Who’s Afraid of Donor Advised Funds?
Four questions — and some answers—about the high-tech “tax hack” that isn’t
Co-blogger Brian Galle and University of Southern California professor Ed Kleinbard have both posted thoughtful responses to my qualified defense of donor advised funds (my initial post here; Brian’s here; Ed’s over at TaxProfBlog). Brian’s concerns about DAFs are essentially two-fold. First, he worries that DAFs — which are not subject to the same minimum payout rules as private foundations — will unduly delay distributions to non-DAF public charities. Second, he notes that DAFs can be used by private foundations for regulatory arbitrage — and in particular, to avoid disclosure requirements. Ed’s primary concern is that DAFs function as “a personal financial asset that compounds at tax-free rates.” He also worries about taxpayers contributing “high flying tech stocks and the like” to DAFs and then claiming “inflated” deductions before the price of the gifted shares crashes.
This is a debate that probably requires a law review-length article (or several) to hash out in full. We can make some progress, however, by looking for answers to four questions: (1) How do payout rates for DAFs and private foundations compare?; (2) Is regulatory arbitrage on the part of private foundations an empirically significant component of overall DAF activity?; (3) How much do donors benefit from the tax-free rate on assets held inside DAFs?; and (4) How easily can donors use DAFs to claim inflated deductions for gifts of stock and other appreciated assets? The answers to these questions, in my view, go a long way toward responding to critics such as Ed who say that DAFs are “a fraud on the American taxpayer.” They do not, however, allay all DAF-related concerns (which is why I ultimately agree with Brian that some additional regulations addressing private foundation gifts to DAFs may be warranted).
Payout rates. At the outset, I should note that I am agnostic as to whether we should encourage charities to accelerate spending. (For further thoughts from Brian on this subject, see his 2016 article in the Washington University Law Review.) Time-value-of-money arguments don’t get us very far. If we apply a social discount rate equal to the rate of return on capital, then we end up in equipoise. For example, assuming that the social discount rate/rate of return on capital is equal to 3%, feeding 100 hungry people today might be “worth” the same, in social welfare terms, as feeding 103 hungry people a year from now, but for precisely the same reason, we can feed 103 hungry people a year from now for the same resource cost as we can feed 100 people today. Which option is preferable? The answer, from a time-value-of-money perspective, is well summarized by the increasingly popular Twitter gesture “¯\_(ツ)_/¯.” (Which isn’t to suggest a shrug at the value of feeding the hungry — rather, it’s to say that both options would make the world a better place, and it’s not clear which one is better.)
Beyond pure time-value-of-money arguments, a number of other considerations are relevant to the timing of charitable contributions, but these considerations point in different directions. Spending sooner, as Brian notes in his 2016 article, gives society “the opportunity to learn from and build on charitable successes and failures.” Spending later, on the other hand, allows philanthropists to wait until rigorous randomized controlled trials shed more light on the efficacy of different interventions. The likelihood that the overall economy — and the philanthropic sector in particular — will grow in the coming years arguably means that money is needed more now than in the future. On the other hand, future generations will have to grapple with problems such as higher temperatures and rising sea levels that are present but less catastrophic today.
Charities and their donors, moreover, are not the only ones who make decisions regarding current vs. future spending on health, education, environmental protection, etc.; governments do too. If policymakers think that society is over-saving, then they can issue more government debt. Many of us are worried about the opposite problem: present-biased politicians have piled on too much debt, and this will put a strain on public schools and universities, health systems, and other vital institutions down the road. (This concern is especially salient here in Greece . . . er, Illinois.) From this vantagepoint, a bit of future bias on the part of the charitable sector doesn’t sound like such a bad thing.
Setting all that aside, and assuming that we share Brian’s concern with delayed payouts, how worried should we be about DAFs? One way to approach this question is to compare the payout rates for DAFs and private foundations. In 2014, according to the National Philanthropic Trust, grants from DAFs were equal to 21.6% of assets held at the end of the prior year. Applying the same methodology (i.e., dividing grants by assets held at the end of the prior year), the payout rate for non-corporate private foundations was 6.3%.
The relatively high payout rate for DAFs might seem surprising on first glance, but there are at least three factors that account for it. First, as discussed below, stuffing assets into a DAF and allowing them to sit there for decades turns out not to be a terribly tax-efficient strategy, so no wonder we don’t see lots of taxpayers doing that. Second, while DAF sponsors aren’t subject to the 5% minimum payout requirement, they must pass section 509(a)’s “public support” test in order to continue to qualify as a public charity. The most straightforward way to do that is to ensure that gifts from the public constitute at least one-third of total support. If DAFs continue to accumulate assets such that interest and dividend income is disproportionate to donation inflows, the DAF sponsor would jeopardize its public charity status. (Most DAF sponsors make sure that they clear the one-third threshold quite comfortably.) Third, the leading DAF sponsors all have relatively stringent minimum-grant-activity requirements. Fidelity Charitable requires that annual grants from each account exceed 5% of average net assets on a five-year rolling basis; if a donor dips below the 5% amount, Fidelity Charitable reserves the right to direct distributions on its own. Schwab Charitable and Vanguard Charitable have similar minimum-activity rules.
In sum, if you share Brian’s concerns about delayed charitable spending, private foundations rather than DAFs are where you should set your sights. The self-interest of DAF donors and self-regulation by DAF sponsors have done a good deal to keep DAF payout rates reasonably high. Form 990 reporting requirements make it relatively easy for members of the public (and particularly, us academics) to monitor DAF payout rates. Perhaps if payout rates drop precipitously in the future, the argument for legislative or statutory intervention might become stronger. For now, if your concern is about speeding up the rate of charitable giving, it seems like your focus should be on making private foundations more like DAFs rather than the opposite.
Regulatory arbitrage. Brian’s second concern with DAFs is related to regulatory arbitrage. “The biggest arbitrage area right now,” according to Brian, “is disclosure.” If a private foundation donates to the NRA Foundation or to the Planned Parenthood Federation of America, it must disclose its contribution on its publicly available Form 990-PF. If the private foundation donates to Fidelity Charitable and then directs Fidelity Charitable to make a grant to the NRA Foundation or Planned Parenthood, all that appears on the private foundation’s Form 990-PF is the gift to Fidelity Charitable. (Note that neither private foundations nor DAFs can contribute to the 501(c)(4) side of the NRA or Planned Parenthood without triggering penalty taxes and potentially jeopardizing their tax-exempt status.)
I don’t stay up at night worrying about this particular loophole, in part because individuals already can donate directly to the NRA Foundation or the Planned Parenthood Federation without any public disclosure. The private foundation-to-DAF-to-public charity route is only attractive to individuals who contribute to private foundations and then decide (or their heirs or their foundations’ managers decide) that they want to make a sub rosa gift to another public charity. In other words, it’s a backdoor way to support politically connected charities without disclosure when the front door is already wide open.
In any event, we can arrive at some sense as to how big this problem is by looking at the numbers. From fiscal year 2014 through fiscal year 2016, a total of 557 foundations — well less than 1% of all private foundations — made at least one grant to a DAF. Donations from private foundations to DAFs in 2014 accounted for approximately 0.4% of all foundation grantmaking that year and less than 1% of all contributions to DAFs. Not nothing, but not enough to tar the entire DAF sector.
There are reasons unrelated to disclosure requirements why we might be concerned about outflows from private foundations to DAFs. Most notably, some private foundations use DAFs as an end-run around the 5% minimum payout rule. For example, a private foundation with $1 billion in assets might give $50 million to Fidelity Charitable, thus (arguably) satisfying the 5% minimum payout requirement, and then might give 5% of that $50 million (i.e., $2.5 million) to non-DAF public charities, thus satisfying Fidelity Charitable’s self-imposed 5% rule. Effectively, the private foundation will have transformed a 5% minimum payout rule into an 0.25% payout floor. (Don’t believe it? Check out the Form 990-PF for hedge fund manager Stephen Mandel’s billion-dollar Zoom Foundation, which in 2016 satisfied its minimum payout requirement entirely through a large gift to an account at Fidelity Charitable.) For those who share Brian’s concerns regarding the timing of charitable spending, that’s worrying. Treasury and the IRS requested public comment this past December on whether they should take regulatory action to close this loophole. If you support the 5% minimum payout rule, then presumably you should support regulatory action.
Tax benefits. One reason why we don’t see more high-net-worth individuals stashing their assets in DAFs is that, from a tax perspective, it’s often not a terribly clever strategy. Here’s why. Let’s say that I hold stock worth $100 today. For arithmetic simplicity, we’ll say that the stock is expected to double in a decade (an annual rate of return of 7.18%). As is the case for Alphabet (Google), Amazon, Berkshire Hathaway, and Facebook, the stock pays no dividends. We’ll say that my marginal tax rate is 37% for ordinary income and 23.8% for long-term capital gains.
— Option 1: I donate the stock to a DAF today. I claim a charitable contribution deduction of $100, which is worth $37. I reinvest the $37. In a decade, the $37 becomes $74. If I sell, I pay a capital gains tax of 23.8% x ($74 — $37) = $8.81. After tax, I’m left with $65.19. If the DAF holds the stock, the DAF now has $200 which it can distribute to the public charity of my choice.
— Option 2: I hold onto the stock and do nothing with a DAF. In a decade, the $100 stock becomes $200. I can donate the stock to charity and claim a charitable contribution deduction of $200, which is worth $74. As above, the public charity of my choice ends up with $200 at the end of the decade, but I have $74 instead of $65.19. I’m $8.81 better off steering clear of the DAF.
This isn’t arithmetic legerdemain. With Option 1 (using a DAF), the stock grows tax-free in the DAF while the value of my charitable contribution deduction grows taxably. With Option 2 (no DAF), everything grows tax-free. Using the DAF means the same amount at the end of the decade for the public charity of my choice, less for me, and more for Uncle Sam.
To be fair, investing inside the DAF becomes more attractive for high-dividend stocks. But one can achieve broad sectoral diversification with zero-dividend stocks if one wants. Moreover, holding zero-dividend stocks outside of a DAF can — as I explain here — allow philanthropically-minded individuals to achieve negative effective tax rates through loss harvesting. In sum, with a small degree of financial sophistication, an individual who wants to make a future gift to charity can do a lot better from a tax perspective outside a DAF than inside.
This is not to say that DAFs are always disadvantageous from a tax perspective. As I noted in my initial post, taxpayers with fluctuating incomes can benefit from contributing to a DAF in their higher-earning years. Upper-middle-income married couples who lack the financial sophistication to implement a zero-dividend stock strategy also might benefit from bunching donations to a DAF once every half-decade or so and claiming the standard deduction in other years. Ed suggests that any taxpayer who wants to “fund[] charitable endeavors today while controlling the long term disbursements of those funds over time” should form a private foundation. But for upper-middle income couples boxed out of the benefits of the charitable contribution deduction on account of the $10,000 SALT cap and the elevated standard deduction, a private foundation is not a viable option.
Insider trading. Finally, as for Ed’s concern regarding taxpayers who donate stock to a DAF at “today’s inflated values” before their shares collapse, it’s worth noting that the same concern exists with regard to gifts of “qualified appreciated stock” to private foundations and almost all gifts of long-term capital gain property to charity. But moreover, this is what securities law is for. While there is no court case precisely on point, lawyers generally advise clients that giving stock to a private foundation, DAF, or other public charity while in possession of material nonpublic information could lead to liability under insider trading rules. The Supreme Court’s “personal benefit” test for insider trading, while formulated in the context of gifts of confidential information, would almost certainly apply to an insider who gives stock to private foundation, DAF, or other public charity and receives a tax deduction in return.
To be sure, there is some evidence that CEOs tend to donate corporate stock to private foundations at peak prices. This may be a function of backdating, which is likely more problematic in the context of private foundations than DAFs (it’s almost undoubtedly easier to get your own family foundation to go along with a backdating gambit than to get Fidelity Charitable to risk its reputation and its multibillion-dollar book of business for the same). It may also be more of a problem in the context of non-DAF public charities than the largest DAFs, because smaller charities will be more dependent on a single donor than Fidelity Charitable, Schwab Charitable, or Vanguard Charitable. In short, backdating of stock donated for charitable purposes perhaps ought to be a securities law enforcement priority, but there is no more reason to make DAFs the focus than anyone else (and maybe even less).
