A CENTERPIECE OF THE 1969 TAX LAW WAS the requirement that foundations distribute a minimum of 6 percent of the previous year’s assets — a “payout” rule that was subsequently cut to 5 percent after lobbying by foundations. Of course, it’s not really 5 percent, a fact that comes as a surprise to most people outside the world of philanthropy. In 1995, for instance, grants by the 13 largest foundations amounted to only 3.5 percent of assets (in part because the payout rule does not apply to some real assets, such as luxury office buildings).
More important, “qualifying distributions” include “reasonable administrative expenses,” which provides no incentive to keep expenses down. Indeed, administrative expenses have jumped from about 6 percent of grant outlays before the law was passed to 15-18 percent ever since. Peter Frumkin, who teaches public policy at the Harvard Kennedy School, argues persuasively that this jump in administrative expenses, and the professionalization of foundation giving, were direct consequences of the 1969 rules.
Ironically, it looks as though the minimum payout requirement may actually have reduced foundation generosity, because managers of the largest foundations have been extremely careful to treat the legal minimum payout as an absolute maximum. Foundations accounted for 8.2 percent of all giving from 1966 to 1968, according to the American Association of Fund Raising Counsel, but only 7.5 percent from 1992 to 1996. The fact that foundations account for a smaller share of total giving than they did before 1969 is rather startling, considering the enormous growth in the number and assets of foundations.
Or perhaps it isn’t so startling. After all, foundation assets grow for two reasons. One is new gifts, mainly to new foundations. But the dominant source of asset growth is, of course, dividends, interest income, and capital gains on assets. The fact is that instead of being used to finance charitable activities, much of that investment income is reinvested, leading to the aforementioned growth in assets. For instance, from 1982 to 1995, foundation assets grew by $18.8 billion a year, while grants grew by only $8.3 billion a year (see figure 1). In fact, grants were not much larger than the infusion of new money alone, which amounted to $5.6 billion a year.
By subtracting new gifts from foundation assets, we can uncover the “net” increase of the assets, the portion due to reinvesting earnings. Aside from new gifts, foundation assets grew by 13.5 percent a year from 1982 to 1995. That is a rough measure of the earnings on investments. With investments growing by 13.5 percent, subtracting a 5 percent payout left 8.5 percent to be reinvested each year. When you start with billions and keep reinvesting 8.5 percent, the magic of compound interest does not take long to show itself. Suppose a foundation starts with $5 billion (five existing foundations already have more than that). If that $5 billion provides an investment return of 8.5 percent, then even after making those pesky grants, the foundation would still grow to $25.6 billion in only 20 years. In 40 years the $5 billion foundation will have become a $130.7 billion foundation. Translation: As shown in figure 2, if reinvested earnings keep paying 13.5 percent, the assets of U.S. foundations can be expected to hit $5.9 trillion by the year 2035, even if nobody gives them another dime.
Of course, stocks and bonds may not keep paying off as well as they have in recent years. A more conservative estimate of future returns may be constructed using the average of the whole postwar era — a period that includes some nasty bear markets. John Cochrane of the University of Chicago has calculated that from 1947 to 1996, the “blue chip” S&P 500 stocks yielded an average “real” return of 9.5 percent, while “growth stocks” of smaller companies paid 11 percent a year — after adjusting for inflation. There is no excuse for foundations earning less than the S&P 500 stock index (although as we will see, some do) since anyone can do that well by simply buying an indexed mutual fund.
Even using the figure of 9.5 percent, and subtracting 5 percent for grants and expenses, the conservative model leaves foundation assets growing at a real, inflation-adjusted rate of 4.5 percent a year. Making the heroic assumption that inflation remains very low, averaging only 2.5 percent a year, gross foundation assets could be expected to grow by 12 percent a year (9.5 percent plus 2.5 percent). Subtracting the 5 percent payout still leaves room for a 7 percent annual increase. Using these (extremely conservative) assumptions, foundation assets can be expected to increase from $226.7 billion in 1995 to $3.4 trillion by 2035 — not even counting new foundations or new gifts to existing foundations.
Organizations of foundation professionals explain rather weakly that their members are just running a sort of king-sized endowment on behalf of charities, a growing stockpile of unsigned IOUs, to permit larger grants at some undetermined date in the future. But using the income from foundation assets to acquire still more assets is not at all comparable to an endowment because with an endowment, all of the income from investments is typically used to pay the expenses of the endowed institution, such as a university or think tank. By contrast, what the largest foundations are doing is accumulating tax-exempt assets to expand themselves — for reasons that are not immediately obvious to most observers.
Moreover, the endowment argument might make some sense if a possible dollar received after two or three decades was just as valuable as an actual dollar today. But any struggling new charity — not to mention the populations they serve — will gladly explain why they would rather have some of that support right now. Universities would also rather have funds distributed to their own endowments today than to leave the money inside a foundation as an uncertain promise to pay at some undefined future date.
To find out what a dollar of donations in the future is worth today, we have to “discount” those future dollars by the expected return on assets. That “discounted present value” tells us that hoarding the treasure within the foundation rather than giving it away would be a good deal for charities only if the foundations have some way to earn an above-normal return on their investments. In fact, foundation investment returns have been worse than just holding an S&P 500 indexed stock fund and/or some ten-year Treasury bonds — both of which paid 17 percent a year from 1982 to 1995.
Foundation apologists also claim that they need to hoard assets to maintain the “real” inflation-adjusted value of their assets and grants. But unless somebody is suggesting that double-digit inflation will be the norm from now on, that excuse doesn’t work either. Recall that Professor Cochrane’s figures show that annual returns on conservative “blue chip” stocks have averaged 9.5 percent a year for five decades in real terms, adjusted for inflation. Recent experience confirms that achieving such average returns is not difficult. Net of new gifts and the 1-2 percent federal excise tax, foundation assets grew by 13.5 percent a year from 1982 to 1995. That still left assets growing by 10 percent a year in “real” terms, after subtracting 3.5 percent for inflation. Foundation grants, by contrast, rose by 4.5 percent a year in real terms. But even the 4.5 percent figure exaggerates the growth of grants from large foundations because smaller foundations devote a larger share of their assets to grants, and because some grants were funded from new gifts rather than from income on old assets.
Regardless of the chosen excuse for minimizing payouts in order to maximize asset growth, a simple political dynamic dooms the asset-maximization approach as a long-term strategy. If major foundations continue to put assets first and charity last, then foundation assets are likely to swell to $4 trillion by 2035, and might end up as large as $5.9 trillion (plus add another $1 trillion to account for new gifts). Before that happens, though, the United States will have faced an unprecedented fiscal crisis as the government tries to pay Social Security and Medicare benefits to retired baby boomers. What is the likelihood, under those circumstances, that Congress will a) fail to notice a tax-exempt hoard of $5 to 7 trillion, and b) decide to leave it alone? After all, Congress had no compunction about slapping a 4 percent tax on foundation assets in 1969, when the prize in increased revenues was much more modest.
At least one Member of Congress — Ways and Means Committee member Rep. Fortney “Pete” Stark — has already discussed the possibility of requiring foundations to “spend down” within a set period. Says a Stark aide: “Foundations should be required to spend all their wealth in a set time period. Donors creating foundations should decide what their goal or mission is, and have a specific timeframe in which to accomplish it. The foundation is there for the taxpayers, not for the purview of some independent trustee.”
When a trend is so obviously unsustainable as the endless accumulation of assets by a handful of corpulent foundations, we can be sure it won’t be sustained. Something will change. But how? The oldest, biggest, and most professionalized foundations seem unlikely to spontaneously begin doing something as bold as, say, donating a few dollars more than the law requires. When compound interest is applied to vast accumulations of old wealth, the result is a powerful force for inertia.
Harvard’s Frumkin proposes several modest changes, the best of which would be to exclude expenses and impose a slightly higher (in the range of 5-6 percent) payout rule. As the 1969 law showed, however, federal rules and regulations have a way of adding to paperwork burden without accomplishing much. Moreover, one problem with any annual payout rule is that it makes grantees too vulnerable to the ups and downs of the stock market. That has made it easier to get foundation funding for “quick fix” research projects, and difficult to get foundation support for operating effective services over the years.
Mr. Frumkin’s most provocative proposal is to eliminate the payout requirement altogether. “[We] might stimulate more giving simply by removing the current 5 percent target and replacing it with a moral responsibility to give as much as possible.” There is little to lose by giving it a try. Foundations were responsible for a larger fraction of total charitable giving before the 5 percent rule than they have been since. The requirement that foundation grants and expenses equal 5 percent of assets has merely provided an “official” sanction for doing just exactly that and no more.
If there were no legal requirement that foundations donate 5 percent of their assets, it is hard to imagine them giving even less than that. On the contrary, without the Scroogian cover of “doing all the law requires,” foundation managers might have to begin explaining a few things — like their expenses, for example, or why they repeatedly retain investment income rather than donating it. The tired pretense of acting as a piggy bank for charities would never stand up to serious scrutiny. In fact, there may be nothing wrong with big foundations that a little old-fashioned public shaming couldn’t fix.
Alan Reynolds is Director of Economic Research for the Hudson Institute. His recent publications include Death, Taxes and the Independent Sector: Reflections on the Past and Future Growth of Private Charities and Foundations, which is available from the Philanthropy Roundtable.