NYSBA Tax Section Seeks More Narrow Scope in Hybrid Regs

June 8, 2020

The Honorable David J. Kautter
Assistant Secretary (Tax Policy)
Department of the Treasury
1500 Pennsylvania Avenue, NW
Washington, DC 20220

The Honorable Charles P. Rettig
Commissioner
Internal Revenue Service
1111 Constitution Avenue, NW
Washington, DC 20224

The Honorable Michael J. Desmond
Chief Counsel
Internal Revenue Service
1111 Constitution Avenue, NW
Washington, DC 20224

Re: Report No. 1438 — Report on Proposed Anti-Conduit Regulation Treating Certain “Hybrid Equity” Transactions as Financings under Treasury Regulation Section 1.881-3

Dear Messrs. Kautter, Rettig, and Desmond:

I am pleased to submit our Report No. 1438 commenting on the proposed regulation issued in April of this year addressing the treatment of certain hybrid equity transactions as financings for purposes of Treasury Regulation § 1.881-3.

Our report addresses issues regarding the scope of transactions treated as financings under the proposed regulation, as well as issues relating to the amount of tax imposed where such transactions are treated as conduit financings and the duty of withholding agents under the “reason to know” standard.

We appreciate your consideration of our Report. If you have any questions or comments, please feel free to contact us and we will be glad to assist in any way.

Respectfully submitted,

Andrew H. Braiterman
Chair
Tax Section
New York State Bar Association

Enclosure

Cc:
Lafayette “Chip” G. Harter III
Deputy Assistant Secretary — International Tax Affairs
Department of the Treasury

Douglas Poms
International Tax Counsel
Department of the Treasury

Brett York
Deputy Tax Legislative Counsel
Department of the Treasury

Brenda Zent
Special Adviser
Department of the Treasury

Erika W. Nijenhuis
Senior Counsel
Department of the Treasury

Peter Blessing
Associate Chief Counsel (International)
Internal Revenue Service

Anne Devereaux
Deputy Associate Chief Counsel (International)
Internal Revenue Service

Daniel McCall
Deputy Associate Chief Counsel (International — Technical)
Internal Revenue Service

Margaret O'Connor
Deputy Associate Chief Counsel (International)
Internal Revenue Service

John Merrick
Senior Level Counsel
Internal Revenue Service

---

Report No. 1438

New York State Bar Association Tax Section

REPORT ON THE PROPOSED ANTI-CONDUIT REGULATION TREATING CERTAIN “HYBRID EQUITY” TRANSACTIONS AS FINANCINGS UNDER TREASURY REGULATION SECTION 1.881-3

June 8, 2020

---

TABLE OF CONTENTS

INTRODUCTION

I. SUMMARY OF RECOMMENDATIONS2

II. SUMMARY OF THE ANTI-CONDUIT REGULATIONS

III. DISCUSSION

IV. Specific Recommendations

---

INTRODUCTION

This report¹ (the “Report”) provides comments on proposed regulations issued on April 8, 2020 (the “Proposed Regulation”),² which would amend Treasury Regulation section 1.881-3 (the “Anti-Conduit Regulations” and, as currently in effect, the “Existing Anti-Conduit Regulations”). Under the Anti-Conduit Regulations, for purposes of determining the amount of U.S. gross basis income and withholding taxes imposed, the Internal Revenue Service (the “IRS”) may recharacterize a chain of legally separate financing transactions as a direct financing between the ultimate provider and the ultimate recipient of the financing. For the IRS to exercise this authority: (1) there must be two or more “financing” transactions linked by one or more “intermediate entities” (i.e., a “financing arrangement”), (2) the participation of the intermediate entity (or entities) must have the effect of reducing tax (measured against the U.S. withholding tax that would apply to a direct transaction between the ultimate provider and ultimate recipient), and (3) the participation of the intermediate entity (or entities) must be pursuant to a tax avoidance plan (various non-exclusive indicia of which are provided in the Existing Anti-Conduit Regulations).

As discussed below, the current definition of a “financing transaction” includes not only debt but also certain non-debt instruments (in particular, certain equity that is redeemable by its terms on or before a date certain). Certain non-redeemable equity that would not meet the definition of a “financing transaction” under the Existing Anti-Conduit Regulations could produce non-U.S. tax consequences that arguably could be viewed as similar to the non-U.S. tax consequences of instruments that do meet that definition.

The preamble to the Proposed Regulation (the “Preamble”) notes, for example, that a financing company may have incentives to form an intermediate entity in a jurisdiction that provides a “notional interest deduction on equity” (whereby an assumed implied interest rate amount is deductible and offsets taxable income from another leg of the financing arrangement). Alternatively, a country may have a purported imputation regime (which unlike a classical tax system seeks to ensure a single level of corporate and shareholder tax) that provides a shareholder or other related person a refund of corporate tax paid (which is somewhat equivalent to allowing a deduction against corporate tax on equity).

The Proposed Regulation would expand the definition of instruments treated as financing transactions to include equity in two additional situations; generally, (1) if a deduction or similar benefit with respect to the equity is provided by the issuer's country of residence or another country in which the equity issuer has a permanent establishment (referred to in the Report as “Category 1,” and such permanent establishment a “PE”) or (2) if the country provides any related person (such as a shareholder or interest holder) a refund (including a credit) of taxes paid to that country by the equity issuer (referred to in the Report as “Category 2”).

The Preamble notes that failing to expand the rules to reach hybrid equity would be inconsistent with the ongoing efforts by the Treasury Department and IRS to address financings that use hybrid instruments. The Preamble also notes that Treasury and the IRS decided not to limit the new rules to classic “hybrid” instruments (treated as equity for U.S. tax purposes but as debt in the other country) because this would not reach instruments treated as equity in the other country but which obtain a tax benefit that is equivalent to an interest deduction.

This Report provides comments on the Proposed Regulation. Part I of the Report summarizes our recommendations. Part II describes in more detail the existing Anti-Conduit Regulations and the provisions in the Proposed Regulation that treat hybrid equity as a financing transaction in the circumstances mentioned above. Part III of the Report discusses general policy considerations relevant to the Proposed Regulation and the relationship of the Anti-Conduit Regulations and Proposed Regulation to certain other tax law provisions such as section 267A. Finally, Part IV of the Report provides our specific, detailed recommendations.

I. SUMMARY OF RECOMMENDATIONS

1. Narrow the Scope of the Instruments to Which the Proposed Regulation Applies

We understand the concern of Treasury and the IRS that base eroding equity returns can achieve results similar to deductible debt financing from a conduit financing perspective. Nevertheless, the distinction between treaty partner tax rules that permit base “erosion” and treaty partner laws that define the corporate base and an appropriate rate of taxation is a matter of degree. At one end of the spectrum of instruments that might be treated as financings on that basis are “classic” hybrid instruments with a limited and preferred return that the intermediate country considers debt for its own purposes but that are equity for U.S. tax purposes. We agree these should be treated as financings as they can be considered to achieve equivalent commercial and tax results to conventional debt instruments, do not raise unmanageable administrability concerns and do not raise significant treaty comity concerns (as a treaty partner can hardly be surprised if the U.S. tax law consistently treats the instrument as a financing for that purpose). At the other end of the spectrum are equity instruments that are subject to notional interest deduction or similar “cost of capital allowance” regimes of general applicability (enacted, for example, to reduce artificial tax incentives to use debt rather than equity capital). We do not think it is appropriate potentially to treat an operating company that issues such equity as engaged in a conduit financing arrangement on the basis that this makes the equity a “financing.” We recognize that addressing cases between those extremes (e.g., instruments treated as debt for foreign tax purposes that, similar to common equity, have the right to residual earnings of the intermediate entity) raises difficult line drawing issues. However, for reasons discussed in the Report, we recommend that the scope of the Proposed Regulations be limited (other than in the case of classic hybrid transactions) to equity that achieves the types of “notional” base eroding benefits described in the Proposed Regulations when such equity is issued by a special purpose company formed only to carry out the related financing transactions. We also think foreign companies subject to regimes equivalent to U.S. REITs and RICs, whose entitlement to treaty benefits under “limitation on benefits” rules has been expressly negotiated with treaty partners, should not be treated as conduits on this basis.

2. Clarify the Definition of Category 2 Equity

We are aware of countries that provide, or in the past have provided refunds or credits of corporate tax (such as an “imputation” credit) to related persons. However, we are not familiar with a regime which provides such a refund tied to equity distributions or deemed distributions wholly unrelated to the liability the beneficiary of that refund would otherwise owe to that country. Nevertheless, to the extent such a regime does exist or might evolve in future and this concerns Treasury and the IRS, we understand how this could be viewed as equivalent to a notional interest deduction. However, we think the definition in the Proposed Regulation is vague and will make it hard for taxpayers, withholding agents and IRS agents to assess what regimes are intended to be covered or excluded. We urge Treasury and the IRS to provide a greater number of more detailed examples that illustrate not only cases intended to be caught but also cases that are not intended to be caught, as discussed in more detail in the Report.

3. Clarify Withholding Regulations and “Reason to Know” Standard

Withholding agents who are not related to the parties to a financing arrangement face significant challenges in applying the Existing Anti-Conduit Regulations because they have limited visibility into the tax planning being conducted by the financing parties and may not even be aware of the underlying sources of capital for the financing legs they see. The current rules provide some protection by excusing them from withholding based on conduit recharacterization absent actual knowledge or a “reason to know” there is a conduit arrangement. This presents complicated questions as to what constitutes knowledge and what diligence is required but most of the factors in the Existing Anti-Conduit Regulations are factual and objective and the withholding rules make clear that knowledge as to these factual indicia does not, without more, constitute reason to know the parties have a withholding tax avoidance purpose. The Proposed Regulation will require knowledge of, or inquiries into, foreign tax laws to even determine if there are related “financings.” At the same time, this is not “subjective” but in a sense relatively objective quasi-factual information which could be diligenced. At a minimum, if the Proposed Regulations are adopted, we think a number of additional examples illustrating what constitutes knowledge or a reason to know of a foreign law regime should be provided. Treasury and the IRS may also wish to consider whether an “actual knowledge” standard would be more appropriate for a withholding agent who is not the US financed party, related to the US financed party or the legal agent of the U.S. financed party (and could protect itself with contractual indemnities).

4. Clarify the Computation and Character of the Amount Subject to Withholding Under the Proposed Regulation

Under the Existing Anti-Conduit Regulations, the computation of the amount of the financed entity's payment that is recharacterized depends upon the relative “principal amounts” of the financing legs. Under the Proposed Regulation, a comparable concept is needed for Category 1 and Category 2 transactions. The Proposed Regulation lacks crucial guidance as how that “principal amount” would be determined. This will require defining what portion of an entity's equity capital should be considered functionally “related” to the U.S. financing leg. We have no made specific recommendations as to how this should be done but it will be essential if our primary recommendation to narrow the scope of the Proposed Regulation is not adopted. We also think more guidance is needed on how the “character” rule should apply upon recharacterization. Under the Existing Anti-Conduit Regulations, if a financing arrangement is recharacterized, generally the resulting direct instrument is characterized based on the financing party-intermediate entity transaction leg unless that leg would give rise to non-deductible payments. If that approach is retained it will be necessary to define how that rule applies to Category 1 and Category 2 transactions. Do U.S. tax law principles govern the question of deductibility for this purpose (notwithstanding the treatment under foreign tax law)? It may be preferable simply to provide that the character of the U.S. transaction leg governs the recharacterized instrument in those circumstances.

5. Revise the Proposed Effective Date

If our primary recommendation to narrow the scope of the Proposed Regulation is adopted, the new conduit regime will not dramatically go beyond what taxpayers might reasonably have anticipated from the prior indications in 2008 that hybrid instruments were under review. If not so narrowed, however, the new regime will represent a dramatic change in how the Anti-Conduit Regulations operate and what they can reach. We would then urge consideration of revising the effective date rule to provide grandfathering (permanently or for a reasonable number of years) of arrangements already in place at the time the Proposed Regulation was issued (and not substantially modified thereafter).

6. Carve Out Partnership Equity Unless Partnership is Hybrid Claiming Treaty Benefits

When an entity the U.S. considers a partnership is treated as a non-transparent resident of another country for its own corporate tax purposes it may be able to claim treaty benefits on that basis. We understand why partnership equity issued by such a treaty hybrid entity raises conduit and base erosion issues if it provides the equivalent of a deduction with respect to the equity distributions. However, we think partnership equity should only be considered a financing in that limited case. In the case of a non-hybrid partnership, the U.S. withholding regime using well-developed principles already looks through a partnership to the partners. Potential inconsistency between how income is allocated for purposes of those withholding rules versus under a conduit “recharacterization” will create confusion and uncertainty while furthering no apparent policy goal. We therefore recommend that you clarify that partnership equity will be treated as a financing only when the relevant partnership is a treaty hybrid entity.

7. Eliminate or Significantly Clarify the PE Portion of Category 1

We recommend that the PE portion of Category 1 be eliminated. While its scope and application are unclear, we have considered the various ways in which it might be intended to apply and have concluded that, if a foreign entity has a PE in a third country, the tax results in the PE's jurisdiction are not an appropriate basis for treating the shareholders of the entity as being engaged in a financing transaction with the entity. If it is retained, we suggest significant clarifications be provided, including as to how the “amount” of the financing transaction and “the rate used to compute the issuer's notional interest deduction” are determined. Finally, if our primary recommendation is adopted, but the PE rule is retained, the PE rule at least should be clarified to apply only when the PE is the equivalent of a special purpose finance entity whose only activities are those relating to the financing arrangement.

II. SUMMARY OF THE ANTI-CONDUIT REGULATIONS

1. The Existing Anti-Conduit Regulations

The Existing Anti-Conduit Regulations were initially finalized in 1995.³ They were amended in 2011 to address disregarded entities.⁴

The Existing Anti-Conduit Regulations apply for purposes of determining the amount of tax due under Section 871(a) or Section 881, including which non-U.S. person is liable for that tax, and the withholding obligations and liabilities of the U.S. payor (or another withholding agent). The Existing Anti-Conduit Regulations empower the IRS to deny the use of an income tax treaty or the portfolio interest exemption to reduce the 30 percent section 871(a) and section 881 gross basis tax rate whenever the conditions set out in the Existing Anti-Conduit Regulations are met, except to the extent that the ultimate financing party would be entitled to a reduction.⁵

The Existing Anti-Conduit Regulations apply if there is a “conduit financing arrangement” (as defined in the Existing Anti-Conduit Regulations) and the IRS director of field operations determines to apply them to disregard the participation of a conduit entity so as to treat the U.S. payor as having made its payment to a different entity involved in the conduit financing arrangement.⁶

There is a “conduit financing arrangement” if there is a “financing arrangement” and there is an “intermediary” that is a “conduit entity” (each, as defined in the Existing Anti-Conduit Regulations and explained below).

A “financing arrangement” exists if there is

“a series of transactions by which one person (the “financing entity”) advances money or other property, or grants rights to use property, and another person (the “financed entity”) receives money or other property, or rights to use property, if the advance and receipt are effected through one or more other persons (“intermediate entities”) and . . . there are financing transactions linking the financing entity, each of the intermediate entities, and the financed entity.”⁷

Thus, there must be a financing transaction at each link in the chain, with one exception under which “related” entities can be combined and treated as a single entity.⁸

A “financing transaction” is an advance of money or other property, or a grant of rights to use property, that is in the form of:

(1) “debt”;

(2) “any lease or license”; or

(3) any other transaction where the transferee who receives the advance of money or other property “is obligated to repay or return a substantial portion of the money or other property advanced, or the equivalent in value”.⁹

Under a special rule, “stock in a corporation (or a similar interest in a partnership, trust, or other person)” may be a financing transaction but only if:

(1) “The issuer is required to redeem the stock or similar interest at a specified time or the holder has the right to require the issuer to redeem the stock or similar interest or to make any other payment with respect to the stock or similar interest;”

(2) “The issuer has the right to redeem the stock or similar interest, but only if, based on all of the facts and circumstances as of the issue date, redemption pursuant to that right is more likely than not to occur;” or

(3) “The owner of the stock or similar interest has the right to require a person related to the issuer (or any other person who is acting pursuant to a plan or arrangement with the issuer) to acquire the stock or similar interest or make a payment with respect to the stock or similar interest.”¹⁰

The Existing Anti-Conduit Regulations provide that a person will not be considered to have a right to cause a redemption or payment solely because the person owns a controlling interest in the issuer.¹¹ The Report refers to equity that meets the requirements of this special rule as “Redeemable Equity”.

Thus, under the Existing Anti-Conduit Regulations, a financing transaction exists only where the money or property advanced is required to be returned. Even in the case of equity, this requirement must be met. The only exception is the collapsing of related persons into one person, whereby the requirement for a financing transaction between them is dispensed with if the IRS determines that one of the principal purposes for the structure of the financing transactions is to prevent the characterization of such arrangement as a financing arrangement. Notably, this collapsing does not dispense with the need for at least two financing transactions.

An intermediary is a “conduit entity” (meaning, its participation can potentially be ignored) if:

(1) its participation reduces the section 881 tax,

(2) its participation is “pursuant to a tax avoidance plan”; and

(3) either —

(a) the intermediate entity is related to the financing entity or the financed entity; or

(b) the intermediate entity would not have participated in the financing arrangement on substantially the same terms but for the fact that the financing entity engaged in the financing transaction with the intermediate entity.¹²

The determination of whether the participation of the intermediate entity in the financing arrangement is “pursuant to a tax avoidance plan” is made under the rules set out in Treas. Reg. section 1.881-3(b). Those rules provide that

“A tax avoidance plan is a plan one of the principal purposes of which is the avoidance of tax imposed by section 881. Avoidance of the tax imposed by section 881 may be one of the principal purposes for such a plan even though it is outweighed by other purposes (taken together or separately). In this regard, the only relevant purposes are those pertaining to the participation of the intermediate entity in the financing arrangement and not those pertaining to the existence of a financing arrangement as a whole. The plan may be formal or informal, written or oral, and may involve any one or more of the parties to the financing arrangement.”¹³

The Existing Anti-Conduit Regulations then list three factors that are among those to be considered in making this determination:

(1) whether the participation of the intermediate entity (or entities) in the financing arrangement significantly reduces the tax that otherwise would have been imposed under section 881;

(2) whether the intermediate entity had sufficient available money or other property of its own to have made the advance to the financed entity without the advance of money or other property to it by the financing entity;¹⁴ and

(3) the length of time between the advances of money or other property, or the grants of rights to use property, by the financing entity to the intermediate entity and the advances by the intermediate entity to the financed entity. A short period of time is evidence of the existence of a tax avoidance plan while a long period of time is evidence that there is not a tax avoidance plan.¹⁵

Withholding tax avoidance need not be a dominant purpose if it is one of the principal purposes. If the financing arrangement involves a financed entity that is unrelated to any of the other entities involved, the Existing Anti-Conduit Regulations can apply even if that financed entity is not aware of or involved at all in the tax avoidance plan.

Setting aside a situation where all the parties are unrelated and instead considering a situation where the financing entity and the intermediary are related but neither is related to the financed entity, the Existing Anti-Conduit Regulations raise serious concern whenever there is an intermediary whose involvement results in a lower tax. Specifically, the intermediary that is related to the financing entity can meet the “conduit” definition if (1) the section 881 tax would be higher if the financing entity had financed the financing entity directly and (2) one of the principal purposes (of the financing entity and/or the intermediary entity) for involving the intermediate entity was to obtain that reduction in tax (which will be determined by considering that there is that reduction, whether the intermediary could have financed the transaction without the advance from the financing entity, and the period of time separating the advances). The purposes and awareness of the financed entity are, essentially, irrelevant.

Withholding Under the Existing Anti-Conduit Regulations

If there is a conduit financing arrangement, there are two consequences: first, the IRS may (but is not required to) determine that one or more intermediaries' participation should be ignored such that the transactions should be treated as if the U.S. payor paid the financing entity directly (as determined by the IRS); and second, the U.S. payor may be obligated to withhold as if the IRS had made that determination (and be individually liable for the tax required to be so withheld if it fails to withhold). This withholding obligation will, if it applies, be imposed at the time the payments are made. A determination by the IRS imposing tax on the financing entity, in all likelihood, would occur long after the payments are made.

Under the withholding tax regulations related to the Existing Anti-Conduit Regulations (the “Withholding Regulations”),¹⁶ a financed entity that is unrelated to the other parties and was not part of the tax avoidance planning does have some protection against liability for not having withheld.

If the arrangement meets the definition of a “conduit financing arrangement”, the financed entity is obligated to withhold as if the IRS had determined, pursuant to Treasury Regulation section 1.881-3(a)(3), that all conduit entities should be disregarded.¹⁷ The financed entity must determine the amount of the tax due applying the rules in Treasury Regulation section 1.881-3(d), which are discussed below.¹⁸ The financed entity is liable for any

amount that should have been but was not withheld if the financed entity “knows or has reason to know that the financing arrangement is a conduit financing arrangement.”¹⁹ The Existing Anti-Conduit Regulations explain that:

“This standard shall be satisfied if the withholding agent knows or has reason to know of facts sufficient to establish that the financing arrangement is a conduit financing arrangement, including facts sufficient to establish that the participation of the intermediate entity in the financing arrangement is pursuant to a tax avoidance plan. A withholding agent that knows only of the financing transactions that comprise the financing arrangement will not be considered to know or have reason to know of facts sufficient to establish that the financing arrangement is a conduit financing arrangement.”²⁰

The Proposed Regulations do not propose any changes to the Withholding Regulations. This is presumably based upon the view that what is required for the withholding agent to “have reason to” know should be the same regardless of the type of “financing transaction” linking the financing party and the intermediary (or intermediaries) — in other words, that the rules should be the same whether those parties are linked by plain-vanilla debt or linked by equity that provides for the types of tax benefit addressed by the Proposed Regulations.

Because the Proposed Regulations are motivated by some of the same hybridity concerns motivating section 267A, it is useful to contrast the withholding test applicable to the U.S. payor under the Existing Anti-Conduit Regulations (and which would apply under the Proposed Regulation) when the U.S. payor is unrelated to the other parties with the rules that would cause such a U.S. payor to be subject to the section 267A deduction disallowance.

Under the section 267A rules, a U.S. payor which makes a deductible payment to an unrelated party is subject to section 267A disallowance only if (i) the arrangement meets the definition of a “structured arrangement” and (ii) the U.S. payor meets the definition of a “participant”. Meeting these requirements means:

(1) “based on all the facts and circumstances (including the terms of the arrangement), the arrangement is designed to produce the hybrid mismatch”,²¹

(2) the U.S. payor “could, based on all the facts and circumstances, reasonably be expected to be aware of the hybrid mismatch”, and

(3) the U.S. payor “shares in the value of the tax benefit resulting from the hybrid mismatch.”²²

Notable is the third requirement. The first two requirements are rather similar to the requirements for a financing arrangement to be a “conduit financing arrangement” and for the U.S. payor to have liability for withholding tax, but the third requirement has no corollary in the Treasury Regulation section 1.881-3 context — in other words, the unrelated party needs to derive a financial benefit from the hybridity in order for section 267A to apply to that unrelated party, whereas Treasury Regulation section 1.1441-3 withholding can apply to an unrelated party which has no financial benefit from the use of the “conduit”.

2. Proposed Regulation

The Proposed Regulation would add two new types of “financing transactions”, both consisting of types of equity instrument. The Proposed Regulation would do this by adding two additional categories to the special rule under which specified types of “stock in a corporation (or a similar interest in a partnership, trust, or other person)” are a financing transaction. The two new categories would be:

“Related” would be defined as it is in the Existing Anti-Conduit by reference to the definitions in sections 267(b) and 707(b)(1), and also include parties that are controlled within the meaning of section 482, and applying the constructive ownership rules of both sections 318 and 267(c).²⁵

The Preamble's description of the proposed two new categories is slightly different and thus worthy of consideration.

In the case of Category 1, the Preamble's description provides additional commentary on the PE aspect of that Category:

“The proposed regulations expand the types of equity interests treated as a financing transaction to include stock or a similar interest if under the tax laws of a foreign country where the issuer is a resident, the issuer is allowed a deduction or another tax benefit for an amount paid, accrued or distributed with respect to the stock or similar interest. Similarly, if the issuer maintains a taxable presence, referred to as a permanent establishment (“PE”) under the laws of many foreign countries without regard to a treaty, and such country allows a deduction (including a notional deduction) for an amount paid, accrued or distributed with respect to the deemed equity or capital of the PE, the amount of the deemed equity or capital will be treated as a financing transaction. See Proposed Regulation §1.881-3(a)(2)(ii)(B)(1)(iv).”²⁶

The text of the Proposed Regulation would appear to require that the deduction or other benefit be allowed with respect to the stock issued by the corporation (or similar interest issued by a partnership or trust or other person), whereas the Preamble by contrast appears to take into account any deduction (including a notional deduction) allowed to a PE “with respect to the deemed equity or capital of the PE”. Indeed, the Preamble also provides that in the case of PE, the amount of the financing transaction would be “the amount of the deemed equity or capital”. This raises a number of questions about how this PE aspect of Category 1 would work, which are discussed later in the Report.

In the case of Category 2, the Preamble's description would appear to be consistent with the text of the Proposed Regulation but perhaps a clearer articulation of what is intended. The Preamble provides:

“The proposed regulations also treat stock or a similar interest as a financing transaction if a person related to the issuer, generally a shareholder or other interest holder in an entity, is entitled to a refund (including a credit) or similar tax benefit for taxes paid by the issuer to its country of residence, without regard to the person's tax liability with respect to the payment, accrual or distribution under the laws of the issuer. See Prop. Reg. §1.881-3(a)(2)(ii)(B)(1)(v).”

Operation of the Collapsing Rule and Comparison to the Operation of the Special Rule For Certain Equity

The difference between the proposed Category 1 and Category 2, on the one hand, and the rule in the Existing Anti-Conduit Regulations for collapsing related parties can be illustrated by an example.

These constitute two financing transactions linking USC to FP2, and FP1's participation can be ignored if FP1 is a “conduit” (which requires that the section 881 tax is lower if the payment USC makes is respected as being made to FP1 than that tax would be if the payment by USC were treated as made to FP2, the involvement of FP1 has as one of its principal purposes the avoidance of tax (under the standards set out in Treasury Regulation section 1.881-3(b)(2)) and FP1 and FP2 are related (which they are in this example).

The collapse-of-related persons rule could not result in a “conduit financing arrangement” because it would result in only a single financing transaction — i.e., one from FP2 to USC.

If, however, there were also a financing transaction between FP2 and another entity (FP3) then FP1 could potentially be disregarded either as a “conduit intermediary” or by virtue of being collapsed with FP2. This outcome would be possible under the Existing Anti-Conduit Regulations.

The rules for collapsing related entities, if applied as written, could also reach results that are likely unintended from a policy standpoint.

If FP1 and FP2 are collapsed on the basis that they are related, there could be a conduit financing because FP2-FP1 has issued hybrid equity and has also entered into a U.S. FDAP generating financing. If the reason for the Proposed Regulation is to prevent base erosion, in the absence of financing linking the related parties, FP1 presumably is subject to tax in its country and does not benefit from the base erosion.²⁷

Relationship to Sections 267A and 245A(e)

While the possibility of expanding the special rule for equity to so-called “hybrid equity” was first introduced in 2008, its re-emergence at this time is motivated by the more recent focus on the base erosion potential of hybrids as seen in the OECD's BEPS project²⁸ and as embraced by the United States in the 2017 enactment of sections 267A and 245A(e) and the following multi-year project to promulgate implementing regulations, which culminated in two sets of final regulations issued on the same day as the Proposed Regulation.

The Preamble addresses the relationship between those rules and the Proposed Regulation as follows:

“The Treasury Department and the IRS have determined that these types of instruments can be used to inappropriately avoid the application of the conduit financing regulations and, therefore, the proposed regulations expand the definition of equity interests treated as a financing transaction by taking into account the tax treatment of the instrument under the tax law of the relevant foreign country, which is generally the country where the equity issuer resides. The Treasury Department and the IRS have determined that, while these types of instruments are characterized as equity for U.S. tax purposes, they still raise conduit financing concerns if they are either indebtedness under the issuer's tax law or provide benefits similar to indebtedness under the issuer's tax law.

For example, a financing company may have an incentive to form a corporation in a country that allows a tax benefit, such as a notional interest deduction with respect to equity, that encourages the routing of income through the intermediary issuer in functionally the same manner as when an intermediate entity issues a debt instrument that is treated as a financing transaction under the current regulations. Similarly, a financing entity may form an intermediate corporation in a country to take advantage of the country's purported integration regime that provides a substantial refund of the issuer's corporate tax paid upon a distribution to a related shareholder, and the shareholder is not taxable on that distribution under the laws of the intermediate country. The Treasury Department and IRS have concluded that these structures raise concerns similar to those Congress intended to address when it enacted sections 267A and 245A(e) regarding arrangements that 'exploit differences in the treatment of a transaction or entity under the laws of two or more tax jurisdictions . . .' See S. Comm. on the Budget, Reconciliation Recommendations Pursuant to H. Con. Res. 71, S. Print No. 115-20, at 389 (2017).

This change essentially aligns the conduit regulations with the policy of section 267A by discouraging the exploitation of differences in treatment of financial instruments across jurisdictions. While section 267A and the final regulations apply only if the D/NI outcome is a result of the use of a hybrid entity or instrument, the conduit financing regulations apply regardless of causation and instead look to whether there is a tax avoidance plan. Thus, this new rule will address economically similar transactions that section 267A and the section 267A final regulations do not cover.”²⁹

These two new categories overlap with the type of foreign tax law tax benefits that can form the basis for an “imported hybrid mismatch” under Treasury Regulation section 1.267A-4, but importantly, these two categories are broader.

The imported mismatch rules treat a foreign law deduction as the basis for denial of US deduction for interest or royalties if the foreign law deduction is for interest or royalties or is a deduction allowed to an issuer of equity with respect to that equity. Under this rule, the types of deductions on equity that are subject to the rule are similar to Category 1 above but are more limited in two ways: first, the only type of tax benefit that qualifies is a deduction; and second, the owner of the equity would be required to include the amount in its taxable income under its own tax law if the amount had actually been treated as interest under such law. The section 267A imported mismatch rules have no corollary to proposed Category 2 tax benefits. We have separately addressed below the different approaches the two sets of rules take when the U.S. payor is not “related” to the other entities involved.

Computation of Tax Due Once Intermediary Is Disregarded

Under the Existing Anti-Conduit Regulations, if the participation of an intermediary is disregarded, the determination of the amount of tax due under section 881 (and sections 871, 884(f)(1)(A), 1441 and 1442) and who is liable for that tax is determined as follows.

First, all or a portion of the amount paid by the financed entity is treated as paid directly to the financing entity. Whether it is the entire amount or a portion thereof is determined based upon the principal amount of all the financing transactions in the chain. If the principal amount of the financing transactions to which the financed entity is a party is equal to or less than the principal amount of the other financing transactions linking any of the parties in the chain, then the entire amount paid by the financed entity is recharacterized. If the principal amount of the financed entity's transaction is greater than any of the other links in the chain, than the portion of the payments made by the financed entity that are treated as if paid to the financing entity is determined by comparing the principal amounts as follows: the payment by the financed entity is multiplied by a fraction the numerator of which is equal to the lowest aggregate principal amount of the financing transaction(s) linking any of the parties to the financing arrangement (other than financing transactions that are disregarded under the rules for collapsing related parties³⁰) and the denominator of which is the aggregate principal amount of the financing transaction(s) to which the financed entity is a party.³¹ We refer to this rule as the “Scale-Back Rule”.

The Existing Anti-Conduit Regulations provide details about how to compute principal amount for this purpose. For debt, a lease or a license, the principal amount equals “the amount of money or the fair market value of other property advanced or subject to the lease or license”. In the case of equity, two rules are provided. If the stock is subject to the current inclusion rules of sections 305(c)(3) or (e), “the principal amount generally will be equal to the issue price.”³² If the equity is a partnership or trust interest, “the principal amount is equal to the fair market value of the money or property contributed to the partnership or trust in return for that partnership or trust interest.”³³ These two rules were written to apply to the Redeemable Equity that is a financing transactions under the Existing Anti-Conduit Rules. Almost all Redeemable Equity would fall into one of these two rules and it is likely that this was the understanding and intent of the drafters of the Existing Anti-Conduit Rules.

Thus, the Existing Anti-Conduit Regulations are designed around a core principle and core focus which is on the amount of money or value of property being advanced from the financing entity indirectly into the financed entity and which are obligated to be returned through the chain from the financed entity to the financing entity. The existence of a deduction in the U.S. or elsewhere and the amount of any deduction in the U.S. or elsewhere is not relevant. This is the case both in determining if the link between two entities is a financing transaction and in determining the amount of the financed entity's payment that is recharacterized as being paid to the financing entity. This is particularly demonstrated by the use of the principal amount of each financing transaction in determining the amount of the financed entity's payment that is recharacterized and the complete irrelevance of the interest rate applicable to each financing transaction or the amount otherwise paid other on each financing transaction.

The Proposed Regulation takes an entirely different approach. First, it defines a financing transaction by reference to the existence of a tax deduction or another type of tax benefit, such a credit or refund of tax paid, with no need for there to be any obligation to return any amount. Second, it also provides a special rule that uses the amount of the intermediary's foreign tax deduction to determine the amount of the financed entity's payment that is recharacterized. This special rule applies, however, only in the case of those Category 1 transactions which entitle the intermediary to a notional interest deduction.

Specifically, Proposed Regulation section 1.881-3(d)(1)(iii) provides that, in the case of a Category 1 transaction providing a notional interest deduction to the intermediary:

“the portion of the payment made by the financed entity that is recharacterized under paragraph (d)(1)(i) of this section attributable to such financing transaction [the Category 1 transaction] will not exceed the financing transaction's principal amount as determined under paragraph (d)(1)(ii) of this section multiplied by the rate used to compute the issuer's notional interest deduction for the taxable year in which the payment is made.”³⁴

The Report refers to this special rule for Category 1 notional interest deduction transaction, setting a limit on the amount of the financed entity's payment that would be recharacterized, as the “NID Limitation Rule”. The Proposed Regulation provides a complex, and helpful, example that illustrates the application of both the Scale-Back Rule and the NID Limitation Rule to a single fact pattern.

In the case of all other Category 1 transactions and all Category 2 transactions, the only limitation on the amount of the financed entity's payment that is recharacterized is the existing Scale-Back Rule, which depends upon the principal amount of all the financing transactions in the chain. Accordingly, applying the Proposed Regulation to compute the amount of the financed entity's payment that is recharacterized requires knowing the principal amount of Category 1 and Category 2 transactions. As discussed below, there are several aspects of these rules that are unclear and potentially would not operate as appears to have been intended.

The Existing Anti-Conduit Regulations also provide that the portion of the payment by the financed entity that is recharacterized as having been paid directly to the financing entity is also recharacterized (for purposed of determining the tax due under sections 871, 881, 884(f)(1)(A), 1441 and 1442) as being the same type of payment as is actually made to the financing entity (unless the payment made to the financing entity would not be deductible if paid by the financed entity, in which case the type of payment by the financed entity will not be recharacterized).³⁵

3. The 2008 Proposed Regulations' Preamble and the 2009 Prior Report

As discussed in the Preamble to the Proposed Regulation, the preamble to proposed regulations issued in 2008 (the “2008 Preamble”)³⁶ stated that Treasury and the IRS were considering whether Treasury Regulation section 1.881-3 should be expanded to apply to financing transactions involving “hybrid transactions” and requested comments. In response, we submitted a report (the “Prior Report”)³⁷ Final Regulations issued in 2011 indicated that the Prior Report was the only comment received. The 2008 Preamble's discussion is short and worth quoting in full:

“Treasury Department and IRS are continuing to study conduit financing arrangements and may issue separate guidance to address the treatment under § 1.881-3 of certain hybrid instruments. Specifically, the Treasury Department and IRS are studying transactions where a financing entity advances cash or other property to an intermediate entity in exchange for a hybrid instrument that is treated as debt under the laws of the foreign jurisdiction where the intermediate entity is resident and is not treated as debt for U.S. federal tax purposes. The issue under consideration is whether such instruments should constitute a financing transaction under § 1.881-3(a)(2)(ii)(A) and part of a financing arrangement within the meaning of § 1.881-3(a)(2)(i)(A). No inference should be drawn from the approaches described in this preamble regarding the treatment of such instruments under current law, including judicial doctrines with respect to conduit financing transactions. One possible approach is to treat all transactions involving such hybrid instruments between a financing entity and an intermediate entity as financing transactions under § 1.881-3(a)(2)(ii)(A). Comments are requested on this approach, including whether and to what extent a connection or relationship between the issuer and recipient of the hybrid instrument (for example, an equity ownership percentage) should be required in order to treat such instruments as financing transactions. Another possible approach is to add additional factors to consider in determining when stock in a corporation (or other similar interest in a partnership or trust) may constitute a financing transaction under § 1.881-3(a)(2)(ii)(B). The additional factors would focus on whether, based on the facts and circumstances surrounding the stock (or other similar interest in a partnership or trust), the financing entity had sufficient legal rights to, or other practical assurances regarding, the payment received by the intermediate entity to treat the stock as a financing transaction. Some possible factors to indicate the presence of a financing transaction might include:

(1) Intent of the parties to pay all or substantially all payments received by the intermediate entity to the financing entity;

(2) History of payment of amounts received by the intermediate entity to the financing entity; and

(3) Precedence of the obligees over other creditors regarding the payment of interest and principal, currently or in bankruptcy.

Comments are requested concerning other possible approaches and any additional factors that the Treasury Department and IRS should consider in expanding the conduit financing regulations under § 1.881-3.”³⁸

The differences between the two approaches suggested in the 2008 Preamble and the approach taken in the Proposed Regulation are noteworthy. Since 2008, there have been many developments and those developments include the OECD BEPS anti-hybrid initiative, the enactment of sections 267A and 245A(e), and the issuance of the final regulations under sections 267A and 245A(e).

III. DISCUSSION

1. General Policy Considerations

The Anti-Conduit Regulations are motivated by and designed to balance various and somewhat conflicting policy considerations. A primary goal is to impede treaty shopping and the use of intermediate entities primarily to obtain withholding tax benefits to which the parties would not otherwise be entitled. However, this is tempered by the desire to have rules that are administrable, clear and treat similarly situated taxpayers consistently. U.S. conduit rules must also contend with existing treaty obligations of the U.S. and the policy issues raised if the rules appear to conflict with (“override”) those obligations.

Anti-conduit doctrine may reflect two overlapping but distinct intuitions about what makes the participation by an intermediate entity a potentially abusive “conduit” arrangement; this Report refers to these as the True Recipient rationale and the Base Erosion rationale. These are best described by considering a paradigmatic conduit financing arrangement involving “back to back” loans with substantially similar terms with the result that the interest payments received by the intermediate entity are largely offset by matching interest payments made by the intermediate entity to the ultimate financing entity.

The True Recipient rationale justifies recharacterization of the arrangement because there was a purposive transactional nexus between the different legs of the financing arrangement. The arrangement is potentially abusive (and appropriately recharacterized) because the intermediate entity is not in an economic sense the true “owner” of that payment. From that perspective, it is not relevant whether the intermediate entity bears significant residence country tax on the net income or can deduct the outbound payments, or whether the parties incur other costs and tax burdens such as non-U.S. withholding taxes on the outbound payments.³⁹ The intermediate entity substantively lacks sufficient dominion and control over the payments received to be considered their “true” recipient for U.S. withholding tax purposes.

The True Recipient rationale is reflected in various features of the Existing Anti-Conduit Regulations. For example, if the intermediate entity is unrelated to both the financing entity and the financed entity, the Existing Anti-Conduit Regulations look to whether the U.S. leg of the financing would or would not have occurred “but for” the financing received by the intermediate entity. (This purposive nexus is “presumed” in the related party context.) The Existing Anti-Conduit Regulations also treat “redeemable” equity as a financing transaction without regard to whether (and even though there is no reason to expect per se that) the dividends on such equity or the equity redemption payments would be deductible. The manner in which conduit recharacterization is applied also reflects this True Recipient theory. For example, the intermediate entity may earn a “spread” if the intermediary receives interest at 7% but pays interest at 6%. From a Base Erosion rationale perspective, one could argue that only 6/7ths of the underlying interest payments should be recharacterized and be denied relief. However, if the intermediate entity should not be considered the “true recipient” of underlying interest payment it received, then it arguably makes sense to deny withholding tax relief entirely (as the Existing Anti-Conduit Regulations would do) rather than recharacterizing only 6/7ths of the underlying interest payments.

The Base Erosion rationale is reflected to a more limited extent in the Existing Anti-Conduit Regulations. The Base Erosion rationale justifies recharacterization of the arrangement because the offsetting payments have the potential to minimize tax in the intermediary country by eroding the tax base in that country. Interest payments on a back-to-back loan are presumptively deductible and, to the extent the rates earned and paid on related financing transactions are similar, result in no net income subject to tax in the intermediary country.⁴⁰ Under the Base Erosion rationale, if the U.S. withholding tax benefit being obtained is treaty-based, it makes no sense to grant that relief based upon a U.S. tax treaty with the intermediary country insofar as treaty benefits are intended to prevent double taxation, not to permit double non-taxation.

The Existing Anti-Conduit Regulations perhaps define “financing transactions” (excepting Redeemable Equity) in a manner that is generally intended to capture the kinds of instruments with the potential for base erosion (because they are of a character that is likely to generate deductible and thus base-eroding interest, rent or royalties under most jurisdictions' tax systems). However, this Base Erosion rationale is not reflected directly in how the Existing Anti-Conduit Regulations actually work. Unlike the “base erosion” rules in treaty “limitation on benefits” (“LOB”) provisions, the Existing Anti-Conduit Regulations do not ask whether the intermediary's offsetting payment is in fact deductible as a matter of intermediary country tax law. The Existing Anti-Conduit Regulations also ignore whether there is a taxable “spread” in the intermediary country, whether the intermediate entity is in a high-tax country or whether other taxes (such as an intermediary country withholding tax) offset any U.S. withholding tax advantage. In other words, the Existing Anti-Conduit Regulations may define a “financing transaction” in a manner likely to coincide with base-eroding deductible payments, but they do not actually seek to determine or even consider whether there is base erosion under the intermediary's tax law in defining a financing transaction or even in assessing whether the arrangement has a tax avoidance purpose.

This presumably is because the anti-conduit doctrine historically has focused narrowly on U.S. withholding tax and not on other tax effects such as income tax rate arbitrage, duplication of deductions in more than one jurisdiction or deductions that are not matched by inclusions. However, there is no statutory “conduit” provision that mandates this narrow focus. The Existing Anti-Conduit Regulations reflect the codification (and expansion) of the common law anti-conduit doctrine under the broad regulatory authority of Code section 7701(l) (which applies to avoidance of any tax and in no way suggests any anti-abuse provision must look narrowly at withholding tax effects).⁴¹ If the Existing Anti-Conduit Regulations focus narrowly on withholding tax avoidance and the general potential for base erosion in principle, rather than testing for or measuring actual base erosion, it is because Treasury and the IRS chose to adopt that approach.

One justification for that decision is that inquiring into details of the actual local tax law treatment of the intermediate entity as a part of a conduit analysis can be inordinately burdensome for both the IRS and withholding agents. Significant administrability benefits result from treating transactions as financing transactions if they are of the type that are in principle likely to give rise to base-eroding deductions whether or not they actually do (avoiding burdensome inquiries into foreign law). However, this justification loses much of its force if the Existing Anti-Conduit Regulations are to be expanded to require the IRS and withholding agents to engage in burdensome inquiries into details of the foreign tax law treatment of the intermediate entity (such as whether it has obtained notional interest deductions) and its shareholders (such as whether they have received credits or other relief based upon the amount of tax paid by the entity, but without regard to the amount of tax they would otherwise owe) and to do so in every case.

Once one concludes that preventing actual base erosion outweighs administrative simplicity and justifies an inquiry into foreign tax laws, it is not clear why that inquiry should apply only to evaluate whether permanent equity instruments give rise to a deduction or similar benefit and focus narrowly only on whether the intermediary's payment is or is not tax deductible. Why, for example, retain the rules treating redeemable equity that produces no deductible payments as a financing when there is no base erosion and the intermediate entity is fully taxed on the inbound payments, resulting in precisely the potential double tax burden its residence country treaty with the U.S. was intended to protect against? Why not treat an intermediate company in a high-tax country that earns a significant taxable spread and thus pays significant foreign tax differently from an intermediary in a low-tax country that pays a negligible amount of tax? Why limit the NID Limitation Rule to Category 1 hybrid equity? Under the Base Erosion rationale, it makes more sense to apply this approach to any financing transaction. If policing base erosion rather than identifying the factual “true recipient” of income is to become the focus and organizing principle of the Anti-Conduit Regulations, adopting this NID Limitation Rule as a general approach would arguably result in a far more calibrated, and less draconian regime. Adopting the Base Erosion rationale only for narrow categories of hybrid equity and relieving its overbreadth through the NID Limitation Rule only in that case has an ad hoc quality.

The policy considerations above also must be balanced by the need to have rules that are administrable and clear and treat similarly situated taxpayers consistently. The administrability of the Anti-Conduit Regulations is of significant concern for both foreign parties who directly participate in multi-party financing arrangements and withholding agents. The U.S. withholding regime relies heavily on withholding agents to administer the applicable withholding rules and collect gross basis tax owed by foreign taxpayers.⁴² However, withholding agents are frequently unrelated to the principal parties to a financing arrangement and typically have limited insight into their motivation for structuring a transaction or the foreign law treatment of the various legs of the transaction. Given that an unrelated withholding agent has little or no financial stake in the relevant financial transactions, faced with any material risk of an adverse conduit recharacterization and liability for having under-withheld, it has every incentive to just withhold.⁴³

Although conduit treatment requires the existence of a “tax avoidance plan,” the Existing Anti-Conduit Regulations as drafted, in practice, grant very substantial discretion to the relevant IRS director of field operations to impute such a plan and recharacterize.⁴⁴ The “tax avoidance plan” test is ultimately subjective, and certain facts may create a presumption of such a plan. Accordingly, even if a taxpayer knows a transaction is not tax motivated, it may be very difficult as an evidentiary matter to disprove that “one of” the principal purposes of an arrangement was avoidance of withholding tax if withholding tax in fact is reduced as a result of participation by an intermediate entity. Whether or not ordinary course transactions ultimately pass muster, the subjective nature of the “tax avoidance” rationale at a minimum involves uncertainty and risk of IRS challenge, and requires a more costly, fact-intensive analysis.

The Existing Anti-Conduit Regulations recognize the practical difficulties faced by withholding agents in identifying conduit arrangements and excuse withholding agents from liability if they fail to deduct or withhold with respect to a conduit arrangement unless the agent “knows or has reason to know” the financing arrangement is a conduit financing arrangement. Financing parties who are unrelated to the other participants face similar concerns and are afforded similar protection.⁴⁵ While this standard provides some comfort, because the “reason to know” standard does not require actual knowledge, withholding agents face concerns about the level of due diligence (if any) they must conduct to ensure that knowledge of a conduit arrangement will not be imputed to them. It is easy to downplay the administrative burden the Proposed Regulation may create by claiming the withholding agent is protected “as long as it does not know or have reason to know” there is a Category 1 or Category 2 equity financing and that the lender qualifies under the Anti-Conduit Regulation's definition of a “conduit”. But we do not think that is realistic in many real-world situations. Consider the following example.

In this example, what does the administrative agent have “reason to know” about whether there is a Category 1 or Category 2 transaction between Country X bank and Special Situations Ltd?⁴⁶ Will the IRS allow it to conclusively rely on a representation from Special Situations Ltd. that it is not a conduit within the meaning of the Anti-Conduit Regulations?

If not, should the administrative agent obtain its own Country Y tax advisor to diligence whether Country Y has a Category 1 notional interest deduction regime or Category 2 refund regime? If an internet search readily indicates that Country Y does have one of these regimes (although it is unclear whether it applies to all equity or what the requirements are for it to apply in any given case) could that change the answer? Should the administrative agent avoid looking into the question at all so that it is not put on notice to inquire further?

What if the administrative agent is a large banking institution and one of its investment banking divisions (uninvolved in this lending transaction) had previously assisted a number of Country Y clients to structure equity offerings entitled to the notional interest deduction regime, would that knowledge be imputed to the administrative agent function for this purpose? Should that function undertake an internal institution-wide diligence inquiry about whether anyone within the institution has knowledge of Country Y's tax law regime for equity just in case? Should that inquiry extend to all affiliates of the legal entity that is the administrative agent? Or rather than relying on particular personnel to conduct ad hoc diligence, should the institution instead develop and institute detailed back office procedures to address these kinds of risks?

To minimize the burden on withholding agents and allow them to comply with whatever their obligations are, it is important that the applicable standards for conduit treatment be as clear and administrable as possible. The challenges third-party withholding agents face, illustrated above, are not created by the Proposed Regulation. They are present under the Existing Anti-Conduit Regulations. However, the Withholding Regulations make reasonably clear that an agent is not lightly imputed knowledge of other financing parties' subjective withholding tax avoidance intent based on knowledge of other conduit indicia, and other conditions for conduit recharacterization under the Existing Anti-Conduit Regulations are factual, objective and susceptible of factual diligence. The Proposed Regulation may require an inquiry into foreign tax laws. As discussed later in the Report, we think that substantially increases the burden and uncertainty for withholding agents.

2. Relationship of the Hybrid Equity Conduit Tests to Other Provisions

a. Relationship to Treaties

We do not believe expanding the Anti-Conduit Regulations to treat hybrid equity instruments as financing transactions would directly conflict with the U.S.'s treaty obligations under treaties that substantially conform to the U.S. or OECD model treaties.⁴⁷

Nevertheless, this expansion could be viewed as inconsistent with the spirit of existing LOB provisions.⁴⁸ The U.S. generally takes the position that domestic law limits on entitlement to treaty benefits (including conduit principles) apply regardless of what the treaty provides.⁴⁹ As applied to an entity that must otherwise satisfy an LOB base erosion test, this new rule would appear to create a second test that conflicts with the premise of the LOB base erosion test and arguably has a “heads we win — tails you lose” quality. If the treaty's LOB base erosion test measures the effective local tax burden for the entity as a whole, this new rule would then apply substantially the same test at the transactional level. One could imagine the treaty party viewing this new, second test as undermining the deal it had cut with the United States. Such a treaty party might well take the position that if the appropriate way to test is transaction-by-transaction, then any transaction that satisfies the test should be eligible for treaty benefits even if the entity-as-a-whole would not meet the test. That said, one can argue that it is not inappropriate to assess base erosion on a transactional as well as an entity-as-a-whole basis where other indicia of potential withholding abuse are present.

We recognize there are also alternative LOB tests that do not take “base erosion” issues into account explicitly, for example the “active business” test or “publicly traded company” test. We do not think this undercuts the point above. If the United States believed “base erosion” should be a general requirement of all LOB provisions in its treaties, it could have negotiated for that. We believe the United States must, to be respectful of its treaty partners and its treaty obligations, be able to articulate how the revised conduit rules reflect a narrow set of factual circumstances and policy concerns (involving highly structured transactions) distinct from more general base erosion concerns reflected in typical treaty LOB provisions.

Prior to the Proposed Regulation, the justification the United States could assert for applying domestic law conduit principles despite negotiated LOB provisions is that these domestic law principles applied in narrow circumstances of abuse, involving structured and related financing transactions which disguised the substantive “true” owner of income. The True Recipient rationale for conduit was distinct from the base erosion concerns of treaty LOB provisions and could be defended as being similar conceptually to (albeit broader in application than) the OECD approach to beneficial ownership which also may deny treaty benefits when the recipient receives interest, dividends or royalties in an intermediary capacity.⁵⁰ To the extent that the Proposed Regulation only capture classic “hybrid” instruments, which the United States considers equity but an intermediate entity's country considers debt, treaty partners are unlikely to complain. They should not be particularly surprised if U.S. conduit financing principles treat as a financing a financial instrument that the country's own tax laws also treat as debt. However, the Proposed Regulation sweeps more broadly and could at the other extreme capture all common equity merely because the country has adopted a cost of capital allowance regime to minimize tax incentives to favor debt over equity capital. United States Treaty partners are likely to be less sympathetic to application of conduit principles based solely on the Base Erosion rationale rather than the True Recipient rationale.

We understand that there may be important policy considerations other than treaty comity at stake in the case of abusive arrangements involving hybrid permanent equity but, unless such abuses are becoming common, treaty comity considerations favor circumspection in expanding our domestic law conduit principles rather than addressing these issues on a country-specific basis in treaty negotiations.

Although few countries (notably Brazil and Belgium) have had notional interest deduction regimes historically, we are aware these are becoming more widespread.⁵¹ Nevertheless, we have not so far experienced in our collective practices wide-spread use of these countries as “conduit” jurisdictions as a result of such tax regimes of general applicability such as notional interest deduction regimes. If Treasury and the IRS are aware of intermediate entities being set up in these countries specifically to avoid U.S. withholding tax, we would be sympathetic to a targeted expansion of the conduit rules to prevent this. However, we would not favor expanding conduit rules based on theoretical concerns that base erosion could in principle make permanent equity equivalent to other kinds of financing because it facilitates base erosion. Actual cases of potential conduit abuse in our experience are more likely to involve the use, in related party contexts, of classic hybrid instruments considered debt in the foreign country but equity by the United States and which narrowly avoid meeting the technical requirements to be Redeemable Equity or, in any event, involve highly structured transactions using special purpose entities. We would support targeting these more narrowly.

b. Relationship to Section 267A

There is overlap with Section 267A and the Proposed Regulation in the case of Category 1 hybrid equity, as explained above, for example because a related “intermediate entity” would likely create an imported deduction/no inclusion mismatch situation if it issued tax deductible equity. We recognize there are other cases where the intermediate entity that enters into a financing transaction with a U.S. entity may have issued Category 1 or 2 hybrid equity that is not subject to section 267A. This is the case if the intermediate entity is not related to the U.S. financed party (subject to the special rules for “structured transactions”), although that is also a less sympathetic situation in which to impose withholding obligations on the unrelated U.S. financed party (as discussed elsewhere). A Category 2 equity hybrid also may not give rise to an imported deduction/no inclusion result technically under section 267A.

One could argue that to the extent Congress was concerned about hybrid equity issued by intermediate entity it has chosen to address this under section 267A by denying deductions for “imported” deduction/no inclusion results. To the extent Congress has chosen to limit the scope of section 267A to situations involving parties related to the U.S. financed party or to interest and royalty deductions, there is no reason to expand the scope of that response to hybridity indirectly through the conduit rules. Moreover, the overlap between section 267A and the Proposed Regulation in the case of Category 1 hybrid equity could result in a loss of interest deductions on the U.S. financing leg as well as, potentially, U.S. withholding tax.

It seems harsh to doubly penalize the same hybrid consequences where there is such overlap. We assume Treasury and IRS may view withholding and deductibility as simply distinct issues.⁵² Once the conduit test for being a financing turns on deductibility as a matter of intermediate country local tax laws and pure base erosion concerns, however, the argument that deductibility for U.S. net income tax purposes and U.S. withholding are just fundamentally different and unrelated issues is not very compelling. The same rationale (that effective net income tax rates and gross basis withholding taxes are not related) could be used to argue that conduit issues should not turn on foreign tax law base erosion results in the first instance.

On the other hand, those cases in which section 267A and the hybrid equity conduit consequences may overlap necessarily involve an intermediate entity that is “related” to the U.S. financed party (other than in the narrow case of structured transactions, as outlined above). Other than in the case of a structured transaction, relatedness of the parties to the U.S. financing leg is a precondition to application of section 267A. Presumably, the related U.S. financed party would not only have access to the relevant tax planning information but some ability to control and plan around the result. If they structure transactions resulting in a double penalty this is not a particularly sympathetic case for relief. That said, application of the conduit recharacterization also is less likely in practice to be necessary to deter that planning.

We have significant concerns about whether addressing concerns over hybridity through conduit rules makes sense. The conduit rules and historical conduit principles evolved in a very different era with different international rules. Sections 267A and 245A(e) and international projects such as BEPS⁵³ and the E.U. ATAD I and ATAD II initiatives⁵⁴ will render obsolete many of the planning techniques and instruments at which the Proposed Regulation takes aim in any event and such regimes are directly focused on, and therefore better designed to target surgically, abusive use of hybrid instruments. However, we accept that the existence of section 267A and these other initiatives will not fully prevent the possibility of that “abuse” and there may therefore be a legitimate policy argument for applying conduit principles. Accordingly, we recognize that Treasury and the IRS may be determined to employ conduit principles to that end regardless. The key policy consideration in that event is to ensure that the administrative complexity and burdens the Proposed Regulation will cause is not disproportionate to the scale of the potential abuse (involving hybrid equity not already captured by the redeemable equity financing transaction definition).

c. Relationship to Domestic Law Withholding Exemptions

The scope of the Anti-Conduit Regulations is not confined to avoidance of U.S. withholding tax based on treaty benefits. The regulations also permit the Commissioner to disregard a conduit whose participation is intended to secure domestic law withholding exemptions that otherwise would be unavailable to the financing provider. For example, the regulations may permit the Commissioner to disregard the participation of a non-bank intermediate entity if the financing party is a bank that would not be entitled to the portfolio interest exemption had it made a loan in the ordinary course of its lending business directly to the financed entity. As noted above, when treaty shopping is at issue, the base erosion rationale for applying conduit recharacterization is more compelling as treaty benefits are premised on preventing double taxation. That is not the premise of the portfolio interest exemption and the application of base erosion concerns to support conduit recharacterization is far less appealing in that context.⁵⁵

IV. SPECIFIC RECOMMENDATIONS

1. Narrow the Scope of the Instruments to Which the Proposed Regulation Applies

We understand the concern of Treasury and the IRS that base-eroding equity can achieve results similar to deductible debt financing. Nevertheless, the distinction between treaty partner tax laws that facilitate or permit base “erosion” and treaty partner laws that merely define the corporate base and appropriate rate of taxation is a matter of degree. At one end of the spectrum are instruments that are “classic” hybrid instruments that the intermediate country considers debt for its own purposes but that are equity for U.S. purposes. We agree these should be treated as financings as they can be considered to achieve equivalent commercial and tax results, raise administrability issues that are manageable and do not raise significant treaty comity concerns (as a treaty partner can hardly be surprised if the U.S. views the instrument in the same manner as the treaty party for this purpose). At the other end of the spectrum, are equity instruments that are subject to notional interest deductions or similar “cost of capital allowance” regimes of general applicability (enacted, for example, to reduce artificial tax incentives to use debt rather than equity capital). We do not think it is appropriate potentially to treat an operating company that is organized in a country that has such a regime as having engaged in a conduit financing arrangement simply because the country has such a regime. We recognize that addressing cases between those extremes raises difficult line drawing issues.⁵⁶ However, for reasons discussed above, we recommend that the scope of the Proposed Regulations be limited to (1) classic hybrid instruments (as discussed directly below) and (2) equity that achieves the types of “notional” base eroding benefits described in the Proposed Regulation if, but only if, such equity is issued by a special purpose company formed only to carry out the subject financing transactions. We also think foreign companies subject to regimes equivalent to the U.S. REIT and RIC regimes, whose entitlement to treaty benefits under LOB rules has been expressly negotiated with treaty partners, should not be treated as conduits on this basis.

We agree that the Proposed Regulation should apply to what we refer to as “classic hybrid instruments” by which we mean instruments that are equity for U.S. tax purposes but are debt for foreign law purposes and provide for a fixed or determinable rate of return. For reasons discussed above and elsewhere in this Report, we think Category 1 and Category 2 instruments otherwise are not financing transaction that are properly the subject of anti-conduit rules except in limited circumstances (i.e., situation (2) identified above and discussed further below). We are not aware in our practices of taxpayers attempting to avoid U.S. withholding tax by utilizing foreign jurisdictions with tax regimes offering Category 1 or 2 tax benefits. Actual cases of potential conduit abuse in our experience are more likely to involve the use, in related party contexts, of classic hybrid instruments considered debt in the foreign country but equity by the U.S. and which narrowly avoid meeting the technical requirements to be Redeemable Equity.

Expanding the Existing Anti-Conduit Regulations in this limited manner is less likely to be perceived as an override of existing obligations to treaty partners. If the Proposed Regulation is finalized in its current form, a treaty partner that agreed with the U.S. to reciprocal reduction of interest withholding tax rates may consider that the U.S. is reneging on its side of that deal solely because that country has, for reasons unrelated to facilitating U.S. withholding tax avoidance, a notional interest or cost of capital allowance regime or a regime that falls within Category 2. Anti-conduit rules that collapse back-to-back financings that the treaty partner views as debt are far less likely to be viewed as contrary to our treaty obligations. Given the statements made in the 2008 Preamble and other developments in the intervening period, expanding the Existing Anti-Conduit Regulations to reach classic hybrid instruments also would result in no surprise to U.S. taxpayers and foreign treaty country residents.

The Proposed Regulation appears in large part to be an attempt to “correct” what Treasury and the IRS view as deficiencies in the scope of the anti-hybrid regime Congress enacted in the form of sections 245A and 267A. Those provisions were enacted by Congress as part of an intentional effort to be consistent with the OECD's anti-base erosion hybrid program and an international consensus. They are carefully targeted at specific abuses and we think that it would be inappropriate to take the base-erosion principles motivating that program and use them to go beyond where Congress went by expanding long-standing rules aimed at back-to-back financing transactions and eligibility for reductions in withholding.

Finally, the approach in the Proposed Regulation would add significant complexity to the already-complex Existing Anti-Conduit Regulations. This complexity combined with the design of the Existing Anti-Conduit Regulations would create significant uncertainty for taxpayers and impose significant burdens on the IRS, which would be tasked with trying to apply the Proposed Regulation in an even-handed, fair and consistent manner across taxpayers. This would require, among other things, detailed investigation into the application of foreign tax laws to foreign entities. Perhaps the greatest burdens, risk and potential unfair application would fall on third-party withholding agents.

In recognition of concerns of abuse, we would, however, support inclusion of Category 1 or Category 2 equity as a financing when the issuer is a special purpose entity substantially all of whose assets are comprised of a single (or series of integrated) financing transactions into the United States. Such a case is much more likely to involve actual withholding tax abuse, is easier to rationalize as a matter of treaty comity (as a narrow anti-abuse rule aimed at highly structured transactions) and by limiting the number of situations in which the foreign tax law must be considered will create far less of an administrative burden for taxpayers, withholding agents and the IRS.

If, however, the basic approach of the Proposed Regulation is to be adopted, we suggest various additional modifications below.

2. Clarify the Definition of Category 2 Equity

The current definition of Category 2 hybrid equity is vague and further elaboration of the requirements (including through examples) is needed.

Stock will constitute a financing transaction under Category 2 only if one of the following conditions is met:

A person related to the issuer is, under the tax laws of the issuer's country of residence, allowed a refund (including through a credit), or similar tax benefit for taxes paid by the issuer to its country of residence on amounts paid, accrued, or distributed (deemed or otherwise) with respect to the stock or similar interest, without regard to any related person's tax liability under the laws of the issuer's country of residence.

This definition has the following requirements

(1) a related person to the issuer (who does not need to be the holder of the stock) must receive a refund (including by way of credit) or “similar tax benefit,”

(2) that tax benefit must be provided by the equity issuer's country of residence and not some other jurisdiction (such as the related person's residence jurisdiction),

(3) receipt of that refund, credit or similar benefit must be tied to an amount distributed, paid, accrued or deemed distributed with respect to the stock by the equity issuer (again, this distribution or deemed distribution is made to the holder of the stock and does not need to be made to the related party that receives the tax refund or other benefit) and

(4) although unclear, it appears the refund, credit or benefit must not be determined or affected by tax liability that such related party (or indeed any related party) otherwise owes to the equity issuer's residence country.

Some examples based on Example 5 of the Proposed Regulation highlight the uncertainties as to what is intended and how this applies. Much of this uncertainty results from the fact that we are not aware of situations where an issuer's country provides a cash refund of the issuer's corporate-level tax paid to a shareholder that is not otherwise subject to taxation by that country (i.e., where requirement 4 is met). Accordingly, examples of regimes that meet these requirements would be helpful in ensuring that stakeholders understand what the requirements are. We begin by setting out the Proposed Regulation's Example 5 in full.

Variations on Proposed Example 5:

Example 5A

(a) Facts. Same as Example 5 except that both FP and FS are resident in Country T.

Example 5 states that FP would not otherwise be taxable by Country T on the dividend but does not specify FP's country of residence and whether FP needs to be a Country T resident to obtain the 90% tax refund. For this Example 5A, assume corporate residents are simply not taxed by Country T on corporate dividends but nevertheless receive what amounts to an imputation credit of 90% of the corporate tax paid on the earnings supporting that dividend. (This may not be a realistic scenario but that reflects our difficulty in understanding what Category 2 is intended to reach.)

(b) Result. FP's stock in FS constitutes a financing transaction because (1) FP is related to FS, and FP gets a refund, credit or similar benefit, (2) the benefit is tied to an amount distributed on equity issued by FS, (3) this benefit is provided by Country T which is FS's residence country and (4) this benefit is provided without regard to any tax liability FP otherwise owes to Country T. We assume the benefit is also provided without regard to any liability any other person related to FS owes to Country T.

(c) Commentary. Because FP and FS are both residents of Country T, this is not a classic treaty shopping scenario and the withholding tax results may not change if the Proposed Regulation applies (i.e., both FP and FS would be eligible for benefits of the US-T treaty). Theoretically, however, it is possible FS could satisfy an LOB requirement which FP does not (although this seems somewhat unlikely). This fact pattern does however illustrate an equity structure where the Proposed Regulations would apply but it would not appear to meet the True Recipient rationale. It is not clear if this is “base erosion”: the result of the fact pattern is a tax paid in Country T on the earnings that get distributed to FP at a rate that is 1/10^th of the normal corporate tax rate, but this is a feature of Country T tax law, not of a manipulation or distortion by the parties. Moreover, Country T could increase or decrease its corporate tax rate resulting in the same Country T tax result and that would not trigger the Proposed Regulation.

Alternatively, what if FP is permitted the benefit only as an offset to corporate tax owed by FP on income from other sources. Is requirement 4 now not met, such that this is no longer a Category 2 instrument (even though the FS earnings are still subject only to a tax of 1/10^th of the Country T corporate tax rate)?

What if instead FS owns FS2, they form a unitary group and part of the tax being credited to FP is imposed on FS2's income. Is the benefit to FP now with regard to a tax liability owed to Country T by a person related to the equity issuer (i.e., by FS2) such that requirement 4 is now not met?

Example 5B.

(a) Facts. Same as Example 5A except that Country T does not exempt dividends from corporate tax. Here, the imputation credit of 90% of the tax paid by FS on the earnings supporting the dividend result in the equivalent of a single layer of corporate tax at 110% of the Country T corporate tax rate.

(b) Result. Unlike Example 5A,the FP stock in FS here is not a financing transaction because requirement 4 is not met (the benefit is not provided “without regard” to FP's tax liability to Country T but instead is available only to offset that liability (and is limited to that such liability)).

(c) Commentary. The total amount of Country T tax paid in Example 5A is much lower than the total paid in this Example 5B. We note however that it is not clear why the difference in the Country T tax regime between Example 5A and Example 5B should matter in determining whether the recipient is eligible for the benefits of the U.S.-T treaty.

(d) What if, alternatively, FP is permitted a cash refund if, after taking into account the Country T taxes on all other income and the imputation credit, it has a net tax “overpayment”. Would one conclude now that requirement 4 is met because FP is permitted the benefit whatever its Country T taxes on its income otherwise would be (i.e., it receives the benefit if its total taxes are more or are less than the refund)?

Example 5C.

(a) Facts. Same as Example 5B except FP is resident in a country other than Country T, that country has no treaty with the U.S., Country T imposes a withholding tax on dividends and the 90% is permitted as a credit or offset to that withholding tax.

(b) Analysis. The FP stock in FS here is not a financing transaction because requirement 4 is not met (the benefit is not provided “without regard” to FP's tax liability to Country T but instead is available only to offset that liability (and is limited to that such liability)).

(c) Commentary: The Proposed Regulation does not apply and the U.S.-T treaty benefits are presumably available to FS. FP has been granted a benefit that results in FP not paying the Country T withholding tax, but this benefit is not considered problematic by the Proposed Regulation.

Example 5D.

(a) Facts. Same as Example 5B except that Country T decides to eliminate the complexity of its imputation regime and just exempt all dividends that are paid out of “previously taxed earnings” entirely while also eliminating the credit for payor corporate tax. Assume this dividend exemption would also apply to withholding tax otherwise owed on Country T sourced income by non-residents, including a non-resident which owns a small amount of the stock.

(b) Result and Commentary. Is the exemption a “similar benefit?” It certainly is a benefit intended by Country T to replace the prior credit for first tier corporate tax. It is triggered with respect to, tied to or measured by the distribution. The exemption is arguably with regard to the tax otherwise owed by a related party insofar as the related shareholder would have owed tax were it not exempt (although the distinction between exempting what would have legally been owed were it not exempt and crediting corporate tax paid on earnings distributed when nothing is technically owed by the shareholders is vanishingly thin). On the other hand, for the unrelated shareholder it would be determined with regard to such holder's liability to Country T not the related party's liability. We believe the result is that this equity is not a Category 2 financing transaction but clarification would be helpful.

To the extent our conclusion immediately above is wrong, and this is a financing, can Country T change that result if it simply exempts dividends outright premised on corporate tax having been paid but with no requirement to actually track corporate tax paid through a previously taxed earnings account?

Finally, we believe that Category 2 is not intended to apply if FP's residence country, which is not Country T, allows FP a foreign tax credit for FS's corporate tax or under that country's controlled foreign corporation regime exempts FP from tax on dividends or deemed inclusions with respect to FS equity to the extent a sufficiently high rate of Country T tax applies. Even if this is a “similar benefit” in some sense, the benefit is not granted by the equity issuer's residence country. However, it would be helpful to confirm this, particularly because this is a much more typical regime than those posited above that would be within Category 2.

3. Clarify Withholding Regulations and “Reason to Know” Standard

For reasons explained above, it is difficult for withholding agents (including financed entities) who are unrelated to the intermediate entity to ascertain whether the arrangement is a potential conduit transaction even under the Existing Anti-Conduit Regulations. However, because the current standards for the existence of a financing transaction and other requirements (other than the subjective “principal purpose of tax avoidance” test) are inherently factual rather than legal they are more susceptible of reasonable diligence by an unrelated withholding agent. The Withholding Regulations also make clear that knowledge of the existence of the financing transactions does not, without more, constitute a “reason to know” the arrangement is a conduit financing arrangement.⁵⁸ This statement could be read to mean that merely knowing that a financing transaction (e.g., back-to-back debt) exists and knowing that the parties are claiming treaty benefits based upon the intermediary country (the withholding agent knows this because that is part of being the withholding agent) does not create any duty to inquire further. It is difficult to imagine that the “more” would be met merely because the withholding agent was also aware that most countries provide a deduction for interest paid on debt. So, does that mean that a withholding agent also does not know “more” if it has knowledge of the intermediary regime's treatment of equity as giving rise to Category 1 or Category 2 benefits?

It would be helpful at a minimum to clarify that knowledge of the existence of a financing, including (in the case of an equity instrument) knowledge that it gives rise to tax benefits of the type described in Category 1 or 2 that cause it to be treated as a financing transaction, does not alone constitute a reason to know that the arrangement is a conduit financing arrangement. Further, knowledge of the existence of such tax benefits without more creates no inference that there is, or that a withholding agent has knowledge of the intermediate entity's tax planning or reason to know there is, a withholding tax avoidance plan within the meaning of section 1.881-3(b) of the Anti-Conduit Regulations. The regulations should also clarify that the “reason to know” standard implies no obligation to undertake diligence into any foreign tax law regime applicable to the instruments that form part of a financing arrangement.

It would also be helpful to revise relevant Forms W-8 to include in the penalties of perjury statement that the provider beneficially owns the income a statement that it is not a conduit with respect to the financing transaction giving rise to the income within the meaning of the Anti-Conduit Regulations. A withholding agent unrelated to the intermediate entity or financed entity should be permitted to conclusively rely on this representation absent actual knowledge to the contrary.

In either case, subjecting unrelated withholding agents to penalties based on an actual knowledge standard rather than a “reason to know” standard seems more appropriate. The latter standard may have been more administrable and equitable when the Existing Anti-Conduit Regulations were focused more narrowly on objective non-legal criteria. However, once the definition of the filters designed to identify potential abuse cases is expanded to require analysis of foreign tax laws, we do not think that “reason to know” standard is generally appropriate.

Treasury and the IRS may wish to consider whether an actual knowledge standard is more appropriate given the evolution in the Anti-Conduit Regulations. It would not be unreasonable in that case to continue to apply a broader “reason to know” standard to a person related to the intermediate entity as (1) it is more reasonable to presume knowledge in that context and (2) they have a much greater practical ability to obtain the necessary information of the intermediate entity-related tax planning if it has occurred. Treasury and the IRS could also apply that standard to a withholding agent who is acting as the agent of such related person or the intermediate entity as a matter of law. If such an agent is in contractual privity with the person charged with knowledge, they can protect themselves contractually with indemnities (and in our experience, in such a context usually would).⁵⁹

4. Clarify the Computation and Character of the Amount Subject to Withholding Under the Proposed Regulation

As discussed above, the computation of the amount of the financed entity's payment that is recharacterized depends upon the “principal amount” of the Category 1 or Category 2 transaction. The Proposed Regulation lacks crucial guidance as how that principal amount would be determined.

As outlined above, the Proposed Regulation does not modify or supplement the Existing Anti-Conduit Regulations' rules for determining principal amount. Those rules provide that the principal amount equals “the amount of money . . . or the fair market value of other property advanced”.⁶⁰ The special section 305 rule issue price rule for equity is unlikely to apply to Category 1 and Category 2 transactions. As a result, in applying the Scale-Back Rule to Category 1 and Category 2 transactions, it would appear the “principal amount” is the amount contributed by the financing entity to the intermediary in exchange for the Category 1 or Category 2 equity. This could result in a significant distortion, however, because it would likely take into account all of the equity held by the financing entity, all the contributions made by the holder of that equity, and not be reduced by any distributions paid by the intermediary to that holder. This could be particularly distortive when the financing entity owns 100% of the intermediary and has made contributions that funded activities of the intermediary unrelated to its financing of the financed entity.

The Existing Anti-Conduit Regulations already reflect an intention to determine the principal amount of the links in the chain in a manner that more appropriately matches them up. Thus, the Existing Anti-Conduit Regulations provide:

“However, if the property advanced, or the right to use property granted, by the financing entity is the same as the property or rights received by the financed entity, the fair market value of the property or right shall be determined as of the close of business on the last date that any of the financing transactions comprising the financing arrangement is entered into. . . .

“In the case of a debt instrument or of stock that is subject to the current inclusion rules of sections 305(c)(3) or (e), the principal amount generally will be equal to the issue price. However, if the fair market value on the issue date differs materially from the issue price, the fair market value of the debt instrument shall be used in lieu of the instrument's issue price. Appropriate adjustments will be made for accruals of original issue discount and repayments of principal (including accrued original issue discount).”⁶¹

In the case of Category 1 transactions that are subject to NID Limitation Rule, the Proposed Regulation appears to intend to prevent such distortion. However, clarification is needed. The NID Limitation Rule utilizes “the financing transaction's principal amount as determined under Treasury regulation section 1.881-3(d)(1)(ii) multiplied by the rate used to compute the issuer's notional interest deduction for the taxable year in which the payment is made.” Example 26, illustrating this rule, utilizes a different method, which might be more appropriate, but it needs to be provided for in the operative rules. Specifically, Example 26 provides as facts:

“FP also contributes $5,000,000 to FS in exchange for FS stock. Pursuant to Country T tax law, FS is entitled to a notional interest deduction with respect to the stock equal to the prevailing Country T government bond rate multiplied by the taxpayer's net equity for the previous taxable year.”⁶²

The Example's analysis treats the $5,000,000 as the principal amount of the Category 1 financing transaction. This leaves unanswered what the principal amount is if there have been multiple contributions to “FS” and if the notional interest deduction is computed based upon a base that differs from the amount(s) so contributed. The Example assumes the “net equity” is $5,000,000, without explanation.

This illustration of the NID Limitation Rule highlights another distortive aspect of the computations in the case of Category 1 and Category 2 transactions. Category 1 applies to any equity where the issuer is allowed, instead of a deduction, “another tax benefit (such as an exemption, exclusion, credit . . . for amounts [actually or deemed] paid, accrued, or distributed . . . with respect to the stock” and Category 2 applies to any equity where a person other than the issuer is “allowed a refund (including through a credit), or similar tax benefit for taxes paid by the issuer to its country of residence on amounts paid, accrued, or distributed (deemed or otherwise) with respect to the stock”. Both of these categories thus provide a benefit that may have a value that is far different than the value an interest deduction paid on the debt of equivalent “principal amount” would have. The NID Limitation Rule demonstrates why the computation should look to something other than the “principal amount” when the benefit is something other than a classic deduction for interest paid on a debt financing. There is no corollary or comparable rule for these other types of Category 1 and Category 2 transactions.

A further illustration of this problem is apparent from the Proposed Regulation's Example 5. Example 5 posits a situation “whereby FP, as the sole shareholder of FS, is allowed a refund with respect to distributions of earnings by FS that is equal to 90% of the Country T taxes paid by FS associated with any such distributed earnings.” The Example does not tell us the “principal amount” of the FS stock held by FP, the amount of FS's earnings or the tax rate that FS has paid, thus there is no indication of whether an application of the Scale-Back Rule standing alone is going to determine the portion of the financed entity's payment that is recharacterized as being paid directly to FP that has a logical connection to the benefit derived by the parties from the Category 2 benefit or the extent to which treating FS as a conduit intermediary represents appropriate tax policy and appropriate application of conduit principles.

The problems raised by trying to determine the correct “principal amount” in the case of Category 1 and Category 2 transactions and in the lack of a corollary to the NID Limitation Rule in the case of all Category 1 and Category 2 transactions not subject to that rule are also a further illustration of the ill-placed nature of these two new categories within the Anti-Conduit Regulations under section 881.

If these categories are retained, then a rule comparable to the NID Limitation should be added and it should utilize a measure of the foreign tax benefit derived by the issuer on a portion of its equity equal in value to the financed entity's financing transaction.

We also think more guidance is required on how the “character” rule should apply upon recharacterization. Under the Existing Anti-Conduit Regulations, if a financing arrangement is recharacterized generally the resulting direct instrument is characterized based on the financing party-intermediate entity leg unless that leg would give rise to non-deductible payments, in which case character is governed by the intermediate entity-financed party leg. In the first instance, if that approach is retained the rules will need to define how that applies to Category 1 and Category 2 transactions. Do U.S. tax law principles govern the question of deductibility for this purpose (notwithstanding the fact that applicability of the conduit rules is premised on deductibility for foreign tax purposes)? It may be preferable simply to provide in that case that the character of the U.S. leg governs the recharacterized transaction. Character is also an issue in the case of certain more classic hybrid instruments where, for example, the intermediate entity-financed party transaction is a lease or license and the financing party-intermediate entity transaction is treated as equity for U.S. tax purposes and interest-bearing debt for foreign tax purposes; in determining the eligibility of the financing party for benefits under its own country's treaty, there may be a different result depending on whether it is treated as receiving rent or royalties as opposed to interest, and if the financing party would be entitled to the portfolio interest exemption on receipt of interest, interest characterization presumably would avoid application of the anti-conduit rules altogether.

5. Revise the Proposed Effective Date

The Proposed Regulation is proposed to apply to payments made on or after the date that final regulations are published in the Federal Register. The approach adopted by the Proposed Regulation, for the reasons explained in this Report, represents a significant departure from the conduit approach that applies under the Existing Anti-Conduit Regulations and common law. Once finalized, the rules would apply to financing structures that taxpayers put in place prior to any indication that they would be potentially open to challenge as conduits. Significantly, the Proposed Regulation goes beyond even what the most sophisticated taxpayers might have expected based on the 2008 Preamble's discussion of whether the Anti-Conduit Regulations should be expanded to apply to financing transactions involving “hybrid transactions.”

It is arguably harsh to apply the rules to all payments, including under existing financing structures, after the Proposed Regulation is finalized (assuming it is finalized in much the current form). That concern would have far less force if our primary recommendation to limit the scope of the Proposed Regulation to classic “hybrid equity” (foreshadowed by the 2008 Preamble) and more highly structured transactions involving special purpose entities.

However, if the broader approach of the Proposed Regulation is retained, we recommend consideration of some form of grandfathering for payments under existing financing arrangements. It would be reasonable to define an existing arrangement as one existing prior to the publication of the Proposed Regulation (and not subsequently materially modified). If Treasury and the IRS are reluctant to permanently grandfather existing arrangements, the exception for payments under existing structures could be limited to some period of time, such as three to seven years, to allow most existing arrangements to “run off” and give time for others to be restructured. Even if the equity leg of the financing arrangements is “permanent,” in unrelated party capital markets transactions it would be unusual for the inbound financing leg of a transaction to be of very long duration.

6. Carve Out Partnership Equity Unless Partnership is Hybrid Claiming Treaty Benefits

The rules defining when equity may be treated as a financing transaction include partnership equity. With one exception, we see no need to treat partnership equity as a financing for conduit purposes, because other than in the case of a treaty hybrid entity (which the intermediate country treats as a tax resident corporation and which claims treaty benefit on that basis despite being a pass-through for other U.S. federal tax purposes), the Withholding Regulations already effectively look through partnerships.⁶³

Although we do not think a foreign country pass-through regime generally should be viewed as resulting in a Category 2 benefit, depending on the mechanics of such a regime and given the breadth of the definition, withholding agents may not be certain of this. For example, if a foreign country had a regime similar to section 1446 where the partnership must withhold tax on a non-resident partner's share of income (which tax payment is a refundable credit to the partner), could a partner be viewed as receiving a Category 2 credit?⁶⁴ If a withholding agent must withhold based on an assumed conduit recharacterization, would this reach the same result as under the Withholding Regulations' “look through” regime? In many cases it may, but as the conduit recharacterization by default is based on the character of the financing (partnership equity leg) unless it is clear that a partnership interest is not the equivalent of a deductible instrument this may be debatable. There also is no clear way to recharacterize the financings as a direct financing when the partners do not share pro rata in income. We see no compelling policy reason for this complexity when the withholding rules look-through a partnership applying existing developed section 704 principles to determine how underlying income is allocable to the partners.

If the intermediate entity is a treaty hybrid (i.e., corporation under foreign law, partnership under U.S. law), we understand under the Base Erosion rationale why recharacterization would seem appropriate. Accordingly, if the approach of the Proposed Regulation is adopted, partnership equity of a treaty hybrid should be treated as a financing transaction, assuming the other conditions of such treatment are met, only if the partnership is a treaty hybrid entity.

7. Eliminate or Significantly Clarify the PE Portion of Category 1

Even if the Base Erosion rationale is accepted as the appropriate governing principle for the Proposed Regulation, we do not think it so clear that Category 1 should take into account a tax benefit granted to the equity issuer by a country that is not its country of residence but is instead “ a country in which the equity issuer has a taxable presence, such as a PE to which a payment on a financing transaction is attributable”.⁶⁵ There are several reasons for this.

First, treaty benefits are bilateral and intended to prevent double taxation only as between the United States and the foreign entity's country of residence (which, here, would be the country of residence of the intermediate entity that is the equity issuer). Indeed, if a third-country PE of an intermediate entity receives a U.S. source payment and is subject to tax in that third country to the same extent as a resident of the third country would have been, neither the intermediate entity nor the PE could claim benefits under a treaty between that third country and the U.S.. Because the third country's treaty with the U.S. would not apply, it could not be the treaty being inappropriately utilized.

As for the treaty between the intermediate entity's residence country and the U.S., benefits under that treaty would be in no way dependent upon whether there was third country taxation or third country base erosion. Thus, it seems misplaced to deny residence country treaty benefits based upon tax consequences in the third country.

The Proposed Regulation is also unclear as to whether the deduction in the third country would need to relate to equity issued by the residence country entity to a shareholder of that entity or could alternatively also be a deduction for a deemed or imputed payment from the branch to the home office. The text of the Proposed Regulation suggests the former,⁶⁶ while the Preamble suggest the latter.⁶⁷

As a practical matter, it seems extremely unlikely that the former would ever exist under any third country's tax law. The latter is problematic for a different reason: it would be treating a deduction permitted under the transfer pricing rules applicable in the third country (its version of the Authorized OECD Approach and of the U.S. rules under Code section 882) as a nefarious abusive base-eroding deduction, which we view as completely unfounded. This would also again have no relationship to the claiming of benefits under a treaty between the U.S. and the residence country. Moreover, it would, as a practical matter, in all likelihood apply whenever a branch lends to a U.S. entity (unless the branch jurisdiction does not have rules that are similar to the Authorized OECD approach, which is unlikely). Finally, the existence of such imputed distributions from a branch to a home office in no way supports treating the holders of the intermediary's equity as being engaged in a financing transaction with the intermediary entity (which is what Category 1 does).We have considered that under some countries' tax regimes the residence country tax may not apply to income of a third country PE that is subject to tax in the third country. In that case, base erosion at the level of the PE in some sense could be viewed as having been imported into the residence country insofar as the exemption assumes taxability of the PE in the third country. Triangular concerns such as this are addressed by the “triangular branch provisions” of Article 1, Paragraph 8 of the 2016 Model Treaty.⁶⁸ We understand that those rules may turn on the effective tax rate in the PE country and it could be appropriate to take into account notional deductions tied to the entity's equity in evaluating that effective rate in applying those rules. We strongly believe any such concern should be addressed in the context of negotiated “triangular branch” treaty provisions, rather than through the mechanism of unilateral U.S. regulations. Even if one embraces the Base Erosion rationale, if the relevant treaty does not include this type of triangular provision, it is not appropriate to impose one through the “back door” of the conduit rules.

We have also considered the possibility that the PE portion of Category 1 might be viewed as applying if the intermediary's “taxable presence” in the third country is actually a separate legal entity (but a PE from the U.S. perspective because it has made a ”check the box” election). If it did apply to that fact pattern, the intermediary's residence country treaty benefits would not be available because of section 894(c) (assuming, as would almost certainly be the case, that the residence country treated the hybrid entity as a separate legal entity). It is possible that the hybrid entity might be eligible for benefits under a treaty between the U.S. and the third country, but in that case the hybrid would be the intermediary and the Anti-Conduit Regulations would apply to that entity with respect to that treaty. In that case, our recommendations that the Proposed Regulation be limited to classic hybrid instruments and structured special purpose entity transactions would apply to the hybrid entity.

For all the foregoing reasons, we recommend that the PE portion of Category 1 be eliminated. If it is retained, however, we believe that significant clarification and additional guidance regarding how it applies will be required.

The “amount” of the financing transaction is extremely important in determining the effect of the financing arrangement being a conduit financing arrangement. This is because the amount of the financed entity's payment that is recharacterized as being paid to the financing entity is dependent upon the principal amount of each of the financing transactions. Accordingly, it is imperative to know if the relevant financing transaction is stock issued by the intermediate entity to a shareholder or, as the Preamble would indicate, “deemed equity or capital” of the PE; and, if it is the latter, how the principal amount of that deemed equity or capital would be determined.

It is equally important to be able to identify “the rate used to compute the issuer's notional interest deduction” because this also is part of the computation of the portion of the financed entity's payment that is recharacterized.⁶⁹ It is not clear how this “rate” is identified if the PE is allowed a deduction with respect to deemed equity or capital and that deduction is not computed by applying a percentage to the amount of that deemed equity or capital.

Finally, if our primary recommendation is adopted, but the PE rule is retained, the PE rule at least should be limited to apply only when the PE is the equivalent of a special purpose finance entity whose only activities are those relating to the financing arrangement.