a fund balance being labeled "restricted" and the fund being truly restrictive. Insofar as the terms of an endowment truly constrain how it may be used, the nonprofit does not have control. But money is fungible, and so a gift that is apparently restricted may, and typically does, leave the nonprofit with wide discretion over expenditures and over saving by building up endowment.
Consider the example of a $1 million donation specifying that it be added to the "true" or "permanent" endowment and that 5 percent of it, $50,000, be devoted each year to student financial aid. It would seem that the college has no control over the deployment of the donation once it is accepted. Not so. It would be so if the school would have given no financial aid but for this gift. But if the school were providing $300,000 of such aid, there would be, in general, nothing preventing it from devoting the new $50,000 to financial aid, thereby releasing that same amount for spending on other programs.
This fungibility of money makes it very difficult to take control away from the school by mandating that total financial aid increase by $50,000 beyond what it otherwise would be, for that requires knowing what would have been spent in future years on a specific program had it not been for the new donation. Only when the gift — in this example the $50,000 per year — exceeds the total of what would otherwise have been spent on the particular activity, would the school lose control over its use of the gift funds. As this is extraordinarily uncommon, the conclusion is clear: Even though the vast majority of endowments are shown on nonprofit organizations' IRS Form 990 returns as "restricted," the fungibility of money permits wide latitude in how endowment funds are spent.
Creating a Rainy Day Fund
There is no simple definition of what constitutes a rainy day, but an overall drop of 10 percent of annual revenues (what Tulane suffered in 2005 as a result of Hurricane Katrina) certainly qualifies as a highly unusual event — in this instance, literally a rainy day. Using a 10 percent drop in annual revenues as the standard metric for a rainy day, we see that if a nonprofit had an annual budget of $100 million and an endowment of $80 million, it would have a rainy day fund that would last for roughly eight years of $10 million transfers from the endowment.
We found, in random samples of 100 IRS Form 990 returns for 2003 (the latest available complete data), that the average rainy day fund for museums was about thirty-seven years, for undergraduate colleges about twenty-one years, and for general hospitals five years. Although the average museum is wealthy in terms of its rainy day fund, nearly 40 percent of museums have no endowment and so no rainy day fund. By comparison, only 10 percent of the general hospitals and 2 percent of the colleges and universities we measured had no endowment.
These three industries are more alike, however, when we compare their median rainy day funds. Half of all museums have rainy day funds that could last more than twenty years, and half of all colleges have endowments that could withstand thirteen years of rainy days, but half of all hospitals could survive only five years of rainy days. The causes of, and justifications for, these differences need more study, but it is likely that they reflect differing volatility and dependability of revenues from various private and public sources.
Wealthier nonprofits generally have larger rainy day funds, but not always. In 2006, Princeton had a 141-year rainy day fund, but tiny Grinnell College, with a far smaller endowment, had the largest rainy day fund of all colleges and universities — an astonishing 191 years. Harvard, Yale, and Stanford have far larger endowments, but they also spend much more, and so their rainy day funds of 96, 92, and 53 years, respectively, do not make even the top ten. Even endowment reductions of 25 percent together with no reductions in program spending would leave their rainy day funds at dozens if not scores of years.
Universities are not the only nonprofit institutions with substantial rainy day funds. In 2008, the Metropolitan Museum of Art had a 90-year rainy day fund. In 2007, the Memorial Sloan-Kettering Cancer Center had a 21-year rainy day fund, and the Boston Symphony Orchestra had a 53-year rainy day fund.
There are also prominent nonprofits that have essentially no endowments. The American Red Cross has only a two-year rainy day fund, relying instead on donations to fund its expenditures during emergencies. In effect, increased donations during disasters are its rainy day fund. But apparently not every nonprofit sees the dependability of increased donations to respond to a fiscal rainy day, whether in the form of diminished revenue or need for increased program expenditures. The greater the positive expected response of donors, private or governmental, to a rainy day, the smaller is the need for a rainy day endowment fund.
Protection against a rainy day need not be the only reason for creating an endowment, but it should surely be a major reason. Real incomes will rise over the years, so nonprofits can, in general, look forward to increased revenue from donations and user fees. Squeezing today's students, patients, and museumgoers to save money for future generations of users is misguided. But building endowment is not misguided if it is used as rainy day insurance, to preserve the stability and long term development of programs central to a nonprofit's mission.
The hard questions are these: "How large an endowment is enough?" and "What is the appropriate balance between spending now and saving for the future?" We do not have the answers to these questions. Indeed, we think there is no single answer, but we do see a number of issues that deserve careful attention by researchers, nonprofit managers, and public policy leaders.
Because nonprofit organizations benefit handsomely from tax breaks on endowments — paying no tax on either the dividend and interest yields or the capital gains — society is justified in asking how much is enough. Should "enough" be measured by size of endowment or, as we prefer, by size of rainy day fund? The substantive question is how much saving by nonprofits should be encouraged.
In any case, regulators should recognize that depending on whether "endowment" is taxed or not, a nonprofit will have a different incentive not only to accumulate or spend down its endowment but also to rename its assets. A college president once said that if the government decided to tax the school's endowment, he could dramatically cut "endowment" in just ten minutes! Whatever the purpose of endowment, it would be poor public policy to encourage nonprofit to spend money on accountants and tax attorneys simply to rename assets.
Now is not the time for hasty decisions and changes in tax laws spurred by today's "rain." It is the time for rethinking conventional policies on financing nonprofit and bringing greater stability to organizations that are providing higher education to our youth, medical care to our sick, and advancement to the cultural fabric of our society.
In today's depressed economy, nonprofits are hurting. Falling college endowments have garnered the lion's share of publicity, but hospitals, museums, foundations, and other nonprofits are suffering in similar ways. What can one learn from these experiences? Nonprofits with sizable endowments must understand that if they succumb to the attraction of riskier and less liquid investments in pursuit of higher returns, they should be prepared to deal with the inevitable fiscal rainy days. Nonprofits with endowments should be willing to spend down their endowment to sustain program expenditures. And those nonprofits that have little or no endowment need to find ways of diverting some of their even limited revenues to creating a rainy day endowment fund, because this is not the last rainy day.