An Adobe PDF version of this document is available here.

December 6, 2011

I write to comment on the Notice of Proposed Rulemaking (REG-128224-06), 76 FED. REG. 55322 (Sept. 7, 2011), re-proposing regulations under section 67(e) regarding the applicability to trusts and estates of the 2-percent floor for miscellaneous itemized deductions.

1.     There is no change in my conclusion that trusts and estates should generally be exempt from the 2-percent floor. In fact, that conclusion is reinforced by the preamble to the proposed regulations.

My view that the appropriate approach of regulations would be to hold that the 2-percent floor generally does not apply to trusts and estates was set out in some detail in my letter of May 27,2008, in response to Notice 2008-32.  A copy of my letter is enclosed.  That view has not changed.  Indeed, the unpersuasiveness of the preamble to the September 7, 2011, Notice of Proposed Rulemaking ("the Preamble") in its attempt to justify a different approach only reinforces my conviction that the 2-percent floor should generally not apply to trusts and estates.

For example, the Preamble acknowledges:

Many of the comments received in response to Notice 2008-32 highlighted the legislative intent of the provision imposing the 2-percent floor for miscellaneous itemized deductions. The commentators noted that the.intent was to simplify recordkeeping, reduce taxpayer errors, ease administrative burdens for the IRS, and reduce taxpayer errors in distinguishing between nondeductible personal expenditures and deductible miscellaneous itemized deductions.

That, of course, is absolutely true, for it is lifted practically word-for-word from the legislative history discussed on pages 3-8 of my 2008 letter and in other comments.  But the Preamble then immediately goes on to state:

The IRS and the Treasury Department recognize the administrative difficulty of determining whether every type of cost incurred by a trust or estate is the type of cost that would be incurred commonly or customarily by individuals owning the same property.  Therefore, the proposed regulations provide simplified rules for the application of section 67(e).

That, of course, does not follow, and it is absolutely clear that it does not follow.  The point of the references to legislative history is to identify thepurposes of section 67, which is the same as identifying the targets of section 67.  Because, uulike individuals acting on their own behalf, fiduciaries generally must keep records of expenditures and must accurately distinguish between trust expenditures and personal expenditures, and often must account to beneficiaries or courts in that regard, the decision compelled by the legislative history is that fiduciaries are not the targets of section 67, not that fiduciaries should be targeted in a simplified manner.  Clearly the simplification that Congress meant for section 67 was to eliminate the need for recordkeeping that the prior law effectively required (H.R. REP. No. 99-426, 99th CONG., 1st Sess. 109 (1985)).  And because fiduciaries must keep such records anyway, that is a simplification that generally can be conferred only on individual taxpayers, not fiduciaries.

It will be obvious to most readers that when the Preamble states that "[t]he IRS and the Treasury Department recognize" something that clearly does not follow, it is probably true that the authors of the Preamble "recognize" very clearly the outcome the legislative history compels, but they have chosen for some reason to deny or avoid it.  This is a judgment that should be reconsidered, for the sake of both faithfulness to the congressional purpose and the credibility and public acceptance of the regulations.

2.     There is no doubt about the authority  of Treasury and the Service to write rules in this case that respect the legislative history.

In Mayo Foundation for Medical Education and Research v. United States, 131 S. Ct. 704 (2011), in a unanimous (8-0, with Justice Kagan not participating) opinion written by Chief Justice Roberts, the Supreme Court confirmed that Treasury regulations in general would be given so-called "Chevron  deference" (see Chevron US.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984)) even if they are promulgated under the general grant of authority to interpret the Internal Revenue Code under section 7805(a).  Moreover, the Court confirmed that if "the plain text of the statute" does not "speak ... with the precision necessary to say definitively whether [the statute] applies ... , such an ambiguity would lead us inexorably to Chevron step two, under which we may not disturb an agency rule unless it is 'arbitrary  or capricious in substance, or manifestly contrary to the statute.'   Household Credit Services, Inc. v. Pfennig,541 U.S. 232,242 (2004) (quoting United States v. Mead Corp., 533 U.S. 218,227 (2001))."  As described on pages 10-11 of my 2008 letter, the Supreme Court, in Knight v. Commissioner, 552 U.S. 181 (2008), also in a unanimous opinion written by ChiefJustice Roberts, directly described its view of section 67(e), "particularly given the absence of regulations," as entailing "some uncertainty" and, indirectly through its citation of Commissioner v. Clark,489 U.S. 726, 739 (1989), embraced a view of section 67(e) as "a somewhat ambiguous exception."

In short, the Supreme Court has confirmed section 67(e) as presenting "such an ambiguity" that courts under Mayo would be obliged to respect any regulation that is not "arbitrary or capricious in substance, or manifestly contrary to the statute."  To be sure, that principle would uphold Treasury's  choice of interpretation, whether harsh or lenient.  But in making that choice, I suggest again that Treasury and the Service should be guided by the purpose of section 67 reflected in the legislative history, just as in Mayo itself the Supreme Court agreed that Treasury had "reasonably determined that [the rule in that case] would further the purpose of the Social Security Act."

3.     Treasury and the Service should therefore lead in this area of tax administration and should not merely react to court decisions.

Frankly, the biggest disappointment with the Notice of Proposed Rulemaking is that, like the 2007 proposed regulations that followed the Second Circuit's  decision in the Knight case, it tries to simply follow the Supreme Court's  decision.  Under Mayo, Chevron, and section 7805 itself, Treasury should promulgate rules that courts will follow, not the other way around.

The case-by-case consideration by judges of such facts as happen to appear in what the Supreme Court bemoaned as "the absence of regulatory guidance" is not a good way to produce rules of general application and to achieve consistency with sound tax policy.  On pages 8-10 of my 2008 letter, I pointed out that these cases seemed to be influenced by a different version of the legislative history foisted by a fairly consistent employment by the Justice Department of what amounted to a misleading 788-page ellipsis.  As I encouraged in 2008, this is the time for the clients - that is, Treasury and the Service - to take control of the case and bring leadership to this issue.

4.     Following the lead of fact-specific court cases has perpetuated an approach that is fundamentally flawed and unworkable.

To attempt to adapt in regulations the "commonly or customarily incurred" standard developed by courts in a regulatory vacuum would just substitute one set of vague and subjective words for another.  The Supreme Court in Knightbelieved that its own formulation "inevitably entails some uncertainty."  Proposed Reg. §1.67-4(b) admonishes that "it is the type of product or service rendered to the estate or non-grantor trust in exchange for the cost, rather than the description of the cost of that product or service, that is determinative.... [C]osts that are incurred commonly or customarily by individuals also include expenses that do not depend upon the identity of the payor (in particular, whether the payor is an individual or instead is an estate or trust)."  That the nature of a cost cannot be changed merely by its "description" is hard to quarrel with.  But absent  from the view reflected  in the Preamble is any acknowledgment of
context,  which  does matter.

For example,  with reference to investment advice, the Preamble speaks  of"a specialized balancing  of the interests of various  parties," without  acknowledging that all actions of a fiduciary involve  a need for "a specialized balancing  of the interests" of beneficiaries that no one in an individual capacity  faces.  The Preamble  goes on to observe  or presume that "[i]ndividual investors  commonly have investment objectives that may require  a balance  between investing for income and investing  for growth and/or a specialized approach  for particular assets."  But that has nothing to do with any kind of fiduciary  responsibility or duty to balance the interests  of different  beneficiaries. It is just an approach  to making  money in one's own self-interest. Whether the money that is made is consumed  or saved generally  has nothing  to do with how it is made.  Indeed,  the most important reason an individual investor "would" even care about notions like "income" and "principal" might  well be to balance the ordinary income  and capital growth that make up that individual's total return, because  ordinary  income  and capital  growth are taxed at different times and at different  rates.  There is nothing improper  about such tax planning, of course, but it would be odd to see it dignified  as a guiding principle of an income tax regulation about the application to fiduciaries of a congressional purpose  of simplification.

The Supreme  Court in Knight wrestled  almost comically with the notion that the trustee had invoked  his fiduciary duty to act as a "prudent investor," a standard that originated  in the notion of a "prudent man," who necessarily is an individual. But the point of over 180 years of fiduciary law is that an individual has no duty to be "prudent" with respect  to his own investments, but a fiduciary always has a duty to be prudent with the investments of the trust or estate.

Nor should  it matter that an individual  might even pay the same amount  for a superficially similar  service.   The fee that a trustee earns compensates that trustee for undertaking a responsibility that is serious and sobering.   The fiduciary relationship represents what Chief Judge  Cardozo  famously elucidated  as "not  honesty  alone,  but the punctilio  of an honor most sensitive." Meinhard v. Salmon, 249 N.Y. 458, 464, 164 N.E.  545 (1928).   An expense-by-expense analysis  would completely miss the point.   For example, empirical  data might indicate  that some institutional fiduciaries charge about the same fees for simply giving investment advice and for serving  in the full role of trustee.   That would tend to encourage  the conclusion that a trustee's fee is entirely  or largely composed of compensation for investment advice.   But such a conclusion would be a leap of faith that would ignore the realities of fiduciary  relationships and of the business world in general.  A trustee  makes  investment decisions, to be sure, but a trustee also assumes  a responsibility for understanding the needs of beneficiaries and for making  investment and distribution decisions with the interests  of all beneficiaries as well as the purposes of the grantor in view.  A trustee  offers "big picture" evaluation and judgment. Often in my experience some ofthe most thoughtful estate planning recommendations come from personnel  employed  by corporate trustees who are able to draw from their wide experience.  Such a trustee offers the availability of specialists, a reputation for imaginative thinking, and, indeed, a "deep pocket" for the assumption of risk.

Ultimately, individuals choose investment advisers to make money, they choose trustees for a whole range of reasons of which making money is only an undivided part, and trustees seek investment advice to fulfill the highest of fiduciary duties.  Any effort to equate those contexts is, in my judgment, futile.

The fact that some such trustees would employ a fee schedule that undertakes all those responsibilities for some clients, but only investment advice for other clients for a similar fee, reflects a business judgment that has no obvious connection with implementing the policy behind section 67.  Similarly, the decision to engage other specialists or to bring all activities "in house" is a business judgment that tax rules should not reward or penalize.

The statute, admittedly unclear, suggests a test of asking whether costs "are paid or incurred in connection with the administration of the estate or trust [which seems always to be conceded] and which would not have been incurred if the property [presumably the property in such trust or estate] were not held in such trust or estate."  So suppose a share of stock was removed from the trust, or, because the trust in question is not likely to be revocable, is distributed out of the trust to a beneficiary, such that the share of stock would no longer be "held in such trust."  Would the investment advice then cease?  Of course it would.  It could be renewed, but that would be a different kind of expense serving a different kind of objective­ one of pure maximization of return rather than of discharge of any duty.  Who could say that the individual beneficiary "would" incur that investment expense?  It might be argued, quite to the contrary, that an individual beneficiary would be most unlikely to incur investment advice expense with respect to one share of stock.  Yet the statute speaks of "property  ... not held in such trust."  If one share of stock is a silly example, would it be more compelling if it were ten shares of stock?  How about a hundred?  Or maybe a thousand?  Would it matter how valuable a share of stock is?  Or what kind of stock it is?  What industry it represents?  How volatile its price is?  Whether or not it is traded on an exchange?  And so forth.  Not to mention all other kinds of property.  Would it really be an improvement - or a "simplification" - to encourage auditing zeal to be directed to such inquiries?  Or, more to the point, does Treasury really want to empower and encourage taxpayers'  advisers like me to spend time constructing and advancing such frivolous and wasteful analyses?

I have seen some of the paperwork that is involved in auditing even a relatively modest and simple trust like that involved in Scott v. United States, 328 F.3d 132 (4th Cir. 2003) ­ particularly when it comes to the necessary tasks of recycling the adjustments through the alternative minimum tax filter, allocating the tax character of specific items to the beneficiaries who receive distributions, and then undertaking the AMT exercise all over again at the individual level. It is impossible to conceive of such activity as a productive and sensible use of resources, much less "simplifying."

5.     The following observations are offered in an effort to be helpful no matter what approach the final regulations take.

While the "commonly  or customarily incurred" standard is unworkable, focusing on what necessarily happens when "property" is removed from a trust suggests that the approach to what might be called "in rem" costs in Proposed Reg. §1.67-4(b)(2) is correct.  In addition, such costs would ordinarily be discrete and therefore would not present subjective "unbundling" challenges.

The approach of Proposed Reg. §l.67-4(b)(3) to various tax returns also seems in the main to be correct.  There might be exceptions in certain cases, such as the gift tax return that relates to the creation or funding of a trust or is prepared and filed by an executor after the donor's death.

In addition to the previous comments about "a specialized balancing of the interests of various parties," I suggest that any approach that treats a "simple trust" (where the mix of ordinary income and capital growth might directly affect the balancing of beneficial interests) more leniently than a "total return trust" or a unitrust would unfairly penalize the use of more modem techniques usually thought to be superior (as the 2003 amendments to Reg. §1.643(b)-1, for example, seemed to acknowledge).

While still viewing it as a mistake to subject a fiduciary's  costs such as the cost of investment advice to the 2-percent floor at all, I believe that to the extent su-ch costs are nevertheless subject to the 2-percent floor, unbundling is a necessary measure to prevent unfair discrimination, especially between large institutions and smaller institutions or individual trustees.  The focus on "hourly rates" might also unfairly discriminate against individual trustees like the trustees in Scott and Knight, who are unable to employ a fee schedule based on experience with many accounts.

I will of course be glad to answer questions or assist in any other appropriate way.

Sincerely,

Ronald D. Aucutt

Enclosure: Copy of May 27,2008, letter

An Adobe PDF version of this document is available here.

May 27, 2008

Dear Ladies and Gentlemen:

I write, pursuant to the invitation in Notice 2008-32, to recommend that the final regulations under section 67(e) exempt the administration expenses of multi-beneficiary trusts from the “2% floor” imposed by section 67. I make this recommendation for the following reasons:

•       Application of the 2% floor to trusts is exceedingly complicated. This is true even of payments by trustees to outside providers of services (such as the payment of investment advisory fees). The complexity would be aggravated if the 2% floor were applied to a portion of a unitary trustee’s fee, referred to as a “Bundled Fiduciary Fee” in Notice 2008-32.

•       The purposes of section 67 and the 2% floor do not require that the 2% floor be applied to multi-beneficiary trusts. Indeed, those purposes are offended by the complexity such application would produce.

•       The recommended exception is permitted by the statute, is consistent with the legislative history, and under case law would be given deference as a reasonable interpretation of the statute.

•       The recommended exception would increase certainty, reduce exasperation, and avoid controversy and likely litigation.

In addition, I recommend that before finalizing the regulations the Service and Treasury schedule a hearing, conference, or other opportunity to discuss these issues with members of the public who respond to Notice 2008-32.

The following discussion will elaborate my recommendation and these reasons. For convenience, I will generally refer in this letter only to trusts, although my recommendation applies only to trusts with more than one beneficiary and applies as well to decedents’ estates.

I.     Background

Section 67 of the Internal Revenue Code, enacted in 1986, allows “miscellaneous itemized deductions” for income tax purposes only to the extent those “miscellaneous itemized deductions” exceed 2 percent of adjusted gross income (the “2% floor”). “Miscellaneous itemized deductions” are defined insection 67(b) and include deductions under section 212 for ordinary and necessary expenses for the production or collection of income, for the management, conservation, or maintenance of property held for the production of income, and in connection with the determination, collection, or refund of any tax.

Section 67(e) provides for the calculation of adjusted gross income of an estate or trust in the same manner as in the case of an individual (thereby clarifying the application of subchapter J, which provides, in section 641(b), that the taxable income of an estate or trust is computed in the same manner as in the case of an individual). Section 67(e)(1) provides that “the deductions for costs which are paid or incurred in connection with the administration of the estate or trust and which would not have been incurred if the property were not held in such trust or estate ... shall be treated as allowable in arriving at adjusted gross income.” The effect is to exempt such costs from the 2% floor.

On January 16, 2008, the Supreme Court decided Michael J. Knight, Trustee v.Commissioner, 552 U.S. ___, 128 S. Ct. 782 (No. 06-1286, Jan. 16, 2008). In a unanimous opinion by Chief Justice Roberts, the Court affirmed the Court of Appeals for the Second Circuit and held that for federal income tax purposes trust investment advisory fees are subject to the 2% floor.

While Knight was pending, the Service released proposed regulations under section 67(e), addressing the application of the 2% floor to trusts. Proposed Reg. § 1.67-4, REG-128224-06 (July 26, 2007). In general, the proposed regulations would exempt from the 2% floor only costs that are “unique” to a trust, including costs of fiduciary accountings, required judicial or quasi-judicial filings, fiduciary income tax returns, estate tax returns, division or distribution of income or corpus to or among beneficiaries, trust or will contests or constructions, fiduciary bonds, and communications with beneficiaries regarding trust matters. As examples of services that are not “unique” to a trust, the costs of which are subject to the 2% floor, the proposed regulationscite the custody and management of property, advice on investing for total return, gift tax returns, the defense of claims by creditors of the decedent or grantor, and the purchase, sale, maintenance, repair, insurance, or management of property not used in a trade or business. The proposed regulations would also require the “unbundling” of unitary fiduciary fees or commissions for fiduciary services, so as to identify the portions attributable to activities and services that are not “unique” and are therefore subject to the 2% floor.

As proposed, the regulations would apply to “payments made after the date final regulations are published in the Federal Register.” Proposed Reg. § 1.67-4(d). The Service received written comments about the proposed regulationsand held a public hearing on November 14, 2007.

On February 27, 2008, the Service issued Notice 2008-32, 2008-11 I.R.B. 1, acknowledging the Supreme Court’s Knight decision, expressing an intention to finalize the regulations consistently with Knight, providing that “unbundling” of a “Bundled Fiduciary Fee” would not be required for taxable years before 2008, and requesting further comments on the proposed regulations by May 27, 2008. Among other things, Notice 2008-32 stated that the Service and Treasury were considering percentage “safe harbors” for unbundling a “Bundled Fiduciary Fee” and requested comments on such safe harbors.

This letter is written in response to Notice 2008-32.

II.     The Legislative History Focuses on Simplification as the Purpose of the 2% Floor

One of the chief criticisms of the Service’s attempt to subject trusts to the 2% floor, and of the cases that have supported that attempt, is that the purposes of the 2% floor – reducing recordkeeping and reducing erroneous deductions of personal expenditures – simply do not apply to trusts, which generally are required to keep accurate records and distinguish personal expenditures anyway.

Admittedly, there are many ways to identify the purposes of a congressional statute, because there are several ways to look at “legislative intent.” The following is a sampling:

•       What Congress “must have” intended, given the mood of the times and the personalities involved.

•       What Congress “actually” intended, which is usually “known” only by “insiders” of the day (and since Congress has 535 Members and many more staff members, such “knowledge” might be accidental and unreliable).

•       What Congress said, typically in committee reports.

•       What the purpose of any given provision should be understood to be, given the provision’s terms and effect and the law-school notion of asking the “reason for the rule” when application to a given set of facts is unclear.

In this letter, I rely, as I believe the drafters of regulations should, on what Congress (and the Administration) said with immediate reference to the 2% floor and also on a “reason for the rule” approach, which leads to the same conclusion.

A.     Administration Proposals

“Tax Reform for Fairness, Simplicity, and Economic Growth” (popularly called “Treasury I”) was published by Treasury on November 27, 1984, just weeks after President Reagan’s landslide reelection. It included a proposal that would subject miscellaneous itemized deductions (along with unreimbursed employee expenses and state and local taxes other than income taxes) to a floor equal to 1% of adjusted gross income. At pages 115-16 of volume 2, Treasury justified this proposal as follows (emphasis added):

Allowance of the various employee business expense deductions and the miscellaneous itemized deductions complicates recordkeeping for many taxpayers. Moreover, the small amounts that are typically involved present significant administrative and enforcement problems for the Internal Revenue Service. These deductions are also a source of numerous taxpayer errors concerning what amounts and what items are properly deductible.

....

Disallowance of a deduction for a normal level of employee business expenses and miscellaneous itemized deductions wouldsimplify recordkeeping, reduce taxpayer errors and ease administrative burdens for the Internal Revenue Service while still providing fair treatment for taxpayers who incur an unusually high level of such expenses.

In 1982, one-half of all itemizers claimed miscellaneous deductions of less than one-half of one percent of their AGI. Fifty-eight percent claimed deductions of less than one percent of their AGI, and 93 percent claimed deductions of less than five percent of their AGI. Thus, introduction of a “floor” or “threshold” of one percent of AGI would substantially reduce the number of returns claiming this deduction. The proposed extension of the miscellaneous deduction to nonitemizers would partially offset the revenue gain from introduction of the floor.

The proposal would broaden the tax base and, thus, contribute to the reduction in marginal tax rates. Any increase in tax liability resulting from this proposal should be more than offset by the reduced marginal rates and the increase in the zero bracket amount and the personal exemption.

 

“The President’s Tax Proposals to the Congress for Fairness, Growth, and Simplicity” (often called “Treasury II”) was published by the White House on May 29, 1985, to communicate the President’s recommendations to Congress. Treasury II included the proposed 1% floor on miscellaneous itemized deductions from Treasury I. On pages 104-105, Treasury II reproduced the same justification quoted above, except that the last paragraph (referring to broadening the tax base) was omitted.

Thus, the original proposal of a special rule for miscellaneous itemized deductions focused on the taxpayer’s recordkeeping burden, the potential for taxpayer errors, and the Service’s administrative burdens. Elaboration in terms of itemizers and nonitemizers confirmed thatindividuals were in view. An initial reference to base-broadening was clearly secondary, in that it appeared at the end of the initial discussion and in subsequent formulations was not mentioned at all.

Under the heading of “Tax Abuses—Income Shifting,” Treasury I and Treasury II also proposed the taxation of the unearned income of children under 14 at the marginal tax rate of their parents, outlined sweeping changes in the income taxation of trusts, and suggested the continuation of a decedent’s taxable year without starting a new taxable year upon death.

The proposal for changing the income taxation of trusts would eliminate the separate rate schedule for trusts (based on the rate schedule applicable to married individuals filing separate returns). At page 105 of volume 2, Treasury I summarized the proposal as follows (emphasis added):

Because all trust income would be taxed to the grantor, taxed to trust beneficiaries, taxed to the trust at the grantor’s marginal rate (during the grantor’s lifetime), or taxed to the trust at the highest individual rate (after the grantor’s death), the proposal would eliminate the use of trusts as an income-splitting device. In this respect, the proposal would reinforce the integrity of the progressive rate structure and thus enhance the fairness of the tax system.

Thus, it was for these proposals, not the special rule for miscellaneous itemized deductions, that propping up revenue was an immediate articulated objective of the Administration proposals.

B.     The House of Representatives

The original House bill that became the Tax Reform Act of 1986 (H.R. 3838, introduced December 5, 1985, and reported by the Ways and Means Committee December 7, 1985) proposed a new section 67 of the Internal Revenue Code, subjecting “miscellaneous itemized deductions” to a floor equal to 1% of adjusted gross income, as in the Administration proposals.

In explaining this proposal, the House Ways and Means Committee stated:

The committee believes that the present-law treatment of employee business expenses, investment expenses, and other miscellaneous itemized deductions fosters significant complexity. For taxpayers who anticipate claiming itemized deductions, present law effectively requires extensive recordkeeping with regard to what commonly are small expenditures. Moreover, the fact that small amounts typically are involved presents significant administrative and enforcement problems for the Internal Revenue Service. These problems are exacerbated by the fact that taxpayers may frequently make errors of law regarding what type of expenditures are properly allowable as miscellaneous itemized deductions.

H.R. REP. NO. 99-426, 99TH CONG., 1ST SESS. 109 (1985) (emphasis added).

The House bill included the following new section 67(c):

(c) DETERMINATION OF ADJUSTED GROSS INCOME IN CASE OF ESTATES AND TRUSTS.—For purposes of this section, the adjusted gross income of an estate or trust shall be computed in the same manner as in the case of an individual, except that the deductions for costs paid or incurred in connection with the administration of the estate or trust shall be treated as allowable in arriving at adjusted gross income.

C.     The Senate

The Senate Finance Committee’s version of the 1986 bill proposed a new section 280I of the Internal Revenue Code, subjecting certain employee expenses to a floor equal to 1% of adjusted gross income. The Senate bill would have added the following new subsection (b) to section 62 (the definition of “adjusted gross income”):

(b) DETERMINATION OF ADJUSTED GROSS INCOME IN CASE OF ESTATES AND TRUSTS.—For purposes of this subtitle, the adjusted gross income of an estate or trust shall be computed in the same manner as in the case of an individual, except that the deductions for costs paid or incurred in connection with the administration of the estate or trust shall be treated as allowable in arriving at adjusted gross income.

With respect to miscellaneous itemized deductions, the Senate bill would have repealed such deductions altogether. The Finance Committee’s explanation of this proposal resembled the Ways and Means Committee’s explanation of the House bill. The Finance Committee began its discussion of “Reasons for Change” with the following:

The committee believes that, as part of the approach of its billto reduce tax rates through base-broadening, it is appropriate to repeal the miscellaneous itemized deductions and to limit deductions for certain employee expenses. The committee also concluded that allowance of these deductions under present law fosters significant complexity, and that some of these expenses have characteristics of voluntary personal expenditures.

S. REP. NO. 99-313, 99TH CONG., 2D SESS. 78 (1986) (emphasis added).

 

Thus, with respect to miscellaneous itemized deductions, in contrast to Treasury and the Ways and Means Committee, the Finance Committee was apparently more concerned with revenue enhancement through base-broadening. In addition, its concern about voluntary personal expenditures was not only that errors might be made, but that some miscellaneous itemized deductions inherently resembled such expenditures. Consistently with those concerns, the Senate bill would have eliminated miscellaneous itemized deductions altogether.

As a result, with respect to the Senate bill, it would be harder to argue that simplification was the dominant concern and base-broadening was only secondary. But then, under the Senate’s approach of total repeal, the identification of “unique” costs, the “unbundling” of unitary fees, and the allocations between the trust and its beneficiaries, which make application of the 2% floor so complicated and burdensome, would not be necessary. Total repeal might, ironically, have been “simpler.” But that is not what Congress chose to do.

D.     The House-Senate Conference

The House-Senate conference refused to go as far as the Senate’s repeal, but it increased the 1% floor of the House bill to the 2% floor now imposed by section 67. It was the House- Senate conferees who added to section 67(e) the words “and would not have been incurred if the property were not held in such trust or estate.” (The Technical and Miscellaneous Revenue Act of 1988 (Public Law No. 100-647) redesignated this statutory provision as section 67(e)(1), added the second “which” in section 67(e)(1), and added a new section 67(e)(2) to clarify that the personal exemption and the distribution deduction are exempt from the 2% floor.)

 

It was the 1986 conference report that first mentioned trusts in committee report language:

Pursuant to Treasury regulations, the floor is to apply with respect to indirect deductions through pass-through entities (including mutual funds) other than estates, nongrantor trusts, cooperatives, and REITs [the rule contained in section 67(c)]. The floor also applies with respect to indirect deductions through grantor trusts, partnerships, and S corporations by virtue of present-law grantor trust and pass-through rules. In the case of an estate or trust [i.e., other than a grantor trust], the conference agreement provides that the adjusted gross income is to be computed in the same manner as in the case of an individual, except that the deductions for costs that are paid or incurred in connection with the administration of the estate or trust and that would not have been incurred if the property were not held in such trust or estate are treated as allowable in arriving at adjusted gross income and hence are not subject to the floor [the rule contained in section 67(e)]. The regulations to be prescribed by the Treasury relating to application of the floor with respect to indirect deductions through certain passthrough entities are to include such reporting requirements as may be necessary to effectuate this provision.

H.R. REP. NO. 99-841, 99TH CONG., 2D SESS. II-34 (1986) (conference report).

 

The single sentence of the legislative history that specifically addressessection 67(e) adds nothing to the statutory language. Thus, it could be argued, as it has been in the ensuing litigation, that the context suggests a congressional concern only with other kinds of passthrough entities, and that the sole purpose of section 67(e) was to relieve estates and non-grantor trusts from the application of the 2% floor.

In any event, Congress did not accept the proposals of the Administration to make sweeping structural changes to the income taxation of trusts and estates. Instead, the ’86 Act simply compressed the rate brackets applicable to trusts, so that the top rate (28%) would be reached at the level of a taxable income of $5,000 (indexed for inflation), rather than $79,500 (indexed for inflation) as under pre-1986 law. (Section 1411 of the ’86 Act did follow through on the Administration proposal regarding the unearned income of children, by enacting the “kiddie tax” now found in section 1(g).)

E.     Summary of the Legislative History

In short, Congress’s stated purposes in subjecting certain deductions to the “2% floor” were simplification (by reducing recordkeeping) and fairness (by removing the opportunity to mix personal expenditures with legitimately deductible expenses). Anyone who has ever administered a trust knows that the trustee’s fiduciary duties to beneficiaries (and sometimes accountability to a court) require careful recordkeeping and identification of the character of expenditures, without regard to tax rules. Congress, judging by its stated purposes, did not aim section 67 at trusts.

F.     The Justice Department’s Treatment of the Legislative History

One of the things about the controversy over section 67(e) that has most exasperated fiduciaries and their advisors is what has been perceived as the Justice Department’s reconstruction of the legislative history to sway the recent decisions of federal courts, where of course Justice Department attorneys enjoy great credibility.

 

For example, on page 34 of the Justice Department’s brief in the Supreme Court Knight case, counsel cited the Senate Finance Committee’s 1986 references to “complexity” and “voluntary personal expenditures” in S. REP. NO. 99-313, 99TH CONG., 2D SESS. 78-79 (1986) (quoted above). In the next paragraph, on the same page, of the Government’s brief, counsel added the following:

Congress also recognized that “[t]he present rules relating to the taxation of trusts and estates permit the reduction of taxation through the creation of entities that are taxed separately from the beneficiaries or grantors of the trust or estate.” 1986 Senate Rep. 867.

Conspicuously, pages 78-79 and page 867 of the Finance Committee report are 788 pages apart. In fact, the Finance Committee’s “permit the reduction of taxation” comment was made in the context of explaining the compression of the income tax rates in section 1(e) applicable to estates and trusts (also described above). The Finance Committee, just two paragraphs later, went on to add:

On the other hand, the committee believes that significant changes in the taxation of trusts and estates are unnecessary to accomplish this result. Accordingly, the bill attempts to reduce the benefits arising from the use of trusts and estates by revising the rate schedule applicable to trusts and estates so that retained income of the trust or estate will not benefit significantly from a progressive tax rate schedule that might otherwise apply. This is accomplished by reducing the amount of income that must be accumulated by a trust or estate before that income is taxed at the top marginal rate. The committee believes that these changes will significantly reduce the tax benefits inherent in the present law rules of taxing trusts and estates while still retaining the existing structure of taxing these entities.

S. REP. NO. 99-313, 99TH CONG., 2D SESS. 868 (1986). Thus, the Finance Committee disclaimed any disposition to implement its “permit the reduction of taxation” objective through any changes to the rules (other than rates) governing the taxation of trusts and estates and in any event gave no indication that it had directed its “permit the reduction of taxation” comment to the treatment of miscellaneous itemized deductions it had addressed 867 pages earlier.

 

I was counsel for the trustees in Scott v. United States, 328 F.3d 132 (4th Cir. 2003), where Justice Department attorneys perpetrated the same 867-page ellipsis, and we pointed that out in our responsive brief. Nevertheless, counsel persisted in obscuring the stated congressional focus on simplification, which makes application of the 2% floor to trusts seem so unnecessary and just plain wrong. While I cannot claim much objectivity in the matter, I respectfully ask Treasury and the Service – Justice’s clients – to consider if public respect for the tax administration system is not worth some caution here. If so, then perhaps Treasury and the Service might step back and take another look at the 2% floor in light of Congress’s stated purposes, rather than merely codifying their lawyers’ judicial successes in collecting a few marginal tax dollars under an unclear statute.

III.     Case Law Leaves the Door Open to a Reasonable Application of the 2% Floor

In Mellon Bank, N.A. v. United States, 265 F.3d 1275 (Fed. Cir. 2001), the Court of Appeals for the Federal Circuit became the first court of appeals to hold a trustee’s investment advisory fees to be subject to the 2% floor. The court stated that section 67(e)(1) “treats as fully deductible only those trust-related administrative expenses that are unique to the administrationof a trust and not customarily incurred outside of trusts.” Id. at 1281. Nevertheless, despite the use of the word “unique,” the court rested its conclusion merely on the observation that “[i]nvestment advice and management fees are commonly incurred outside of trusts.” Id.

In Scott, the Court of Appeals for the Fourth Circuit reached the same result. The court quoted the reference to “unique” expenses in Mellon Bank, but immediately added that “[p]ut simply, trust-related administrative expenses are subject to the 2% floor if they constitute expenses commonly incurred by individual taxpayers.” Id. at 140.

Writing for a unanimous Court in Knight, Chief Justice Roberts adopted an approach of “hypothetical” “prediction” – the Court’s words. Rejecting the notion (entertained by the Second Circuit) that “would not” means “could not,” the Court seemed more inclined to the tests employed by the Federal Circuit and the Fourth Circuit. The Court quoted the statement in Mellon Bank thatsection 67(e)(1) “treats as fully deductible only those trust-related administrative expenses that are unique to the administration of a trust and not customarily incurred outside of trusts” and said “[w]e agree with this approach.” 128 S. Ct. at 789. Nevertheless, like the Federal and Fourth Circuits, the Supreme Court did not rest its decision on the concept of “uniqueness.” The Court reduced the operation of the statute to a simple question: “whether a particular cost would have been incurred if the property were held by an individual instead of a trust.” Id. at 787 n.4. The Court’s approach is to imagine, hypothetically, that the property in question were not held in trust and then ask if the expense in question “would have been incurred” by the individual owning it.

But the Court stopped far short of viewing the statute as clear and unambiguous and compelling any particular result. To the contrary, the Court said that “[w]e appreciate that the inquiry into what is common may not be as easy in other cases, particularly given the absence of regulatory guidance.... Congress’s decision to phrase the pertinent inquiry in terms of a prediction about a hypothetical situation inevitably entails some uncertainty, but that is no excuse for judicial amendment of the statute.” Id. at 791.

Moreover, the Court supported its view of section 67(e)(1) by quoting the statement in its 1989 opinion in Commissioner v. Clark, 489 U.S. 726, 739 (1989) (a case involving the treatment of “boot” received in a “triangular merger” as a dividend rather than capital gain under the exception in section 356(a)(2)) that “[g]iven that Congress has enacted a general rule ..., we should not eviscerate that legislative judgment through an expansive reading of a somewhat ambiguous exception.” 128 S. Ct. at 789.

These references to “a somewhat ambiguous exception,” “some uncertainty,” and “the absence of regulatory guidance” leave the door open for Treasury to provide definitive practical guidance. The Supreme Court mandates that the courts give wide deference to Treasury’s interpretation of its own ambiguous regulations. Chevron v. Natural Resources Defense Council,Inc., 467 U.S. 837 (1984); Auer v. Robbins, 519 U.S. 462 (1997). As a result, when a court reviews a construction of a statute, the court must determine only whether the regulation is based on a permissible construction of the statute. Thus, courts need not conclude that the regulatory construction was the only permissible construction, or even the construction the court would have reached if it examined the statute in the first place. Chevron, 467 U.S. at 843-44.

Although Congress did not explicitly delegate rulemaking authority with respect to section 67(e)(1), rulemaking authority is derived from the general delegation in section 7805(a). As a result, if Treasury’s administrative interpretation of the statutory provision is reasonable, courts will grant the regulation deference and uphold it. Id. Courts may not simply impose their own construction of section 67, but instead must defer to Treasury’s reasonable, and thus permissible, regulatory construction of the statutory provision.

In other words, Treasury and the Service are free to publish final regulations providing a reasonable interpretation of section 67(e)(1) and a reasonable application of the 2% floor. Such regulations would be consistent with Knight, as Notice 2008-32 forecast.

IV.    It Would Be Reasonable to Exempt Multi-Beneficiary Trusts from the 2% Floor

A.     Exemption Would Carry Out the Purpose of the Statute

As described above, the dominant purposes identified in the legislative history of the 2% floor are to reduce recordkeeping, avoid disproportionate administrative efforts, and reduce the occasions for errors of law in distinguishing legitimately deductible expenses from personal expenditures. Even if it is contended that the 1986 House-Senate conferees did not consciously intend a broad exemption for trusts when they added what is now the last clause of section 67(e)(1), it would certainly be reasonable to view simplification as the primary “reason for the rule” in determining the limits of section 67(e)(1) when application to a given set of facts is unclear. 

As stated above, the “reasons for the rule” of reducing recordkeeping and reducing erroneous deductions of personal expenditures generally do not apply to trustees, which are required to keep accurate records and distinguish personal expenditures in any event. Moreover, as other commentators have no doubt demonstrated, application of the 2% floor to trusts would be disproportionately complicating, not simplifying.

Although I write this letter on my own behalf and not on behalf of any client or organization, I am familiar with the practices and challenges of fiduciaries, particularly corporate fiduciaries responsible for large numbers of fiduciary income tax returns each year. I am convinced of the burdens the 2% floor in general, and particularly the “unbundling” requirement set forth in the proposed regulations, will impose. The comments that your office has received and is likely to receive from fiduciaries are not whining to secure a tax benefit for those fiduciaries’ clients. These administrative burdens are real. It is also important to remember that recordkeeping and other duties imposed on fiduciaries already protect against the confusion between legitimate deductions and personal expenditures – there are no “abuses” or “tax shelters” here.

For additional confirmation, I recommend consultation with revenue agents in the field. It is hard to believe that many would view it as an efficient use of resources to sift through a trustee’s admittedly legitimate expenses, coordinate the 2% floor with distributable net income, determine the correct allocations among beneficiaries, ensure the proper flow-through to K-1s, and arrange for integration with the beneficiaries’ own 2% floors and the trust’s and beneficiaries’ alternative minimum tax profiles, all in pursuit of a doubtful congressional mandate and often in the context of small numbers. The calculations and allocations involved even for a discrete payment to a third-party service-provider can be quite intricate, and probably, by reason of their complexity, they themselves introduce an element of arbitrariness into the result.

This complexity is only compounded in the case of unitary fiduciary fees that must be unbundled, where first the unbundling must be done and then all of the same intricate calculations and allocations must still follow.

Corporate fiduciaries spend many thousands of hours each year on preparing tax returns that are thorough, accurate, and understandable. Even in the environment of low audit rates experienced for fiduciary income tax returns, those fiduciaries perform yeoman service on the front line of compliance and make an important contribution to the integrity of the selfassessment system. It is counterproductive to incur the risk of exasperating and demoralizing those professionals by imposing complex requirements that serve questionable ends.

Some have suggested consideration of safe harbors in the form of percentages, and Notice 2008-32 specifically asked for comments on such percentage safe harbors. In my view, percentage safe harbors will not work. Besides the fact that such safe harbors would retain much of the complexity that offends the legislative purpose and, on the thesis of this letter, would be implementing a flawed principle, percentage safe harbors might actually add to complexity. Even with safe harbors available, trustees held to the high standards of fiduciary duties might be obliged to attempt a more precise allocation anyway, so as not to harm the trust and the beneficiaries by accepting an overly conservative or otherwise inappropriate short-cut. A low safe harbor percentage (measured in terms of the amount that is exempt from the 2% floor) would not be accepted and would not achieve its purpose. A high percentage would only highlight the lack of proportion between the required compliance effort and the marginal difference it makes. Either way, additional controversy would be likely.

Moreover, safe harbors can be abused, whether intentionally or inadvertently, such as by segregating clearly “unique” costs into separately identified payments and applying a percentage safe harbor to the balance of largely non-unique costs (or whatever nomenclature is used in the final regulations). Anti-abuse rules could be prohibitively complex. For example, any effort to deny or limit the use of safe harbors when there are separated costs would be arbitrary and could penalize trustees who outsource. Again, additional controversy would be likely.

On the other hand, in the words of Notice 2008-32, “safe harbors [that] reflect the nature or value of the assets” could be written to limit the 2% floor to “in rem” expenses, and “safe harbors [that] reflect ... the number of beneficiaries” could be written to limit the 2% floor to single-beneficiary trusts that are the equivalent of outright individual ownership – views that are both embraced in this letter.

The dominant stated purpose and principal logical reason for the 2% floor is simplification. That purpose is not served, but is clearly frustrated, by imposition of the 2% floor in the context of trusts. Despite the foregoing comments about judicial deference to any reasonable interpretation, it could easily be concluded that exempting trusts from the 2% floor is the most reasonable interpretation, because it alone would meaningfully serve the objective of simplification.

To the extent a purpose for the 2% floor is fairness, exemption of trusts would not defeat that purpose in any way. I assume for purposes of this analysis that it promotes the objective of fairness to level the playing fieldamong individual itemizers by removing what had become an occasion (and perhaps sometimes even a temptation) to commingle personal and deductible expenditures in an environment of relatively small numbers where an examination is unlikely to occur and would be disproportionately burdensome to the Service if it did occur. In that light, I also assume that it promotes the objective of fairness to prevent the use of trusts to achieve benefits not available to an individual. That interest of fairness would be fully protected by exempting only trusts with multiple beneficiaries. Subjecting a trust for a single beneficiary – such as a “2503(c) trust” (or 2642(c)(2) trust) created for a minor as a substitute for an outright gift – to the 2% floor would also be complicating and burdensome in some cases, particularly those involving “unbundling,” but that might nevertheless represent a reasonable balancing of the objectives of simplification and fairness.

B.     Exemption Would Be Consistent with the Language of the Statute

Finally, the case law has shown that the statute is difficult, even though it has been seen clearly – but differently! – by various courts. In fashioning a workable “reasonable” interpretation, it will be necessary to respect the words “would not have been incurred if the property were not held in such trust or estate.” The words “would not have been incurred” are unusual in the Internal Revenue Code. The only analogs are the definition of acquisition indebtedness in the context of unrelated debt-financed income in section 514(c) and the similar restriction in section 2031(c)(4) added in 1997 in the context of the estate tax treatment of conservation easements. In the long-standing unrelated debt-financed income rules, it is clear that the words “would not have been incurred” are susceptible of a simple single-taxpayer balance sheet analysis (see the examples in Reg. § 1.514(c)-1(a)(2)), and presumably the new conservation easement estate tax rules can be applied in the same way. There is no known precedent for the behavior-predictive analysis contemplated by the Supreme Court.

Against that background, the standard of reasonableness for a regulatory interpretation seems quite broad. Surely it would be reasonable for Treasury and the Service to interpret such quixotic language with a view to its simplification objective.

It also must be acknowledged that the “two prong” approach to section 67(e)(1) has been overworked. Certainly we must start with an effort to give meaning to both the clause “are paid or incurred in connection with the administration of the estate or trust” and the clause “would not have been incurred if the property were not held in such trust or estate” – so as not to “render part of the statute entirely superfluous, something we are loathe to do.” Knight, 128 S. Ct. at 788-89, quoting Cooper Industries, Inc. v. Aviall Services, Inc., 543 U. S. 157, 166 (2004). Indeed, the Supreme Court in Knight applies that principle in both directions, because, in rejecting the Second Circuit’s “could not be incurred” approach, the Court states that “[w]e can think of no expense that could be incurred exclusively by a trust but would nevertheless notbe ‘paid or incurred in connection with’ its administration.” 128 S. Ct. at 788. And here is the nub of the matter: the Court does not explain how that dilemma is avoided merely by changing “could” to “would.”

In between these two observations, the Court cites Bogert on Trusts for the proposition that “the payment for expenses must be reasonably necessary to facilitate administration of the trust.” Id., citing G. BOGERT & G. BOGERT, LAW OF TRUSTS AND TRUSTEES §801, at 134 (2d rev. ed. 1981). Thus, it seems inevitable that the so-called first prong ofsection 67(e)(1) will always be met and therefore arguably will always be superfluous, and Treasury and the Service should feel free to finalize the regulations in a manner that reasonably deals with that inevitability.

The way to deal with the “would” standard is to do what most courts have seemed reluctant to do, but which regulations surely can do and some Justices in the Knight oral argument found intriguing – and that is to look at the context and occasion for incurring the expense. This would be a natural extension of the analysis that often supports deductibility in the first place, which, after all, is the framework in which miscellaneous itemized deductions arise and the 2% floor operates.

For example, I might ask: Can I get an income tax deduction for what I pay someone to mow my lawn? The obvious answer is no – that’s a personal expenditure. But what if I am a landlord, the lawn is associated with a residence I rent to tenants, and the lease obligates me to maintain the lawn? That surely is different. Or what if the lawn is associated with the converted residence I use for a business? That is different still. Same lawn, same mowing, different income tax results. Likewise, I might ask: Can I deduct the rent I pay for a safe deposit box? That depends on what I keep in the box. Again, same box, different income tax results.

In the context of the 2% floor, the fiduciary relationship is just as significant. While the grass grows the same at the rental residence as it does at the personal residence, fiduciary expenditure decisions are more likely to always be informed by fiduciary duties. They really are different from an individual’s expenditure decisions. And in advising a fiduciary about fiduciary relationships and duties (including investment advice), the fiduciary relationship sometimes might not matter to the advisor, but for the reasons set forth in this letter and by others, the final regulations should indulge the reasonable simplifying presumption that the pervasive fiduciary relationship always matters. 

As suggested above, an exception for “in rem” expenses that truly run with the property (such as the condo fee mentioned by taxpayer’s counsel in the Knight oral argument) should be an acceptable compromise that truly respects the words of both clauses of section 67(e)(1) – distinguishing in a logical way between expenses that solely follow “the property” in the second clause and expenses that relate to “the administration of the ... trust” in the first clause.

V.    Recommendation

Thus, in light of the Supreme Court’s treatment of the statute, in effect, as ambiguous, affirming the discretion of Treasury and the Service to address these issues in regulations, I recommend that the final regulations clarify the application of the statute in a bold, practical, palatable, and statesmanlike manner. The following considerations should inform that process:

•       As described above, the stated purposes of section 67 (alleviating a recordkeeping burden and removing the temptation to deduct personal expenses) generally do not apply to fiduciaries.

•       The “which would not have been incurred if the property were not held in such trust or estate” clause in section 67(e)(1) has been overworked as an alleged “second prong” of the statutory test. In the acknowledged absence of any authoritative articulation of congressional intent, there is no reason to view it as anything more than a completion of the overall thought of a relationship to estate or trust administration.

•       To the extent that the test of section 67(e)(1) nevertheless suggests elements of both context (“in connection with”) and motivation or occasion (“would not have been incurred”), the interests of tax administration demand the simplifying and realistic assumption that a fiduciary’s actions (including requests of an investment advisor or other service provider) are always informed by the unique standards of fiduciary duties.

•       Even if it is thought necessary to give greater independent effect to the “would not have been incurred” “second prong” of the section 67(e)(1)test, then that effect should reflect the reference in the statute to “property,” suggesting that it is the nature of the property that is critical, not the circumstances of the holder, and that therefore an appropriate carve-out would be limited to incremental “in rem” expenses that run with the property.

•       Administration of a test such as that reflected in the proposed regulationswould require disproportionate expenditure of compliance and audit resources and would inevitably lead to widely divergent results, especially in the complex task of reflecting an overall correct approach in the fiduciary’s K-1s and the beneficiaries’ individual returns – just the opposite of the simplification that was Congress’s stated purposes.

•       Unitary or “bundled” fees are welcomed by trust grantors and beneficiaries and reflect not only à la carte services but also the fiduciary’s availability, reputation, big-picture judgment, and assumption of risk. While “unbundling” fees may be a superficially appropriate way to encourage similar treatment of similar taxpayers, it might only add complexity and might in any event operate imperfectly in the marketplace of negotiated fee structures (which could include negotiated unbundling methods), and it would represent one more administrative burden in conflict with Congress’s stated purposes.

VI.    The Benefit of a Further Hearing or Conference

While Notice 2008-32 reopened the period for comment on the proposed regulations, it did not schedule a further public hearing. Because the intervening event of the Supreme Court’s Knight decision has been viewed as so fundamental, many will view a second public hearing as a good idea. I share that view. Alternatively, a less formal conference could be scheduled with those who have provided comments pursuant to Notice 2008-32. Because of the intensely practical nature of the issues and practical consequences of the way those issues are addressed, that kind of dialogue could be extremely useful, both to the personnel who are responsible for preparing the final regulations and to the fiduciary and professional communities whose acceptance is important to tax administration.

In any event, I am prepared to offer any additional input or assistance that you might find helpful.