Tax Group Suggests Changes to Proposed Hybrid Regs

February 26, 2019

CC:PA:LPD:PR (REG–104352–18)
Room 5203
Internal Revenue Service
P.O. Box 7604
Ben Franklin Station
Washington, DC 20044

Re: Comments on proposed §§ 245A(e) and 267A regulations in REG–104352–18

Dear Sirs or Madams,

The Silicon Valley Tax Directors Group (“SVTDG”) hereby submits these comments on the above-referenced proposed regulations issued under §§ 245A(e) and 267A of the Internal Revenue Code of 1986, as amended, in REG–104352–18, 83 Fed. Reg. 67612 (December 28, 2018) (the “Proposed Regs”). SVTDG members are listed in the Appendix to this letter.

Sincerely,

Robert F. Johnson
Co-Chair, Silicon Valley Tax Directors Group
Capitola, CA

---

I. INTRODUCTION AND SUMMARY

A. Background on the Silicon Valley Tax Directors Group

The SVTDG represents U.S. high technology companies with a significant presence in Silicon Valley, that are dependent on R&D and worldwide sales to remain competitive. The SVTDG promotes sound, long-term tax policies that allow the U.S. high tech technology industry to continue to be innovative and successful in the global marketplace.

B. Summary of recommendations — changes that should be made to the Proposed Regs

We recommend that Treasury and the IRS make certain specific changes to, and reconsider certain aspects, of the Proposed Regs. Here we summarize our main recommendations.

[A] Prop. § 1.245A(e)-1 should be narrowed and clarified

Prop. § 1.245A(e)-1 provides rules disallowing the § 245A dividends received deduction for hybrid dividends, which are dividends for which the paying CFC was allowed a deduction under local law. To track hybrid deductions, U.S. shareholders are required to maintain hybrid deduction accounts for each share of CFC stock. The Proposed Regs have a retroactive rule whereby a dividend paid on non-hybrid shares (i.e., shares treated as equity in both jurisdictions) can still be treated as a hybrid dividend if the U.S. shareholder has a hybrid deduction account for hybrid shares in the CFC. Because the rules track hybrid deductions on a share-by-share basis, dividends should only be treated as hybrid dividends if paid on shares with hybrid deduction accounts. We therefore recommend this rule be eliminated. If the rule is retained, we recommend it be prospective only, as applying it retroactively to taxpayers who have already paid dividends would be unfair.

There's a tiered hybrid dividend rule in Prop. § 1.245A(e)-1(c) that treats as subpart F income a hybrid dividend from one CFC to another. We suggest a minor clarification of the language in that rule to insure — consistent with our understanding of Treasury and the IRS's intent — that the rule require an income inclusion only for domestic corporations.

Prop. § 1.245A(e)-1(d)(5) has rules for determining hybrid dividend accounts upon a transfer of stock in a CFC. These determinations are done as of the close of the date of the transfer, applying principles of § 1.1502-76(b)(2)(ii). We suggest Treasury and the IRS include an example illustrating this rule, particularly in the case of post-transfer distributions.

[B] Prop. § 1.267A-2 is overbroad, applying to transactions that shouldn't result in deduction disallowance

Prop. § 1.267A-2(a)(2) defines “hybrid transaction,” a critical term that determines the scope of the disallowance rules. While the Proposed Regs generally define hybrid transactions as those that give rise to deductions without a corresponding income inclusion (referred to as a “D/NI” outcome), the Proposed Regs also include in their scope other types of transactions. One such type of comprises transactions that result in temporary timing differences rather than permanent non-inclusions. The Proposed Regs treat as a hybrid transaction any transaction in which the income inclusion occurs more than 36 months after the end of the taxable year in which the corresponding deduction is taken. This rule is unnecessary and unfair to taxpayers because the payment ultimately is included in the recipient's income — that is, the rule leads to “inclusion but no deduction” outcomes. The rule will create problems if the payer is on the accrual method and the recipient is on the cash method, particularly for agreements that don't require payments within the 36-month window. We therefore suggest the 36-month rule be removed. If the rule is retained, we suggest extending the “safe harbor” period to around 10 years. Deferral of income for three years doesn't constitute long-term deferral and isn't equivalent to non-inclusion of income.

Prop. § 1.267-2(a)(2) applies if a party gets a deduction for amounts treated as interest or royalties in the U.S. (or that party's jurisdiction) but not “so treated” in the recipient's jurisdiction. We suggest that Treasury and the IRS clarify that transactions that are treated as royalties for the payer and sales for the recipient are not hybrid transactions. The recipient still includes the payment in income, and the differing characterization (sale vs. license) isn't cause for denying a deduction.

There's a deemed branch payment rule in Prop. § 1.267A-2(c) that disallows deductions for amounts of interest and royalties deemed paid, pursuant to an income tax treaty, by a PE to its home office if the deemed payment is taken into account by the home office under the home office's tax law. This rule is inconsistent with U.S. treaty obligations, which require treating a PE as if it were a separate corporation. We therefore recommend eliminating the deemed branch payments rule.

[C] The rules in Prop. § 1.267A-3 relating to income inclusions should be broadened in some respects and narrowed in others

Prop. § 1.267A-5(a)(19), in defining “specified recipient,” states that there may be more than one specified recipient with respect to a specified payment. This has the effect of denying a deduction for a payment if less than all specified recipients include the payment in income. An example in Prop. § 1.267A-6(c) Example 1(iii) illustrates this point. It involves a payment to an entity that includes the payment in income but that is fiscally transparent for the entity's owner's jurisdiction. The deduction is denied if the owner doesn't also take the payment into income. This rule goes beyond neutralizing a D/NI outcome by requiring a double (or more) inclusion. This blanket rule — which applies regardless of the tax rates of the specified recipients — should be removed. If Treasury and the IRS are concerned with the potential insertion of low-tax intermediate entities, an anti-abuse rule should suffice.

The Proposed Regs don't treat payments subject to withholding tax as included in income (unless otherwise included in income), although the preamble requests comments on the issue.¹ We recommend Treasury and the IRS reconsider. Payments subject to withholding are subject to tax; a D/NI outcome is therefore avoided. Moreover, the withholding tax rate (30 percent in the U.S.) can be greater than the ordinary income rate (21 percent in the U.S.).

The proposed regulations also treat a payment as not included in income if it's offset by an exemption, exclusion, deduction, or credit “particular to such type of payment.”² Generally applicable deductions, such as for depreciation, or other tax attributes don't prevent an income inclusion. We recommend clarifying that other forms of basis recovery — such as upon disposition by sale or upon repayment of principal for a debt instrument — are likewise treated as generally applicable deductions or tax attributes.

We also suggest a technical change to Prop. § 1.267A-3(a)(1)(i), which provides that a specified payment is included in income if it's included “at the full marginal rate imposed on ordinary income.” Some jurisdictions impose tax on certain items at higher than ordinary income rates. We therefore suggest this regulation apply to inclusions “at a rate equal to or higher than the full marginal rate imposed on ordinary income.”

[D] Miscellaneous rules — in particular, the important mismatch rule of Prop. § 1.267A-4 should be reconsidered

The imported mismatch rule of Prop. § 1.267A-4 disallows deductions for non-hybrid payments if the income attributable to the payment is directly or indirectly offset by a hybrid deduction. This provision applies the disallowance rule to non-hybrid transactions if the income inclusion is offset by a hybrid deduction on a purely foreign-to-foreign transaction. We believe this rule will prove burdensome and difficult to administer, as it will require taxpayers to track numerous foreign transactions and match them with non-hybrid transactions. The rule also significantly expands the scope of the statute. We therefore recommend removing it.

Prop. § 1.267A-5(b)(1) provides that § 267A applies “after the application of any other applicable provisions of the Code and regulations.” But Prop. §§ 1.163(j)-3(b)(1) and (2) indicate that § 267A applies before § 163(j). We recommend Prop. § 1.267A-5(b)(1) be modified to provide that § 267A applies before § 163(j). This result is consistent with the proposed § 163(j) regulations and is sensible: otherwise a deduction that will be disallowed under § 267A could “consume” a taxpayer's § 163(j) limitation.

II. SVTDG CONCERNS WITH, AND RECOMMENDATIONS FOR CHANGES TO, THE PROPOSED REGS

A. Prop. § 1.245A(e)-1 should be narrowed and clarified

1. Background

Prop. § 1.245A(e)-1 disallows the § 245A dividends received deduction for “hybrid dividends.” Under § 245A(e), hybrid dividends are dividends for which a § 245A(a) dividends received deduction would be allowed but for § 245A(e), and for which the CFC is or was allowed a deduction or other tax benefit under relevant foreign tax law (such deduction a “hybrid deduction”). Because the dividend and hybrid deduction may occur in different taxable years — and may not arise from the same payment — U.S. shareholders are required to maintain “hybrid deduction accounts” to track hybrid deductions.³ A dividend paid by a CFC to a shareholder that has a hybrid deduction account is treated as a hybrid dividend to the extent of the shareholder's balance in all its hybrid deduction accounts with respect to each share of stock of the CFC.⁴

2. Prop. § 1.245A(e)-1 should be prospective only with respect to disallowing dividends on non-hybrid shares

Hybrid deduction accounts are maintained with respect to each share of CFC stock.⁵ A dividend paid by a CFC to a U.S. shareholder that has a hybrid deduction account with respect to the CFC can be a hybrid dividend, even if the dividend is paid on a share that hasn't had any hybrid deductions allocated to it. In Prop. § 1.245A(e)-1(g) Example 1, dividends with respect to both hybrid shares (treated as debt under the CFC's tax laws) and non-hybrid shares (treated as equity for both U.S. and foreign purposes) are hybrid dividends, as long as the shareholder has a large enough hybrid deduction account with respect to the CFC. All of Prop. § 1.245A(e)-1 applies to distributions made after December 31, 2017.

Because hybrid deductions are allocated on a share-by-share basis, dividends should be considered paid out of hybrid deduction accounts only to the extent paid with respect to those shares that have been allocated hybrid deductions. If this approach isn't adopted, the proposed rule potentially treating dividends on non-hybrid shares as hybrid dividends should be made prospective only. This rule isn't in § 245A(e), and taxpayers weren't anticipating such a rule. Moreover, the proposed rule can snare non-abusive transactions: not all payments of dividends on non-hybrid shares are intended to circumvent the Proposed Regs. Thus, some taxpayers may have engaged in non-abusive transactions without anticipating this rule, and it would be inordinately harsh to penalize them by denying a § 245A deduction. This rule should therefore be prospective only.

3. The tiered hybrid dividend rule of Prop. § 1.245A(e)-1(c) should be clarified to apply only to domestic corporations, not domestic partnerships

Under Prop. § 1.245A(e)-1(c), if a CFC gets a tiered hybrid dividend (that is, a dividend that would be a hybrid dividend if the receiving corporation were a domestic corporation) from another CFC, and a domestic corporation is a U.S. shareholder with respect to both CFCs, the tiered hybrid dividend is treated as subpart F income of the receiving CFC and the U.S. shareholder must include in gross in income an amount equal to its pro rata share of such subpart F income. Evidently this rule applies only to U.S. shareholders that are domestic corporations, not domestic partnerships. A partnership could have non-corporate partners, who wouldn't be entitled to a § 245A deduction — which applies only to corporations — and thus shouldn't have subpart F income under the tiered hybrid dividend rule. We ask that Treasury and the IRS clarify that “[t]he United States shareholder” referenced in Prop. § 1.245(e)-1(c)(ii) is the domestic corporation referred to in the introductory language.

4. The rules in Prop. § 1.245A(e)-1(d)(5), relating to the determination of hybrid deduction accounts upon a transfer of stock, should be clarified

Prop. § 1.245A(e)-1(d)(5) provides that if there's a transfer of stock with an associated hybrid deduction account, the determinations and adjustments necessary with respect to that account are done as of the close of the date of the transfer. The principles of § 1.1502-76(b)(2)(ii), which allow an electable, ratable allocation method to be used, apply.

The principles of § 1.1502-76(b)(2)(ii) appear to apply in situations in which an accrual (for example, of interest expense) occurs for a year, but stock is transferred during the year. This rule should be clarified as to how it applies in situations in which a dividend is paid on the transferred stock during the portion of the taxable year after the transaction. That is, do post-transfer distributions in the year of the transfer have any effect on the hybrid dividend accounts?⁶ It would be helpful if an example were provided applying Prop. § 1.245A(e)-1(d)(5) both to the accrual of income and to post-transfer distributions.

In addition, Treasury and the IRS should clarify that a § 338(g) election eliminates the hybrid deduction account. If a § 338(g) election is made, a purchaser of stock is treated as purchasing assets, not stock. Because a hybrid deduction account only attaches to stock, it's appropriate for such account to disappear when only assets are purchased.

B. Comments related to Prop. § 1.267A-2 and the transactions to which the proposed regulations apply

1. Background

Section 267A disallows deductions for “disqualified related party amounts” paid pursuant to a hybrid transaction or by or to a hybrid entity.⁷ The statute and the § 267A Proposed Regs are targeted at financial instruments and transactions that result in D/NI outcomes. The Proposed Regs refer to these transactions as “hybrid transactions” and disallow deductions for certain payments (“disqualified hybrid amounts”) under such transactions.⁸ The definition of “hybrid transaction” in Prop. § 1.267A-2(a)(2) is therefore critical and the subject of several comments below. Other key definitions include a “specified party” (a tax resident of the U. S., a CFC that has at least one 10 percent or greater U. S. shareholder, and a U.S. taxable branch);⁹ a “specified payment” (the amount paid or accrued by the specified party);¹⁰ and “specified recipient” (the tax resident that derives, under its tax law, a specified payment or a taxable branch to which the specified payment is attributable).¹¹ Stated simply, the Proposed Regs apply to deny deductions only with respect to certain specified payments¹² made by specified parties.

2. The 36-month rule in Prop. §§ 1.267A-2(a)(2) and 1.267A-3(a)(1)(i) should be eliminated

Prop. § 1.267A-2(a)(2) defines a “hybrid transaction” as “any transaction, series of transactions, agreement, or instrument one or more payments with respect to which are treated as interest or royalties for U.S. tax purposes but are not so treated for purposes of the tax law of a specified recipient of the payment.” Prop. § 1.267A-2(a)(2) also deems a specified payment as made pursuant to a hybrid transaction if the taxable year in which a specified recipient recognizes the payment under its tax law ends more than 36 months after the taxable year in which the specified party would be allowed a deduction for the payment under U.S. tax law. A similar provision in Prop. § 1.267A-3(a)(1)(i) treats a specified payment as not included in income of the recipient unless the recipient will include the payment in income during a taxable year that ends no more than 36 months after the end of the specified party's taxable year (the “36-month period” and each such rule the “36-month rule”).

We recommend the 36-month rule should either be eliminated or the period should be lengthened. The rule itself appears unnecessary and burdensome. The rule for temporary timing differences isn't necessary because all amounts will eventually be included in income. This leads to harsh results for taxpayers as it permanently disallows a deduction due to temporary timing differences, leading to “inclusion but no deduction” outcomes. Under the rule as proposed, if the recipient includes the payment in income after the end of the 36-month period — even by one day — the payer's deduction remains disallowed. In addition, the rule will be difficult to administer. Taxpayers will be required to estimate whether payments will be made within the 36-month period, and then ensure such payments are in fact made with the 36-month period.

The 36-month rule therefore expands the hybrid deduction disallowance rule from permanent D/NI situations to temporary timing differences. This can create problems in particular if the payer is on the accrual method of accounting and the recipient is on the cash method. Taxpayers may accrue payments under an agreement that doesn't require payment within the 36-month period. It's unclear how the 36-month rule will apply in situations in which the relevant agreement doesn't require payment within the 36-month period. If Treasury and the IRS choose not to eliminate the 36-month rule, or lengthen the period as recommended above, we recommend that taxpayers be permitted to rely on their expectation that payment be made within the 36-month period. For example, taxpayers could certify that they intend to make payment within the 36-month period.

The 36-month rule isn't in the statute. While, as noted in the preamble to the Proposed Regs, the Senate Committee on Finance's Explanation of the Bill — although not the TCJA Conference Report — expresses concern with “long-term deferral,”¹³ a 36-month period is overly restrictive. Long-term deferral can be thought of as indefinite deferral, such as taxpayers were able to achieve on their foreign earnings prior to the enactment of subpart F.¹⁴ For debt instruments, long-term can mean 10 years or longer.¹⁵ Thus, if a rule applying to temporary timing differences is retained, it should apply only to timing differences that are truly long-term — i.e., those lasting 10 years or more.

3. Treasury and the IRS should clarify that the definition of “hybrid transaction” doesn't apply to transactions treated as a sale in one jurisdiction and a license in the U.S. (or other jurisdiction)

The definition of “hybrid transaction” should also be modified to prevent application to transactions that are treated as a license or lease in one jurisdiction and a sale in the other. As noted above, a “hybrid transaction” is a transaction in which “one or more payments with respect to which are treated as interest or royalties for U.S. tax purposes but are not so treated for purposes of the tax law of a specified recipient of the payment.”¹⁶ A hybrid transaction could therefore include a transaction treated as a license or lease for U.S. purposes but a sale for a foreign jurisdiction's purposes (a “sale/license hybrid”). Including sale/license hybrids in the definition of “hybrid transaction” makes the regulation over-broad by including in its ambit transactions that aren't abusive. Sale/license hybrids don't present the same potential for abuse as other hybrid transactions because the payment is still included in the specified recipient's income. Whether the recipient includes the payment as sales income or royalty income shouldn't be relevant for determining whether a deduction for the payment is allowed.

4. The deemed branch payments rule of Prop. § 1.267A-2(c) should be withdrawn because the rule is inconsistent with U.S. treaty obligations

Under Prop. § 1.267A-2(c), if a specified payment is a “deemed branch payment,” the payment is a disqualified hybrid amount if the tax law of the home office provides an exclusion or exemption for income attributable to the branch. Deemed branch payments are payments deemed made, under an income tax treaty, by a U.S. branch to its home office, but the payment isn't regarded or otherwise taken into account under the home office's tax law. An example is an amount allowed as a deduction in computing the business profits of a U.S. PE for a deemed royalty paid to the home office for use of the home office's intellectual property.

The proposed rule conflicts with U.S. treaty obligations, which treat a PE as if it were a separate entity. For example, under Article 7(1) of the United States Model Income Tax Convention (February 17, 2016), a contracting state can tax the profits attributable to a PE in that contracting state. Under Article 7(2), such profits attributable to a PE are the profits the PE “might be expected to make, in particular in its dealings with other parts of the enterprise, if it were a separate and independent enterprise engaged in the same or similar activities under the same or similar conditions.” Similar provisions are in most major U.S. treaties. For a PE to be treated as a separate enterprise, it must be allowed the same deductions a separate entity would be allowed. Disallowing a deduction for royalty and interest payments deemed made by a PE to its home office is inconsistent with the obligation under income tax treaties to treat the PE as a separate enterprise for purposes of computing the profits attributable to the PE. We accordingly recommend the rule in Prop. § 1.267A-2(c) be withdrawn.

C. Comments and recommendations related to inclusions in income

1. Background

A specified payment is only a “disqualified hybrid amount” to the extent a specified recipient of the payment doesn't include the payment in income (to such extent, a “no-inclusion”).¹⁷ Thus, whether a specified payment is included in income of a specified recipient is crucial to determining whether there's a prohibited D/NI and thus whether a deduction for such payment is allowed. Prop. § 1.267A-3 provides rules as to whether a specified recipient takes a specified payment into income. Under Prop. § 1.267A-3(a)(1), a specified payment is included in income of a tax resident or taxable branch if the 36-month rule is met and “[t]he payment is not reduced or offset by an exemption, exclusion, deduction, credit (other than for withholding tax imposed on the payment), or other similar relief particular to such type of payment.”

2. Treasury and the IRS should eliminate the third-country rule in Prop. §§ 1.267A-5(a)(19) and 1.267A-6(c) Example 1(iii)

Deductions for specified payments can be disallowed if the payment isn't included in income of a specified recipient.¹⁸ Prop. § 1.267A-5(a)(19) states that there may be more than one specified recipient with respect to a specified payment. This means that if even one specified recipient includes the specified payment in income, the deduction for the payment can still be denied if another specified recipient doesn't include the payment in income. The regulations include an example, Prop. § 1.267A-6(c) Example 1(iii), to illustrate this point. In the example, US1, a domestic corporation, makes a payment to FZ, a Country Z entity, under an instrument. FZ is wholly owned by FX, a Country X corporation, and FZ is fiscally transparent under Country X's law but not Country Z's law. Because FX doesn't include the amount in income, the fact that FZ includes the payment in its income has no bearing on the disallowance of US1's deduction. This is true regardless of whether FZ has a lower tax rate than FX. This concept also appears for reverse hybrids in Prop. § 1.267A-6(c)(5) Example 5(iv).

This rule should be withdrawn. The rule is overbroad and applies to situations in which payment to an intermediate entity isn't made for tax avoidance purposes. In the example, FZ could have a higher tax rate than FX. As long as one specified recipient includes the specified payment in income, a D/NI outcome is avoided. A double (or more) inclusion shouldn't be required to get a deduction. This rule is also contrary to U.S. tax principles in § 894(c), which generally allows treaty benefits to a recipient entity as long as the item of income is taxed to the recipient under the tax laws of the recipient's jurisdiction. This rule moreover isn't in § 267A, nor is targeted in the grant of regulatory authority in § 267A(e). If Treasury and the IRS are concerned with taxpayers avoiding § 267A by inserting low-tax intermediate entities, this concern is more appropriately addressed through an anti-abuse rule or rate test rather than a blanket disallowance rule that can apply to even high-tax entities.

3. Treasury and the IRS should modify Prop. § 1.267A-3(b) to treat payments subject to source withholding tax as included in the income of the recipient

Under Prop. § 1.267A-3(b), specified payments aren't disqualified hybrid amounts to the extent the amounts are included or includible in a U.S. tax resident's or U.S. taxable branch's income. The regulations don't treat specified payments subject to withholding tax as included in income (unless the payment is otherwise included in income), and the preamble specifically provides that “[s]ource-based withholding tax imposed by the United States (or any other country) on disqualified hybrid amounts does not neutralize the D/NI outcome and therefore does not reduce or otherwise affect disqualified hybrid amounts.”¹⁹ “Nevertheless, the Treasury Department and the IRS request[ed] comments on the interaction of the proposed regulations with withholding taxes and whether, and the extent to which, there should be special rules under section 267A when withholding taxes are imposed in connection with a specified payment.”²⁰

We recommend Treasury and the IRS treat specified payments subject to source withholding tax as included in the income of a U.S. tax resident or U.S. taxable branch for purposes of determining whether such payments are disqualified hybrid amounts. Section 267A was enacted to target hybrid arrangements that “'exploit differences in the tax treatment of a transaction or entity under the laws of two or more tax jurisdictions to achieve double non-taxation, including long-term deferral.'”²¹ A specified payment subject to withholding tax doesn't achieve double non-taxation; such payment may in fact be subject to higher taxation — at a 30 percent U.S. withholding rate — than under the U.S. corporate income rate of 21 percent. In other words, the proposed § 267A regulations aren't needed to neutralize the D/NI result because the withholding tax has already done so.

We recommend the final regulations provide that to the extent tax has been deducted and withheld in the source country on a specified payment, such payment doesn't constitute a disqualified hybrid amount. Such a result would be consistent with the proposed base erosion and anti-abuse tax rules, which exclude from the definition of “base erosion tax benefit” such benefits attributable to base erosion payments upon which tax has been deducted and withheld under §§ 1441 or 1442.²²

4. The definition of “income inclusions” in Prop. § 1.267A-3(a)(1) should be clarified

Under Prop. § 1.267A-3(a)(1), a specified payment is included in income of a tax resident or taxable branch if the 36-month rule is met and “[t]he payment is not reduced or offset by an exemption, exclusion, deduction, credit (other than for withholding tax imposed on the payment), or other similar relief particular to such type of payment.” “A specified payment is not considered reduced or offset by a deduction or other similar relief particular to the type of payment if it is offset by a generally applicable deduction or other tax attribute, such as a deduction for depreciation or a net operating loss.”²³

The rules should clarify that recovery of basis in the disposition of a capital asset constitutes a “generally applicable deduction or other tax attribute.” Depreciation deductions are mechanisms for recovery of basis over time, and recovery of basis upon disposition of a capital asset should likewise be a “generally applicable deduction or other tax attribute.” Similarly, a payment treated as return of principal on repayment of debt — and thus not subject to tax — should likewise be a “generally applicable” tax attribute and not an exclusion particular to a certain type of payment. Return of principal is also a type of basis recovery.

5. The rule in Prop. § 1.267A-3(a)(1) for income inclusions should clarify the tax rate at which amounts are treated as included in income

Prop. § 1.267A-3(a)(1)(i) provides that a specified payment is included in the income of a tax resident or taxable branch if, in part, the resident or branch includes the payment in its income or tax base “at the full marginal rate imposed on ordinary income.”²⁴ (Emphasis added.) Because some jurisdictions, such as Ireland, impose tax on certain types of income (e.g., capital gains) at rates higher than the ordinary income rate, items subject to these higher tax rates could technically be excluded by the regulatory language, which applies only to items taxed at the ordinary income rate. We therefore request the language in Prop. § 1.267-3(a)(1)(i) be changed to “at a rate equal to or higher than the full marginal rate imposed on ordinary income.”

D. Comments and recommendations regarding other provisions

1. The imported mismatch rule of Prop. § 1.267A-4 should be reconsidered

Through Prop. § 1.267A-4, the proposed § 267A rules apply not only to hybrid arrangements involving the U.S. but also to hybrid arrangements between two foreign corporations outside the U.S. taxing jurisdiction if that arrangement is “imported” to the U.S. jurisdiction through a non-hybrid arrangement. Prop. § 1.267A-4 disallows deductions for “disqualified imported mismatch amounts.” A disqualified imported mismatch amount is a non-hybrid specified payment for which the income attributable to the payment is directly or indirectly offset by a hybrid deduction of a foreign tax resident or taxable branch.²⁵

We believe this rule will prove difficult to administer. Taxpayers will be required to track numerous purely foreign transactions and match a U.S.-involved non-hybrid transaction with them. Taxpayers may not have access to this information, especially when considering the broad application of the indirect offset rules.²⁶ The preamble recognizes that the chosen imported mismatch rule “involves greater complexity than the alternatives.”²⁷ This rule also is overbroad and exceeds the authority granted by the statute.

2. The ordering rule of Prop. § 1.267A-5(b)(1) should be clarified

Prop. § 1.267A-5(b)(1) provides that “[e]xcept as otherwise provided in the Code or in regulations under 26 CFR part 1, section 267A applies to a specified payment after the application of any other applicable provisions of the Code and regulations under 26 CFR part 1” (emphasis added). But Prop. § 1.163(j)-3(b)(1) and (2) have a similar rule, indicating that the §163(j) rules apply after the § 267A rules.²⁸ These ordering rules should be clarified to state that the proposed § 163(j) regulations apply after the § 267A rules. Interest deductions that are disallowed under § 267A shouldn't be included in determining the amount of interest subject to the § 163(j) limitation. That is, if § 163(j) were to apply before § 267A, it's unclear that interest deductions disallowed would be the same interest deductions disallowed under § 267A. For example, assume a corporation has $110 in interest expense deductions for a taxable year. Under §163(j), $100 of the $110 is allowed as a deduction and $10 is disallowed and carried forward. Assume further that $20 of the interest expense would be disallowed under § 267A. It isn't clear that the $10 interest expense disallowed under § 163(j) is part of the $20 interest expense disallowed under § 267A. In other words, the corporation should be allowed for the year an interest deduction of $90 ($110 – $20) rather than $80 ($110 – $10 – $20). It isn't clear that this result obtains if § 163(j) were applied prior to § 267A. We therefore recommend the ordering rule be clarified to apply § 267A first.

---

Appendix — SVTDG Membership

Accenture

Activision Blizzard

Acxiom

Adobe

Agilent

Airbnb

Amazon

AMD

Ancestry.com

Apple

Applied Materials

Aptiv

Arista

Atlassian

Autodesk

Bio-Rad Laboratories

BMC Software

Broadcom Limited

CA Technologies

Cadence

CDK Global

Chegg, Inc.

Cisco Systems, Inc.

Dell Inc.

DocuSign

Dolby Laboratories, Inc.

Dropbox Inc.

eBay

Electronic Arts

Expedia, Inc.

Facebook

FireEye

Fitbit, Inc.

Flex

Fortinet

Genentech

Genesys

Genomic Health

Gilead Sciences, Inc.

GitHub

GLOBALFOUNDRIES

GlobalLogic

Google Inc.

GoPro

Grail, Inc.

Guidewire

Hewlett-Packard Enterprise

HP Inc.

Indeed.com

Informatica

Ingram Micro, Inc.

Integrated Device Technology

Intel

Intuit Inc.

Intuitive Surgical

Keysight Technologies

KLA-Tencor Corporation

Lam Research

Marvell

Maxim Integrated

MaxLinear

Mentor Graphics

Microsoft

NetApp, Inc.

Netflix

NVIDIA

Oath, A Verizon Company

Oracle Corporation

Palo Alto Networks

PayPal

Pivotal Software, Inc.

Pure Storage

Qualcomm

Red Hat Inc.

Ripple Labs, Inc.

Rubrik

salesforce.com

Sanmina-SCI Corporation

Seagate Technology

ServiceNow

Snap, Inc.

Stripe

SurveyMonkey

Symantec Corporation

Synopsys, Inc.

The Cooper Companies

The Walt Disney Company

TiVo Corporation

Trimble, Inc.

Twitter

Uber Technologies

Velodyne LiDAR

Veritas

Visa

VMware

Western Digital

Workday, Inc.

Xilinx, Inc.

Yelp