The old bromide “whatever goes up must come down” applies as much to markets as to objects tossed in the air, which is to say that while economic downturns may not be wholly predictable, they are absolutely inevitable. That inevitability poses a challenge for philanthropic organizations whose financial resources are in an endowment: when a downturn happens, should they reduce spending to preserve capital for future grantees and beneficiaries, or should they maintain spending to preserve the work of present grantees and beneficiaries?
The onset of COVID-19 has led some observers to up the ante still more. When a big crisis hits, they say, just maintaining spending isn’t enough: philanthropic funders should spend more. In the United States, some backers of this position have even gone to Congress, pushing legislation that would require foundations to double their payout rate for at least the next three years.
The public debate has been largely one-sided, with skeptics of increased payout remaining quiet. This is partly because, while some proponents of giving more have been thoughtful and balanced, others disparage the values and belittle the motives of funders who aren’t taking their advice—jeering sarcastically that they lack “boldness” and cling to “archaic” concepts by mindlessly letting inertia govern their decisions. Such funders, they say, put payout over people and “math over mission” by prioritizing the size of their endowments rather than the needs of dedicated nonprofit organizations and the communities they serve. Thinking one should preserve capital for future needs, we are told, is “a paternalistic, narcissistic notion” held by people who “live in a bubble of privilege wrapped in another bubble of delusion.”
Strong stuff, though (sadly) of a piece with public discourse generally today. As I have written elsewhere, the possibility that people might reasonably disagree has fallen out of fashion, as has the idea that someone who thinks differently could have honorable motives for doing so. Rather than recognize counterarguments or acknowledge competing concerns, we just assume bad motives and cut straight to the insults.
With some trepidation, then, I will try to explain why a funder might credibly think it wiser not to increase payout during an economic downturn, even a severe one. I hope to show that this is so not only because people might reasonably disagree about the right thing to do, but also because there might actually be more than one right thing to do. What is appropriate for one funder may not be appropriate for another, and what is apt for a funder at one time may not be fitting even for the same funder at a different time.
It doesn’t follow that any or all regulation is misguided: minimum payout requirements that leave the option for longevity are important, and government may well want to encourage spending through the use of tax and other incentives. Devices like these make sense because they leave funders options to maximize the good they can do given their varied goals and objectives. The thoughtful proposals developed by the Initiative to Accelerate Charitable Giving, which the Hewlett Foundation has endorsed, fit this description.
The point is not that the issue is complex, though certainly it has complexities, but that funders differ in what they are trying to accomplish and how they are trying to accomplish it, and these differences matter when it comes to calibrating payout under different circumstances. I do not believe anyone is holding back because they care more about a swollen bank account than helping people. They are balancing differently the needs of present and future and the relative amount of good they can do by spending differently.
Present Versus Future
There has been a lively debate in recent years about whether it is ever appropriate for foundations to allocate capital with an eye on preserving their resources in perpetuity. “Perpetuity” is something of an affectation, of course, as no one believes these institutions will literally last forever. But the language signals an organization’s aim to preserve its real spending power over time, which requires achieving annualized endowment growth equal on average to payout plus inflation.
Not all philanthropists approach their giving this way. Many plan from the outset to expend their resources in some limited or defined period of time. Some, perhaps many, of the commentators and critics pushing for increased foundation spending today actually favor this approach, seeing the current economic crisis as an expedient device to push their more basic conviction that foundations should be required to spend down. Not everyone arguing for increased payout takes that position, however, and some advocates for spending more now propose this as an occasional action, appropriate during economic downturns even for funders generally committed to existing in perpetuity.
The debate over managing an endowment to exist in perpetuity is nevertheless relevant, because it was in this context that philanthropists developed the argument for limiting payout even when spending more money now could accomplish more good now. Of course, a funder can almost always accomplish more good now by spending more money now. But that’s not the right question: The question is whether a funder can accomplish more good overall by spending more money now, which means taking into account the opportunity costs of not being able to spend as much money in the future. Seen in that light, whether to spend more now depends on what one is trying to achieve.
Certainly, there are social problems that can be solved by shorter-term infusions of cash and that do not require cultivating a grantee ecosystem or developing enduring relationships and deep expertise. For foundations that focus on such problems, spending down may make sense. Physical diseases can be cured by inventing a vaccine, and some social problems, like bail reform or possibly even mass incarceration, could be fully or largely addressed by the enactment of a discrete policy solution. In such cases, it makes sense to keep spending now until the next marginal dollar yields no return.
But many social problems—and especially the big ones—are intractable in ways that call for foundations to employ long-term strategies and sustained efforts. There are no silver bullets when dealing with poverty, racism, education, climate, or human rights (to name only a few of the significant issues philanthropy addresses), and working seriously on these kinds of problems requires a different mindset altogether. Social problems like these are bred and live inside obdurate systems of clashing interests; causes and solutions alike depend on people, with all the unpredictability and confounding behavior they bring. Progress is slow not because foundations are unwilling to take chances or invest enough money, but because the problems are hard—more, they are mutable and continuously changing. Problems like these are not “solved” so much as mitigated or reduced or contained.
Addressing such challenges takes time and patience—a great deal of both, in fact—and a commitment by philanthropists of human as much as financial resources. It takes willingness to learn from others and to develop real partnerships with grantees, while hearing from (and listening to) intended beneficiaries. Headway is made not by pushing cash out the door while looking for speedy results, but by becoming part of and helping to nurture an ecosystem of grantees, beneficiaries, and other funders and organizations whose efforts, cumulatively and over time, help make progress. Success comes from doing the hard work to understand a problem deeply, developing a thoughtful approach to address it, observing what happens, changing one’s understanding based on evidence and experience, and—most important—being willing to do that over and over and over. Sometimes, subsequent developments undo progress and new efforts may be needed just to sustain what has been accomplished. For all these reasons, a foundation’s commitment, consistency, and presence over time are essential.
Which brings us to the nub of the spending problem. Because when an economic downturn comes and an endowment decreases in value, foundations face an ineluctable choice between present and future grantees and beneficiaries. Either cut spending now to preserve future spending or continue spending at the same level on a smaller endowment and have less later on. Go farther by increasing spending now, as some argue is necessary, and have even less money later on.
It does not automatically follow that a foundation should not spend more now—a question I return to below—but one should not make this choice without acknowledging the inescapable dilemma it presents: do more to help grantees and communities now, have less to help grantees and communities later, and vice versa. That much is unavoidable.
Having Our Cake and Eating It Too
Most proponents of spending more today have ignored this predicament and just argue that today’s greater needs require greater spending. But a few, more attentive commentators have challenged its logic. We can have our cake and eat it too, they contend: we can spend more today without hurting the prospects of future grantees and communities. Would that this were true, because it would make everything so much easier. Unfortunately, the arguments do not withstand scrutiny.
Earning it back | Phil Buchanan, president of the Center for Effective Philanthropy and among the sector’s most thoughtful commentators, argues that “foundations’ relentless focus on endowment size is misguided” because they’ll “be able to grow their endowments later” and needn’t “worry that they won’t be able to claw their way back to the same asset levels afterward.”
Putting aside the intimation that foundations care about endowment size as an end in itself (as opposed to their capacity to support future grantees), there are a number of flaws in Buchanan’s argument. The first is a simple matter of arithmetic. If I have a $10 billion endowment now, I can do more good than if I have an $8 billion endowment now. If I reduce my $10 billion to $8 billion so I can do more now, it is certainly true that I can regrow it to $10 billion in a few years. But, of course, had I not overspent (by increasing my grantmaking) in the first place, I would have more than $10 billion at that time—I’d have, say, $12 billion—and so be able to do more good then when it will cost more to do so. And while I might get to $12 billion in a few years more, but for the initial reduction I would have more than $12 billion at that time when it will cost still more. And so on.
That much is inescapable: if I spend more now, I will have less later. How much less depends on how much one increases present spending and for how long, but the math is irrefutable. The David and Lucille Packard Foundation was significantly larger than the William and Flora Hewlett Foundation in 2000 and would have been larger even with the addition of Bill Hewlett’s estate, which came a few years later. But the Packard Foundation handled the dot.com downturn differently than the Hewlett Foundation and ended up approximately 20 percent smaller, which it has been ever since. More telling, the head of another large foundation (who asked to remain anonymous) relayed to me that their foundation chose to overspend during the Great Recession, with the result that its grant budget today is more than one third smaller than it would otherwise have been. The foundation’s board regrets the earlier decision and does not plan to repeat the action.
In response to this argument, proponents of counter-cyclical spending reply that foundations have over time been able to earn returns greater than what is needed merely to sustain their spending power. Even if spending more during a downturn leaves them smaller than would otherwise have been the case, they maintain, it does not deprive future grantees of any support to which they are entitled—not unless one thinks future grantees are entitled to greater relative support than present ones (which no one does).
The problem with this rejoinder, already suggested by the examples above, is that it’s not supported by the facts. Remember, foundations must still pay out 5 percent each year, so just staying even requires earning returns of inflation plus 5 percent. Growth in real terms thus begins only after one climbs out of a 6-8 percent hole, which is why it took the Hewlett Foundation a full decade after the 2008 downturn just to reach the endowment size it had at the end of 2007—and this despite the fact that the Hewlett Foundation cut its grantmaking budget at the time. Remember, too, that recovery from any particular downturn will be interrupted by subsequent downturns, which will continue to recur on their irregular, unpredictable timeline.
Contrary to common belief, the 5 percent payout rule is not lower than needed to make perpetuity possible. Most foundations aim for a risk/return that is equivalent to a conventional investment mix of 70 percent equity and 30 percent fixed income, and annualized returns for this passive index mix over the 15-year period ending December 31, 2019 were only 6.3 percent—which is less than the 5 percent required payout plus inflation (which has averaged 1.5 percent over that period of time). Lengthen the time period, and the results are even worse. According to publicly available data compiled by the Hewlett Foundation’s investment team, annualized returns over the past 90 years have been just 5.3 percent, with an average inflation rate of 3 percent. So if foundations had done only as well as a 70/30 index fund, they would have lost ground—even during the longest period of economic expansion in US history.
But foundations have not done that well—not even the largest foundations, nearly all of which have produced subpar returns in recent years. There are many reasons for this, including the need to manage for volatility as well as returns. Whatever the explanation, since the Great Recession 12 years ago, the mean annualized returns of 27 of the largest foundations in the United States have been an anemic 5.8 percent, with only three garnering annualized returns of 7 percent or higher, and only one above 7.1 percent. (These data are based in information shared confidentially.) Nor should we expect foundations to suddenly start doing better going forward, because most economists had identified a secular trend of slower growth in the global economy even before COVID-19. Just staying even will be increasingly difficult.
Lastly, these figures likely understate the real decline in foundation spending power, as analysts use the consumer price index (CPI) to measure inflation. But the basket of goods that goes into the CPI inflates more slowly than the basket of goods on which foundations spend and make grants (which consist predominantly of labor costs). A foundation-specific inflation rate would almost certainly be higher than the CPI, and with it, the returns needed to regrow diminished spending power would also have to increase.
Issuing debt | Foundations have not traditionally used debt financing. They have not needed it to fund grants, and they pay substantial UBIT (unrelated business income tax) if they issue debt and invest the proceeds. In 2020, a number of foundations sold bonds to generate extra grantmaking capacity for struggling grantees—asserting that today’s low interest rates enable them to avoid the conundrum of choosing between present and future. How so? By issuing debt instead of liquidating assets, these foundations are gambling that they will earn returns on the money they will still have in their endowments that, compounded over the term of the bond, are greater than the interest they pay plus the discounted value of what they must pay when the bonds mature.
To illustrate, suppose that a foundation with a $10 billion endowment issues 30-year bonds worth $1 billion. The likely interest rate on a bond of that duration today is approximately 2.5 percent—which means the foundation has $1 billion to spend now but must pay $25 million in interest each year for the next 30 years, at the end of which it must also pay bondholders back the original $1 billion they invested.
Earning returns greater than those costs would be a safe bet if the foundation’s endowment were a passive vehicle for growth from which the foundation did not have to pay anything other than the bond interest. This is, indeed, what people usually have in mind when they say money is “cheap” because interest rates are low: they can borrow capital at the low interest rate and invest it for higher returns. But foundations don’t have that option. They cannot invest the proceeds of the bonds, but instead must spend them in ways that cannot produce a financial return while continuing to pay out at least 5 percent each year with the money that remains in their endowment.
By issuing bonds, then, the foundation has not escaped the present versus future tradeoff, and adding debt, no less than liquidating assets, will reduce the foundation’s future grantmaking capacity. Three interrelated features of the bond issue produce this outcome: that the foundation must pay interest (amounting to $750 million over the life of the $1 billion in bonds); that it loses the compounding benefit of the interest paid, which accumulates and grows larger over time; and that it must still pay back the $1 billion when the bonds mature.
Take the example above, in which a foundation with a $10 billion endowment issues $1 billion in 30-year bonds at an interest rate of 2.5 percent. For the sake of simplicity, assume that the bond proceeds are used in year one, that payout throughout the 30-year term is always 5 percent of the prior year’s gross asset value, and that the portfolio’s annualized return is 6.5 percent of gross asset value. In terms of expected outcomes at the end of the 30-year term, the foundation will have paid out $370 million more in grants than it would have paid had it not issued the bonds, but it’s endowment will be approximately $2 billion smaller.
The additional grant payments will have been made in the first few years, with progressively lower payout amounts each year (relative to business as usual); and going forward, the $2 billion shrinkage means a grant budget that is permanently smaller by $100 million, with that gap continuing to grow larger over time due to lost compounding. The foundation may decide that this future reduction in spending power is worth taking to pay out more now—a matter discussed earlier and to which I return below—but it has in no way avoided the tradeoff between present and future.
There are, moreover, several other consequences that flow from issuing bonds that seem to have escaped notice in the public discussion of this option. (I am sure the issues were vetted inside the foundations that chose to issue bonds.) First, bondholders now have a claim on foundation assets that has legal priority over the foundation’s non-contractual obligations to grantees. In good times, this likely will not matter, but when the next downturn occurs, especially if it is severe, the foundation may find itself constrained in the actions it can take.
Second, and more important, as an additional claim on the endowment, the bond issue increases the risk associated with the foundation’s investment portfolio. The foundation has basically taken a leveraged position on its endowment. That’s a gamble the foundation’s board and investment team may want to take if their tolerance for risk is increased by a desire to make more grants today, but it has potentially significant consequences.
In the hypothetical example above, I assumed that market returns were steady every year. In the actual world, there will inevitably be years when capital markets decline, and in these years, the foundation will feel the impact of paying interest on top of grants from a diminished pool of assets. There will also be good years, of course, in which the foundation benefits from the assets it retained by using debt. Through a risk-adjusted lens, the value is the same—but only at the cost of reducing stability in cash flow (and so payout rate) by accepting higher volatility and greater variability in annual returns.
The increased risk is hardly trivial. In my example, the likelihood of a 20 percent drawdown is 10 percent higher, while the likelihood of a 30 percent drawdown increases by 20 percent and that of a 40 percent drawdown by 30 percent. Which is precisely why foundations whose only source of revenues is an endowment ordinarily avoid leveraging it—a general practice that may prove particularly desirous in the coming years, which are likely to be economically challenging.
Having decided to subordinate future needs to present ones, a separate question is whether doing so by debt financing is better than simply liquidating assets. The main point for present purposes is the same: by either means, increasing payout in the present reduces payout capacity in the future. But the effect on spending differs in the two cases:
- In both cases, liquidating assets or issuing bonds, the foundation increases its payout in the first year by $1 billion over business as usual (BAU).
- But liquidating assets to the tune of $1 billion begins meaningfully to affect payout in year two, whereas the effect on payout comes more slowly if the money was raised through bonds. Over the same 30-year period, a foundation that liquidates assets will pay out $1.2 billion less in grants than one that issues bonds.
- Because the foundation need not pay interest if it finances the $1 billion by liquidating assets, however, its endowment regains lost ground over time. At the end of the 30-year term, the endowment would be $1.5 billion smaller than BAU (as compared with $2 billion smaller in the bond case), meaning its grant budget going forward will be permanently less by $75 million per year than BAU, as compared to $100 million per year less in the bond case. Remember, too, that in both cases compounding causes the reduction in grant budget to become continually larger over time.
In terms of grant dollars awarded, then, issuing debt is identical to liquidating assets in the short run, better in the medium term, and worse over the long term. To which one must add the increased risk and volatility and potential conflict between bondholders and grantees that come with issuing debt.
So which is better, issuing debt or liquidating assets? I don’t think there is a clear “right” answer. The consequential decision is to spend more in the present, which we have seen has substantial consequences for future spending regardless of whether one liquidates assets or issues bonds. Once that decision is made, balancing the different costs and benefits of these options depends on things like how optimistic or pessimistic one is about the economy, how much flexibility the foundation has to adjust grant spending as markets shift, and how disciplined it thinks it can or will be in using that flexibility.
New funders | A final reason sometimes offered for funders to worry less about the future is that there will be new wealth and new money to take up the slack. David Callahan had good reason to subtitle his book, The Givers, “Wealth, Power, and Philanthropy in a New Gilded Age.” For a huge (some say obscene) amount of new wealth has been generated in the past several decades, much of which is likely to make its way into philanthropy, where it will outstrip the sector’s present resources in much the same way new wealth from John D. Rockefeller, Andrew Carnegie, and others did a century ago.
That may be true, though it also bears observing that nearly all the new big funders say they plan to spend down. And since “gilded ages” are not continuous—there was some growth in philanthropic resources between the original one and today, but nothing close to either of these—we may be looking at a limited bump, albeit a large one, in philanthropic spending. Note, too, that while there is a lot of new money, the scope and scale of the needs have similarly grown and continue to grow, as have philanthropy’s ambitions, which today increasingly aspire to work on a global scale.
But the real problem with depending on new money lies elsewhere. As I wrote earlier, philanthropy’s biggest contributions require developing enduring relationships and deep expertise alongside grantmaking. Progress is made through consistent funding of long-term strategies and efforts, and the groups and communities foundations hope to serve depend on their commitment, consistency, and presence over time.
Consider the outsized role the legacy foundations continue to play today, even with new funders bringing larger resources. Imagine how diminished the current philanthropic scene would be if the Rockefellers, Carnegies, and Fords had all spent down or significantly diminished their scope during the Great Depression. Where would philanthropy have been in the 1960s and 70s when these foundations took the lead in advancing the Civil Rights Movement, the Green Revolution, Head Start, and much more?
Or what if these legacy funders had spent down then? A great deal would have been lost that new money does not and cannot replicate. There really is such a thing as institutional culture and wisdom, and what is enabled in philanthropy through the combination of longevity and substantial ongoing investment from a committed institution isn’t easily transferred or replicated.
Let me illustrate the point with a couple of examples. The Hewlett Foundation has worked in the environmental field for more than 50 years, with a goal of conserving biodiversity and protecting the ecological integrity of half the North American West. We have made significant progress, but it has come in fits and starts, and we have had as many setbacks as victories. Our strategy and approach evolved as we learned and as circumstances changed, and progress has depended on our capacity to utilize our experience, reputation, relationships, and networks, which are a product of steady funding and staunch commitment. Other funders have come and gone, but very few have shown the same steadfastness or willingness to stay the course that is necessary for success. If the Hewlett Foundation halted or significantly reduced its support and commitment, it is all but certain that progress would diminish in equal measure.
Or consider the famous Rosenwald Schools, funded and built with support from Julius Rosenwald and Booker T. Washington in the early 20th century. Seeking to promote educational opportunities for Black children, the Rosenwald Rural Schools Initiative developed a matching grant program that, between 1913 and 1931, facilitated the construction and operation of nearly 5,000 schoolhouses for Black children living in the rural American South.
The effort was wildly successful: by 1928, more than one-third of all Black children were being educated in Rosenwald Schools, with marked improvements in school attendance, literacy, years of schooling, and cognitive test scores. But the Rosenwald Fund stopped building schools in 1932 and was spent down by 1948. Other funders showed little interest in Rosenwald’s and Washington’s remarkable success, having shifted their focus to integration. Within less than a generation, every single Rosenwald School had closed. Might it not have been better had the fund spread its program out over a longer period of time—building fewer schools at any one time but persisting long enough to preserve them in the face of a changing environment and persistent efforts by state governments to undermine them?
Hard Choices
The point of these examples—of the last section as a whole, in fact—is not that it is always better for a foundation to allocate capital with an eye on existing in perpetuity. It is that not doing so has costs for the future, just as not spending more in the present has costs, and there is no escaping a choice between them. In “normal” times, funders’ regular spending practices (presumably) reflect how they have decided to balance these costs. When an economic downturn happens, they must choose how to rebalance their spending: whether to preserve their plans for the present at the expense of their efforts in the future, or vice versa. The answer to that question should be whether the amount of good one will create or preserve in the present by spending more now is greater than the diminished good one will be able to accomplish in the future as a result.
Some advocates of increasing payout during economic downturns say that this is a false choice. There is no comparison, they maintain, between the clear reality of present needs and the hypothetical possibility that future needs will be as great. This is just another way of stating the argument for spending down generally, because it is neither more nor less true when choosing whether to change normal spending practices in response to a downturn than it is when establishing these practices in the first place. Societal needs and philanthropic opportunities to address those needs may grow during a downturn, but that will be true in the future too, because future economic downturns are inevitable.
It might be different were there a future in which philanthropy was unnecessary or in which philanthropic resources were going to be greater than required to address society’s needs, but that’s a possibility we can comfortably set aside. When it comes to big, long-term problems—poverty, racism, educational opportunity, human rights, and the like—we can be confident of two things: first, present needs are greater than philanthropy could address today even if funders spent every penny they have; and, second, even if spending more today addressed more of those needs, there still will be more unmet needs than philanthropy can address in the future.
How, then, should foundations and other philanthropists decide what to do? Let’s not pretend there is an analytically rigorous or statistically quantifiable way to answer this question. The most one can do is make an educated guess—which should, perhaps, cause everyone to pause before too strongly denouncing others who reach a different conclusion. A still better reason to hesitate before opting for reproach is that there isn’t one answer that applies across the whole field of philanthropy. Decisions will differ—and appropriately so—depending on such things as the nature, size, and expected duration of a downturn; whether and in what form public agencies and for-profit actors respond; and a particular funder’s goals, methods, and opportunities.
Just as funders legitimately differ over the most important problems to tackle, just as they legitimately differ over the best ways to tackle those problems, just as they legitimately differ over whether to spend down or conserve resources to spend over time generally, they can (and will) legitimately differ over whether their ability to achieve their goals is better served by spending more during an economic downturn.
So how should a funder who is generally committed to spending over time decide whether to overspend? Some problems need to be addressed today to head off future harms that could be exponentially worse, though even here the choice about spending can be difficult. Take climate change, for example. If ever there were a problem that justified overspending, this would seem to be it. The Hewlett Foundation spends between $120 million and $130 million each year to mitigate global warming, more than a quarter of its total grant budget. There was nothing scientific or even particularly systematic in the choice of that amount: it is as much as we felt we could put into the effort given other demands and priorities and our standard practice of capping overall spending at around 5 percent.
But why hold to the cap in the face of such a problem? After I published an editorial criticizing other funders for ignoring climate change, David Callahan acerbically tweeted, “meanwhile, [Hewlett] won’t tap its endowment principal to fight what it sees as a time-urgent existential threat.”
But “won’t” is too strong a word. The Hewlett Foundation’s payout policy reflects a pragmatic judgment about the best way to accomplish the most good over time. It’s not holy writ, and the foundation would spend more were we confident this would make a sufficiently meaningful difference relative to the opportunity costs. But for us to just spend more doing more of what we are already doing on climate—even a lot more, even our whole endowment—wouldn’t do that. It might capture a few more emissions, but only while curtailing our ability to work on other problems that also matter, and without significantly changing the overall climate calculus.
That’s because what’s needed to keep global warming below 2 degrees Celsius is both vastly larger resources and the energy and ingenuity of a wider community of funders. If, on the other hand, overspending by us could help produce that—could galvanize new funders to join the effort at scale, for example, or catalyze critical new work capable of moving the needle—we would certainly consider it.
How about the COVID-19 crisis, which has roused so many calls for increased foundation spending? For most nonprofit organizations, there is nothing particularly different or unique about this downturn. It looked at first as if the economic consequences might be extreme—that we were headed toward not just a severe recession like in 2009, but maybe even a depression. So far, however, financial markets have held firm, and the threat faced by most grantees is the one that accompanies any downturn, namely, that funders may trim their grant budgets and individual donations may decline. Nonprofits whose funders do cut their grantmaking will need to slow down or cut back their work temporarily, but they will regrow with the economy and—per the conjecture underlying the 5 percent payout policy—funders will be able to support them and other nonprofits better and more over time.
What is unique in the current situation is the economic shutdown forced by directives to shelter-in-place. No one anticipated that, and it saddles additional costs on a subset of nonprofit organizations: those whose revenue depends to a meaningful extent on providing services that can only be delivered in person, such as performing arts organizations. For this particular subset, sheltering-in-place means losing what may be a major source of income wholly apart from grants and donations. Paying more now to rescue key organizations that may fail because sheltering has made it impossible for them to earn essential revenue makes sense. Funders might even enlarge this subset to include any grantee that is important to a program or strategy and unable to survive the current downturn.
(In the case of the Hewlett Foundation, this subset of organizations turned out to be relatively small, concentrated mainly in our performing arts portfolio. We were able to provide support to these without changing our normal spending practice by using the unallocated funds described in the final section below. We did so knowing that we could face a bigger challenge in 2021. If the economic shutdown continues or repeats in waves, organizations that are managing to scrape by now may increasingly find themselves unable to continue—especially as markets come to terms with the cascading economic consequences, and what has so far been a manageable downturn becomes extreme. At that point, we will need to reassess, though if the situation is too extreme, funders’ resources may be too little to matter, in which case it may end up making sense to just hunker down, weather the storm, and plan on rebuilding.)
Moral Hazards
In the climate and COVID-19 examples, the decision to overspend responds to a distinct, unique, and manifest need or opportunity in which the relative costs and benefits of encumbering future spending are specific, identifiable, and reasonably clear-cut. This matters because of a feature of the decision-making process that seems to have escaped notice in the current debate—namely, that the relevant decision makers’ incentives are inherently skewed in favor of the present.
In economics, a “moral hazard” exists when an actor stands to benefit from taking action while some or all of the risks associated with that action are borne by someone else. Payout decisions present a clear instance. The decision whether to increase spending in a downturn must be made by a foundation’s current board, president, and program officers—all of whom have strong incentives to say yes. By spending more, they can do more. They will earn the gratitude of their grantees and beneficiaries, who are understandably eager for support, while being praised by journalists and commentators. With that on one side of the scale, why hold back? In weighing present versus future opportunities, favoring the present is only natural given that all the benefits accrue today, while all the costs will be borne by unknown successors and nameless, faceless future grantees and beneficiaries.
The simple truth is that deciding to overspend is the easy thing to do—not because anyone is acting in bad faith, but because that’s how human psychology works. Hence all the well-meaning observations about how we don’t know for sure what future needs will be, and the wishful thinking that future resources will be more than enough to handle them. Diminished resources will not be an abstraction to the people who run the foundation in later years, much less to their grantees and beneficiaries, but they understandably feel that way to present decision makers (not to mention present grant recipients and beneficiaries). Rules to constrain payout are thus a device for countering the built-in bias that inevitably and inherently predisposes current boards and staff to favor authorizing themselves to spend more.
Such rules need not and should not be wholly inflexible. As we have seen, there are instances in which it makes sense to overspend in response to an economic downturn. But thoughtful boards and CEOs will want to check themselves to ensure they are fairly balancing the needs of future grantees and beneficiaries given their wholly natural but nevertheless self-serving incentive to discount those needs. Requiring clear evidence that overspending is justified by a specific, identifiable, and (most important) unique need or opportunity is a useful device for this purpose.
Anticipating the Problem
Entirely overlooked while we have been debating whether foundations should increase their payout has been the question of what we might have done to avoid or mitigate the problem in the first place. We cannot prevent economic turmoil, but surely we can prepare for it better. Below, I touch briefly on two measures funders and nonprofits could take that might address all or most of their needs during a normal downturn and appreciably moderate them in a severe one.
General operating support | My father was a small businessman. He owned a neighborhood grocery store, then a deli, and then a small hot dog restaurant. In all his businesses, he took care to keep what he quaintly called a rainy day fund, by which he meant a three-month reserve (though he hadn’t gone to college and never learned that language). I mention this simply to underscore that it doesn’t take an MBA or a lot of sophistication to see the necessity of keeping a healthy reserve. Yet few nonprofit organizations do so, which is why they are invariably and immediately in trouble when their sources of funding decline.
So why don’t most nonprofit organizations have adequate reserves? Funders bear much of the blame. Rather than provide general operating support (GOS), which nonprofits could use to build reserves (and resiliency more generally), the vast majority of funders offer only tightly controlled project grants. Worse, these project grants seldom cover even the full costs of the projects, forcing nonprofits to use what little GOS they do receive to make up the shortfall. (The Hewlett Foundation does not fund this way: approximately 70 percent of our grant dollars are flexible, and we work hard to fund the true costs of our project grants.) But the absence of reserves is not only the fault of funders, because even when nonprofits have available GOS, they seldom use it to build reserves; instead, they use it to grow operations to increase their immediate impact.
This would all be fine if the economy always grew and downturns never happened. But they do happen, and the failure of both funders and nonprofit organizations to anticipate and plan for them epitomizes classic short-term thinking: “let’s maximize impact today and not worry about that inevitable future event that will strain our grantees’ or our organization’s ability to function.” Then, when the inevitable event happens—as inevitable events inevitably do—yet more short-term thinking drives the call for a solution: funders are asked to increase their payout to help nonprofit organizations now, even though this means having less with which to support them and the communities they serve in the future.
It’s not as if the situation were unforeseeable. I don’t mean a profound economic stoppage of the kind produced by the pandemic; I mean a bad downturn that would severely affect nonprofit finances and call upon a need for reserves. For that, we had what amounts to a dress rehearsal in 2009, and coming out of the Great Recession, one might have thought funders and nonprofits would recognize the need to plan for that wholly predictable next rainy day. Yet few funders or nonprofit organizations developed specific plans for downturns drawn from the lessons of 2009, and almost none changed their funding or savings practices. So when funders who do provide GOS are asked to increase their payout—in effect, being asked to further subsidize the shortsighted funding approach of others—it’s legitimate for them to wonder whether anything will change. Or will everyone just go right back to the same bad habits and put nonprofit organizations in the same vulnerable position next time around?
There is a ray of hope. In the immediate wake of the COVID-19 shutdown, hundreds of funders pledged to loosen restrictions on their grants, including “converting project-based grants to unrestricted support” and “making new grants as unrestricted as possible.” Many more took these steps without signing a formal pledge. If nothing else comes from the present crisis, let’s hope funders continue these practices after it is over, learning what funders that emphasize unrestricted support have long understood—that providing GOS enhances the efficiency, effectiveness, and resiliency of the organizations they are supporting, and so advances their goals better. (There are, of course, many instances in which project support is appropriate, such as when first exploring a new partnership to see if there is alignment. But project funding ought to be the exception, rather than the rule, and funders should take care to cover their projects’ true costs.)
The availability of robust reserves would enable most nonprofits sensibly to manage most downturns. But COVID-19 could turn out to be bigger than most downturns—though, as noted above, markets have so far held up shockingly well—and it could last longer. In a well-functioning system, this is where government should come into play. The scale of resources needed to manage a society-wide crisis, which affects for-profits as much as nonprofits and touches every sector in society, are immense. These are the kinds of circumstances that only public institutions have the capacity to address, and we should be able to turn to our governments for help. Ideally, nonprofits (like for-profits) would rely on reserves to get them through the first few months until public resources became available to help carry them going forward.
Certainly, the more than $2 trillion that the US Congress initially allocated offered much needed relief. As the pandemic drags on, however, it’s unclear whether our governments will continue to provide resources adequate for the countless businesses and organizations in need. And even if they do, it’s unclear what proportion of those public resources will make their way to our sector. Additional government aid will surely provide some relief, but we cannot know how adequate it will be. So while reserves and government aid can be significant first-line defenses, it is important to acknowledge that they may not be enough, which will depend on what happens in the coming months.
A grantmaking reserve | In 2018, the board of the Hewlett Foundation approved a plan that laid out how it would deal with future economic downturns. I described the plan publicly in a blog post at the time, and reiterated its main points in an open letter to grantees published soon after we began sheltering-in-place. In short, our plan aims to balance the needs of current and future grantees by maintaining spending during the year of a downturn, resetting the grant budget the following year to around 5.1 percent of the endowment’s reduced net asset value, and absorbing the necessary cuts in our grant budget by reducing the portion that is unallocated at the beginning of each year.
We developed this plan based on lessons learned in 2008. Having been forced to significantly reduce our grantmaking during that downturn, as the endowment recovered its value we did not commit all the regrowth to programs. Instead, we assigned approximately 40 percent of each year’s new growth to unallocated “flexible” funds. These serve, in effect, as a kind of grantmaking reserve. In good years, we can use these flexible funds to launch time-limited initiatives, explore new work, and respond to unexpected needs or opportunities. But since none of our programs or grantees are relying on these specific resources, when a downturn occurs we can use them to absorb the decrease in our endowment’s value without needing to reduce support for ongoing work. This, in turn, preserves the endowment’s spending power for future grantees and beneficiaries.
Our plan does not evade the choice between present and future grantees, which (as noted throughout) is inescapable. Rather, it addresses the conflict in a way that enables us to maintain full support for existing grantees and beneficiaries, while also preserving our spending power for the future. The plan is limited by the amount of our unallocated funds, but we have built a sufficient pool to absorb as much as a 20-25 percent drop in financial markets before needing even to think about reducing support for ongoing programs and initiatives. What we give up is our flexibility in the present to do more or new things, at least until the endowment begins to grow again. But if something has to give—and it does—this seems on balance the most sensible tradeoff to make, at least absent unique or extraordinary circumstances like those discussed earlier.
* * *
Among the more baffling statements made about a 5 percent payout rule is that it “means that 95 percent [of a foundation’s resources] remains untapped.” Meant to be provocative, this oft-heard dictum is just spurious, for there is an obvious difference between leaving assets inactive and using them to produce the surplus income that will enable a foundation to continue doing good work year after year in the future. There is, as I have acknowledged, a legitimate question whether spending that way will ultimately promote more good than spending the money now. But it’s not a choice, as the naysayers would have it, between helping people and gazing fondly at a pile of gold kept in a backroom. It’s a choice between different views of what will help the most people in the most meaningful way over the long term.
Support SSIR’s coverage of cross-sector solutions to global challenges.
Help us further the reach of innovative ideas. Donate today.
Read more stories by Larry Kramer.