Menu
Tax Notes logo

NYSBA Tax Section Comments on Proposed PFIC Regs

SEP. 9, 2019

NYSBA Tax Section Comments on Proposed PFIC Regs

DATED SEP. 9, 2019
DOCUMENT ATTRIBUTES

September 9, 2019

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. 1422 — Report on the Proposed “PFIC” Regulations under Sections 1291, 1297 and 1298

Dear Messrs. Kautter, Rettig, and Desmond:

I am pleased to submit our Report No. 1422 commenting on the recently proposed regulations under Sections 1291, 1297 and 1298. The proposed regulations provide important and helpful clarity, and we welcome the issuance of this long-needed guidance.

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,

Deborah L. Paul
Chair
New York State Bar Association
Tax Section

Enclosure

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

Douglas L. Poms
International Tax Counsel
Department of the Treasury

Peter Blessing
Associate Chief Counsel (International)
Internal Revenue Service

Daniel M. McCall
Deputy Associate Chief Counsel (International-Technical)
Internal Revenue Service

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

John J. Merrick
Senior Level Counsel, Office of Associate Chief Counsel (International)
Internal Revenue Service


New York State Bar Association Tax Section

REPORT ON THE PROPOSED “PFIC” REGULATIONS
UNDER SECTIONS 1291, 1297 AND 1298

September 9, 2019


TABLE OF CONTENTS

1. Introduction

2. Summary of Recommendations

3. PFIC Ownership “Attribution”

A. “Top-down” approach for ownership attribution through intermediate partnerships

B. Extending top-down attribution to intermediate non-look through corporations

4. General Income Tests

A. Clarification of cross-references

B. Section 954(h) “banking and financing” exception

C. Gains (Netting)

D. Related Person Look-through Rules

5. General Asset Tests

A. Treatment of Working (Cash) Capital

B. Change in 1297(e) status during year

C. Classifying Shares of Non Look-through Subsidiaries

6. Look-through Subsidiaries

A. General approach

B. Characterization of Look-through Subsidiary Items

C. Elimination of Look-through Subsidiary Items

7. Look-through Approach to Partnerships

8. Stapled Entities

9. Effective Date and Interim Reliance Issues


1. Introduction

This Report1 comments on the proposed regulations (the “Proposed Regulations”) issued on July 11, 2019 by the Department of the Treasury (“Treasury”) and the Internal Revenue Service (“IRS”) under Sections 1291, 1297 and 1298.2 The Proposed Regulations provide welcome clarification and guidance regarding passive foreign investment companies (“PFIC”s). In particular, the Proposed Regulations provide guidance on the application of the tests pursuant to which a foreign corporation may be classified as a PFIC under Section 1297.

Because the Proposed Regulations focus primarily on PFIC classification issues (addressing in detail the income and asset tests, attribution of ownership, look through rules and certain exceptions), our comments do not address other aspects of the PFIC regime that were not covered by the Proposed Regulations. We have addressed some of these other issues in prior Reports.3 A number of aspects of the PFIC regime merit guidance in future regulatory projects. In particular, guidance is urgently needed under the mark-to-market regime of Section 1296 and the qualified electing fund (“QEF”) regime of Section 1293 in the case of structures involving multiple tiers of foreign corporations. This Report also does not address the qualifying insurance corporation exception under Section 1297(f) and the Proposed Regulations.

The Report will not dwell on the many difficulties when applying the PFIC rules to real world situations.4 However, in summary we believe there is broad consensus that: (1) the PFIC regime not infrequently entraps active businesses that likely were not intended by Congress to be treated as PFICs,5 (2) PFIC rule compliance imposes major administrative burdens (in particular, because foreign corporations do not typically maintain books and records based on U.S. tax principles), and (3) major difficulties with compliance arise due to lack of access to necessary information as small shareholders may be unable to compel a foreign corporation to provide factual information needed to assess the active or passive status of corporate income and assets. It is rare in our experience in practice to encounter a company that Congress likely would have viewed as a PFIC but that technically is not a PFIC. That experience informs our comments in this Report.

To the greatest extent possible within the parameters of the statutory scheme, we believe the regulatory rules should prevent inadvertent PFIC classification and limit PFIC status to cases Congress clearly had in mind (i.e., passive investment vehicles). Generally, we believe that the Proposed Regulations provide important and helpful clarity, and we welcome the issuance of this long-needed guidance. However, a few of the proposed rules may not fully achieve these goals and the Report suggests certain revisions and additions to the Proposed Regulations.

2. Summary of Recommendations

The Report's primary recommendations are the following:

A. We support a “top-down” ownership attribution approach to partnerships whereby a partner is not attributed PFIC shares through a non-PFIC corporation owned by the partnership if the proportionate (direct and indirect) interest of the partner in the intermediate non-PFIC corporation is less than 50%. We recommend, however, that to reach that result final regulations adopt an “aggregate” approach to top-down attribution of ownership through partnerships rather than adopting a 50% threshold analogous to that of Section 1298(a)(2)(A). We do not support extending the “top-down” approach to intermediate corporations.

B. We support the general approach of the Proposed Regulations with respect to the exceptions to foreign person holding company income under Section 954(c) that should be incorporated by cross-reference into the definition of “passive income” for purposes of Section 1297(a)(1). Specifically, we commend the incorporation of the active financing exception under Section 954(h) for this purpose, and believe this should continue to be applicable even after the promulgation of final regulations under the active banking exception of Section 1297(b)(2)(A).

C. A foreign corporation should be permitted to net gains and losses of its 25%-or-more owned “look-through” subsidiaries in testing the foreign corporation's PFIC status.

D. We generally support the approach of the Proposed Regulations to the related person look-through rules under Section 1297(b)(2)(C) for dividends, interest, rent and royalties, but with certain modifications in the case of dividends. We believe the active or passive characterization of related person dividends under Section 1297(b)(2)(C) should not be determined by current year earnings and profits. We recommend instead that the dividends be characterized based on the active to passive gross income ratio of the related dividend payor over some rolling look-back period such as three or five years. Alternatively, for those foreign corporations that do maintain earnings and profits accounts based on U.S. principles, we suggest the characterization rule could look first to current year E&P and then accumulated earnings and profits but only for the period the dividend payor and recipient were related persons for purposes of Section 1297(b)(2)(C).

E. We continue to believe the per se treatment of cash as passive even if it is working capital is not appropriate given the purposes of the PFIC rules. We suggest at a minimum that working capital cash be treated as partly active and partly passive in the same proportions as the total gross income (other than any income from the cash assets) of the foreign corporation is active or passive.

F. We recommend that the value method rather than the basis method under Section 1297(e) should apply under the PFIC asset test if the foreign corporation is publicly traded at any time during the year. We also request clarification as to how Section 1297(e) applies to tiered foreign corporations when the parent foreign corporation is publicly traded.

G. We recommend that shares of related corporations (within the meaning of Section 954(d)(3)) that are not also 25%-or-more owned “look through” subsidiaries be characterized based on the active or passive character that a hypothetical distribution of earnings would have been subject to under our approach to characterizing such distributions (namely, based on the ratio of gross active to passive underlying income over a rolling look-back period).

H. We recommend broadening the approach relating to “attribution” of activities conducted by 25%-or-more owned look-through subsidiaries and partnerships. We suggest that this apply not only if the tested foreign corporation owns, directly or indirectly, more than 50% by value of the subsidiary corporation but also if it owns more than 50% by vote, provided in the latter case that the tested foreign corporation materially participates in the same or a complementary line of business to that of the subsidiary corporation. As an alternative, the rules could permit activity attribution but on a scaled back basis proportionate to the ownership percentage in the activity conducting subsidiary.

I. We generally support the approach of the Proposed Regulations to eliminating inter-company items of a tested foreign corporation and its 25%-or-more owned look-through subsidiaries to prevent duplication. However, we recommend eliminating all intercompany dividends received from a 25%-or-more owned look-through subsidiary even if paid out of pre-acquisition earnings and profits. However, if Treasury and the IRS decide that dividends out of pre-acquisition earnings and profits should not be eliminated, we suggest either a “last-in-first-out” approach to attributing dividends to earnings or that the rules limit the length of the period a tested foreign corporation would be required to look back. We recommend that the Proposed Regulations' general approach to eliminating interest — eliminating only a proportionate amount of the interest and debt when a 25%-or-more owned look-through subsidiary is the debtor — be extended to rents and royalties.

J. The approach of the Proposed Regulations to “look-through partnerships” raises complicated questions. As explained in the Report, we do not believe the Proposed Regulations achieve equivalent treatment of partnerships and corporations for purposes of Section 1297(b)(2)(C). For this and other reasons, we suggest the Treasury and IRS reconsider the approach. Nonetheless, we acknowledge the difficulty of developing an approach consistent with the purposes of the PFIC rules, administrable and not unduly burdensome to taxpayers. Accordingly, we do not recommend a specific alternative but have instead summarized the advantages and disadvantages of the Proposed Regulations' approach and alternative approaches for further consideration by the Treasury and IRS.

K. We generally support the “single entity” approach of the Proposed Regulations to stapled entities. If the single entity approach is intended to apply in cases where more than a de minimis quantum of the shares of one or both entities is not stapled, however, we recommend issuing specific rules applicable to holders of unstapled shares requiring the use of separate entity PFIC testing as to them.

L. We appreciate that the Proposed Regulations will permit taxpayers to reasonably rely on the principles of the Proposed Regulations on a consistent basis. However, because of the “once a PFIC always a PFIC” rule, there may be taxpayers who in practice are unable to rely on the Proposed Regulations because of positions they have adopted regarding PFIC status under prior law. We request clarification of the interaction of the “once a PFIC always a PFIC” rule and the reliance allowed by the Proposed Regulations. We discuss certain simplifying presumptions that could be allowed when determining if the “once a PFIC always a PFIC” rule prevents reliance on the Proposed Regulations to limit the administrative burden on the IRS and taxpayers of analyzing PFIC status retrospectively over multiple years.

3. PFIC Ownership “Attribution”

A. “Top-down” approach for ownership attribution through intermediate partnerships

A U.S. taxpayer may be treated as owning stock of a PFIC the person does not directly own under attribution rules in Section 1298. In the case of indirect ownership through a non-PFIC foreign corporation, Section 1298(a)(2)(A) provides that stock in a lower-tier PFIC owned by such a foreign corporation is attributed to a U.S. owner of the non-PFIC corporation only if the U.S. owner owns 50% or more (by value) of the non-PFIC foreign corporation stock. However, a different rule applies to attribution through partnerships. Section 1298(a)(3) provides that, for purposes of attributing stock of a PFIC to a U.S. person, stock “owned, directly or indirectly, by or for a partnership” is considered owned proportionately by its partners. These rules potentially conflict when a U.S. person owns a partnership which owns stock of a non-PFIC foreign corporation.

Example 1

image 1

In Example 1 shown in the chart above, U.S. Person indirectly owns less than 50% (i.e., 50% of 50%, or 25%) of a foreign corporation (FC 2) through a partnership. Applying the rules from the “top down” (or treating the partnership as an aggregate rather than an entity), U.S. Person would be deemed to own 25% of FC 2 and therefore is not attributed the FC 3 PFIC shares owned by FC 2. However, one could also apply the rules by attributing 50% of the FC 3 shares owned by FC 2 to the partnership, and then treating 50% of those PFIC shares (i.e., 25% of the shares of FC 3) as owned by U.S. Person (i.e., “bottom-up” attribution). The statute does not provide an ownership threshold below which stock owned or deemed owned by a partnership may not be reattributed to the partners. A bottom-up approach would also arguably be consistent with the literal language of Section 1298(a)(5) which treats stock owned by a person under the attribution rules as actually owned for purposes of re-attributing that stock. The statute is not clear as to whether the attribution rules should be applied using a “top-down” or “bottom-up” approach when there is an intervening partnership.

The Proposed Regulations take a “top-down” approach.6 Moreover, in testing whether a U.S. person owns, directly or indirectly, 50% or more of the value of a foreign corporation that is not a PFIC, the U.S. person only looks through a partnership if the U.S. person owns 50% or more of the ownership interests in the partnership that directly or indirectly owns the stock of the non-PFIC, foreign corporation.7 Thus, under the Proposed Regulations there is no attribution of the shares of FC 3 through FC 2 on the facts of Example 1 because U.S. Person is only deemed to own 25% (50% of 50%) of FC 2.8

We support the results reached by the Proposed Regulations on the facts of Example 1. The statute does not speak directly to the question of aggregate or entity treatment of partnerships. However, there is nothing to indicate that Congress intended an intermediate partnership to be treated as consolidating ownership of lower-tier corporate shares at its level. Such consolidation at the partnership entity level would have the effect of causing the “directly or indirectly” attribution rule applicable to partnerships to override the 50% “cut off” rule in Section 1298(a)(2)(A) in cases where an intermediate partnership owns, directly and indirectly, the shares of an upper-tier non-PFIC and a lower-tier PFIC. However, because the plain statutory language of Section 1298(a)(5) generally reattributes ownership up (rather than down) the chain, we believe reaching the appropriate results reflected in the Proposed Regulations should be predicated on applying an aggregate theory of partnerships. The preamble to the Proposed Regulations (the “Preamble”) acknowledges that the “top-down” approach could be viewed as consistent with an aggregate theory of partnerships.9 However, the “top-down” approach in the Proposed Regulations, in particular the provision limiting attribution to a 50% or greater partner, deviates from a true aggregate approach, as illustrated by the next example.

Example 2

image 2

Prior to the Proposed Regulations, because FPRS owns 95% of FC 1, FPRS arguably would be deemed to own 57% (95% of 60%) of FC 3, which it is deemed to actually own for purposes of re-attribution and U.S. Person then arguably owns 27.93% (49% of 57%) of the FC 3 stock through FPRS.

Under the Proposed Regulations, it appears that ownership of FC 1 stock by FPRS would not be attributed to U.S. Person as U.S. Person owns less than 50% of FPRS.10 U.S. Person also owns less than 50% of FC 1 directly, and no attribution of FC 3 stock therefore applies through FC 1 on account of U.S. Person's direct ownership of FC 1. Thus, the Proposed Regulations appear to reach “equivalence” (as to the 50% cut off) whether U.S. Person owns shares of FC 3 through a partnership (“FPRS”) or a corporation (although an intermediate corporation in place of FPRS would have to be a non-PFIC for this rule to apply).11

If the rules adopted a true “aggregate” approach to partnerships, the result would be that U.S. Person owns 46.55% (49% of 95%) of FC 1 indirectly and owns 5% directly, resulting in U.S. Person owning 51.55% of FC 1 and thus being attributed the FC 3 shares owned by FC 1 proportionately. As discussed below, we believe this reaches more appropriate results given the purposes of the PFIC rules.

Given aggregate treatment of partnerships for many purposes under the Code, we can discern no good policy reason to treat a U.S. person who owns the same interest in an upper-tier non-PFIC corporation that owns a lower-tier PFIC entity through a partnership less favorably than if that person owned a proportionate amount of the corporate shares of an upper-tier non-PFIC directly. Assuming 50% ownership is the appropriate threshold at which attribution of ownership of a lower-tier PFIC through a non-PFIC corporation is cut off (as Congress concluded), that cut-off should apply whether the U.S. person owns that quantum of shares directly or indirectly through a partnership or in part directly and in part indirectly through a partnership. Though the statute does not speak to this point directly, Congress would have been aware that partnerships may be treated as aggregates, not entities, for many purposes, and we generally agree with the specific examples in the Proposed Regulations which reach results consistent with this approach. However, we believe that in the situation illustrated by Example 2, an aggregate approach reaches more appropriate results than the 50% by-value attribution requirement the Proposed Regulations adopt for partnerships.

Accordingly, we recommend that an aggregate approach to partnerships apply for this purpose. As noted above, we support the results of the “top-down” approach to ownership attribution reflected in the Proposed Regulations when the effect is to prevent the partnership (unlimited) attribution rule overriding the corporate ownership limits on attribution. A general 50% by-value limit for attribution through partnerships, however, does not have direct support in the statute and we see no compelling policy reason for adopting it and we suggest eliminating it in final regulations.

B. Extending top-down attribution to intermediate non-look through corporations

Treasury and the IRS have requested comments as to whether a “top-down” approach or some alternative method should apply under Section 1298(a) in a purely corporate structure. We do not support extending the “top-down” attribution approach applicable to partnerships to corporations for purposes of ownership attribution, even though this may result in inconsistent treatment of a U.S. shareholder in similar structures.

Example 3

In the example below, if a “bottom-up” approach applies, F Parent is attributed 50% of FC 3's shares through FC 1 and 25% of FC 3's shares through FC 2, which it is deemed to actually own for purposes of reattribution. Thus, U.S. Person then is attributed 37.5% of FC 3's PFIC shares. By contrast, under a “top-down” approach, because U.S. Person owns less than 50% of FC 2 indirectly, U.S. Person is not attributed the FC 3 shares owned by FC 2.

img 3

The statutory language in Sections 1298(a)(1)(A), (a)(2)(A) and (a)(5) seems clear in this context. Under Section 1298(a)(2)(A), F Parent (because it owns 50% or more of FC 2), is deemed to own the shares of FC 3 owned by FC 2 (proportionately by value), and is then treated as actually owning those shares for reattribution purposes by Section 1298(a)(5).

Congress chose to distinguish between partnerships and corporations under the PFIC regime by providing a 50% (by value) ownership attribution cut off for indirect ownership through corporate (non-PFIC) companies. It provided no equivalent cut-off at the partnership level for ownership through partnerships. There is no basis to think Congress may have anticipated an “aggregate” approach to corporations or intended equivalent treatment. We believe it would be questionable to distort the statutory language to achieve “equivalence” between a partnership (aggregate pass-through entity) and corporate parent company when the basic statutory attribution rules on their face treat partnerships and corporations differently and we see no compelling policy reason for reaching an “equivalent” result.

We understand that this may mean that different PFIC outcomes may result depending on whether a “check the box” (“CTB”) election has been made for a particular entity as compared to an otherwise identical entity that has not made a CTB election. In isolation, given the purposes of the PFIC rules, it may seem anomalous to reach different attribution results merely because of a CTB election. However, the U.S. federal income tax system is replete with similar inconsistencies. These are a consequence of the policy decision to permit elective entity classification and the presumption in favor of aggregate treatment for partnerships.

To provide only one example in the context of attribution rules, Section 875(1) attributes U.S. trade or business activity to passive limited partners and small shareholders treated as partners solely because of a CTB election. Conversely, even the controlling shareholder of an entity treated as corporate is not deemed engaged directly in the business activity of the corporation. Whatever the policy merits of this particular result given the purposes of Sections 864 and 875 in isolation, the result clearly follows from the general aggregate approach to partnerships, including by operation of a CTB election.

Congress adopted different attribution rules for partnerships and corporations (which may reach different results). We think the reattribution rules of Section 1298(a)(5) operate differently for partnerships and corporations (and Congress would have anticipated potential aggregate treatment for partnerships but not corporations). We do not see a compelling policy reason (1) to ignore the plain statutory language to achieve equivalence between corporations and partnerships or (2) to provide equivalence for CTB partnerships but distinguish actual legal partnerships. Inconsistent treatment of entities depending on whether a CTB election is made is a much broader policy issue posed by elective entity classification generally and not something that sensibly can be addressed in the isolated context of the PFIC rules.

We therefore do not recommend adoption of a “top-down” approach to ownership attribution through intermediate non-look through corporations.

4. General Income Tests

Passive income for purposes of the PFIC regime generally is any income of a kind that would be foreign personal holding company income (“FPHCI”) under Section 954(c). Section 1297 modifies Section 954(c)'s definition of passive income in certain respects, however. As noted in a prior Report, it is unclear whether the Section 954(c) and Section 954(h) exceptions to FPHCI should be treated as incorporated by reference (completely, partially or not at all) for purposes of Section 1297(b). We recommended in that Report that Treasury and the IRS clarify the scope of the cross-reference to Section 954(c) in Section 1297 and the application of Subpart F's other FPHCI exceptions in making PFIC determinations.12

A. Clarification of cross-references

Under the Proposed Regulations, Treasury and the IRS take the general approach that it is appropriate for income derived by any foreign corporation whose status is being tested under Section 1297(a) (a “Tested Foreign Corporation”) to be eligible for the exceptions to FPHCI, including the Section 954(h) exception, unless there is a specific PFIC rule (e.g., Section 1297(b)(2)(C) relating to related person income) intended to supersede the equivalent Section 954(c) exception.13 Under the Proposed Regulations, PFIC income classification would not take into account Section 954(c)(3) (income from related persons), Section 954(c)(6) (income from related CFCs) and Section 954(i) (active insurance income).14 Conversely, the following sub-sections of Section 954(c) would apply for PFIC income classification purposes: Section 954(c)(2)(A) (active rents and royalties from related persons), Section 954(c)(2)(B) (export financing income), Section 954(c)(2)(C) (dealer income), Section 954(c)(4) (sales of partnership interests), Section 954(c)(5) (commodity hedging), and Section 954(h) (banking, financing or similar business).15 This clarification on the interaction between the Section 954(c) and Section 954(h) exceptions and PFIC passive income is welcome and appreciated. We generally agree with the approach and conclusions adopted by the Proposed Regulations as to which Section 954(c) exceptions should be incorporated by reference to determine “passive” PFIC income.

B. Section 954(h) “banking and financing” exception

We commend the inclusion of Section 954(h) in the list of cross-referenced exceptions and recommend that Treasury and the IRS continue to apply Section 954(h) to non-bank active financing businesses even after final regulations under Section 1297(b)(2)(A) are released.

In the Preamble, Treasury and the IRS noted that the PFIC regime is not meant to apply to foreign corporations engaged in active businesses as support for the exclusion of active banking, financing and insurance income from the definition of passive income.16 We agree with the conclusion of the Proposed Regulations that the Section 954(h) banking and financing company exception should be applicable in determining whether income is passive for PFIC purposes. We also agree it is reasonable and appropriate to distinguish between the exception in Section 954(h) and the insurance exception in Section 954(i) when Congress clearly intended a superseding PFIC rule in Section 1297(f).

Many entities that are not regulated as banks and do not take deposits nonetheless conduct active financing businesses earning gross revenue consisting primarily of fixed income, like interest. The proposition that a financing business may generate active income is well accepted in other areas of the tax law. The relevant distinction is the nature of the activities conducted by the business to earn those revenues (active versus passive). Accordingly, equivalent exceptions should apply for non-bank active financing businesses as apply for bank active financing businesses. As we noted in our 2010 Report, Section 954(h) appropriately distinguishes such businesses from passive investment or trading businesses and imposes a high standard.17

While in some cases, active non-bank financing businesses may be a core activity of a foreign company or group, the potential for inadvertent PFIC treatment also frequently arises in practice in the case of “captive” financing companies of non-financing business groups. Many active businesses conduct financing activities ancillary to their core active business. For example, in many countries retailers or auto dealers provide customers with financing through an in-house finance company as a necessary ancillary service of the primary business. Because such finance businesses are often highly leveraged, the gross assets of such a supplementary financing business may be disproportionately large relative to the gross assets of the supported retail business. In either case, given the purposes of the PFIC rules, we believe that it is inappropriate to treat an activity that fundamentally is not a passive investment activity as passive for PFIC purposes.

We further believe that Section 954(h) should continue to apply to non-bank active financing activities even after regulations under Section 1297(b)(2)(A) addressing the active banking exception are finalized. Alternatively, the final regulations addressing the PFIC active banking exception could include specific rules for other active financing businesses under the implied authority of the Section 954(h) exception to FPHCI as there should be consistency between the definition of “passive income” for Subpart F and PFIC purposes.

C. Gains (Netting)

PFIC income characterization is based on the FPHCI definition.18 Certain categories of FPHCI are treated as such only to the extent gains exceed losses in the relevant category.19 However, in applying these FPHCI rules for PFIC purposes it is necessary to address the inter-play of these rules with the look-through rules of Section 1297(c) (the “Related Person Look-through Rules”).20 Under Section 1297(c), if a Tested Foreign Corporation owns (directly or indirectly) 25% or more (by value) of another foreign corporation (i.e., a “Look-through Subsidiary”),21 for purposes of applying the income and asset tests the Tested Foreign Corporation is treated as if it earned its share of the underlying Look-through Subsidiary's gross income and owned its share of the underlying Look-through Subsidiary's gross assets directly. Thus, a question arises, for example, if a Tested Foreign Corporation has losses from a particular commodity and a Look-through Subsidiary has gains; may these be netted, or does the Tested Foreign Corporation pick up its pro rata share of the commodity gains as FPHCI? The Proposed Regulations adopt a “separate company” approach to this question.

Under Proposed Treasury Regulation § 1.1297-1(c)(1)(ii), items of income under Section 954(c) that are determined by netting gains against losses are calculated separately with respect to the Tested Foreign Corporation and each Look-through Subsidiary and the latter's net (separate company) gains are then picked up under the look-through rules. A Look-through Subsidiary determines its net gain (taking into account only its own losses) in the relevant category and its 25% owner or owners would include this net FPHCI item when testing their own PFIC status even if that owner has corresponding category losses of its own. Accordingly, gains and losses of different foreign corporations are not netted even if they are Look-through Subsidiaries of the same Tested Foreign Corporation or are gains and losses of the Tested Foreign Corporation and its Look-through Subsidiary.

Given the purpose of the PFIC regime, the separate company approach may have the effect of potentially overstating FPHCI of an integrated business conducted through multiple subsidiaries. For example, securities, commodities or derivatives dealers may have (for regulatory reasons) separate legal companies in different geographic locations in which they operate that otherwise conduct an integrated dealer business (but not all of the income from which is technically “active” income). The inability to net losses against gains of different group members may in our view overstate the FPHCI of the Tested Foreign Corporation parent or other members. This is not the correct approach given Congress' intent to avoid treating foreign corporations that own subsidiaries primarily engaged in active business operations as PFICs with the Look-through Rule.22

We see no compelling policy reason favoring a separate entity approach to netting these gains and losses, but Treasury and the IRS may be concerned about regulatory authority. We believe that Treasury and the IRS have broad authority under Section 1298(g).23 Self-evidently, the look-through regime under Section 1297(c) cannot operate sensibly (e.g., will on its face result in duplication of items and other anomalous results that cannot have been contemplated by Congress) absent regulatory guidance to reconcile items of the Tested Foreign Corporation and its Look-through Subsidiaries. For example, the legislative history indicates that intercompany items, such as dividends and interest, received from a Look-through Subsidiary are to be eliminated from the Tested Foreign Corporation's income in applying the PFIC income test.

The legislative history clearly contemplates that regulations will fill these gaps given the legislative intent of this rule. Accordingly, we believe that strong arguments exist for the view that netting gains and losses in determining income is within the ambit of that implied delegation of authority and the broad authority of Section 1298(g). If Treasury and the IRS believe that they do not have authority, then we recommend that they seek a statutory amendment.

Based on the language of Section 1297(c), which treats the owner as if it owned and earned directly the assets and income of a Look-through Subsidiary, we propose allowing a parent Tested Foreign Corporation to net gains and losses of its directly or indirectly owned Look-through Subsidiary/ies (or partnership/s).24

Furthermore, as a policy matter, if an integrated business is being conducted arguably “netting” of parent gains and losses (or even sister Look-through Subsidiary gains and losses) should also be permitted when the Look-through Subsidiary rather than a parent is the Tested Foreign Corporation. However, we do not recommend this broader “netting” approach for three reasons. First, it is more difficult to square with the statutory language of Section 1297(c) because the subsidiary Tested Foreign Corporation does not directly or indirectly own the relevant gain or loss assets and economically such gains or losses are not even indirectly realized by the subsidiary.25 Second, such a broad “netting” system would create significant complexity. If a Look-through Subsidiary has more than one 25%-or-greater owner, presumably the subsidiary should not be permitted to net 100% of both parents' gains and losses but it is unclear on what proportionate basis these gains and losses should be attributed to the Look-through Subsidiary. Finally, if the parent Tested Foreign Corporation is not a PFIC (after applying our suggested “netting approach”) ownership of a lower-tier Tested Foreign Corporation generally would not be attributed to less-than-50% U.S. owners of the foreign parent in any event. Accordingly, in practice such a broader “netting approach” is likely to benefit only minority shareholders of the subsidiary Tested Foreign Corporation, a situation in which “netting” seems least appropriate.

D. Related Person Look-through Rules

i. Section 954(d)(3) “related person” definition

Under the Related Person Look-through Rules of Section 1297(b)(2)(C), passive income does not include any income which is interest, a dividend, or a rent or royalty, which is received or accrued from a related person (within the meaning of Section 954(d)(3)) to the extent such amount is properly allocable (under regulations prescribed by the Secretary) to income of such related person which is not passive income. The Proposed Regulations provide helpful guidance on the application of the Related Person Look-through Rules.26 As an initial matter, the Proposed Regulations clarify that the determination of whether interest, dividends, rents and royalties were received or accrued from a person that is a related person is made on the date of the receipt or accrual, as applicable based on the recipient's method of accounting, of the interest, dividend, rent, or royalty.27

ii. Interest

Under the Proposed Regulations, interest that is received or accrued, as applicable based on the recipient's method of accounting, from a related person is allocated to income of the related person that is not passive income in proportion to the ratio of the portion of the related person's non-passive income for its taxable year to the total amount of the related person's income for the taxable year that ends with or within the taxable year of the recipient.28 The Proposed Regulations thus replace the “cream skimming” approach that applies for purposes of Section 904 — which treats interest paid to a related person as passive income to the payee to the extent that the payor has any passive income (i.e., as if it was paid first out of passive income of the payor) — with a pro rata approach.29 We agree with the Proposed Regulations that the purposes of the PFIC regime are not served by the “cream skimming” approach, which addresses policy concerns specific to the foreign tax credit rules, and welcome the pro rata approach adopted by the Proposed Regulations.30

iii. Dividends

The Proposed Regulations provide that the character for purposes of the Related Person Look-through Rules of dividends paid is determined based on the portion of the related payor's current year earnings and profits (“E&P”) for the taxable year that ends in or with the taxable year of the recipient that are attributable to non-passive income.31 The Preamble requests comments on alternative methods of determining the portion of dividends treated as properly allocable to non-passive income of a related person (including if the payor has no current E&P).32

The approach of the Proposed Regulations — looking only to current year E&P in characterizing a dividend — has certain advantages. Foreign corporations frequently do not maintain E&P accounts based on U.S. principles and these may be extremely burdensome (or impossible) to reconstruct. For example, a subsidiary that is now related under Section 954(d)(3) principles may have accumulated pre-acquisition E&P. The acquiring Tested Foreign Corporation may not have known that the acquiring Tested Foreign Corporation's U.S. owners would be affected by failing to make a Section 338(g) election to eliminate historic E&P of the dividend payor. Avoiding the need to determine accumulated E&P by looking only to current year E&P has the virtue of administrative simplicity. Moreover, even if historic accumulated E&P accounts are available or can be reconstructed, in some cases it may be questionable whether looking to historic E&P is indicative of whether the shares in the dividend payor are “passive.” For example, if a group acquires an interest in a subsidiary that historically conducted an activity in a passive manner but changes how it operates post-acquisition to be active, it is not obvious from a PFIC policy perspective that looking to legacy E&P pools is an appropriate proxy for whether any dividends paid currently should be classified as active or passive.

Looking to current-year E&P first has the virtue of emphasizing recent activities over distant legacy activities in characterizing related party dividends and minimizing the administrative burden of reconstructing the character of E&P in prior years. On the other hand, if a related subsidiary has no current year E&P, it seems inappropriate to treat a dividend distribution to the Tested Foreign Corporation as per se passive. Giving zero weight to recently accumulated E&P seems arguably inappropriate even if the related dividend payor has sufficient current year E&P. Assume, for example, that a related person earns $100 of active income in year one, $100 of passive income in year two and $0 of income in year three. If the related person pays a $200 distribution in year two, then under the Proposed Regulations, the entire amount of the distribution would be treated as passive (whereas a $100 distribution in year one and a $100 distribution in year two would result in receipt of $200 of aggregate dividend income that is 50% passive).33 If, however, the related person pays the same dividend in year three, when it has no current-year E&P, under the Proposed Regulations the dividend may be treated as entirely passive. Treating the entire distribution in year two or year three as passive (or vice versa) is not consistent with economic reality and places undue emphasis on the timing of dividend distributions.

Given the purpose of the PFIC rules, it is debatable whether any approach based rigidly on E&P accounts and the ordering rules under Section 316 makes policy sense. Further, even if one believes that characterizing a distribution based on the character of E&P is appropriate for purposes of the income test under Section 1297(a)(1), a further problem arises. The treatment of the dividend as passive may have a “knock on” effect because treating the dividend as passive means that the shares arguably are held to produce passive income and to that extent are characterized as passive for purposes of the asset test under Section 1297(a)(2).34 It seems inappropriate in the example above that the shares of the related dividend payor (whose earnings on average for the period are half active and half passive) could be treated as 100% active or 100% passive for purposes of the asset test in a given year depending on the timing of dividend distributions.

As a general matter, the PFIC regime at a fundamental level does not rigidly follow E&P constructs. In particular, the excess distribution regime applies to distributions that technically are not dividends and economically constitute (at least based on E&P constructs) a return of capital. Section 316, on the other hand, follows corporate law accounting conventions to ensure that distributions give rise to taxable income to the extent of corporate-level earnings before being treated as a return of capital. The “income first” approach may make sense for that purpose. However, when the rule is applied for purposes of associating a dividend with underlying income of the payor that is either passive or active, it is not clear that tracking Section 316 principles makes policy sense, as illustrated by the simple example above.

The Proposed Regulations also gloss over the complexities that arise from the “net” nature of E&P when the PFIC rules hinge on gross income. The related person dividend must under the PFIC rules be attributable to gross passive income. To 'trace” the dividend under any reasonable method to gross income of the related dividend payor, expenses must necessarily also be characterized as either active or passive. Under the PFIC rules, “passive” income is not limited to income from passive portfolio investments in securities with minimal associated expenses. Using an E&P approach may therefore require the use of a complicated methodology (e.g., an approach similar to the allocation and apportionment of expenses under Treas. Reg. § 1.861-8) in order to determine the extent to which E&P is “attributable” to passive versus active gross income.

To reconcile the concerns discussed above, we propose two alternative approaches for characterizing dividends received from a related person. It is important to emphasize that whatever approach is adopted in this context applies only for purposes of characterizing the dividends and associated shares under the Related Person Look Through Rules as active or passive. The approach adopted does not affect the amount or character of income included by any U.S. taxpayer (except in the very attenuated sense that it may affect whether a Tested Foreign Corporation is or is not a PFIC).

We believe that the most rational method is a pro rata approach that looks at the ratio of underlying passive to active gross income over some fairly recent period (without regard to E&P). This is highly preferable from an administrative perspective as compared with an approach based on E&P, because in practice most foreign corporations do not maintain E&P calculations based on U.S. tax principles, and it is inordinately burdensome to reconstruct such accounts. Because recent earnings tend to be more indicative of the economic character of a distribution as derived from either passive or active income, we suggest this pro rata approach use a reasonable rolling look-back period (e.g., three or five years). This could also incorporate an anti-abuse rule in the event there was a material change in the business of the related person dividend payor shortly prior to or during the look-back period that was intended to recharacterize any dividends as active. However, in the limited situations in which the Related Person Look-through Rules are even relevant (i.e., when the payor is not a Look-through Subsidiary but is nonetheless related under Section 954(d)(3)),35 it seems unlikely in practice that a passive investment vehicle would assume the risks of entering into an active business in an attempt to cleanse the payee's PFIC status for the benefit of remote U.S. shareholders.

If, however, Treasury and the IRS are reluctant to depart entirely from E&P constructs, we suggest a bifurcated approach. If a foreign corporation does not otherwise maintain E&P accounts on U.S. tax principles, it is subject to the potentially taxpayer unfavorable presumption under current law that all distributions it pays are considered dividends (even if any such distributions constitute a return of capital under Section 316). Thus, there is a “price” to failing to maintain E&P accounts. If a payor corporation consistently treats all distributions as dividends (i.e., per se income items), we recommend that it be allowed to characterize such dividends as passive or active using the pro rata gross income ratio approach described above. If, however, the payor corporation does otherwise maintain E&P calculations under U.S. tax principles, the rules could provide that the character of a dividend as active or passive be based on the active-passive ratio for current year E&P up to the amount of positive current year E&P, with the balance based on the active-passive ratio for accumulated E&P. Given, as discussed above, that the nature of earnings accumulated before the payor became related may be less relevant to the character of a current dividend, this approach could incorporate a limitation on the relevant years for accumulated E&P (e.g., the years during which the dividend payor has been a related person under Section 954(d)(3)).

However, we do not favor this second, bifurcated approach, as we see no policy reason to reach different results in the unusual case in which the lower-tier “related” foreign corporation has maintained U.S. E&P accounts.

iv. Rents and royalties

The Proposed Regulations provide that rents and royalties are allocable to income of the related person which is not passive income to the extent the related person's deduction for the rent or royalty is allocated to non-passive income under the principles of Treas. Reg. §§ 1.861-8 through 1.861-14T.36 The application of these rules will be difficult and complicated, even if sufficient information is available. Necessary information may not even be available in this context. However, unlike dividends, rents and royalties do not economically represent a distribution of the results of the aggregate business of the payor. Although a pro rata approach would be simpler and more administrable, it therefore seems less appropriate in this situation and potentially open to abuse.

5. General Asset Tests

A. Treatment of Working (Cash) Capital

The approach to cash taken by Notice 88-22 — classifying cash as a per se passive asset even if it is working capital — reaches results that we believe are inappropriate. As we and others have noted, treating working capital that is reasonably maintained by a company engaged in an active trade or business as a passive asset generally undermines the PFIC regime's objective of distinguishing between investments in active businesses and passive investment companies.37 We regret that the Proposed Regulations did not address this issue and believe that treating working capital as passive merely because it may produce interest (although very little interest at current market rates) reaches the wrong result.

Because this issue has been raised many times and Treasury and the IRS have had ample time to consider the question, this Report will not restate the arguments against the Notice 88-22 approach and possible solutions. At a minimum, however, given the dual asset approach applied by the Proposed Regulations in other situations,38 we believe working capital could be bifurcated into an active and passive asset in the same ratio as the overall gross active to passive income (other than any income from the cash assets) of the company. The “working capital” subject to this rule could be limited to bank deposits and other short-term investments that generate little or no interest as a proxy for whether the cash is held for the reasonable needs of the business rather than as an investment. We do not believe that most passive investment vehicles (that would in any event be PFICs under the tests regardless of this relief) are likely to maintain as a significant portion of a portfolio low-or-no yielding cash assets.39

B. Change in 1297(e) status during year

Section 1297(e) prescribes the method for measuring assets for purposes of the asset test. If the Tested Foreign Corporation is publicly traded or if the Tested Foreign Corporation is a non-CFC (and does not elect out of the fair market value method), the asset test applies based on fair market value. If the Tested Foreign Corporation is a CFC, the test is applied based on adjusted tax basis of the assets unless the CFC is publicly traded (in which case the fair market value method applies). Effectively, only non-publicly traded CFCs are required to use adjusted tax basis to “value” assets for purposes of the asset test. Publicly traded Tested Foreign Corporations must use fair market value. Non-publicly traded, non-CFCs may elect whether to use fair value or tax basis. The statute does not address what method applies to a Tested Foreign Corporation that becomes publicly traded during a year and is traded for less than the entire year.

The Proposed Regulations adopt a “majority of days” approach in which assets are measured on the basis of value for the entire taxable year if (i) shares of the foreign corporation were publicly traded under Section 1297(e)(3) for the majority of days during the taxable year or (ii) Section 1297(e)(2) did not apply to the foreign corporation on the majority of days during the taxable year.40 Otherwise, assets must be measured on the basis of adjusted basis for the entire taxable year. The Preamble states that Treasury and the IRS sought to minimize shareholder optionality in choosing either method with respect to a Tested Foreign Corporation in order to minimize avoidance of the PFIC rules.41

We question whether this “majority of days” approach makes sense as a policy matter. At stake is whether goodwill or other self-created business intangibles may be taken into account under the asset test, as such intangibles typically have no tax basis. Generally, the existence of such self-created intangibles is indicative of an operating business and ignoring such intangibles tends to recharacterize operating businesses inappropriately as passive.

We believe that using fair value rather than tax basis more appropriately reflects the purposes of the PFIC rules in general. The tax basis method is mandatory only for CFCs that are not publicly traded. It appears that Congress concluded that CFC's would already need to calculate tax basis under U.S. principles for other reasons and it would be burdensome to require valuation absent public trading for items like intangibles that are difficult to value. It is questionable, however, why CFC status resulting from the U.S. residence status of 10% shareholders should adversely affect smaller U.S. shareholders who are not subject to the CFC regime, particularly after the adoption of the PFIC/CFC overlap rule of Section 1297(d). As a result of the overlap rule, in most cases PFIC characterization no longer has any impact on those U.S. taxpayers whose ownership caused the Tested Foreign Corporation to be a CFC. It also is not apparent as a PFIC policy matter why CFCs should be denied the flexibility afforded non-publicly traded non-CFCs to use fair market value unless they elect otherwise. In any event, non-traded, non-CFC taxpayers are expressly afforded “optionality” by the statute. The plain language of Section 1297(e) perhaps may require this use of the basis rather than value method when it unambiguously applies by its terms.42 However, there is no policy reason to extend this result any more broadly than is strictly required by the statute.

Thus, we recommend that final regulations require the use of the value method if shares of the Tested Foreign Corporation are publicly traded at any time during the taxable year (at least for a sufficient period of time during which price quotations would be available). We do not believe this creates any greater “optionality” in a practical sense than the Proposed Regulations' approach. No company in the real world in our experience times its initial public offering based on PFIC considerations. We recognize that, because the asset test looks at quarterly averages, valuing based on a year-end IPO share price may be an imprecise basis on which to value assets during the first quarter of the company's taxable year. However, the relative valuation of assets derived from share price of a company's net equity is an imprecise measurement as well, and asset basis is even less precise. The PFIC regime already lives with imperfect valuation for non-publicly traded non-CFCs (based, if anything, on more subjective valuations). We believe that public share price is a better proxy from which to value business intangibles like goodwill than assuming zero value for such intangibles as under the tax basis method.

A separate question under Section 1297(e), not addressed by the Proposed Regulations, arises when a parent Tested Foreign Corporation is publicly traded but subsidiary Tested Foreign Corporations are not. The application of Section 1297(e) to tiered Tested Foreign Corporations is not entirely clear under the Code and Proposed Regulations. We ask that Treasury and the IRS clarify how Section 1297(e) applies to tiered Tested Foreign Corporations. Assume that a publicly traded foreign parent corporation owns a Look-through Subsidiary whose shares are not traded, and that the Look-through Subsidiary has one or more minority U.S. shareholders and is technically a CFC.43 May the parent Tested Foreign Corporation use value rather than basis when looking through to the assets of the Look-through Subsidiary? May the Look-through Subsidiary itself then use the value method rather than basis when the Look-through Subsidiary is the Tested Foreign Corporation?

For the reasons explained above, if the parent Tested Foreign Corporation is publicly traded, we believe the foreign subsidiary's assets should be measured under the value method for purposes of testing the parent's status as a PFIC. Indeed, it seems reasonably clear that, because the publicly traded Tested Foreign Corporation is deemed by Section 1297(c) to own directly the Look-through Subsidiary's assets, the value rather than the tax basis of the look-through assets should be used for purposes of testing the parent's PFIC status. We recognize that the value of publicly traded parent shares may be an imperfect proxy by which to value subsidiary assets (particularly if the parent has other assets and is not a pure holding company). On the other hand, as noted above, non-CFC subsidiaries are permitted to use value rather than basis (presumably, based on subjective appraisals of such values, untethered to any public trading of shares).

It is more questionable whether a non-traded subsidiary of a publicly traded parent corporation should be allowed to use the value method when it is the Tested Foreign Corporation. As a technical matter the subsidiary shares are not publicly traded (except perhaps in an indirect sense) and the use of the value method in practice primarily affects U.S. minority shareholders of the subsidiary's non-traded shares. On the other hand, if the value rather than basis of the Look-through Subsidiary's assets is used for the purposes of testing the parent company's PFIC status, one could argue that the same value should be applicable in the latter case as well.

C. Classifying Shares of Non Look-through Subsidiaries

As a general matter, although this is not expressly stated in the Code, Notice 88-22 treats shares of subsidiaries that are not Look-through Subsidiaries as passive because stock generally produces passive dividends. This approach is arguably implicit in the Look-through Subsidiary exception. It would be odd for the system to pro rate shares of more-than-25% owned subsidiaries based on underlying income but treat portfolio shares as active. The approach is also supported by a statement in the 1986 Bluebook which provides that such shares are passive because the dividends that such shares produce are passive.44

Despite the general approach of Notice 88-22, the Proposed Regulations include a special rule for shares that produced dividends governed by the Related Person Look-through Rules in prior years but do not produce any current year dividends.45 Such shares are subject to “dual asset” treatment. That is, the value (or adjusted basis) of the shares is allocated between an “active asset” and “passive asset” in proportion to the average percentage of dividends that were characterized as passive and non-passive, respectively, income for the previous two taxable years under the Related Person Look-through Rules.46 The Proposed Regulations do not explicitly discuss the scenario in which shares pay a current year dividend that is at least in part active under the Related Person Look-through Rules. We believe it must be the case that the Proposed Regulations intend that the shares in that case would be characterized as partly active based on that current year dividend (i.e., that the dual asset approach applies in this case also). The two-year lookback rule otherwise would make little sense.47 In general, we welcome and agree with the Proposed Regulations' treatment of shares of a related person as potentially active to the proportionate extent that those shares would generate active dividends under the Related Person Look-through Rules. As a policy matter, it seems anomalous to treat such shares as per se passive for purposes of the asset test.

As discussed above in Part 4(D)(iii), we recommend that E&P principles not be rigidly applied in determining whether dividends paid by a related person are active or passive under the Related Person Look-through Rules. If our recommended gross income approach is adopted, the character of dividends would depend on the ratio of active to passive gross income for a rolling look-back period. If this approach is adopted for dividends, we recommend that the underlying shares be characterized as passive or active in a given year based on the manner in which any dividends (regardless of whether any such dividends are actually paid) would be treated if paid in such year.

Under this approach, any related person shares would be characterized on the basis of the type of dividend income that they would be expected to produce. Thus, in the simplest case, if a subsidiary has had zero passive income since being acquired, and any dividends received from such subsidiary would be active income under the Related Person Look-through Rules, the shares of such subsidiary would be considered active assets under the asset test (even if no dividends are actually paid). This seems consistent with Notice 88-22 with respect to assets that have not produced income, as such assets generally are to be characterized on the basis of the income that is reasonably expected to be generated in the reasonably foreseeable future. Combined with our proposed approach to characterizing dividends, we believe this reaches sensible and consistent results without regard to the happenstance of dividend policy (and will be less easily manipulated by taxpayers).

If our recommended gross income approach is not adopted with respect to the characterization of related person dividends and related person shares, a look-back period somewhat longer than two years may be appropriate for purposes of the special rule under Proposed Treasury Regulation § 1.1297-1(d)(2)(iii). Arguably, though, this should not include any period prior to the time that the subsidiary became a related person under Section 954(d)(3). For example, the final regulations could provide for a look-back period of the lesser of (i) five years or (ii) the period the subsidiary has been a related person under Section 954(d)(3). We recognize that a longer testing period imposes administrative burdens, as the current share owner may be required to look back to more prior years, as to which it may not have been tracking underlying income. However, the current share owner would not be required to look back to years prior to the date on which the dividend payor became a related person. It is more likely that a Tested Foreign Corporation has reasonable access to information from a related person. Further, the practical effect of a short look-back period is that, if no dividends were paid during that short period, the shares are per se passive. The inability of the Tested Foreign Corporation to show that a prior dividend was active under a longer look-back period due to a lack of historic information in practice leaves it no worse off. Final regulations could make clear that the burden is on the Tested Foreign Corporation to justify, based on available information, the extent to which prior year dividends were active, failing which they are presumed to be passive for purposes of the average.

6. Look-through Subsidiaries

A. General approach

Under Section 1297(c), if a Tested Foreign Corporation owns (directly or indirectly) 25% or more (by value) of another foreign corporation (i.e., a Look-through Subsidiary and such ownership threshold the “25%-By-Value Test”), for purposes of applying the income and asset tests the Tested Foreign Corporation is treated as if it earned its share of the underlying Look-through Subsidiary income and owned its share of the underlying Look-through Subsidiary assets directly.

The Proposed Regulations apply the 25%-By-Value Test to lower-tier subsidiaries on a “top-down” basis.48 For example, if TFC 1 owns 25% of LTS 1 which owns 25% of LTS 2, LTS 2 is not a Look-through Subsidiary as to TFC 1 as TFC 1 indirectly owns less than 25% (i.e., 25% of 25%) of LTS 2. LTS 2 would be a Look-through Subsidiary as to LTS 1 if LTS 1 were the Tested Foreign Corporation. Under the Proposed Regulations, as discussed in more detail below, the shares of LTS 2 that TFC 1 is deemed to own upon operation of the look-through to LTS 1 income and assets are characterized under general principles (rather than based on how LTS 1 treats those shares when it is the Tested Foreign Corporation).49

We agree with the implicit conclusion of the Proposed Regulations that the “directly or indirectly” language of Section 1297(c) is most consistent with a top-down approach to the question whether to treat LTS 2 as a Look-through Subsidiary of TFC 1. Accordingly, we agree that in the example above LTS 2 should not be treated as a Look-through Subsidiary of TFC 1 even if it is a Look-through Subsidiary as to LTS 1.

However, even if LTS 2 is not treated as a Look-through Subsidiary as to TFC 1 but is a Look-through Subsidiary as to LTS 1, a separate characterization question arises. When TFC 1 “looks through” to the gross assets and income of LTS 1, including the shares of LTS 2 that LTS 1 owns, should the character of the LTS 2 shares and dividends (as active or passive) be based on general principles or follow the implied characterization of such shares in the hands of LTS 1 (i.e., effectively pro rating the character based on the underlying active to passive ratio of LTS 2)? Section 1297(c) does not address this question.

B. Characterization of Look-through Subsidiary Items

i. General separate entity approach to characterization

The Proposed Regulations generally characterize items by reference to their character in the hands of the Look-through Subsidiary, subject to the special rule discussed in paragraph (iii) below permitting attribution of activities among members of a “50% group.”50

Under the Proposed Regulations, the determination of whether an item is received from a “related person” for purposes of Section 1297(b)(2)(C) is made based on whether the payor is related (under Section 954(d)(3)) to the Look-through Subsidiary, not whether the payor is related to the Tested Foreign Corporation.51 This generally reaches the appropriate results for reasons discussed below in paragraph (ii).

ii. Multi-tier structures

As noted above, however, the Proposed Regulations adopt an approach to characterizing shares of a lower-tier subsidiary as active or passive that arguably is not consistent with this general approach of characterizing items at the Look-through Subsidiary level. This may reach questionable results as illustrated in the next diagram.

Example 4

image 4

The Proposed Regulations appear to adopt a “top-down” approach both (1) to determining which entities are Look-through Subsidiaries and (2) to characterizing non-Look through Subsidiary shares at the level of a higher-tier Tested Foreign Corporation.52 For purposes of characterizing TFC 2 and TFC 4, the Look-through Subsidiary rule applies to LTS 1 and LTS 2 because each of TFC 2 and TFC 4 owns directly or indirectly 25% or more of LTS 1 and LTS 2, respectively. However, for purposes of testing TFC 1 or TFC 3, the Look-through Subsidiary rule only applies to TFC 2 and TFC 4 as each of TFC 1 and TFC 3, respectively, indirectly owns less than 25% (25% of 90% and 90% of 25%, respectively) of the relevant LTS company. As a result, TFC 1, is, for example, deemed to own 25% of TFC 2's assets and income, including only 22.5% (25% x 90%) of the LTS 1 shares.53

It is not clear a top-down rather than a bottom-up approach reaches better policy results given the purposes of the PFIC regime on the facts above. TFC 2 and LTS 1 in the example conduct an integrated operating business (and file consolidated financial statements presenting themselves commercially to the world as an integrated operating business). Given the purpose of the PFIC regime, it is not evident why it should matter whether LTS 1 and LTS 2 are 100% owned or 90% owned in testing TFC 1 and TFC 3 PFIC status.54 It may make sense to effectively view TFC 2 and LTS 1 as quasi-tax consolidated (single entity) solely for PFIC purposes. On the other hand, such a bottom-up (roll-up) approach arguably would reach a less justifiable result in the case of TFC 4 and LTS 2. In the abstract, a 50% test for rolling up lower tiers might make more sense as it is a better proxy for control and conduct of an integrated business.

We believe sensible results may be reached in most cases under the top-down approach, based upon the interaction of the Look-through Subsidiary rules and the Related Person Look-through Rules, provided that our proposal above as to how shares in non-Look-through Subsidiaries should be characterized is adopted.

LTS 1 and LTS 2 shares are presumptively, but not necessarily, passive under this top-down approach. The character of those shares also depends on the characterization of the dividends under the Related Person Look-through Rules. If TFC 2's ownership of LTS 1 exceeds the Section 954(d)(3) (50%) threshold for relatedness, the Related Person Look-through Rules may treat the dividends in whole or part as active and that would potentially make those shares active.55 However, at least under the approach of the Proposed Regulations, related party dividends are characterized by current year E&P making the asset test dependent on dividend policy.56 On the other hand, if our recommendation in Part 5(C) for characterizing such shares is adopted, sensible results would be reached in Example 4.

Our recommendation in Part 5(C) would characterize the LTS 1 shares if the Related Person Look-through Rule applies based on the ratio of active to passive underlying gross income over some specified rolling look-back period even if no dividends are paid. Accordingly, because LTS 1 is more-than-50% “related” to TFC 2, when TFC 1 looks-through TFC 2 under the Look-through Subsidiary rules, the LTS 1 shares may nevertheless be (at least in part) active. By contrast, LTS 2 is not “related” to TFC 4 and when TFC 3 looks through TFC 4 under the Look-through Subsidiary rules, the LTS 2 shares are passive.57 The “top-down” approach to Look-through Subsidiary testing when combined with our proposed approach to non-Look through Subsidiary share characterization seems to reach more appropriate results than either a pure “top-down” or “bottom-up” approach.

Assuming the related party dividend characterization approach we recommend in Part 5(C) is adopted, we therefore support the “top-down” approach of the Proposed Regulations. However, if the related party dividend characterization approach we recommend in Part 5(C) is not adopted, we believe that Treasury and the IRS should reconsider whether a “bottom-up” approach to characterizing non-Look-through Subsidiary shares reaches better policy results.

iii. Activity “Attribution”

Treating a Tested Foreign Corporation as if it directly owned the income and assets of a Look-through Subsidiary, as required by Section 1297(c), directly supports taking into account the Tested Foreign Corporation's own activities in determining whether Look-through Subsidiary income and assets are active or passive. It is less clear whether activities of a Look-through Subsidiary should be taken into account in characterizing income and assets directly owned by the parent Tested Foreign Corporation. However, nothing in the statutory language precludes this.

Operating groups frequently separate functions by legal entity and centralize employees and officers in one subsidiary but have property owned in others (e.g., in real estate or hospitality businesses). Testing the active or passive nature of a company's business solely based on activities conducted by the relevant Tested Foreign Corporation's own officers, employees and directors could cause the assets and income of a property owning company to be treated as passive even if the group actively conducts a business. The Proposed Regulations address this concern in the case of the active rent and royalty exceptions to FPHCI58 by allowing a Tested Foreign Corporation to take into account the activities of certain Look-through Subsidiaries in its economic group when determining whether an item of rent or royalty income is passive (“Activity Attribution”).59 The Proposed Regulations limit Activity Attribution to Look-through Subsidiaries with respect to which the Tested Foreign Corporation owns (directly or indirectly) more than 50% by value.60 The rationale for this threshold is to limit the number of owners to whom Look-through Subsidiary activities can be attributed.61

We agree that otherwise equally active operating businesses ideally should not be treated differently for PFIC purposes solely due to centralization of management or employment functions within a particular management subsidiary. Limiting the rents and royalties FPHCI exception to activities performed by each group member's own employees would frustrate the intention of the PFIC regime, which is to distinguish active operations from passive investment activities.

There is no particular policy reason, however, to limit the activities that may be “attributed” to those that are intended to earn affiliated rent or royalty income. In practice, these issues most frequently arise in scenarios involving rents and royalties, because the “active rent and royalty” test explicitly considers officer and employee involvement. Nonetheless, consideration should be given to extending the attribution exception to the Section 954(h) and commodity producer tests, as financial businesses (e.g., securities, derivatives or commodities dealers) typically segregate assets and operations that are part of an integrated business in different group entities for regulatory and other non-tax reasons.

The Proposed Regulations request comments on the adoption of the more-than-50% by value ownership requirement as a condition for Activity Attribution. While we recognize that a bright line value test has the benefit of simplicity and ease of administration and avoids the difficult, subjective and inherently factual analysis of the relationship between functional activities conducted by different entities, we question whether the more-than-50% by value threshold is the right test for permitting Activity Attribution.

According to the Preamble, the more-than-50% by value test was adopted to ensure that the “same” activities are only attributable within one economic ownership group.62 Unlike assets or income, both of which can be attributed proportionately on some basis, activities are qualitative. On one hand, we understand the concern over “double counting” the same activities. On the other hand, it is possible for more than one owner to materially participate in the underlying activity (particularly in a joint venture context). In such a case, it does not seem abusive for both owners to take the same subsidiary activities into account, as each owner would still only count their proportionate share of expenses for purposes of the active rent and royalty safe harbors in determining passivity. Accordingly, given the purpose of the PFIC regime is to distinguish passive from active conduct of activities, we do not think it is per se problematic that the rule could permit more than one economic group to take the same activities into account.

A test that looks to the extent to which the business is actually operated on an integrated basis by the group as a whole (perhaps incorporating “material participation” standards or even looking to tangible factors such as financial consolidation) likely reaches the most “correct” results in theory. However, we acknowledge that this approach by itself involves vague, factual and inherently subjective determinations and would be difficult to administer.

To reconcile the foregoing concerns, in lieu of the proposed more-than-50% by value threshold, we recommend an ownership threshold that allows Activity Attribution at a threshold of more than 50% ownership by the Tested Foreign Corporation whether by vote or value (i.e., the Section 954(d)(3) “relatedness” threshold), with the further requirement that, if the Tested Foreign Corporation owns more than 50% by vote but 50% or less by value of the “activity conducting” subsidiary, it materially participates in the same or a complementary line of business conducted by the activity conducting subsidiary. This would track the Related Person Look-through Rules under Section 1297(b)(2)(C) while still limiting the number of “beneficiaries” of Activity Attribution.63 A Section 954(d)(3) threshold also allows a more accurate determination of “control,” which is a better proxy for whether the activities of one company are in fact integral to earning income of another.

If, despite our discussion above, Treasury and the IRS remain concerned about inappropriate duplicative attribution of activities where the related person conducting the activities is not wholly owned despite material involvement, an alternative approach would be to pro rate in some way the extent to which the activities could affect the status of the Tested Foreign Corporation. Thus, the interest in the income generating and asset owning entity that could be favorably affected by attribution of activities from a different subsidiary could be limited based on the percentage ownership of the “activity conducting” subsidiary. For example, if the Tested Foreign Corporation owns only 30% of a subsidiary directly conducting real estate management activities, and also owns 50% of a property owning subsidiary (that is passive considered on a stand-alone basis), it might be permitted to characterize gross rental income and the related gross real estate assets of the property owning Look-through Subsidiary as active based on those activities under the active rental safe-harbor but only as to 30% of those income and assets. Such a proportionate limitation could apply at any ownership level or apply only when the Tested Foreign Corporation owns a 50 percent or less equity interest in the activity conducting entity.

C. Elimination of Look-through Subsidiary Items

The regime for Look-through Subsidiaries, which treats the Tested Foreign Corporation owner as if it owned and earned the assets and income of the underlying Look-through Subsidiary, would on its face potentially double count assets and income if, for example, the shares of the Look-through Subsidiary (and dividends thereon) are not eliminated. Legislative history indicates that Congress expected Treasury to develop rules to address this.64

i. Dividends

The Proposed Regulations would eliminate intercompany dividends between a Look-through Subsidiary and Tested Foreign Corporation only to the extent that such dividends are attributable to income of the Look-through Subsidiary that the Tested Foreign Corporation included in its gross income for purposes of determining its PFIC status.65

While we recognize that the Proposed Regulations' approach of eliminating only dividends that have been included in the Tested Foreign Corporation's gross income reaches arguably appropriate policy results by eliminating only those dividends that would otherwise “double count” the same income, we have several concerns with the approach.

The Proposed Regulations would not eliminate dividends arising from pre-acquisition E&P. The view that “double counting” does not arise when distributing pre-acquisition E&P is premised on the assumption that tax E&P accounts correlate with shareholder economic income. Economically, however, it is questionable whether a distribution “attributable” to E&P accumulated prior to the time shares were acquired is true “economic” income, as it reflects indirectly purchased E&P.66 Thus, we question whether there is any “abuse” that needs to be prevented.

In addition, the determination whether a dividend is “attributable” to income already included by the Tested Foreign Corporation can be complicated. The PFIC rules apply to “gross” items of income while E&P is measured on a net basis. For example, even if a Look-through Subsidiary has negative E&P in a given year, the Tested Foreign Corporation that owns the Look-through Subsidiary may take into account its share of the Look-through Subsidiary's gross income for PFIC purposes. Further, information required to reconstruct E&P and income from past years may be difficult to ascertain and is typically harder to establish the more time has passed. Because guidance on these rules has been lacking for many years, taxpayers may have to look far back into their financial history to implement the proposed “tested” double counting rule.

Depending on the method under which E&P must be traced to gross income, the intended benefit of the dividend “elimination” could in practice be nullified if a Tested Foreign Corporation is unable to establish the amount of E&P that had accumulated in years prior to acquiring at least 25% of the Look-through Subsidiary. For example, assume that a Look-through Subsidiary pays a dividend in a year in which it has no current E&P. If the dividend is deemed to be supported pro rata by the accumulated E&P pool, the Tested Foreign Corporation arguably would need to demonstrate that it has no pre-acquisition E&P (or the specific amount of pre-acquisition E&P) to be eligible to eliminate the dividend in whole or part. If the historic pre-acquisition E&P information for an acquired Look-through Subsidiary is not available, the Tested Foreign Corporation may in practice not be able to eliminate the dividend. It is often (perhaps usually) the case that foreign corporations are unable to reconstruct their entire prior year E&P accounts.

We therefore recommend removing the double counting requirement and simply eliminating dividends of Look-through Subsidiaries entirely. We believe that the “abuse” that the double counting rule is intended to prevent does not really exist.

However, if Treasury and the IRS are not inclined to remove the limitation as a general matter, to prevent de facto nullification of the dividend elimination for taxpayers with lengthy Look-through Subsidiary holding periods, we believe that the taxpayer should not be required to look back to 1986 to establish their entitlement to eliminate Look-through Subsidiary dividends.

One option would be to allow taxpayers a simplifying assumption as a safe harbor if they apply it consistently. In determining whether a dividend is “attributable” to income that has been picked up for PFIC purposes: (1) a dividend up to current year E&P would be deemed attributable to income that has been included in the Tested Foreign Corporation's income for purposes of determining its PFIC status and (2) to the extent the dividend exceeds current year positive E&P, if there was enough E&P looking to prior years in reverse serial order back to the 25% ownership start date year to cover that excess, then the Tested Foreign Corporation could treat the income as having been included in the Tested Foreign Corporation's income.

For example, if a Tested Foreign Corporation received a distribution of $100 in 2019 only $80 of which was attributable to current year E&P, then assuming the Tested Foreign Corporation owned at least 25% of the Look-through Subsidiary in 2018, the taxpayer would look to 2018 to see if there was sufficient total E&P after reduction by any 2018 dividend such that its share of the undistributed 2018 E&P was at least $20. If there was not sufficient E&P in 2018, the taxpayer would continue looking back to 2017 and other previous years in which the 25% ownership threshold is met.

As an alternative safe harbor, taxpayers could be allowed to pro rate the amount of dividend deemed attributable to pre-acquisition E&P based on the ratio of pre-acquisition to post-acquisition E&P but over a limited look back period (e.g., no more than five years). Under this approach, a Look-through Subsidiary holding period in excess of five years would essentially mean that dividends would be deemed to be paid from included E&P.

ii. Partial ownership limitation for eliminating debt and interest items

In the case of a Tested Foreign Corporation that owns less than 100% of a Look-through Subsidiary, under the Proposed Regulations only a proportionate amount of any debt receivable owing by the Look-through Subsidiary and interest thereon would be eliminated.67 For example, if the Tested Foreign Corporation owns 30% of a Look-through Subsidiary which owes the Tested Foreign Corporation $100, on which $10 of interest accrues for the year, then the Tested Foreign Corporation eliminates $30 of the debt for purposes of the asset test and $3 of the interest for purposes of the income test. Conversely, if the Tested Foreign Corporation was the debtor and the Look-through Subsidiary was the creditor, the Tested Foreign Corporation may entirely eliminate the debt and interest arising from its own obligation which it would otherwise pick up under the Look-through Subsidiary rule.68 We agree that the Tested Foreign Corporation should not take into account its share of its own obligations or interest thereon as it does not take into account any of the direct interest expense for purposes of the gross income test or any of the liability for purposes of the asset test regardless of its percentage ownership of the Look-through Subsidiary. We also agree with the Proposed Regulations' approach of limiting the elimination of debt receivables and interest owed to the Tested Foreign Corporation based on its proportionate direct and indirect ownership (by value) of the Look-through Subsidiary.

iii. Approach to rents and royalties and other items

We believe that the “proportionate” elimination approach for interest and debt described in paragraph (ii) above should be extended to intercompany rents and royalties (and any associated intangible asset inherent in the lease or license).

One could argue that the elimination rule should be extended beyond interest, rent, royalties and the related assets, because Section 1297(c) supports disregarding intercompany transactions involving Look-through Subsidiaries. For example, a sale of property by a Tested Foreign Corporation to a Look-through Subsidiary could be disregarded for purposes of the income and asset tests on the basis that Section 1297(c) treats the property that was sold as still owned by the Tested Foreign Corporation. The transaction could be treated as if it were a transfer to a disregarded entity of the Tested Foreign Corporation. Alternatively, the separate existence of the Look-through Subsidiary could be respected under a “quasi-consolidation” approach perhaps under a regime comparable to Treasury Regulation § 1.1502-13. Such an approach may raise complexities particularly where income and the corresponding item of tax expense are not reflected in the same year. For example, if a Tested Foreign Corporation sells depreciable property to a Look-through Subsidiary, one could eliminate the gain on the sale consistent with the fact that the depreciation associated with the basis step-up (an above-the-line deduction, not an exclusion) does not reduce gross income of the Look-through Subsidiary in future years. However, if the basis method rather than value method applies under Section 1297(e), the basis step-up would be taken into account for purposes of the asset test unless the basis step-up is also eliminated for that purpose (as would be the case if the intercompany transaction is disregarded altogether).

7. Look-through Approach to Partnerships

One of the more controversial features of the Proposed Regulations is likely to be the treatment of entities classified as partnerships for purposes of the income and asset tests.

The Proposed Regulations distinguish between “Look-through Partnerships” and other partnerships. A Look-through Partnership with respect to a Tested Foreign Corporation means a partnership at least 25% of the value of which is owned (as determined under Proposed Treasury Regulation § 1.1297-2(b)(1) as if the partnership were a corporation) by the corporation, in the case of the asset test on the relevant measuring date and in the case of the income test on the date the income is received or accrued.69 The principles applicable to corporate Look-through Subsidiaries generally then apply mutatis mutandis to Look-through Partnerships.

A Tested Foreign Corporation is treated as holding directly its proportionate share of the assets of a Look-through Partnership, applying subchapter K principles to determine this proportionate share of assets or tax basis, as applicable. In characterizing the underlying partnership assets, the assets are treated as producing passive or active income based on whether they are held to produce income of that character in the hands of the partnership (under Proposed Treasury Regulation § 1.1297-1(c)(2)). A Tested Foreign Corporation is treated as earning directly its allocable share of the income of a Look-through Partnership, characterizing the underlying income (including applying the exceptions in Section 1297(b)(2) and relevant exceptions to FPHCI in Section 954(c)) based on its character in the hands of the partnership and taking into account partnership-level activities only (subject to the special Activities Attribution principles discussed elsewhere). The same principles that apply under the rules eliminating look-through “items” with respect to a Look-through Subsidiary apply to a Look-through Partnership (generally, as if it was a corporation).

Income from a less-than-25% partnership is treated per se as passive and the partnership interest also is per se passive.70

The Preamble states that Treasury and the IRS determined that income earned through a partnership should be treated similarly to income earned through a corporate subsidiary and that for purposes of the asset test, partnerships should also be characterized in a manner similar to corporations. The Preamble notes that different treatment of partnerships for PFIC purposes is warranted because of the flexibility entities have in electing entity classification and that treating a subsidiary as a partnership may not have collateral income tax consequences as it does for a CFC. Implicitly, Treasury and the IRS appear to have concluded that elective entity classification may be inappropriate when the sole consequence is that it achieves a different result in classifying a Tested Foreign Corporation under the income and asset tests. Accordingly, the Proposed Regulations seek to achieve “equivalence” between partnerships and corporations for that purpose.

The proper approach to partnerships in applying the PFIC characterization tests (in particular look-through rules) raises difficult issues, as discussed below.

The plain language of the statute is not clear and contains no specific look-through rules for partnerships in applying the income and asset tests. As the PFIC statute was originally enacted in a pre-CTB world, it is of course possible that Congress simply thought the question of applying look-through rules to pass-through entities would rarely arise. However, the specific ownership attribution rule in Section 1298(a)(3) enacted for partnerships suggests Congress understood that PFIC ownership structure might include partnerships. It seems doubtful that Congress specifically intended that partnerships would be treated just like corporations for purposes of the look-through rules in that Section 1297(c) could have so provided. More likely, Congress may simply have assumed an aggregate approach applied. That would be consistent with the ownership attribution approach of Section 1298(a)(3) and, at least for purposes of the income test, Section 701 under which partnership level income passes through to the partner for all subchapter K purposes. It is less clear given Section 741 how Congress may have anticipated the asset test would apply to partnership interests.71 We believe many market participants have, now for many years, reasonably been applying an aggregate approach to entities classified as partnerships. Others may have followed some form of modified aggregate approach. However, we do not believe taxpayers believed or expected that partnerships were subject to the same look-through rules as corporations. We expect that the proposed approach, if adopted, may well result in many structures now being treated as non-PFICs becoming PFICs.

In light of regulatory silence on this question for over 25 years, we think adopting this approach should be undertaken with great caution given its potential to upend existing non-abusive structures and planning.

We would also note that the intended “equivalence” of the proposed approach is imperfect. As one example, because Congress did not treat partnerships and corporations equivalently there is no related party partnership distribution exception equivalent to the related party dividend exception under Section 1297(b)(2)(C). Consider a case in which TFC 1 owns 25% of LTS 1 which owns 90% of LTS 2. LTS 2 is not a Look-through Subsidiary as to TFC 1. However, as discussed above in Part 5(C), because LTS 2 is a related party as to LTS 1 under Section 954(d)(3), its dividends may be active and under the dual asset rule the shares in LTS 2 may be “active” so that when TFC 1 looks through to LTS 1, the LTS 2 shares do not tend to make TFC 1 a PFIC.72 However, in an identical situation, in which LTS 1 owns 90% of a partnership, PRS, rather than LTS 2, the PRS interest and PRS income allocable to LTS 1 is per se passive under the Proposed Regulations.73

As another example of the equivalence problem, in the example above, assume that both LTS 2 and FPRS conduct a 100% “passive” activity and each earns $100 of net profit. Although the dividend from LTS 2 is passive under the related party dividend rule, because dividends are determined by “net” E&P, TFC 1 would reflect a $100 passive income item when looking through LTS 1. By contrast, we assume that the passive item reflected with respect to FPRS is 90% of the underlying “gross” passive income of FPRS allocable to LTS 1 under subchapter K, which if LTS 2 and FPRS are highly leveraged could be a much larger amount.

It is somewhat peculiar in this and the preceding example, that the intended “equivalence” approach would in some sense treat the partnership as more “opaque” or passive than the similarly situated lower-tier corporation.74 We suspect there may be other anomalies like this we have not yet identified.

A further discontinuity may arise from the application of Section 954(c)(4), which contains the look through rule for sales of partnerships interests in determining FPHCI when the selling partner owns more than 25% of the capital or profits interests in the partnership (with the application of specified attribution rules similar to Section 958(b)).75 This will not necessarily correspond to the 25% by value threshold adopted by the Proposed Regulations for Look-through Partnerships.76 As one example, a Tested Foreign Corporation may own a more–than-25% profits interest but much less of the capital and less than 25% by value of the partnership. The partnership will not be a Look-through Partnership and will generate passive income and be a passive asset unless and until the Tested Foreign Corporation sells the interest, at which point a look-through approach applies.77 Section 954(c)(4) represents Congress's fairly recent view of an appropriate look-through threshold in a fairly analogous context. If the general approach of the Proposed Regulations is retained in final regulations, we recommend that Treasury and the IRS at least consider whether the applicable threshold for partnerships should conform to Section 954(c)(4) rather than Section 1297(c).

The Report summarizes below the advantages and disadvantages of the proposed approach and also evaluates the advantages and disadvantages of alternative approaches:

The primary advantages of the proposed approach to 25% Look-through Partnerships are (1) certainty and ease of administration, in that it provides a relatively clear, bright line and objective test and limits the need to obtain information about the active versus passive nature of underlying assets, which may be difficult for small partners to obtain, and (2) greater equivalence of result as between a lower-tier entity that has or has not elected to be treated as a pass-through under the CTB regime. For the reasons explained above, however, this equivalence is imperfect and the approach may treat partnerships less favorably than corporations in practice although likely in a relatively small number of situations. While the desire for equivalence may be driven by the desire to reach the same result regardless of whether the CTB election (which has limited non-tax relevance) is made, the solution imposes the same results on entities that are partnerships for tax purposes by default,78 as distinguished from being classified as a partnership by reason of a CTB election (or else the regime must distinguish “CTB” elective partnerships from default partnerships in applying the rules, an approach that would create its own discontinuities). While we appreciate the intuition against having tax results depend on a fictive tax election given the purposes of the PFIC rules, the application of the CTB regime in the international context raises this concern in many other areas.79 We question the utility of trying to achieve equivalence in this very narrow context.

Fundamentally, we question equivalence as a worthwhile goal unless the corporate subsidiary test of Section 1298(a)(2) reaches appropriate results most of the time. In our view, based on experience in applying this test in real world situations, the 25% by value ownership test has not proven to be a particularly effective proxy by which to distinguish actively conducted underlying businesses from passive activities or investment. Of necessity, a bright line value threshold must be set at a high enough level to ensure that passive investors will rarely accumulate such a large position. However, this then treats as passive some ownership stakes in businesses that are not motivated by passive investment. In some countries, for example, it is more usual than in the United States for a supplier to take an equity stake in an important customer (or vice versa) to cement important active business ties. In other countries, large conglomerates routinely are structured with complex cross-ownership arrangements that (whatever the corporate governance merits) do not reflect a passive investment intent. A 10% general or managing partner may be very involved in controlling the underlying business in which many smaller passive partners are invested. For better or worse, the 25% by value test is prescribed by the statute for Look-through Subsidiaries. However, because in our experience the 25%-by-value test functions imperfectly in distinguishing businesses that are intuitively active from those that are passive, we see no compelling policy reason to extend that approach to entities treated as partnerships. Nevertheless, there is an administrative advantage to limiting the need to obtain information as to the active versus passive nature of underlying assets which may be difficult for a small partner to access.

An alternative approach would be to provide that, solely for purposes of, but for all purposes of the PFIC classification tests the “CTB” regime does not apply and a company is treated as a corporation. This has the virtue at least of achieving true “equivalence” and preventing a purely tax driven election from affecting the PFIC status of a Tested Foreign Corporation. However, that approach raises authority questions. More importantly, it would treat legal partnerships and CTB partnerships differently and resurrect the complexities of applying the multi-factor Kintner regulations of prior law.

Another approach is to treat any pass-through entity as an aggregate without regard to the ownership threshold, involvement of the partner in the business or other factual indicia. This objective approach achieves a degree of certainty and ease of administration. On the other hand, small partners would need to obtain information as to the active versus passive nature of underlying assets which may be difficult. A pure aggregate approach also provides flexibility to taxpayers to minimize misclassifying active group businesses with complex structures as PFICs and thereby minimizes the likely unintentional overbreadth of the PFIC statute.80 We also believe it would be consistent with the statutory structure and the approach many taxpayers have been following. Obviously, however, it may result in very different results depending on whether a Tested Foreign Corporation is able to make CTB elections for less-than-25% owned lower-tier entities. The same flexibility this approach provides could, however, enable taxpayers to achieve results arguably contrary to the purposes of the PFIC regime.

As an example, assume a Cayman corporation owns less than 10% by value in each case of shares in ten or more non-publicly traded foreign companies in a variety of different businesses and (because, fortunately for the Tested Foreign Corporation, none of these are per se entities or have other U.S. shareholders who might object) a CTB election can be made on all of them. The Cayman corporation can avoid PFIC status by causing a CTB election to be made to classify the non-publicly traded foreign companies as partnerships if those companies are generally active. The Cayman corporation would clearly be a PFIC absent such CTB election. For the reasons described above, we do not believe that discontinuity in the treatment of entities treated as corporations rather than as pass-through entities is problematic in and of itself. The appropriate question is whether this kind of self-help planning could allow taxpayers to reach results that are contrary to the purposes of the PFIC regime. There are many circumstances in which the flexibility an aggregate approach affords permits planning to avoid PFIC classification in circumstances Congress likely did not anticipate would result in PFIC status. However, that same flexibility permits planning that may allow taxpayers to avoid PFIC status where doing so is arguably inappropriate, including perhaps the case above.81

We have considered whether an aggregate approach combined with an anti-abuse rule could be appropriate. The anti-abuse rule could, for example, take into account factual indicia such as whether the Tested Foreign Corporation syndicates its shares, holds a diversified portfolio of portfolio companies or has other features indicative of investment funds.82 Ultimately, we do not favor such an approach as it will introduce considerable uncertainty and more fundamentally involves significant circularity. It is not possible to define what cases are abusive without a clear consensus on the fact patterns Congress intended to be subject to the PFIC regime. The PFIC statute itself can be viewed as an “anti-abuse” rule that speaks to what Congress intended to capture. Applying a separate anti-abuse rule premised on features Congress has taken into account in other contexts as indicia of passivity,83 but which Congress did not choose to import into the statutory definition of a PFIC seems problematic.

Another approach may be to require a look-through approach if the Tested Foreign Corporation itself, directly and with the application of Activity Attribution principles, materially participates in the underlying business of the partnership. For example, a Tested Foreign Corporation that is the general partner of a partnership could then apply a look through approach even if the economic interest of the Tested Foreign Corporation is much smaller than 25%. Many permutations of such an approach present themselves. The rules could provide for per se Look-through treatment for a 25% or greater partner but permit a look-through at lower thresholds of economic ownership if a material participation test is met, only apply Look-through principles if a material participation test is met, or permit a Look-through at some lower level of minimum economic ownership (e.g., at least 5% of the capital or profits) if a material participation test is met. Where such material participation is present, it is also more likely that the materially participating partner has ready access to underlying information about the partnership's assets, income and activities.

The primary advantage of those approaches, in principle, is their consonance with intuitions about whether the Tested Foreign Corporation owns the partnership interest as part of an active business or as a passive investment and thus those approaches would seem to reach results more consistent with the purposes of the PFIC regime. The primary disadvantage is that such tests are inherently more factual and potentially subjective and therefore less administrable.84 For example, if the material participation test permits a look-through approach if the Tested Foreign Corporation materially participates in the underlying business based on all the facts and circumstances it will be unclear what facts are most indicative of “active” participation and how much participation is material. Such a test would impose significant audit burdens on the IRS. In practice, the rule could devolve into a de facto pure “aggregate” regime for more aggressive taxpayers. It is also questionable whether the kinds of factors that apply in arguably similar contexts like Section 469 in determining material participation work in this context. An individual taxpayer's hours to be spent on given activities during a year are finite and therefore the proportion of time spent on the activity may be a reasonable proxy for active participation. A Tested Foreign Corporation by contrast has non-finite resources and in theory can devote whatever employee effort is required to meet a given test.

One metric which may be worth considering, however, is the expense incurred directly (or by application of Activity Attribution principles) by the Tested Foreign Corporation with respect to its position in the foreign partnership. Generally, the more “passive” an investment, the more critical it is for an investor to minimize expenses to carry the investment in order to maximize investment returns. Passive mutual funds with high expense ratios, for example, must compete with even more passive funds employing indexing strategies. A similar intuition is behind the active rental “substantiality” safe harbor in Treasury Regulation § 1.954-2(c).

Thus, Regulations could provide, for example, that a Tested Foreign Corporation may treat a partnership as a Look-through Partnership despite owning a less than 25 percent interest if certain defined expenses incurred to manage the interest during the relevant period exceed a specified percentage of the fair market value of the partnership interest held by the Tested Foreign Corporation. We express no view on what the appropriate percentage would be as this is not a legal question. But we suspect it would be rare, for example, for a “passive investor” to acquire an interest when it would incur annual expenses to carry the investment of more than a small percentage of the value of the interest. An alternative would be to compare the direct expenses incurred to the net partnership income allocable to the Tested Foreign Corporation under Section 701 for some defined rolling period (e.g., the shorter of the preceding three years or the tested foreign corporation's ownership period). Alternatively, the safe harbor could compare the direct expense to the allocated gross expense items from the partnership for that period. We doubt, for example, that a partner with a small “by value” percentage interest that incurs substantial direct expenses relative to its allocable share of the underlying partnership's business expense is holding the interest as a passive investor. As in the case of the active rental safe harbor under Section 954, it would be necessary to identify expenses at the Tested Foreign Corporation level appropriate to take into account. For example, management fees paid to unrelated persons to manage the position and interest expense should not count but officer and employee expenses (taking into account Activity Attribution principles) should count.

On the other hand, such a material participation standard does not encompass situations like those mentioned above — such as a supplier taking an equity stake in an important customer (or vice versa) to cement important active business ties rather than for passive investment purposes. Accordingly, such a material participation standard would not necessarily reach appropriate policy results in light of the purposes of the PFIC rules in every situation.

8. Stapled Entities

The Proposed Regulations85 provide that when equity interests in two or more foreign entities are stapled within the meaning of Section 269B(c)(2), they are treated as a single entity in applying the PFIC asset and income tests. Section 269(B)(c)(2) generally treats two entities as stapled if more than 50% in the beneficial ownership interests in each entity are stapled interests (i.e., subject to ownership restrictions preventing transfers of one interest unless the other interest is also transferred).

If foreign structures using stapled entities involve close to 100% overlap (i.e., all of the equity of both entities is stapled), single entity treatment is consistent with Treasury's rationale.86 As a matter of general federal income tax principles, and given the purposes of the PFIC regime, it seems appropriate to treat the stapled entities as a single entity for PFIC classification purposes in such cases.

However, stapling need not involve 100% overlap, and the rule creates planning opportunities as well as potentially unintended results.

The stapled entity regulations87 are arguably overbroad in that they may encompass pairs of entities that do not economically represent a single economic interest for their shareholders. As an example, consider two corporations that each have two classes of shares where 75% of one class is stapled to 25% of the other. Assume, for example, that TFC 1 has issued class A shares representing 75% by value of all of the TFC 1 equity which are stapled to TFC 2's class B shares which represent 25% of the value of TFC 2 equity, and that the TFC 1 class B shares are stapled to the TFC 2 class A shares. Under the regulations TFC 1 and TFC 2 are stapled entities.88 Yet, the “stapled entity” is economically not really a single entity and the owner of the stapled TFC 1 A - TFC 2 B stapled shares may have a fundamentally different economic investment than the owner of the TFC 1 B - TFC 2 A stapled shares.

Further, the Proposed Regulations do not address how a holder of a non-stapled interest in an entity that, on a stand-alone basis, would or would not be a PFIC, should be treated. That is, a question arises as to whether the non-stapled interest's status as an interest in a PFIC should be tested on a separate or a combined basis if: (1) both entities would be PFICs on a stand-alone basis, so presumably the combined entity will be a PFIC; (2) neither entity would be a PFIC on a stand-alone basis, so presumably the combined entity will be a non-PFIC; (3) one entity would be a PFIC on a stand-alone basis and the other would be a non-PFIC but the combined entity would be a PFIC and (4) one entity would be a PFIC on a stand-alone basis and the other would be a non-PFIC but the combined entity would be a non-PFIC. In the first and second scenarios, the question of separate versus combined testing of the unstapled interest may not matter. However, in the third and fourth scenarios, the combined entity approach to PFIC characterization is more problematic as to the holders of unstapled shares. For example, a holder of unstapled shares in the stand-alone PFIC could be treated as holding non-PFIC shares based on the combined entity classification, and vice versa. This may not be appropriate.

The simplest solution may be to limit the single entity (combined PFIC classification) approach to situations in which all but a de minimis quantum of the outstanding equity interests in both entities are stapled to each other. In the absence of a specific provision in the PFIC rules mandating the use of the Section 269B definitions, there is nothing preventing a more restrictive definition of “stapled entity” for PFIC purposes.

However, if Treasury and the IRS do not wish to limit the rule to situations involving close to 100% stapling, an alternative solution may be to clarify how the anti-abuse rule of Section 1298(b)(4) applies in this context. That provision allows separate interests in a single corporation to be treated as PFIC interests where necessary to prevent abuse89 and could potentially apply to stapled interests. The Proposed Regulations should clarify (although it is implicit) that the holders of the stapled interests in a deemed single non-PFIC entity are not automatically subject to Section 1298(b)(4) but rather may be subject to Section 1298(b)(4) if unusual features exist (e.g., stapled preferred interest in the separate PFIC disproportionate in value to the relative value of the PFIC and non-PFIC stapled entities) and the arrangement otherwise would allow such a holder to avoid PFIC treatment.

A U.S. person that owns a non-stapled interest in the entity that would be a PFIC on a stand-alone basis generally should be treated as owning a PFIC interest under Section 1298(b)(4). From a policy perspective, a holder of unstapled shares in an entity that would be a PFIC on a stand-alone basis should be treated as a PFIC shareholder. But equally, there is no policy reason to treat a holder of unstapled shares in an entity that would not be a PFIC on a stand-alone basis as a PFIC shareholder merely because the combined entity would be a PFIC.90

The treatment of stapled entities under the Section 1298(b)(4) approach may require additional guidance, e.g., on how to determine pro rata shares under the QEF regime and how the mark-to-market election should apply to the non-stapled portion of a class of shares the stapled portion of which is traded.

9. Effective Date and Interim Reliance Issues

The Proposed Regulations are proposed to apply to taxable years of U.S. persons that are shareholders in Tested Foreign Corporations beginning on or after the date of publication of the Treasury Decision adopting the Proposed Regulations as final regulations in the Federal Register.91 However, the Proposed Regulations and Preamble provide that, until the Proposed Regulations are finalized, taxpayers may apply the Proposed Regulations (other than Proposed Treasury Regulations §§1.1297-4 and 1.1297-5) in their entirety to all open taxable years as if they were final regulations provided that taxpayers consistently apply the Proposed Regulations (“Reliance Relief”).92

We believe the intention of Reliance Relief is to permit taxpayers to amend their tax returns for open taxable years and, presumably, to file for refunds. If the Tested Foreign Corporation would not have been a PFIC based on the Proposed Regulations in those open years, and also would not have been a PFIC in any closed years under the Proposed Regulations, the taxpayer clearly would be permitted to treat the Tested Foreign Corporation as not being a PFIC and could obtain the intended Reliance Relief.

An issue not addressed in the Proposed Regulations — and one that will need to be addressed either when final regulations are issued or as supplemental guidance to the Reliance Relief — is the interaction of Reliance Relief with Section 1298(b)(1). Section 1298(b)(1) provides that if at any time during the holding period of a taxpayer with respect to the stock of a foreign corporation such corporation was a PFIC, then unless a QEF election has been in effect the stock continues to be treated as PFIC with respect to that taxpayer regardless of the corporation's ongoing classification under the general rules (commonly referred to as the once-a-PFIC-always-a-PFIC rule or the “OPAP Rule”). The OPAP Rule may interfere with the taxpayer's ability to avail itself of Reliance Relief.

Assume that, based on Notice 88-22 or general principles of statutory interpretation, a U.S. person as a matter of prudence, and in light of the long-standing lack of clarity regarding the PFIC rules, has taken a conservative position that a corporation was a PFIC and has reported the stock of such corporation on its tax returns on that basis. Assume further that if the U.S. person were to apply the principles of the Proposed Regulations consistently, the Tested Foreign Corporation would not have been a PFIC in any open or closed taxable year of the U.S. person. The taxpayer cannot amend its closed year returns. The Proposed Regulations do not address whether a taxpayer is entitled to Reliance Relief in that situation.

A related problem could arise if the shareholder took the position in the past under prior law for closed years that the corporation was a PFIC, the taxpayer did not make a QEF election, and now under the Proposed Regulations the corporation would not be a PFIC during the past periods (at least in the open years), yet in the future the corporation becomes a PFIC even under the Proposed Regulations. The shareholder might at such time as the corporation becomes a PFIC under the Proposed Regulations want to make a QEF or mark-to-market election without a purging election or else not make such an election and have the Section 1291 look-back period not include the closed years.

We believe that some relief from the operation of the OPAP Rule is appropriate in either such case. Failing to provide such relief from the OPAP Rule otherwise may negate the relief intended under the Reliance Relief for many such taxpayers. It may also have the perverse effect of penalizing taxpayers who, in closed taxable years, took conservative positions regarding the PFIC status of foreign corporations in which they held stock compared to taxpayers who took more “aggressive” positions that with the benefit of hindsight have now been validated by the Proposed Regulations. Accordingly, we believe the Reliance Relief clearly should be available even if a foreign corporation has been treated as a PFIC under prior law with respect to the taxpayer in earlier closed years if it would not have been a PFIC under the Proposed Regulations. To minimize the burden of a multi-year redetermination, the relief could also provide that, if the taxpayer can demonstrate with respect to any open years and the first closed year in which the taxpayer affirmatively took the position the Tested Foreign Corporation was a PFIC that such corporation would not have been a PFIC under the principles of the Proposed Regulations, then the taxpayer may assume it was not a PFIC for all closed years solely for purposes of the OPAP Rule.

In any event, we believe that such taxpayers should be permitted to benefit from Reliance Relief without an affirmative requirement to file amended tax returns. For many taxpayers who have held stock in a foreign corporation and taken the position that the OPAP Rule was applicable, the requirement to file an amended return may involve significant administrative burden. The actual cash taxes may not have been material (e.g., if the taxpayer has been subject to the excess distribution regime but no material distributions or gains have been realized). Some form of statement attached to a current-year tax return stating that the taxpayer is adopting the position that the OPAP Rule is not applicable and explaining in reasonable detail how the Proposed Regulations support that change in position should suffice.

Conversely, it is theoretically possible that a taxpayer that has reasonably treated a Tested Foreign Corporation as not a PFIC based on prior law in certain closed years (e.g., applying an aggregate theory of partnerships) would have been required to treat the corporation as a PFIC had the Proposed Regulations been in effect. If, due to a change in circumstances, the corporation now would not be a PFIC under the Proposed Regulations, the OPAP Rule would apparently preclude the taxpayer from relying on the Proposed Regulations to treat the Tested Foreign Corporation as a non-PFIC even if the taxpayer applies the Proposed Regulations consistently to open years and going forward. On the one hand, it could be argued that the taxpayer should not have the best of both worlds. On the other hand, in light of the many years since the enactment of the PFIC statute during which there has been limited guidance on the proper application of key aspects of the PFIC classification regime, this result seems harsh if the prior positions were a reasonable, good faith application of the law prior to the Proposed Regulations.

A similar issue will arise after final regulations are promulgated. Presumably, taxpayers will not be required to retroactively redetermine for all prior open and closed years that the Tested Foreign Corporation was not a PFIC to prove that the OPAP Rule does not apply. The IRS could of course challenge unreasonable applications of prior law for open years, but it would arguably be unfair to challenge a reasonable interpretation of prior law and apply the OPAP rule on the basis that final regulations require a different result.

Reliance Relief could be burdensome if a taxpayer must demonstrate that the Tested Foreign Corporation would not have been a PFIC applying the principles of the Proposed Regulation for any year in the taxpayer's holding period (including closed years). As a practical matter information for much earlier years needed to “prove” non-PFIC status under the principles of the Proposed Regulations (or future final regulations) may not even be available.

There is no simple solution to this problem given the retrospective nature of the OPAP rule. One reasonable compromise would be to provide that if, by application of the principles of the Proposed Regulations, a Tested Foreign Corporation is not a PFIC and also would not have been a PFIC during any open taxable year of a U.S. person's holding period, the OPAP Rule will not be asserted with respect to the U.S. person's ownership of stock in such Tested Foreign Corporation based on closed years. This relief could be subject to the proviso that the closed year position was a reasonable interpretation of the law as then in effect, although this would create some uncertainty as to the application of the relief and it is also doubtful in practice that the IRS will ever wish to audit multiple closed years to determine if a foreign corporation would have been a PFIC under prior law.

FOOTNOTES

1The principal drafter of this Report was Andrew Walker with helpful comments from Andy Braiterman, Kim Blanchard, Bob Cassanos, Ed Gonzalez, Shane Hoffman, Alan Kaden, Stephen Land, Jiyeon Lee-Lim, Lea Li, John Lutz, Debbie Paul, Eschi Rahimi-Laridjani, Rick Reinhold, Yaron Reich, Mike Schler, Michael Shulman, David Sicular, Joseph Toce, Joseph Tootle and Gordon Warnke. This Report reflects solely the views of the Tax Section of the New York State Bar Association and not those of the Executive Committee or House of Delegates of the New York State Bar Association.

284 Fed. Reg. 133, 33120-33161 (July 11, 2019). Unless otherwise indicated, all Section (and §) references are to the Internal Revenue Code of 1986, as amended (the “Code”), or the Treasury Regulations promulgated thereunder. Defined terms not defined in this Report have the meaning given in the Proposed Regulations.

3See NYSBA Tax Section Report, Select Issues with Respect to the Passive Foreign Investment Company Rules, dated May 8, 2010 (the “2010 Report”); NYSBA Tax Section Report, Proposals for Guidance with Respect to Passive Foreign Investment Companies, dated May 22, 2001 (the “2001 Report”).

4See the 2010 Report and 2001 Report, supra n. 3; N.Y. City Bar Ass'n,, Report Offering Proposed Guidance Regarding the Passive Foreign Investment Company, dated September 21, 2009 (the “NYCBA 2009 Report”).

5See, e.g., 2001 Report at 2-8.

6See Prop. Treas. Regs. §1.1291-1(b)(8)(iii)(A).

7Prop. Treas. Regs. §1.1291-1(b)(8)(iii)(A).

8See Prop. Treas. Regs. §1.1291-1(b)(8)(iv)(B), Ex. 2.

984 Fed. Reg. 133 at 33121.

10See Prop. Treas. Regs. §1.1291-1(b)(8)(iii)(A).

11See Prop. Treas. Regs. §§1.1291-1(b)(8)(iii)(A) and (b)(8)(iv).

12See 2010 Report at 23, 25.

13See Prop. Treas. Regs. §§ 1.1297-1(c)(i)(A)-(C).

14Prop. Treas. Regs. §1.1297-1(c)(1)(i)(B).

15Prop. Treas. Regs. §1.1297-1(c)(1)(i)(A).

1684 Fed. Reg. 133 at 33123.

17See 2010 Report at 24.

18Section 954(c)(1).

19See Sections 954(c)(1)(B) (gains from certain property), 954(c)(1)(C) (commodities transactions), 954(c)(1)(D) (foreign currency transactions), 954(c)(1)(f) (notional principal contract income).

20As discussed in Part 6(C), to prevent duplication certain intercompany items (such as dividends and interest paid by the Look-through Subsidiary to the Tested Foreign Corporation) are eliminated. Under Section 1297(b)(2)(C), passive income does not include any interest, dividend, or rent or royalty income which is received or accrued from a related person (within the meaning of Section 954(d)(3)) to the extent such amount is properly allocable to income of such related person which is not passive income (i.e., the Related Person Look-through Rules). In practice, because of the elimination of Look-through Subsidiary items, the Related Person Look-through Rules generally will be relevant only when the “related” payor is not also a Look-through Subsidiary of the Tested Foreign Corporation.

21See Prop. Treas. Regs. § 1.1297-2(g)(1).

222010 Report at 34 and See H.R. Conf. Rep. No. 99-841 at II-644 (1986); Staff of J. Comm. on Taxation, 99th Cong., General Explanation of the Tax Reform Act of 1986 at 1026 (Comm. Print 1987) (the “1986 Blue Book”).

23Section 1298(g) provides that the Secretary shall prescribe such regulations as may be necessary or appropriate to carry out the purposes of this part.

24Whatever approach to “look through” partnerships is adopted (as discussed below), given that Subchapter K allocates partnership gain and losses to partners on an aggregate/look-through basis, we believe that a partner's allocable gains and losses should be netted.

25See Section 1297(c)(2).

26In practice, because of the elimination of Look-through Subsidiary items, the Related Person Look-through Rules generally will be relevant only when the “related” payor is not also a Look-through Subsidiary of the Tested Foreign Corporation. See note 20, supra.

27Prop. Treas. Reg. § 1.1297-1(c)(3)(iv).

28Prop. Treas. Reg. § 1.1297-1(c)(3)(i). We assume that the Proposed Regulations intend the allocation be made based on the ratio of gross non-passive income to gross total income. However, it would be helpful to clarify this in final regulations.

29See Prop. Treas. Reg. § 1.1297-1(c)(3)(i).

30The Preamble recognizes that because the PFIC income test is applied on the basis of gross income, interest expense cannot be used to reduce the amount of passive income and, therefore, the PFIC regime does not raise the policy concerns addressed by the “cream skimming” rule in the foreign tax credit and subpart F contexts. 84 Fed. Reg. 133 at 33124.

31Prop. Treas. Reg. § 1.1297-1(c)(3)(ii).

3284 Fed. Reg. 133 at 33125.

33See Prop. Treas. Reg. § 1.1297-1(c)(3)(ii).

34See Prop. Treas. Reg. § 1.1297-1(d)(2) (adopting a dual asset approach, as discussed further below).

35Necessarily, the Tested Foreign Corporation must own “directly or indirectly” less than 25% of the payor (from a “top-down perspective”) or the payor would be a Look-through Subsidiary. In the great majority of such instances, it is unlikely the affected U.S. shareholders of the Tested Foreign Corporation will be in any position to exercise effective practical control over the related dividend payor for planning purposes.

36Prop. Treas. Reg. § 1.1297-1(c)(3)(iii).

372010 Report at 24-28 and 2001 Report at 12-17 and 23-41. See also NYCBA 2009 Report at 34-35.

38Cf. Prop. Treas. Reg. § 1.1297-1(d)(2) (treating shares subject to the Related Person Look-through Rules as partly passive and partly active assets in proportion to the active versus passive income treatment of dividends paid on the shares).

39A possible exception may be offshore money market funds (where most of the assets could be relatively low yielding fixed income investments). In practice, these vehicles generally pay current dividends and, at least at current market rates, would require a very long holding period before a material tax deferral benefit was achieved in any event. However, if there is concern that potentially abusive structures could evolve, rules could impose additional requirements. For example, rules could limit the percentage of assets that could rely on the working capital exception to no more than 25 percent of the aggregate assets of the Tested Foreign Corporation. However, this would present additional complexities such as how to apply Look-through Subsidiary Rules if treasury functions are centralized in one group company.

40Prop. Treas. Reg. § 1.1297-1(d)(1)(v)(A).

4184 Fed. Reg. 133 at 33125.

42Among other (likely unintended) consequences of the repeal of Section 958(b)(4), downward attribution of foreign subsidiary shares to U.S. subsidiaries may now cause subsidiaries of groups with a foreign parent company to be technically treated as CFCs even if there are no inclusion U.S. shareholders who own 10% or more of the CFCs. Section 958(b)(4) repeal makes it more likely that small U.S. shareholders must use basis rather than value in determining PFIC status, in turn, making PFIC status more likely for reasons explained above. There is no policy reason for this result, which cannot have been intended by Congress. We believe Treasury and the IRS should provide relief at least in this case.

43As noted in the prior footnote, the subsidiary may be a CFC because there is concentrated indirect U.S. ownership or merely because, after repeal of Section 958(b)(4), the foreign subsidiary is deemed to be owned by a U.S. subsidiary of the foreign parent even if there are no U.S. “inclusion” shareholders.

44See 1986 Blue Book at 1024.

45Prop. Treas. Reg. § 1.1297-1(d)(2)(iii).

46Id.

47If the approach of the Proposed Regulations is retained in final regulations, rather than our suggested approach below, it would be helpful for the final regulations to make this explicit.

48Prop. Treas. Reg. § 1.1297-2(b)(1).

49See Prop. Treas. Reg. § 1.1297-2(b)(3), Ex. 1.

50Prop. Treas. Reg. § 1.1297-2(d) and (e).

51See Prop. Treas. Reg. §1.1297-2(d)(1) (relatedness tested at the Look-through Subsidiary level not Tested Foreign Corporation level).

52Prop. Treas. Reg. § 1.1297-2(b)(1) and (g)(1).

53See Prop. Treas. Reg. § 1.1297-2(b)(3), Ex. 1.

54Whatever approach is adopted may cut both ways. In the example above, assume LTS 1 is a highly levered finance entity that does not meet the tests to be an active bank and TFC 2 is a pure holding company. Treating the LTS 1 shares as passive when testing TFC 1 may be taxpayer favorable if TFC 1 has other active assets as only 22. 5% of the value of the LTS 1 shares is treated as passive rather than 22.5% of the value of the gross assets of LTS 1.

55See Prop. Treas. Reg. §1.1297-2(d)(1); discussion above in Part 5(C).

56Prop. Treas. Reg. § 1.1297-1(c)(3)(ii).

57It is important to note that this approach differs from applying the look-through approach under Section 1297(c), because the value or basis of the LTS 2 shares (a number inherently net of debt or other liabilities at the LTS 2 level) is characterized but TFC 1 does not look through to the gross asset basis or value of the underlying LTS 2 assets.

58See Treas. Reg. §§ 1.954-2(c)(1) and 1.954-2(d)(1).

59See Prop. Treas. Reg. §1.1297-2(e)(1).

60See Prop. Treas. Reg. §1.1297-2(e)(1).

61More than one Tested Foreign Corporation may in effect benefit from Activity Attribution. For example, if TFC 1 owns 100% of LTS 1 which owns 95% of LTS 2, LTS 2's activities may be attributed both to TFC 1 and LTS 1 when TFC 1 or LTS 1 are the Tested Foreign Corporations. However, Activity Attribution is permitted only within a single (more than 50% related) economic ownership group. Conversely, as we read the Proposed Regulations, a lower-tier Tested Foreign Corporation may not attribute activities conducted by its more-than-50% owner. In other words, activities can only be attributed from an entity that is directly or indirectly owned by the Tested Foreign Corporation.

6284 Fed. Reg. 133 at 33139.

63It would also be similar to the “active business” prong of the limitation on benefits tests that apply under U.S. income tax treaties. See United States Model Income Tax Convention (2016), Article 22(3)(c).

64See 1986 Blue Book at 1025 (“Under this look-through rule, amounts such as interest and dividends received from foreign or domestic subsidiaries are to be eliminated from the recipient's income in applying the Act's income test and the shareholder's stock investment is to be eliminated from its assets in applying the Act's asset test. For example, a foreign holding company that receives dividends or interest from a subsidiary is not treated as receiving passive income unless the subsidiary derives passive income”); H.R. Rep. No. 100-795 at 268 (1988) (“Amounts such as interest and dividends received from foreign or domestic subsidiaries are eliminated from the shareholder's income in applying the income test and the stock or debt investment is eliminated from the shareholder's assets in applying the asset test.”)

65See Prop. Treas. Reg. § 1.1297-2(c)(2).

66The issue would not arise if a Section 338(g) election, which would eliminate legacy E&P pools, has been made as to the acquired Look-through Subsidiary. Often, however, an acquiring Tested Foreign Corporation may not realize that its small U.S. shareholders will be affected by the acquisition of another foreign corporation. There will also be many Tested Foreign Corporations who acquired foreign subsidiaries prior to publication of the Proposed Regulations with no reason to expect such an election would be beneficial. Not eliminating dividends attributable to pre-acquisition E&P will primarily be a trap for the unwary.

67See Prop. Treas. Reg. §1.1297-2(c)(1).

68See Prop. Treas. Reg. § 1.1297-2(c)(2).

69Prop. Treas. Reg. § 1.1297-1(f)(1). The same approach to attributing ownership “directly or indirectly” in the case of corporations appears to apply in the case of partnerships. For example, if TFC 1, owns 25% by value of FPRS 1 which owns 50% of FPRS 2, FPRS 1 is a Look-through Partnership as to TFC 1 but because TFC 1 indirectly owns less than 25% of FPRS 2, FPRS 2 is not a Look-through Partnership as to TFC 1.

70Prop. Treas. Regs. §§ 1.1297-1(c)(2)(ii) and 1.1297-1(d)(3).

71Cf. Section 751 (supporting a look-through approach in the case of a disposition of a partnership interest under Section 741 regardless of the partner's ownership interest); Section 864(c)(8), Treas. Reg. § 1.864-4(c), and Rev. Rul. 91-32, 1991-1 C.B. 107 (treating gain on sale by a foreign partner of a partnership interest as effectively connected income to the extent that a sale of partnership assets would have given rise to effectively connected income regardless of the partner's ownership interest percentage).

72See Prop. Treas. Reg. § 1.1297-2(d)(1) (relatedness tested at the Look-through Subsidiary level).

73See Prop. Treas. Reg. § 1.1297-1(c)(2)(ii) and (d)(3)(ii).

74Another potentially counter-intuitive result could arise because of Section 704(c). If a business asset is contributed by a Tested Foreign Corporation to a less than 25% owned partnership (for example, in the foreign equivalent of an UPREIT structure) the income subsequently specially allocated under Section 704(c) would be passive even though the Section 704(c) taxable income allocation reflects gain economically accrued with respect to the property while it was directly in the hands of the contributing partner.

75The Proposed Regulations, correctly in our view as an interpretive matter, would apply this exception for PFIC classification purposes. See Prop. Treas. Reg. § 1.1297-1(c)(1)(i)(C).

76See Prop. Treas. Regs. § 1.1297-1(d)(3) and (f)(1).

77This inconsistency will also arise in situations other than actual sales. See, e.g., Section 731.

78See Treas. Reg. § 301.7701-3(b)(1)(i).

79Cf. Prop. Treas. Reg. § 301.7701-3(h), 64 FR 66591 (Nov. 29, 1999) (addressing “extraordinary” transaction including “check and sell” transactions); Notice 2003-46, 2003-2 C.B. 53 (Jul. 14, 2003) (withdrawing proposed regulations); Dover Corp. & Subsidiaries v. Commissioner, 122 T.C. 324 (2004).

80For example, in the situation mentioned above — where a supplier acquires a less than 25% equity stake in an important customer (or vice versa) to cement important active business ties rather than for passive investment — electing to treat the customer equity stake as a partnership interest may avoid the (arguably inappropriate) “passive” treatment of the shares.

81Whether the particular example is such an inappropriate result depends on one's view of the purposes of the PFIC regime. Under a narrow view of that purpose — that Congress intended PFIC to apply to entities similar to offshore mutual funds — the example might not be objectionable. Because of the publicly traded partnership rules under Section 7704, such “CTB” self-help planning is possible only for relatively illiquid portfolio positions unlike those in which passive vehicles like mutual funds typically invest. Under a broader view of Congress's purpose, allowing CTB planning in this situation inappropriately could allow a somewhat “passive” vehicle like a foreign venture capital fund to avoid PFIC status.

82There is some indication in the legislative history that Congress was focused on investment funds. Cf. H.R. Rep. 99-841, 99th Cong., 2nd Sess. (1986) (“The conferees believe that eliminating the economic benefit of deferral is necessary to eliminate the tax advantages that U.S. shareholders in foreign investment funds have heretofore had over U.S. persons investing in domestic investment funds.”)

83See, e.g., Section 856(c) (imposing various income, asset and diversification tests for REIT qualification purposes, which Congress presumably considered indicative of “passivity”); Section 851(b) (similar requirements to be treated as a regulated investment company).

84The fact that this may reach different results depending on whether the entity can make a CTB election, is an issue discussed above. For the reasons discussed, we do not view “equivalency” as a compelling policy objective in this context.

85Prop. Treas. Regs. § 1.1297-1(e).

86See Preamble, 84 Fed. Reg. 133 at 33127-8 (“Given that stapled interests represent a single economic interest to their shareholders, the Treasury Department and the IRS have determined that it is appropriate, for purposes of determining whether a stapled entity is a PFIC, to treat them as such”).

87Treas. Reg. § 1.269B-1.

88See Treas. Reg. § 1.269B-1(b)(3). More extreme cases — such as stapling a class of shares issued by one corporation representing 99% of its aggregate equity value to a class issued by another corporation representing 1% of its equity value, and vice versa — are possible that would result in even greater divergence between stapling and single entity economics.

89The 1986 Blue Book provides that regulations under Section 1298(b)(4) may provide that a separate class of stock is treated as creating a separate corporation, which, because all its investments were passive, would be a PFIC (and shares of a lower-tier PFIC could be attributed to the holders of the separate class of stock).

90We recognize that because Section 1298(b)(4) is an anti-abuse provision it is arguably more directly applicable when the use of the stapled arrangements allows a shareholder to avoid PFIC status inappropriately.

91Prop. Treas. Reg. § 1.1297-2(h).

92In addition, taxpayers may continue to rely on Notice 88-22 until the Proposed Regulations are finalized.

END FOOTNOTES

DOCUMENT ATTRIBUTES
Copy RID