NYSBA Members Suggest Changes to Proposed PTI Regs

I am pleased to submit the attached report (the "Report") of the Tax Section of the New York State Bar Association. The Report comments on certain proposed regulations under sections 959 and 961 of the Internal Revenue Code of 1986 that were issued in 2006 (the "Proposed Regulations").

Pursuant to the authority granted by section 959, the Proposed Regulations provide rules for tracing distributions from controlled foreign corporations ("CFCs") to previously taxed earnings and profits ("PTI") when a CFC makes a distribution that would otherwise be treated as a dividend by employing a system of earnings & profits ("E&P") accounts. The Proposed Regulations also provide rules for making the basis adjustments required under section 961 to prevent double taxation of PTI in transactions involving dispositions of CFC shares. The guiding principle adopted by the Proposed Regulations is that the purpose of section 959 is to prevent double taxation of amounts that have been previously included in gross income by a United States shareholder (a "US Shareholder") under section 951(a) and, importantly, to prevent such double taxation at the earliest possible time. A corollary of that guiding principle is that the rules generally treat distributions as coming "first" from the PTI account rather than other earnings.

We support the framework adopted by the Proposed Regulations and our comments are primarily directed towards extending the "PTI first" approach adopted for distributive transactions to dispositions of CFCs.

Our primary recommendations are as follows:

This report ("Report")¹ of the New York State Bar Association Tax Section (the "Tax Section") comments on proposed regulations regarding the exclusion from gross income under Section 959 of previously taxed earnings and profits and related adjustments under Section 961 to the basis of stock (and certain other interests) in controlled foreign corporations ("CFCs") issued by the Internal Revenue Service (the "IRS") and Treasury Department ("Treasury") on August 29, 2006 (the "Proposed Regulations").²

Subpart F of the Code establishes an anti-deferral regime that taxes United States shareholders of a CFC on certain earnings of the CFC currently, whether or not the CFC makes any distributions.³ Current income inclusions arise if the CFC earns "subpart F income" ("subpart F inclusions") and if the CFC invests its earnings in "United States property" ("Section 956 inclusions").⁴ The purpose of subpart F is to prevent U.S. persons from deferring U.S. tax on certain income earned through foreign corporations, but it is not intended to impose duplicative U.S. tax on the earnings. Because U.S. shareholders are taxed on their pro rata share of certain earnings of a CFC in advance of distributions of those earnings, a mechanism is required to ensure that those earnings are neither taxed again when they are distributed to the U.S. Shareholder or to another CFC in the ownership chain nor excluded from income more than once (for example, in connection with subsequent investments in "United States property").

Sections 959 and 961 establish the mechanism by which double taxation of the earnings of a CFC is to be avoided. Section 959(a) provides that earnings and profits ("E&P") of a CFC that have already been taxed to a U.S. Shareholder under Section 951(a) (such previously taxed income, "PTI") will not be taxed again when distributed to that shareholder (or, in certain circumstances, to a successor in interest) or would otherwise result in Section 956 inclusions to that shareholder or a successor. Section 959(b) excludes such PTI from the gross income of an upper tier CFC to which the PTI is distributed for purposes of determining the subpart F income of the U.S. Shareholder who was previously taxed on the earnings (or its successor in interest).

Section 961 provides for adjustments to the basis of stock in the CFC and certain other property. It is intended to ensure that the U.S. Shareholder who was previously taxed on the earnings of a CFC does not suffer double taxation (in the form of gain attributable to PTI) if it sells its interest in the CFC (or in a foreign entity through which the CFC is owned) prior to receiving a distribution of the PTI. Section 961(a) generally requires an increase in a U.S. Shareholder's basis in the shares of the CFC (or a foreign entity through which the shares of the CFC are owned) to reflect subpart F income or Section 956 inclusions previously taxed to the U.S. Shareholder. Section 961(b) reduces the basis of the CFC stock (or of the foreign entity through which the CFC is owned) by the amount of a distribution excluded from income as PTI under Section 959(a).

Section 961(c) was added to the Code in 1997 to prevent duplicative tax that could otherwise arise if an upper tier CFC sells a lower tier CFC for an amount that reflects PTI of the lower tier CFC. Absent Section 961(c), gain recognized by the upper tier CFC that is attributable to the lower tier CFC's PTI would be subpart F income and could be taxed again to U.S. Shareholders. To prevent this result, section 961(c) provides for increases and decreases in the basis of lower tier CFCs that parallel the adjustments pursuant to Section 961(a) and (b). However, these basis adjustments apply only for purposes of determining the amount included under subpart F in the income of a U.S. Shareholder or a successor in interest. The Report will refer to this limitation on the effect to be given to basis increased by Section 961(c) as the "Solely for Subpart F Limitation."

Pursuant to the authority granted by Section 959, the Proposed Regulations provide rules for tracing CFC distributions to PTI when the CFC makes a distribution that would otherwise be treated as a dividend by employing a system of E&P accounts. The Proposed Regulations also provide rules for making the basis adjustments required under Section 961 to prevent double taxation of PTI in transactions involving dispositions of CFC shares. The guiding principle adopted by the Proposed Regulations is that the purpose of Section 959 is to prevent double taxation of amounts that have been previously included in gross income by a U.S. Shareholder under Section 951(a) and, importantly, to prevent such double taxation at the earliest possible time. A corollary of that guiding principle is that the rules generally treat distributions as coming "first" from the PTI account rather than other earnings.

The rules of subpart F, including the rules relating to PTI and related basis adjustments under Section 961 (collectively, the "PTI regime"), have evolved over many decades. Unresolved issues and problems have often been addressed piecemeal with the result that the PTI regime is not in all respects entirely consistent with the policies underlying subpart F. The Proposed Regulations represent a major step towards clarifying remaining gaps in the PTI regime. We therefore support the general approach of the Proposed Regulations. However, in certain respects the Proposed Regulations do not extend the "PTI first" approach they adopt for distributive transactions to dispositions of CFCs. This inconsistency probably reflects a basic tension in the fundamental design of subpart F and the PTI regime between the treatment of CFCs as a form of pass-through entity with respect to PTI and the "classical" subchapter C regime for corporations, including foreign corporations like CFCs, which generally treats each corporation as a distinct taxable entity.

The subpart F regime is essentially a "hybrid" regime, which retains many features of a classical subchapter C regime while at the same time functioning as a form of pass-through regime for previously taxed U.S. Shareholders of a CFC with respect to PTI (but only with respect to PTI and generally only with respect to those previously taxed shareholders and their successors in interest). Specifically, U.S. Shareholders are taxed currently on undistributed subpart F income (contrary to "classical" subchapter C principles) but are permitted to recover the resulting PTI without being subjected to further, duplicative U.S. federal income tax. Yet, at the same time, the PTI regime generally preserves the subchapter C approach to E&P that are not PTI ("non-PTI") with respect to those U.S. Shareholders and as to all E&P with respect to shareholders not subject to subpart F. That subchapter C approach creates the potential for "duplication" of both E&P and basis in various circumstances in ways that conflict with the "pass-through" nature of the PTI regime.

One such potential conflict, which the Proposed Regulations address, is the approach to allocating basis and, in the case of the PTI regime, allocating PTI attributes including basis arising under Section 961, among different shares, blocks and classes of shares.⁵ The regular subchapter C rules generally follow a "tracing" approach, which treats shareholders as having basis associated with particular shares. This may mean that a shareholder has different basis in different shares, blocks of shares or classes of shares rather than "averaging" its basis among all shares the shareholder owns. Adopting a consistent approach under the PTI regime would mean that, potentially, a distribution out of E&P with respect to a particular block of shares (for example, in a redemption treated as a distribution) or with respect to a particular class of shares is taxable, even though the amount of the distribution does not exceed the aggregate PTI attributes of the U.S. Shareholder who receives it. The Proposed Regulations resolve this tension between the general pass-through approach to PTI and the subchapter C "tracing" approach to basis by allowing the U.S. Shareholder, effectively, to reallocate PTI attributes from shares with unused PTI attributes to shares with deficient PTI attributes. This somewhat resembles basis "averaging" except that it does not require reallocation of PTI attributes from shares with disproportionate PTI attributes to other shares in the converse situation when a distribution is received with respect to those shares. However, this is consistent with the principle adopted by the Proposed Regulations that PTI attributes generally should be recovered as early as possible.

As discussed in the Report, we generally support the "hybrid" pass-through/subchapter C approach followed by the Proposed Regulations, including the "sharing" rules for PTI attributes because we believe this achieves the "pass-through" treatment of PTI that Congress intended while preserving subchapter C treatment in other respects.

Yet the Proposed Regulations do not similarly modify the otherwise applicable subchapter C basis rules as applied to Section 961 Basis in the case of dispositions and transfers that represent an indirect recovery of PTI. Foreshadowing the more detailed discussion in the Report, the basic problem can be illustrated by a schematic example.

 

 

L-CFC has PTI of $80 and USS has $80 of basis increase under Section 961(a) in its U-CFC shares as a result of the subpart F inclusion. The more difficult question is what the correct basis adjustment should be in U-CFC's stock in L-CFC. If the basis increase is limited to the $80 of subpart F income actually taxed to USS, upon a sale of L-CFC stock, there would still be $20 of gain attributable to the $100 of E&P that arose from the passive income. A pro rata share of this gain ($16) would be taxable as subpart F income to USS creating PTI of $16 and the remaining $4 would be regular E&P. This seems inconsistent with the premise of the PTI regime that USS should not be duplicatively taxed on PTI (although duplicative taxation of earnings when there are multiple tiers of corporations is an inherent feature of a "classical" corporate tax system). Conversely, if U-CFC's basis is increased by $100, there would be no gain attributable to the $100 of passive income (and thus no E&P attributable to that passive income) at the U-CFC level from selling L-CFC. This would prevent USS including the PTI again as subpart F income, and also ensure that a subsequent distribution of $80 of proceeds indirectly attributable to the $100 of passive income would not be a taxable dividend to USS. However, it would also mean that a distribution of $20 of the proceeds to persons other than U.S. Shareholders (who in other cases may include less-than-10% shareholders that are U.S. taxpayers) would also not be taxed as a dividend. This is not only inconsistent with general subchapter C principles but seems clearly inappropriate as a PTI matter as those shareholders would not have been previously taxed on the L-CFC earnings.

The Proposed Regulations adopt an approach which (although not entirely clear) may be read to interpret the "Solely for Subpart F" limitation on Section 961(c) Basis narrowly to mean that the basis increase is $100 for the sole purpose of determining whether USS recognizes subpart F income from the disposition. For all other purposes, U-CFC's basis in L-CFC is not adjusted. Thus, although U-CFC's sale does not result in current duplicative taxation to USS, it does result in $100 of non-PTI at U-CFC attributable to the $100 of previously taxed passive income. This ensures that minority shareholders are appropriately taxed on a subsequent distribution of this E&P by U-CFC. But it also means that USS would be taxable on an $80 distribution as a dividend. One could argue this is simply the result of the potential for duplication of earnings and basis inherent in a classical corporate tax system. On the other hand, the result appears inconsistent with the "pass-through" nature of the subpart F regime (and, moreover, inconsistent with the result that would have obtained if U-CFC and L-CFC were a single corporation, for example, if L-CFC had elected to be a disregarded entity rather than a corporate subsidiary of U-CFC).

Thus, as illustrated in greater detail in the Report, under the Proposed Regulations, duplication of E&P (and thus of tax) arguably may arise upon a sale of a lower tier CFC. Similarly, regular SubC Basis rules may result in duplication of 961(a) Basis in certain contribution and reorganization transactions that could benefit a person other than the previously-taxed U.S. Shareholder or its successor, when the contributed shares (in what is now a lower tier CFC) are disposed of, by reducing the E&P of the selling CFC. As discussed below, the Report questions whether these results are consistent with the policies of subpart F. In particular, it is unclear that there is a policy justification for treating dispositive and distributive transactions differently (in terms of E&P and basis duplication), especially when a U.S. taxpayer effectively can elect the result it prefers by engaging in self-help distribution transactions prior to a disposition. For example, if a U.S. taxpayer wishes to avoid duplication of earnings in connection with the disposition of a lower tier CFC, it can cause a distribution of the PTI up the chain (to reduce gain on sale) prior to the disposition of the lower tier CFC. We believe the primary effect of the rules without further modification, would be to encourage costly and therefore wasteful "self-help" planning rather than to raise revenue. Forcing the taxpayer to engage in a self-help distribution may, in addition to transaction costs, result in foreign withholding taxes on the distribution, which may even be disadvantageous to the fisc to the extent the taxes in question are creditable.

The primary focus of the Report is on the conceptual relationship between the general rules for earnings and basis under subchapter C and the PTI regime. Part II summarizes the Report's recommendations. Part III lays out the statutory structure and legislative history of the PTI regime. Part IV summarizes in greater details the approach adopted by the Proposed Regulations. Part V illustrates and analyzes the essential features of the Proposed Regulations and their approach to PTI through various examples. Part VI analyzes some alternative general approaches to basis adjustments under Section 961 that could be adopted in the face of the discontinuities illustrated by Part V and Part VII contains the Report's recommendations.

II. Summary of Recommendations

Conversely, if the PTI regime adopts without modification the regular rules that apply in determining "carryover" SubC Basis, the rules may result in duplication of 961(a) Basis in certain contribution and reorganization transactions that could benefit a person other than the previously-taxed U.S. Shareholder or its successor, when the contributed shares (in what is now a lower tier CFC) are disposed of, by inappropriately reducing the E&P of the selling CFC.

The statute and Proposed Regulations suggest that the character of basis as 961(a) or 961(c) Basis depends on the circumstances at the time the basis adjustment occurred. However, given the purpose of Sections 959 and 961 it is questionable whether this makes policy sense from a subpart F perspective. Purely from that perspective, the better approach would be that 961(a) Basis should upon contribution not carry over under regular subchapter C principles but instead "morph" into 961(c) Basis. This would ensure both that the U.S. Shareholder is not taxed duplicatively on the earnings generated by the subpart F income but also prevent other shareholders benefitting from a reduction in E&P attributable to earnings on which only the U.S. Shareholder has actually paid tax. Conversely, the better policy from a subpart F perspective would seem to be that 961(c) Basis should "morph" into 961(a) Basis in an inbound liquidation scenario.

In light of these considerations, we believe that, although subchapter C serves as the "default" regime applicable to foreign corporations and their shareholders, as a general matter the rules should adopt a regime that achieves an effect substantially comparable to a pass-through regime from the perspective of the previously-taxed U.S. shareholders with respect to PTI, regardless of the number of tiers of CFCs through which this income is earned, but only with respect to PTI. Conversely, the consequences of classical subchapter C taxation (including the potential for duplication of earnings and basis) generally should apply to CFC investors that are not U.S. Shareholders of the CFC and to U.S. Shareholders with respect to non-PTI.⁷

To achieve this, we recommend that a specific rule be adopted that, upon a disposition of a lower tier CFC, would treat PTI of the lower tier CFC in a manner analogous to the treatment of non-PTI under Section 964(e), which effectively deems the E&P to "tier up" in connection with the disposition to the extent of any gain recognized. Under this approach, 961(c) Basis would only be taken into account in measuring subpart F income of the U.S. Shareholder (not in measuring the gain for E&P purposes), but an appropriate portion of the PTI at the lower tier CFC (equal to the additional E&P arising from ignoring 961(c) Basis for other purposes) would "tier up" to the upper tier CFC. This would ensure appropriate taxation of shareholders not subject to the subpart F regime while preserving pass-through treatment of PTI to the U.S. Shareholders subject to the subpart F regime.

We further recommend that in tax-free contribution and distribution transactions otherwise requiring "carry over" 961(a) Basis or 961(c) Basis, between upper tier and lower tier CFCs, that 961(a) Basis be converted into 961(c) Basis in contribution situations and vice versa when lower tier CFC shares become upper tier CFC shares. For example, if in the example above in the Introduction, USS and F1 were to contribute their shares in U-CFC to a new holding company CFC in exchange for new CFC stock, the $80 of 961(a) Basis in the hands of USS should become $100 of 961(c) Basis (with respect to USS) in the hands of the new holding company CFC, subject to the "Solely for Subpart Limitation" (but applied as described immediately above).

The Report also makes a number of subsidiary recommendations.

It is unclear whether the Proposed Regulations intend to permit a U.S. Shareholder to share PTI attributes in shares it is deemed to own in a CFC under Section 958(a), but actually owns indirectly, with other directly owned shares in the CFC and vice versa. We believe the IRS and Treasury should clarify whether sharing will be permitted in this situation. This kind of sharing would be consistent with the manner in which subpart F ignores the "classical" subchapter C approach to corporations as separate taxpayers by taxing a U.S. Shareholder directly on subpart F income earned by a lower tier CFC it owns indirectly and would treat PTI as an attribute of the taxpayer who bears the tax on the subpart F income that gave rise to the PTI attributes. However, there are good arguments for and against this kind of sharing, and this Report does not make a recommendation on this point.

We believe further that strong arguments can be made in favor of, and we generally support permitting, sharing of PTI attributes within a consolidated group. However, to the extent this result is appropriate, it may not be based on subpart F policy considerations, but rather because it is consistent with the "single entity" treatment of consolidated groups under the policies and approach adopted by the consolidated return regulations. Consideration should be given to including the sharing rules for PTI attributes of consolidated return members in a separate regulation under Section 1502 rather than including those rules in PTI regulations.

III. Background Statutory Provisions and Legislative History

Section 959 establishes certain tracing rules. Under Section 959(c), actual distributions by a CFC are allocated first to earnings that have resulted in Section 956 inclusions, then to earnings that resulted in subpart F inclusions and finally to non-PTI. Potential Section 956 inclusions are allocated first to E&P previously taxed as subpart F income and then to non-PTI pursuant to Section 959(f). Section 959(f) goes on to provide that adjustments to a CFC's E&P are made first to account for actual distributions and then for Section 956 inclusions.

As noted above, Section 959 provides that a "successor in interest" to a U.S. Shareholder can benefit from the exclusion of PTI. Any person that acquires all or a portion of the U.S. Shareholder's interest in the CFC succeeds to the shareholder PTI accounts of the transferor. If such successor (including a person who succeeds to the PTI accounts of a previous non-U.S. successor) is a U.S. person, it is treated as a "successor in interest" that may take advantage of the related PTI attributes, provided it can provide documentary proof with respect to the acquired interest, as required by Treasury regulations.⁸ There is no requirement that the acquisition involve a taxable disposition. Accordingly, it appears a person that acquires shares in a Section 351 transaction or other corporate reorganization may be a successor to PTI attributes of the acquired shares. Under the Proposed Regulations, a successor in interest is a U.S. person who acquires ownership of stock within the meaning of Section 958(a), which includes stock owned directly or indirectly in a CFC through another foreign entity (although not stock owned constructively under Section 958(b)), so it appears that a person who acquires a lower tier CFC (i.e., an indirect interest of a U.S. Shareholder) may be a successor, although this could be clearer.⁹

Section 961 provides for adjustments to the basis of stock in a CFC and certain other property. It is intended to ensure that the U.S. Shareholder does not suffer double taxation of the retained PTI of the CFC if the U.S. Shareholder sells its interest in the CFC (or in a foreign entity through which the CFC is owned) prior to receiving a distribution of that PTI. Section 961(a) calls for regulations pursuant to which a U.S. Shareholder's basis in the shares of the CFC (or a foreign entity through which the shares of the CFC are owned) is increased to reflect subpart F inclusions or Section 956 inclusions previously taxed to the U.S. Shareholder. Section 961(b) calls for regulations pursuant to which the basis of the CFC stock (or of the foreign entity through which the CFC is owned) is reduced by the amount of a distribution excluded from income as PTI under Section 959(a). The basis reduction applies to the original U.S. Shareholder and any other U.S. person holding the relevant CFC stock or interests in a foreign entity.

Section 961(c) in turn calls for regulations that provide for increases and decreases in the basis of lower tier CFCs that parallel the adjustments to be provided for pursuant to Section 961(a) and (b) at the level of the U.S. Shareholder's direct holding. As discussed above, that basis adjustment applies only for purposes of determining the amount included under subpart F in the income of the U.S. Shareholder or a successor in interest (the "Solely for Subpart F Limitation").

The original statutory provisions did not, however, include any provision to adjust basis in lower tier CFCs equivalent to current Section 961(c), which was only enacted in 1997. The 1997 Conference Report noted that the purpose of the provision was to parallel the basis adjustments provided for under Section 961 with respect to the shares of upper tier CFCs.¹³ The new provision was not described as providing a basis increase, however, but instead as granting regulatory authority to reduce a U.S. Shareholder's subpart F income in the year the shares of a lower tier CFC are sold by the upper tier CFC. The House Report notes that this "in effect, allows for a step-up in the basis of the stock of the second-tier CFC."¹⁴

Only E&P of a foreign corporation attributable to the shares of stock sold or exchanged result in dividend treatment under Section 1248. The amount attributable to a share is generally the pro rata portion of E&P of the foreign corporation accumulated during the period the share was held by the shareholder while the foreign corporation was a CFC. E&P attributable to shares that are retained by the shareholder are not taken into account. Regulations under Section 1248 prescribe methods for computing a CFC's E&P attributable to shares that are sold. Calculations can be made for a "block" of shares sold or exchanged in a single transaction, rather than on an individual share basis.¹⁷ To qualify as a "block," each share in the block generally must be sold in the same transaction, have the same amount realized, basis and holding period and (where there was an income inclusion under Section 951 during the holding period) have identical amounts of remaining PTI.¹⁸ In complex cases when there are multiple classes of shares, Section 1248 applies the principles of the rules that under Section 951 determine a U.S. Shareholder's pro rata share of subpart F income.¹⁹ Those rules determine pro rata shares by looking to which shares would have received a hypothetical distribution of the E&P. In the case of subpart F income, this pro ration is made when the subpart F income is includible (which is also when the E&P is earned by the CFC). By contrast, Section 1248 attributes E&P to shares based on the ownership position of the U.S. Shareholder during the relevant taxable year in which the E&P arose (adjusted for deficits from operations and distributions) and not by looking to how remaining E&P would be distributed in a hypothetical distribution in the taxable year when the disposition that results in the Section 1248(a) dividend occurs.

Section 964(e) recharacterizes gain realized by one CFC on the sale of shares of a lower tier CFC as a dividend "to the same extent that it would have been so included under Section 1248(a) if such CFC were a United States person." A deemed dividend under Section 964(e) cannot be excluded from subpart F income under the same country exception and thus generally will give rise to a subpart F inclusion for a U.S. Shareholder of the selling CFC.²⁰ The provision was enacted at the same time as Section 961(c) in 1997. The legislative history to Section 964(e) indicates that the provision was intended to give U.S. Shareholders access to foreign tax credits in connection with income inclusions resulting from gain on the sale of a lower tier CFC's shares by an upper tier CFC.²¹ The legislative history does not address the interplay between Section 964(e) and the PTI regime.

IV. Overview of the Proposed Regulations

Existing regulations under Sections 959 and 961 were published in 1965 and have not been updated to reflect changes in law since the early 1980s. Treasury and the IRS issued the Proposed Regulations in 2006. The preamble to the Proposed Regulations states as a guiding principle that Section 959 is intended to "[effect] the relevant gross income exclusion at the earliest possible point" and that Section 961 is intended to "[ensure] that [PTI] is not taxed twice if the stock in the foreign corporation is sold before [PTI] is distributed."²²

A distribution out of a PTI Account is not taxable to the U.S. Shareholder or its successor in interest and is not treated as a dividend.²⁵ It does, however, reduce the CFC's E&P in the relevant E&P PTI Subaccounts. Distributions in excess of a shareholder's PTI Accounts in stock it owns (within the meaning of section 958(a)) can be treated as dividends only to the extent of corporation's C3 E&P Subaccount.²⁶

If PTI is distributed through a chain of CFCs, the entire amount of E&P of a lower tier CFC with respect to which a U.S. Shareholder had a subpart F inclusion is treated as PTI at the level of an upper tier CFC in order to ensure that the U.S. Shareholder is not treated as having a second subpart F inclusion with respect to the same E&P.²⁷ Put another way, the PTI "tiers up" up along with the E&P. The following example in the Proposed Regulations illustrates this rule:

As noted above, accounts are maintained at the shareholder level and parallel accounts are maintained at the level of the foreign corporation (in its functional currency). Shareholder level accounts are share specific, but the Proposed Regulations permit a U.S. Shareholder to maintain PTI Accounts with respect to "blocks of shares".³¹ Blocks of shares for this purpose are shares treated as forming part of a "block" for purposes of the Section 1248 regulations, provided the PTI attributable to each share in the block is the same.

In addition, the Proposed Regulations provide for "sharing" of PTI Accounts among blocks of shares owned (within the meaning of Section 958(a)) by the same shareholder.³² The following example in the Proposed Regulations illustrates this inter-block sharing:

Generally, the rules above permit PTI attributes to be shared only among shares in the same CFC owned (within the meaning of Section 958(a)) by the same U.S. Shareholder. However, a comparable PTI sharing approach applies to shares held by different members of the same consolidated group of corporations.³⁵ The following example from the regulations illustrates PTI sharing within a consolidated group:

By allowing "sharing" of PTI attributes in consolidated groups, the Proposed Regulations extend the "single entity" approach that is generally reflected in the current consolidated return regulations to PTI attributes. Corresponding basis adjustments must be made under Treasury regulations Section 1.1502-32.³⁸

PTI is reduced by foreign income taxes imposed on distributions of PTI by a CFC through a chain of foreign corporations.⁴⁷ Expenses other than such foreign taxes are not allocated or apportioned to PTI.⁴⁸ Similarly, deficits in E&P are not taken into account to adjust the PTI Accounts (and related E&P PTI Subaccounts). Instead, deficits are applied only to reduce non-PTI.⁴⁹

Similarly, basis in directly held CFC shares or other property held by a U.S. Shareholder or other covered shareholder is reduced at the time, and to the extent, the PTI Accounts are reduced pursuant to the account adjustment rules described above.⁵³ Basis is also reduced by the dollar amount of any foreign income taxes allowed as a credit under Section 960(a)(3) with respect to the related E&P. As with basis increases, basis decreases are to be made on a pro rata basis with respect to each share of stock or ownership unit of other property.

Adjustments are only made to the basis of property held by an upper tier CFC. This will not necessarily include basis of all stock or other interests in the corporate ownership chain through which a U.S. Shareholder may receive distributions. For example, if part of the Section 958(a) ownership interest is held through a non-CFC, that non-CFC does not adjust its basis in lower tier CFC interests or other property through which the lower tier CFC is owned. Presumably, because Section 961(c) Basis applies only to measure a U.S. Shareholder's subpart F inclusions, it was thought unnecessary for a non-CFC to step up basis since a sale of shares by a foreign corporation while it is not a CFC would not result in subpart F inclusions. However, adjusting or not adjusting basis does affect E&P and therefore may have collateral consequences.

If a U.S. Shareholder owns less than 100% of the lower tier CFC, the increase to the upper tier CFC's basis in the shares of the lower tier CFC (or in the units of property through which it owns the lower tier CFC) equals the amount that would be excluded from the upper tier CFC's gross income if the amount giving rise to the adjustment to the U.S. Shareholder's PTI Accounts with respect to the lower tier CFC were actually distributed through the chain of ownership to the upper tier CFC.⁵⁵ This amount may exceed the amount actually included by the U.S. Shareholder. However, this is a corollary of the rule that a distribution is excluded from treatment as a dividend for subpart F purposes up to the entire amount of the subpart F income that gave rise to an inclusion. In connection with a reduction in a U.S. Shareholder's PTI Accounts with respect to a share of a lower tier CFC that makes a distribution, downward adjustments are made to the basis of the distributee in the distributing CFC's stock equal to the amount of PTI excluded from the distributee's gross income. The following example from the regulations illustrates these rules:

V. Features of the PTI Regime and Basis

The following section of the Report compares some of the key features of the PTI regime to a "classical" subchapter C regime. The schematic examples used to illustrate the differences are intended to exaggerate the essential consequences of the rules for 961 Basis. For simplicity, the examples generally assume (1) there is no basis or E&P other than the specific items described, (2) no foreign taxes are incurred on the earnings, (3) there are no foreign currency translation issues, (4) there are no dividends-received deductions and (5) unless otherwise specified, there is no consolidation. The application of other Code provisions in the examples is intended merely to be illustrative of the Section 961 rules and the stipulated consequences under those other provisions may not be, but should be assumed to be, correct.

As a general rule, in "distributive" transactions (i.e., transactions in which assets are returned by the corporation to its shareholder, including redemptions treated as distributions under Section 302(d)), E&P is recovered before basis (the "E&P first" principle). Under subchapter C, this means that in distributive transactions basis in corporate shares is recovered only after E&P (i.e., "basis last" as a corollary of the E&P first principle). However, in dispositions of corporate shares treated as a sale or exchange (i.e., "dispositive transactions"), basis in corporate shares conceptually is recovered pro rata to reflect the fact that the proportionate ownership interest of the selling shareholder in the underlying corporate E&P pool and assets has been reduced. However, for policy reasons specific to subchapter C, or as an artifact of history, basis is traced to particular shares and is not generally averaged when a shareholder owns different classes of shares, or even blocks of the same class of shares acquired at different times. Thus, changes in a shareholder's share basis are imperfectly correlated with indirect changes in its proportionate ownership interest in the underlying corporate assets and E&P pool of a C corporation.

Example 1 -- Subchapter C Distribution and Recontribution

 

 

Assuming that the companies are treated as domestic C corporations (and ignoring corporate level tax), Company 2 initially has $100 of E&P, reduced to $20 by the $80 distribution treated as a dividend, and then increased to $40 when Company 2 earns $20 of non-passive income. Company 1 has E&P of $80 as a result of the dividend from Company 2, reduced to zero by the Section 302(d) distribution to USS. USS increases its basis in Company 1 as a result of the $80 recontribution to $160 and Company 1 increases its basis in Company 2 as a result of the recontribution to $180. Company 2 is taxable on the $100 of passive income, Company 1 is taxable on the dividend of $80 of that income and then USS is taxed on the dividend of that income.

If, instead, prior to any distribution of earnings by Company 2, Company 1 sells Company 2, gain attributable to the E&P at the Company 2 level is potentially taxable to Company 1, generating additional E&P at that level which, when distributed to the ultimate shareholder as a dividend, is taxable again to the shareholder. If instead the shareholders sell Company 1, they pay only one level of tax, but there still is built in gain at the Company 1 level, and E&P at the Company 2 level to be taxed again at some point when each is recognized or distributed (and the taxes on which will be borne economically by the successor shareholder).

On the other hand, if Company 1 sells Company 2 after the $80 distribution and recontribution, Company 1's increased basis (resulting from USS's contribution of after-tax funds) reduces the gain recognized on the portion of the amount realized that is attributable to the assets generated by the recognition of the passive income. This reduces the E&P that would otherwise have been generated at the Company 1 level. However, there is still the potential for gain recognition that is attributable (1) to the $20 of untaxed, undistributed Company 2 passive income (representing F's indirect interest) and (2) the $20 of untaxed, undistributed business income, and this will be taxable to Company 1. This creates potential E&P at the Company 1 level of $40. If subsequently the sale proceeds are distributed, USS is taxed not only on its 80% of the $20 of E&P attributable to the active business income but also on 80% of the $20 of E&P related to the passive income. USS therefore may be taxed duplicatively on a portion of the passive income. At the same time, F benefits economically from the reduction in gain resulting from USS's prior recontribution of $80 of after-tax funds, which restored the initial 80/20 ownership split between USS and F. Subchapter C has no mechanism analogous to the Section 704(c) rules under subchapter K to specially allocate built-in gain attributable to the untaxed passive income to F and restrict to USS the benefit of reduced E&P resulting from basis that arises from contributed funds on which USS (rather than F) incurred U.S. income tax.

Thus, in its pure form the "classical system" under subchapter C, in which each corporation is a separate taxpayer, results in "duplicated" earnings and basis.

Pass-through regimes like subchapter K generally adopt the opposite approach to recovery of earnings and basis. Assume that in the example above, Company 1 and Company 2 are pass-through entities. The $100 of income is taxed to USS and F whether or not distributed and regardless of its character and the outside basis of both partners is proportionately increased. The $80 distribution and recontribution merely reduces then increases USS's basis in Company 1. Because all income is taxed to all of the owners, there is no need to trace distributions to previously-taxed versus untaxed income (a situation which arises under subpart F because only some kinds of undistributed income are currently taxed and only to certain shareholders). Effectively, a pass-through regime applies a "basis first" rule. If F subsequently contributed property with a built-in gain, F receives an outside basis equal to the post-tax funds it has already invested (i.e., its basis in the property becomes its basis in the partnership interest and the partnership takes a carryover basis). However, in contrast to subchapter C, Section 704(c) generally should ensure that the gain that accrued in F's hands (or any untaxed gain that accrued economically to other partners prior to F's admission) is specially allocated to the appropriate partner. Thus, despite the carryover of basis to the partnership, there is no duplication of basis in a subchapter C sense. In a pass-through regime, the result to the owners does not differ fundamentally regardless of how many tiers of pass-through entities there are.

Assume in the example above that Company 1 and Company 2 are foreign corporations and therefore CFCs. USS would be taxed on $80 of the Company 2 passive income under subpart F. While Company 2 has $100 of E&P, $80 of the E&P is treated as PTI. The $80 distribution to Company 1, would be excluded from income as a distribution of PTI, creates $80 of E&P at the Company 1 level all of which would be allocable to Company 1's C2 E&P Subaccount (and the distribution would correspondingly reduce Company 2's C2 E&P Subaccount). Company 1 can then distribute $80 to USS without USS incurring additional U.S. federal income tax.

Thus, subpart F departs from classical subchapter C principles by currently taxing the U.S. Shareholder on certain corporate earnings. Similarly, the PTI regime then permits distributions of the PTI through a chain of CFCs in a manner that avoids duplication of the earnings by permitting the distributee CFC at each level to treat the E&P as PTI, albeit only with respect to the previously taxed U.S. Shareholder. The distribution of earnings up a chain of CFCs nevertheless increases the E&P of each distributee CFC (correspondingly reducing the E&P of each distributor CFC), which ensures that distributions by the upper tier CFC to shareholders who did not incur the subpart F income taxes are taxable to those shareholders. Put another way, PTI "tiers up" the chain of corporate ownership along with the associated E&P. Thus, the Section 959 treatment of PTI ensures that earnings that are distributed to a U.S. Shareholder who has previously been taxed on the earnings will suffer only a single level of tax regardless of how many tiers of CFCs there are in the chain of ownership, thereby effectively functioning as a form of pass-through regime from the U.S. Shareholders' perspective in the context of distributions of PTI while preserving "classical" subchapter C consequences in other respects. This bifurcated treatment of U.S. Shareholders and other shareholders is not necessary in a true pass-through system like subchapter K because all earnings are currently taxable and there is no need to trace distributions to previously taxed and non-previously taxed earnings.

By contrast, under "classical" subchapter C principles, if the earnings (otherwise subject to subpart F) were distributed to the U.S. Shareholder as a dividend and then recontributed as illustrated by the earlier example, the distribution of earnings would reduce E&P at each level in the chain of ownership (and the recontribution creates additional stock basis in the relevant CFC shares). While the result is similar to Section 959 if the U.S. Shareholder owns 100 percent of the CFC,⁶¹ it differs if there are minority shareholders when a lower tier CFC is sold. As the earlier example shows, gain from a sale of Company 2 that is in part attributable to the passive income on which USS has been previously-taxed would give rise to E&P with respect to the portion indirectly owned by the other shareholders. As no mechanism specially allocated the E&P generated to those other shareholders, USS may be taxed again when the E&P is distributed while the other shareholders are arguably under-taxed on the distribution.

Although the PTI regime adopts a form of pass-through regime when PTI is recovered by way of distributions, somewhat incongruously, it is not clear that the rules function like a pass-through regime when, instead of distributing PTI, a U.S. Shareholder disposes of its indirect interest in a lower tier CFC with undistributed PTI.

Example 2 -- Disposition of Lower-tier CFC

 

 

L-CFC has PTI of $80, U-CFC has $100 of 961(c) Basis increase and USS has $80 of 961(a) Basis increase as a result of the subpart F inclusion. The sale of L-CFC does not result in additional subpart F income to USS because the 961(c) Basis is taken into account for this purpose (and sale proceeds do not exceed the aggregate basis of $200). However, at least if read narrowly 961(c) Basis is taken into account solely for that purpose. Consequently, U-CFC will be treated for general purposes as having sold shares with a basis of only $100 for $200, giving rise to $100 of E&P. Because U-CFC's sale does not generate subpart F income taxable to USS, there is no PTI account at the U-CFC level. Consequently, a subsequent distribution of $100 to the shareholders is not only taxed to F1 as a dividend (as it clearly should be) but also to USS. Effectively, treating 961(c) Basis, under the narrow reading of the rule, as relevant solely to determine the subpart F income from the disposition preserves the potential for duplication of income and gain.

This is inconsistent with the treatment of E&P of L-CFC that is not PTI. For example, if L-CFC prior to the sale had also earned $20 of non-subpart F income and U-CFC sold L-CFC for $220, Section 964(e) would treat $20 of the gain as a deemed dividend, resulting in $20 of E&P at the U-CFC level but also causing further subpart F income that results in a PTI account of $20 at the U-CFC level.

If our proposals were adopted, upon the sale of L-CFC, $100 of PTI would tier up to U-CFC (increasing the E&P PTI Subaccounts at that level), so that USS would not be taxed again on the subsequent distribution.

If Congress intended to limit Section 961(c) Basis to reducing subpart F income from a disposition of a lower tier CFC in order to preserve the potential for earnings duplication, the PTI regime does not in fact ensure that result either. The PTI regime does not adjust prior 961(c) Basis increases at upper tier entities back down again after the CFC with the related PTI has been sold. Imagine that in the example above, U-CFC is instead owned by USS and F1 through another CFC, Parent CFC.

Example 3 -- Disposition of Lower Tier CFC With Multiple Tiers

 

 

The conflict in the PTI regime between the treatment of distributive and dispositive transactions is particularly pronounced in the case of redemption transactions that qualify as exchanges under Section 302(a).

Example 4 -- Redemption Base Case

 

 

In year 2, CFC redeems USS's entire interest for $100 in a transaction treated as an exchange.⁶² USS recognizes gain of $50. Under Section 1248, $35 of the gain is treated as a dividend of USS's share of the $50 of non-PTI. The $100 distribution in redemption (while not a dividend to the extent of the $50 of PTI) nevertheless reduces E&P under Section 312(n)(7) and the Proposed Regulations. Under the Proposed Regulations, E&P is reduced by the PTI attributable to the redeemed stock (but without accounting for PTI that would otherwise have arisen from the Section 1248 deemed dividend) (i.e., by $50) and a ratable portion of the non-PTI (i.e., by a further $35).⁶³ The result is that CFC's E&P is reduced to $36.50. Under the Proposed Regulations, USS's PTI Accounts attributable to the redeemed shares cease to exist (even if they had a positive balance).

If, instead USS had sold its shares to USX, a U.S. person, for $100, USS would have been taxed on $50 of gain ($35 of which would be treated as a deemed dividend under Section 1248(a)). CFC's E&P PTI Subaccount of $50 would be increased by the $35 of Section 1248 deemed dividend and USX would succeed to shares with PTI Account attributes of $85. The Section 1248(a) deemed dividend does not reduce E&P (as it is treated as PTI).⁶⁴ CFC's E&P (including PTI) would be $121.50. However, the buyer, USX, as a successor in interest would also inherit the PTI attribute.

The result of a redemption could be viewed as the worst of both worlds, with the redemption treated as a distribution from CFC's PTI perspective (reducing its PTI attributes) but as dispositive from USS's perspective. This result is arguably unavoidable, however, given the general subchapter C treatment of redemption distributions that qualify as exchanges from an E&P standpoint. Because Section 312(n)(7) eliminates the E&P (which includes PTI), it would arguably be inappropriate to preserve the E&P PTI Subaccounts as this effectively would allow a subsequent distribution of other E&P to be treated as if it were from PTI. In the case of an upper tier CFC, one can also argue that the result is not inappropriate. USS recovers the earnings without additional tax because the 961(a) Basis reduces gain (and in that sense provides the benefit of PTI attributes) while the shareholder receives the cash attributable to the E&P. However, the results may be less obviously appropriate when the redeeming CFC is a lower tier CFC.

Example 5 -- Redemption by Lower-tier CFC

 

 

In the redemption of U-CFC's interest in L-CFC for $100, U-CFC recognizes $100 of gain for general purposes, $35 of which is a deemed dividend under Section 964(e) (applying Section 1248 principles) which under Section 964(e) is not eligible for the same country dividend exclusion.⁶⁵ However, for purposes of determining USP's subpart F income, taking into account 961(c) Basis, the gain is only $50. Presumably, although not entirely clear from the Proposed Regulations, $35 of this is attributable to the Section 1248 amount previously taxed (i.e, the amount deemed realized as exchange gain is reduced by the deemed dividend) and $15 is treated as gain that is subpart F income.⁶⁶ U-CFC's E&P is for general purposes increased by $100 (corresponding to $35 of Section 964(e) deemed dividend and $65 of net gain), of which $35 is PTI (corresponding to the $35 subject to tax as subpart F income because of Section 964(e)), $15 is PTI (corresponding to the $15 subject to tax as a subpart F gain) and $50 is non-PTI.

If U-CFC distributes the $100 to USP, USP is taxed on $50 as dividend income. U-CFC's C2 PTI Account would also have been $50 if instead it had sold its interest in L-CFC to a third party. In this case, however, the Proposed Regulations treat the distribution as reducing PTI at the L-CFC level. Because this is not a distribution under Section 302(d), the $50 of PTI account does not tier up to U-CFC. Yet, the Proposed Regulations expressly provide that USP's C2 PTI Account with respect to the redeemed shares is eliminated.⁶⁷ The result is arguably harsh even as compared to the duplication that results from a sale of a lower tier CFC to a third party. In that case the buyer would at least, as successor in interest to U-CFC, inherit the $50 C2 PTI Account at L-CFC.

If our proposals were adopted, $50 of L-CFC's PTI would tier up to U-CFC in connection with a redemption treated as an exchange, preventing additional tax on a subsequent distribution to USP.

Example 6 -- Lower-tier CFC Redemption

 

 

Now assume L-CFC redeems U-CFC1's entire interest for $50 in a Section 302(a) redemption treated as an exchange. For subpart F purposes, there is no gain. However, U-CFC1 has $50 of E&P which is not PTI (as the 961(c) Basis is ignored for that purpose). Under the Proposed Regulations, L-CFC's E&P is reduced to $50 as are its E&P PTI Subaccounts. Following these transactions, the status of the shareholders and their CFCs is as follows.

 

 

If our proposals were adopted, $50 of L-CFC's PTI would tier up to U-CFC1 upon a redemption treated as an exchange, so that a subsequent distribution of $50 to USP would not be taxed again.

As subpart F fundamentally departs from classical subchapter C principles in any event by requiring current taxation of undistributed earnings, it is somewhat unclear why, as a policy matter, Congress through Section 961(c) would have preserved the potential for duplication of earnings in one case (disposition transactions) but not the other (distributions). If preserving the potential for duplication of earnings was a policy goal under subpart F, the rules could equally have treated distributions that "tier up" a chain of CFCs as dividends for all purposes (including increasing E&P) other than for purposes of determining whether the distributed amount is subpart F dividend income to the U.S. Shareholder. Ultimate distribution to the U.S. Shareholder of the earnings that had borne tax when earned by a lower tier CFC could be taxed again as a dividend. This approach would similarly defer current recognition of duplicated income under subpart F, but without eliminating the duplication. Yet, that is not how subpart F and Section 959 work in the case of distributions. On the other hand, Section 961(c) defers, but not does not prevent, double taxation of PTI in the case of dispositions.

A possible rationale for this inconsistent result is that, notwithstanding the fundamental classical scheme under subchapter C, there are superseding rules in many situations (such as the dividends-received deduction or under the consolidated return rules) that effectively avoid or reduce the consequences of duplication of income when earnings are distributed but do not prevent duplication in the case of dispositions. Further, as a practical matter the duplicated income resulting from a disposition of a lower tier CFC with PTI generally will not be taxed until it is ultimately distributed to the U.S. Shareholder in a non-redemptive, non-liquidating distribution. Given the ability and propensity of U.S. Shareholders to maintain deferral indefinitely with respect to other earnings and therefore to avoid repatriating earnings, the potential for duplicative taxation may in most cases be theoretical rather than real. Nevertheless, the rules if applied in this manner force U.S. Shareholders to engage in self-help transactions and planning, which may be inefficient.

It also is not clear that Congress fully appreciated this potential of effective duplication of earnings from dispositions of lower tier CFCs when it adopted Section 961(c) in 1997. The limited discussion in the Conference Report of the Taxpayer Relief Act, which enacted Section 961(c), does not directly address the effect of the basis adjustment on E&P:

The simple example given in the 1997 Conference Report sidesteps the more difficult questions presented by the Solely for Subpart F Limitation on the scope of 961(c) Basis increases. Assume in the Conference Report example above that upper tier CFC owned an intermediate CFC which owned a lower tier CFC and it is the intermediate tier CFC that disposes of the lower tier CFC but then distributes the $300 proceeds to the upper tier CFC. It is doubtful, given the discussion in the Conference Report above about avoiding duplicative subpart F tax, that Congress expected the distribution to upper tier CFC of the $100 attributable to PTI to have the potential to be taxed a second time to the U.S. Shareholder as a subpart F dividend.⁶⁹ Indeed, in this situation, one can read Section 961(c) of the Code to require that E&P be computed taking into account the 961(c) Basis to the extent this is necessary for the purpose of determining "the amount included under Section 951 in the gross income" of the U.S. Shareholder. Section 961(c) does not say 961(c) Basis is relevant only for purposes of determining the amount included under Section 951 with respect to the disposition.

Thus, an alternative view of the purpose of the Solely for Subpart F Limitation on the relevance of 961(c) Basis is that Congress really intended, more narrowly, to prevent the basis adjustment resulting from a U.S. Shareholder's tax on undistributed subpart F income from benefitting other shareholders who had not been taxed on the subpart F income, but because Congress was focused on a fairly simplistic factual example, it did not focus on the fact that the limitation, if read narrowly to apply only to subpart F income resulting from a disposition of the lower tier CFC, effectively preserved the potential for duplicative taxation of the PTI to the U.S. Shareholder as well.

Congress could have expressly treated PTI in the same manner as non-PTI is treated under Section 964(e). Applying a similar approach, upon a disposition of lower tier CFC shares, a proportionate amount of PTI would be deemed to have been distributed immediately prior to the disposition, reducing lower tier CFC PTI, and causing a concomitant downward adjustment in 961(c) Basis. In addition to avoiding current taxation under subpart F of the PTI, under the exclusion in 959(b) for distributions of PTI, this approach would effectively cause lower tier CFC PTI to "tier up" into the E&P PTI Subaccounts of the upper tier CFC ensuring that subsequent distributions of disposition proceeds attributable to PTI would not be taxed again when ultimately distributed to the U.S. Shareholder. At the same time, the treatment preserves the potential for appropriate taxation with respect to other shareholders who have not been previously-taxed on their share of the earnings. In light of the fact that Sections 961(c) and 964(e) were enacted by the same legislation and at the same time, the failure by Congress to extend the deemed dividend construct to PTI could call into question whether Congress intended this result. As discussed later in the Report, however, Section 964(e), as drafted, literally does not cause PTI to tier up only because of the manner in which Section 1248 technically avoids double taxing PTI (i.e., by excluding it from the definition of E&P for Section 1248 purposes rather than treating all E&P as tiering up in the deemed distribution and then excluding the amount attributable to PTI). As Congress was focused when it enacted Section 964(e) on the impact of the deemed distribution from a foreign tax credit perspective, it is far from clear Congress even appreciated this subtlety.

Section 961 directly suggests an association of PTI attributes with particular shares. Section 961(a) provides for an increase in the "basis of a United States shareholder's stock" in the CFC by the amount required to be included "with respect to such stock". Section 961(c) is to the same effect. This language is not unambiguous, but could be read to attribute the PTI attributes to the specific shares with respect to which Section 951(a) treats the U.S. Shareholder as receiving an inclusion. However, Sections 961(a) and 961(c) also provide for an increase in basis "of property of a United States shareholder by which he is considered under Section 958(a)(2) as owning stock" of a CFC by the amount the shareholder included with respect to such property. This can be read to suggest that the basis adjustment is spread over all stock or property through which the CFC is directly or indirectly owned. The Proposed Regulations adopt the latter approach for purposes of Section 961.

An alternative approach that could have been adopted by the Proposed Regulations would be to treat PTI attributes as personal to the U.S. Shareholder for so long as it retains an interest in the CFC (rather than allocated among specific shares) and treat the attribute as transferred to a successor only following a complete disposition. However, such an approach when combined with the rule providing for a transfer of PTI attributes upon a complete disposition could in theory enable the U.S. Shareholder to create shares with "super-charged" PTI attributes.

Assume that USS owns 100% of CFC, which earns $100 of subpart F income. USS sells 99% of CFC to F for $99. Effectively, USS's proportionate interest in the underlying E&P pool has been reduced to $1. If the hypothetical PTI regime allows USS to treat the PTI as a personal attribute, presumably USS must recognize gain of $99 on the sale and reallocate its $99 of 961(a) Basis to the retained shares. Those retained shares give USS the right to treat the next $100 of distributions as out of PTI even though in an economic sense $99 of the distribution would be made out of E&P generated by future active business activities that is non-PTI.

The result above may seem relatively unobjectionable insofar as USS has been taxed on $100 of earnings and also on the subsequent sale (which in effect represents a tax at the shareholder level on those future earnings USS may receive in subsequent distributions). However, USS may be subject to a relatively low rate of tax on the subpart F income (e.g., if it can offset subpart F inclusions with net operating losses or is a tax-exempt U.S. Shareholder). USS would now own shares that could be sold to a third party (for $1) that would entitle that buyer as a successor in interest to the same result (i.e., up to $100 of tax-free distributions). In extreme cases, the successor in interest would receive shares that effectively allow it to repatriate unrelated future earnings tax-free. Possibly, the need to prevent the creation of shares with "super-charged" PTI attributes is another reason the Proposed Regulations seek to pro rate PTI attributes on some reasonable basis among all of the shares owned by the U.S. Shareholder.

Although neither Section 959 nor the Proposed Regulations thereunder contain a rule prescribing separately how PTI Account attributes are allocable to specific shares, it also must be the case that the PTI Account attributes are allocated to the same shares to which 961 Basis is allocated. If that were not the case, the account and basis adjustment mechanics would not work appropriately.

Example 7 -- Multiple Share Classes

 

 

If, hypothetically, the $40 C2 PTI Account were allocable to USS's preferred shares in CFC but the $40 of 961(a) Basis was pro rated across all shares through which USS is deemed to own CFC under Section 958(a), less than $40 of the basis increase would be attributed to the preferred shares held by USS and the balance to the common shares. Upon a distribution of $100 on the preferred shares, $40 to USS, USS would be required by Proposed Treasury regulations Section 1.959-3 to reduce its basis to zero and any excess would be taxable as gain.⁷⁰ The "inter-block" sharing of PTI, discussed below, does not resolve this problem because it reallocates PTI from shares with an excess PTI Account to shares with a deficient PTI Account. It does not reallocate basis to shares with inadequate basis to support the block's PTI Account. In order to avoid an incongruous result, it must be the case that the Proposed Regulations intend there to be a consistent allocation of PTI Accounts and basis among different shares of stock.

The Proposed Regulations require 961 Basis adjustments be made "pro rata" among all of the shares (or property that causes the shareholder to be deemed to own the CFC) owned by the U.S. Shareholder.⁷¹ To achieve consistency, the PTI Accounts must therefore be apportioned among the shares in a corresponding manner. However, the Proposed Regulations provide no guidance as to how that is to be done (i.e, what is meant by "pro rata") when there are different classes of shares.

It would be nonsensical to "pro rate" the 961 Basis adjustments and PTI Accounts based on the number of shares outstanding when the share classes may have vastly different economic terms. In the example, above, there might be 100,000 shares of common stock and 100 shares of preferred stock. It would be more sensible to pro rate the PTI and related basis adjustments based on the relative fair market value of the classes. However, this raises valuation and administrability concerns. It would also lead in complex cases, somewhat anomalously, to an inconsistent allocation of PTI for PTI regime purposes and non-PTI for purposes of Section 1248.

Section 1248 in complex cases generally allocates E&P by reference to the manner in which E&P would be distributed in a hypothetical distribution, applying the principles of Treasury regulation Section 1.951-1(e). Thus, the $100 of non-PTI would be allocated entirely to the preferred shares ($40 to USS's shares) for Section 1248 purposes in connection with a disposition.

Assume in the example above common shares and preferred shares each represent 50% of the aggregate net equity value of the corporation. A pro ration of PTI and 961(a) Basis based on relative fair market value would result in USS having a C2 PTI Account of $20, and 961(a) Basis of $20 with respect to the preferred shares it owns (the balance of each would be allocated to the common shares it owns). By contrast, $40 of the non-PTI would be attributed to USS preferred shares for Section 1248 purposes. A redemption of a portion of the USS preferred shares for $40, would (assuming for illustrative purpose it was treated as an exchange) result in USS recognizing $20 of gain, all of which would be a Section 1248 dividend. By contrast, if 961(a) Basis is allocated in the same way as non-PTI, there would be no gain and no dividend inclusion under Section 1248.

While there is no fundamental reason the rules cannot allocate different subcategories of E&P differently among shares for PTI and Section 1248 purposes, at a minimum it seems counter-intuitive and inconsistent with the general principle of the Proposed Regulations that PTI should be recovered as early as possible. Moreover, there would seem to be no compelling policy reason not to conform the PTI regime's approach to the detailed approach that already exists under Section 1248 for allocating E&P among shares in complex cases. While there may be other sensible approaches to allocating PTI, it does not seem necessary or appropriate to create yet further complexity by establishing a separate system of allocation for PTI only. The inter-block sharing rule below makes the allocation of PTI attributes and related basis adjustments largely irrelevant for practical purposes in distribution transactions, so this allocation is primarily relevant in the case of dispositions of CFC shares which is also when Section 1248 allocation of the E&P is relevant (and 961(a) Basis is taken into account in measuring gain, which in turn, determines how much of the non-PTI will be treated as a dividend under Section 1248).

As mentioned above, the Proposed Regulations allow "sharing" of PTI attributes of different shares in the case of distributions, which substantially reduces the import of per share allocations of PTI attributes in that context.

Example 8 -- Inter-Share "Sharing" of PTI Attributes

 

 

If instead, CFC distributed $100 pro rata, under the general rule associating the PTI attributes with specific shares that were owned at the time of the inclusion of subpart F income, USS would potentially be taxed on the $80 received with respect to the subsequently-purchased shares to which no PTI attributes attached. However, under the inter-share sharing rule, the Proposed Regulations would effectively reallocate the excess $80 of PTI from the originally owned shares to the subsequently-purchased shares and USS is not taxed. This would be permitted among shares, different blocks of shares and different classes of shares owned by USS. Thus, although the rules allocate PTI to specific shares, in the case of subsequent distributions, the "sharing" rule substantially reduces the consequences of that original allocation.

If the scope of the "sharing" rule is limited only to shares of the same CFC owned directly by the same taxpayer, this may reach results that are counter-intuitive in certain situations.

Example 9 -- Non-consolidated Inter-taxpayer Sharing (Redemption)

 

 

At a time when L-CFC assets have appreciated by $400, L-CFC redeems USP's 20% interest for $100. Because USP is deemed to own 100% of L-CFC as a result of the attribution rules under Section 318, the redemption is treated as a dividend. The C2 PTI Account (and 961(a) Basis) associated with the shares redeemed is only $20. USP has 961(a) Basis of $80 with respect to the shares in U-CFC. The Proposed Regulations state that the sharing rules apply "if a covered shareholder owns (within the meaning of section 958(a)) more than one share of stock in a foreign corporation."⁷² Thus, the Proposed Regulations can be read to treat USP as having a PTI account of $80 with respect to the L-CFC shares it owns indirectly through U-CFC within the meaning of Section 958(a), in which case USP could access PTI in its indirect L-CFC account to shelter the redemption distribution on its directly held L-CFC shares. If that is not the case, USP will be subject to tax on the $80 dividend. Sharing in this situation appears to be a fairer outcome because USP was taxed on the entire $100 of subpart F income, because the attribution rules of Section 958(a) treat USP for this purpose as if it did directly own the shares of L-CFC that are actually owned by U-CFC, and USP receives a distribution of $100 in cash in connection with the redemption. Moreover, if USP sold the directly owned shares in L-CFC to U-CFC for $100, in a sale that would be recharacterized as a dividend under Section 304, it appears USP would not be taxed because to the extent treated as a dividend from L-CFC, the distribution is deemed to be made first out of PTI.⁷³ Given the result in the substantively identical Section 304 sale, the absence of sharing in the redemption scenario may not make much sense as a policy matter.

The appropriate answer seems less clear in a modified fact pattern.

Example 10 -- Non-consolidated Inter-taxpayer Sharing (Pro Rata Distribution)

 

 

USP has a direct PTI account of $20 in L-CFC, so $20 of the $100 dividend paid to USP is clearly excluded from income. It is less clear, however, how the remaining $80 distributed to USP should be treated. One approach would be that USP is deemed to have a C2 Account of $80 with respect to the shares it owns indirectly through U-CFC and can use that account to characterize the remaining $80 as PTI. Another approach would be to treat the distribution of $80 to USP as a dividend. The $400 distribution to U-CFC would then be treated as a distribution of $80 of PTI and $320 of non-PTI and USP could recover the remaining $80 of PTI upon a subsequent distribution from U-CFC. This example demonstrates the tension between following what would essentially be a tracing approach with respect to PTI (as cash moves up through tiers of corporations) and an approach that seeks to implement more fully the principle of recovering PTI first.

If the final regulations do not to allow "sharing" of PTI attributes among shares of the same corporation held indirectly pursuant to Section 958(a) by the same person, the PTI attribute would not necessarily be wasted. If U-CFC later distributes $80 to USP, that distribution would be out of PTI. However, in the absence of a sharing rule that encompasses shares indirectly owned under section 958(a), tax to USP would be accelerated and USP may not be permitted to recover PTI "as early as possible." The result may be inconsistent with the "pass-through" approach of subpart F and the guiding principle of the Proposed Regulations that PTI be recovered as early as possible.

If Treasury and the IRS decide to apply PTI sharing to shares in a lower tier CFC indirectly owned (within the meaning of Section 958(a)) by a U.S. Shareholder, whether in the context of redemptions only or more generally, Treasury and the IRS should consider extending it to situations in which a lower tier CFC is owned through an upper tier entity that is not a CFC. Assume in the Example 9, that USP owns 49% by value and F, a foreign person, owns 51% by value of F Corp (100% of the voting common shares). USP's interest in F Corp is in the form of non-voting convertible cumulative preferred. Assume, solely for illustrative purposes, that because of the accrued preferential right to dividends, USP's pro rata share of the $100 of subpart F income is treated as 100% and that the $100 distribution is a Section 302(d) distribution.

Example 11 -- Non-consolidated Inter-taxpayer Sharing (non-CFC)

 

 

While this Report does not express a position on whether inter-taxpayer sharing should be permitted with respect to shares that are owned within the meaning of Section 958(a), we believe in any event that the foregoing examples are distinguishable from the situation in which the intermediate company is a U.S. corporation. Assume in the prior example, that instead of owning L-CFC through F Corp, USP owns L-CFC through USX.

Example 12 -- Non-consolidated Inter-taxpayer Sharing (US Co)

 

 

Nevertheless, "sharing" of PTI attributes may be appropriate if USP and USX are members of a consolidated group. A consolidated group is treated for many purposes as a single taxpayer. To achieve "single entity" treatment typically members for most purposes may share favorable tax attributes, including net operating losses and foreign tax credits. Those shared attributes may determine the extent to which the actual U.S. Shareholder is in fact taxed on a subpart F inclusion. For example, in a case in which USP and USX are consolidated, it would be somewhat anomalous for USX's net operating losses to reduce the tax paid by the group (including with respect to USP's direct interest) on the subpart F income inclusion but preclude USX from using USP's PTI resulting from the same subpart F inclusion. It therefore seems consistent with this general approach to favorable tax attributes to follow a single entity model for PTI also. It is true that the consolidated return regime is a "hybrid" with complex rules for intercompany transactions and use of attributes intended to preserve the location of particular items. In particular, the consolidated return rules do not generally permit basis sharing but follow a subchapter C basis "tracing" approach (although with various basis adjustments resulting from the mechanics of the consolidated regime). However, 961 Basis is different from Sub-C Basis. The PTI regime effectively ties tax-free recovery of PTI to 961 Basis through the adjustment mechanics under Proposed Regulation Section 1.959-3 (e.g., treating a distribution of PTI in excess of basis as taxable gain). Indeed, although Congress chose to denominate the attribute as a form of "basis", functionally 961 Basis in most respects is more like a tracking mechanism that is part of the PTI account construct. Sharing of 961 Basis is therefore necessary to permit sharing of PTI account attributes.

As discussed above, SubC Basis generally arises only from an investment of post-tax funds (and by analogy, upper tier CFC basis in a lower tier CFC that has recognized subpart F income would be limited to the amount actually taxed to U.S. Shareholders). However, limiting the basis adjustment of an upper tier CFC in a lower tier CFC to the amount actually taxed to the U.S. Shareholder does not achieve the purpose of preventing double taxation of PTI when the U.S. Shareholder has a less than 100% interest in the underlying E&P and PTI pools. Because subchapter C has no analog to 743/754 partnership basis and 704(c) allocation of gain, subpart F income and E&P would arise when an upper tier CFC sells a lower tier CFC if 961(c) Basis was limited to the actual amount of the earnings that were taxable to the U.S. Shareholder.

Example 13 -- 961(c) Basis

 

 

Conversely, if the 961(c) Basis adjustment were treated as basis for all purposes (whether it is limited to $80 or treated as $100) this would reduce E&P from the sale at the U-CFC level, which benefits the other shareholders. For example, if F1 sells to USX, on subsequent distributions to the shareholders of U-CFC of $200, $80 or $100 (as the case may be) less would be treated as a taxable dividend saving USX from being taxed on $16 or $20 of income as the case may be. Accordingly, the statute and Proposed Regulations treat this "grossed up" basis adjustment as applying "solely for purposes" of determining USS's subpart F income, and may be read narrowly as applying "solely for purposes" of determining such subpart F income on a sale of L-CFC by U-CFC. Treating the 961(c) Basis like any other basis (at least to the extent of $80 in the example above) would benefit the minority shareholder, which does not appear to be the correct result. That result would, however, reflect an inherent problem with subchapter C rather than a problem specific to subpart F.

Assume in the example above that prior to the sale to F2, L-CFC distributed $80 to U-CFC which distributed it solely to USS (e.g., by way of a redemption treated as a distribution), and then USS recontributed the amount down the chain of ownership in exchange for newly issued shares. USS would be able to recover the $80 without further tax under subpart F and convert $80 of 961(c) Basis into $80 of SubC Basis at both the U-CFC and L-CFC share levels. When the sale to F2 occurred, this amount would reduce E&P and therefore the potential for taxation when the proceeds are later distributed to USX. However, that result arises from the fact that subchapter C does not apply principles analogous to Section 704(c) to duplicated cost basis and the problem is not peculiar to the PTI regime.

The Proposed Regulations (and, arguably, the statute) imply that the character of basis as 961(a) or 961(c) basis depends on the circumstances at the time the basis adjustment occurred. However, given the purpose of Sections 959 and 961 it is questionable whether this makes policy sense purely from a subpart F perspective.

Example 14 -- Duplication of 961(a) Basis

 

 

If U-CFC's basis in L-CFC is $100 for all purposes (i.e., there is no carryover of 961(a) Basis), if it subsequently sells L-CFC for $200, there will be $100 of subpart F income at U-CFC, $80 of which is includible by USS giving rise to additional E&P at that level of $100 ($80 of which would be in the U-CFC's C2 E&P subaccount). Effectively, USS would have been taxed twice on the PTI. If U-CFC's basis in L-CFC is $180 (i.e., includes carryover 961(a) Basis of $80), there is still $20 of gain, of which USS would include $16, again resulting in double taxation of the PTI, albeit only a portion of it. Similarly, a distribution of $200 up the chain would result in $20 gain to the extent the PTI Account reduction exceeds the available basis. In order to prevent double taxation, the basis in L-CFC needs to be $200 (i.e., what it would have been under Section 961(c) had the L-CFC subpart F income arisen under the new corporate configuration). On the other hand, if the subpart F income had arisen under the new corporate configuration the $100 adjustment would have been 961(c) Basis subject to the Solely for Subpart F Limitation.

Conversely, if the original $80 961(a) Basis is treated as basis for all purposes and it carries over to U-CFC shares in L-CFC, it effectively reduces the E&P recognized by U-CFC on the sale of L-CFC which would ultimately benefit the other shareholders who were not taxed on the $80. (If the amount were "grossed up" to equal what could be excluded by U-CFC upon a distribution of PTI by L-CFC, the benefit to the other shareholders would be even greater.)

Purely from the perspective of subpart F, the better approach would be that the 961(a) Basis should upon contribution not carry over under regular subchapter C principles but "morph" into 961(c) Basis. Specifically, on these facts, the $80 of 961(a) Basis should morph into $100 of 961(c) Basis. This would ensure both that USS is not taxed duplicatively on the earnings generated by the subpart F income but also prevent other shareholders benefitting from a reduction in E&P attributable to earnings on which only USS has actually paid tax.

The reverse situation could also apply. Assume that in the example above the facts are the same except that U-CFC is a holding company from inception and owns L-CFC.

Example 15 -- Potential "Wasting" of 961(c) Basis

 

 

L-CFC recognizes $100 of subpart F income, of which $80 is taxed to USS. USS has additional 961(a) Basis in U-CFC of $80 (total basis of $160), and U-CFC has additional 961(c) Basis in L-CFC of $100 (total basis $200). USS now buys F's 20% interest for $40. Subsequently, U-CFC is liquidated under Section 332. For simplicity, assume U-CFC has no E&P and L-CFC has no non-PTI so that there are no Section 1248 amounts and the transaction is not taxable under Section 367(e).⁷⁴

If USS has a carry-over basis of $200 (SubC Basis from the originally contributed assets of $100 plus $100 of 961(c) Basis) this seems inappropriate in two respects. First, the amount taxed to USS (and the PTI that can now be distributed directly to USS without tax) was $80 not $100. It is true that USS is losing outside basis of $200, but $40 of that is SubC Basis attributable to the purchase of shares from F. U-CFC's inside asset basis should not be stepped up under subchapter C principles by that $40. (Any intuitive disquiet about the loss of outside basis is really an objection to an artifact of subchapter C and how it treats liquidations and basis under Section 332). Thus, the amount of USS's basis in L-CFC should be $180 not $200. However, whether it is $180 or $200, the basis in excess of $100 is 961(c) Basis subject to the Solely for Subpart F Limitation. However, if USS receives a distribution from L-CFC or sells L-CFC, the income or gain is just income or gain, not subpart F income or gain. Literally therefore the 961(c) Basis is not taken into account, with the result that USS is double taxed on $80 of subpart F-related E&P.

Once again, the better policy answer from a subpart F perspective would seem to be that 961(c) Basis of $100 should "morph" into 961(a) Basis of $80.

Example 16 -- Section 362(e) Interaction

 

 

Under the Proposed Regulations, the $50 loss does not reduce the C2 PTI Account or the C2 E&P Subaccount. It creates a separate deficit in the C3 E&P Subaccount. Consequently, if in year 3 L-CFC recognizes $50 of non-subpart F income, in year 4 L-CFC can distribute $200 up the chain to USS without USS incurring tax.⁷⁵ If that were not the rule, USS's C2 PTI Account would have been only $50 and $50 would have been a taxable dividend.

However, now assume that at the end of year 2, U-CFC1 contributes L-CFC to newly formed U- CFC2. At the time, assume L-CFC is only worth $150 as a result of the $50 loss. Consequently under Section 362(e), U-CFC2's carryover basis in L-CFC which would otherwise be $200 ($100 of SubC Basis and $100 of 961(c) Basis) is reduced to $150.⁷⁶

If L-CFC again recognizes $50 of non-subpart F income in year 3, and then in year 4 distributes $200 up the chain of ownership, U-CFC2 would recognize $50 of gain (as the required basis reduction under Proposed Treasury regulations Section 1.959-3 of $100 exceeds the available 961(c) Basis). This would be taxed to USS as subpart F income, creating additional PTI at that level (and additional 961(c) and 961(a) Basis up the chain) so that the further distribution up the chain of the $200 would not result in additional tax (although it would in turn reduce the $50 of 961(a) and 961(c) Basis so created).

Effectively, Section 362(e) overrides the rule in the Proposed Regulation that insulates PTI attributes from non-PTI earnings deficits. To the extent one believes that is the correct policy approach from a Section 959 perspective, applying the Section 362(e) basis reduction to 961 Basis would seem the wrong result from a subpart F policy perspective.

An alternative approach would be to ignore carry-over 961(a) Basis and 961(c) Basis in determining whether there is a built-in loss and only look to SubC Basis when applying Section 362(e). In the example above, since the SubC Basis in L-CFC shares is $100 in aggregate and the value is $150, there would be no basis reduction. Insofar as the additional 961(c) Basis exists solely to determine USS's subpart F income in any event, arguably it does not raise the kind of policy concern over duplication of built in losses that should cause the transaction to be subject to the adjustment. While carryover of 961(a) Basis could present such a concern if it is treated like SubC Basis in the hands of U-CFC2, if our recommendation is adopted and 961(a) Basis morphs into 961(c) Basis upon a contribution, carryover of 961(a) Basis similarly should not implicate the policy concerns underlying Section 362(e). Moreover, the fact that a distribution of PTI up a corporate chain to U.S. Shareholders will result in reductions of 961 Basis at each level (eliminating the duplication) further limits the potential for the kind of abuse with which Section 362(e) is concerned and thus supports ignoring 961 Basis in applying Section 362(e).

Example 17 -- Triangular Reorganization

 

 

 

 

However, no carryover 961(c) Basis arises in U-CFC's shares in L-CFC. Instead, under Treasury regulation Section 1.958-6, U-CFC adjusts its basis in L-CFC to achieve the same effect as if CFC had merged into U-CFC, which then contributed the assets and liabilities acquired to L-CFC. A distribution up the chain of $100 would result in taxable subpart F income at U-CFC unless U-CFC has other basis to support the distribution and the downward basis adjustments required by Proposed Treasury regulations Section 1.959-3. Presumably the transaction that generated the subpart F income originally resulted in an asset of CFC with an equivalent amount of cost basis that will be replicated in the L-CFC shares owned by U-CFC. However, that basis may have been reduced by unrelated losses. As discussed earlier in the Report, that may result in a distribution of PTI becoming taxable in a manner arguably inconsistent with the premise of the Proposed Regulations under Section 959 which "firewall" PTI from unrelated losses from an E&P perspective.

Query whether this result is appropriate from a subpart F policy perspective? In particular, does it make for 961(c) Basis, which as discussed elsewhere in this Report is not basis in the regular subchapter C sense, to disappear as a result of the application of a rule (Treasury regulation Section 1.958-6) focused specifically on SubC Basis? Does the answer change if instead of being owned by USP, U-CFC is owned by USX, an unrelated U.S. Shareholder prior to the merger?

USS and F form CFC and contribute $100. USS owns 80% and F owns 20% of CFC. In year 1, CFC earns $100 of subpart F income, of which $80 is taxed to USS creating PTI of $80 and increasing USS's 961(a) Basis in CFC by $80 (to $160). In year 2, when CFC is worth $200, USS and F contribute CFC to a newly formed foreign partnership, FPRS.

Presumably, USS takes a substituted basis of $160 in the FPRS interest and F a substituted basis of $20. FPRS presumably takes a carry-over basis of $180 in CFC. In year 3, CFC distributes $100 which is a dividend to the extent of the E&P of $100. However, presumably USS may exclude from income its shares of the $100 distribution as from PTI. That would presumably require FPRS to reduce its carry-over 961(a) basis in CFC by $80. However, this does not reduce USS's outside basis in FPRS. Indeed, if the portion of the distribution the PTI regime treats as a distribution that is not a dividend were treated as "tax-exempt income" (which may not be the correct characterization), this would increase USS's outside basis by $80 (to $240). Also, upon a distribution of $80 to USS, its outside basis would be reduced to $160 resulting in the potential for a "phantom" capital loss of $80.

VI. Alternative Approaches to Basis under the PTI Regime

If this is the appropriate approach, we believe the Proposed Regulations generally reach results consistent with this approach. General subchapter C principles and rules generally favor the recovery of taxable earnings as fast as possible, which is the converse of the Proposed Regulations' approach to recovering PTI as early as possible. While the statute and legislative history evidence a general Congressional intent to treat distributions as first from PTI, it is far less clear how far that general intention extends. Congress similarly clearly intended, based on the successor in interest rule, to treat PTI attributes as associated with shares in the form of "basis" and nothing in the statute or legislative history directly supports reallocating PTI attributes (overriding general subchapter C basis tracing) to achieve a "strong" version of the PTI first principle. Accordingly, PTI attribute and basis "sharing" in any situation other than among consolidated return group members seems inconsistent with "pure" subchapter C principles, not expressly prescribed by Congress and therefore inappropriate under such an approach. The rules for sharing of PTI attributes by consolidated return members might even under this view be retained, although perhaps more appropriately as a separate regulation under Section 1502, not for reasons for subpart F policy, but because the result is appropriate under the single entity theory applicable to consolidated groups under Section 1502.

Purely from the perspective of subpart F policy goals, this approach has little to recommend it. To the extent the statutory language precludes a more sensible policy approach as to duplication of earnings, it also must preclude a more sensible policy approach to the potential for duplication of 961 Basis. If Congress when it said "basis" really meant SubC Basis (albeit in the case of 961(c) Basis for the limited purpose of measuring subpart F inclusions), we see no compelling argument that this would not require accepting the potential for "duplication" in both instances. In other respects the statute is more ambiguous, for example, in what it intends about how 961 Basis is to be allocated among particular blocks and classes of shares although the PTI regime clearly presupposes some form of allocation by which PTI attributes are associated with particular shares. In yet other respects, a pure subchapter C model seems inconsistent with Congressional intent and the statute, as the legislative history generally supports the view that Congress intended distributions to be traced first to PTI (although it is less clear about precisely how far this "PTI first" principle was supposed to extend.)

This approach would maximize the likelihood that a U.S. Shareholder would fully recover the benefit of PTI attributes attributable to previously-taxed earnings. It would avoid potentially unintended interactions with other subchapter C basis rules like Section 362(e). Effectively, the CFC regime would function like a pass-through regime as to PTI but only with respect to the U.S. Shareholder that had actually been taxed on the earnings under subpart F.

Whether or not this approach reflects a regime Congress should have adopted as a policy matter, in its pure form it is more difficult to reconcile with the statutory language Congress actually enacted. Among other concerns, it would effectively nullify the concept of the "successor in interest" as there would never be (except perhaps in the case of a complete disposition) any PTI attributes to which a successor could succeed. Conversely, if a successor did succeed to PTI attributes upon a complete disposition, in various circumstances these attributes could be grossly disproportionate to the successor's interest in the underlying E&P pool of the CFC (i.e., reflect "super charged" PTI attributes). The latter concern is not insuperable. The rules could provide that upon a complete disposition of the U.S. Shareholder's interest in a CFC, a buyer would succeed to the remaining PTI attributes but subject to a cap equal to the buyer's pro rata share of the underlying E&P pool at the time of acquisition (applying principles of Treasury regulations Section 1.951-1(e) to determine that pro rata share). The rules could also provide that pure pass-through treatment applies with respect to PTI attributable to shares a U.S. Shareholder retains, while PTI attributable to transferred shares would move to the successor in interest. Once such an approach is taken, however, the resulting PTI regime would be moving closer to the regime we describe below as Modified Subchapter C.

Nevertheless, it is somewhat unclear whether such a regime in its pure form would be consistent with the statutory scheme for PTI or consistent with Treasury and IRS regulatory authority.

This modified approach is basically the approach adopted by the Proposed Regulations when it comes to distributions. If it is the correct approach, then the rules for basis sharing (and PTI account sharing) among different share blocks and classes are appropriate because they ensure pass-through treatment with respect to PTI. The question, however, is whether the Proposed Regulations go far enough in modifying the otherwise applicable SubC Basis rules in the case of dispositions and transfers. For example, one could argue that an appropriate modified regime should adjust 961(a) Basis to match indirectly owned E&P PTI Subaccount attributes following a contribution of the shares to an upper tier CFC or following a corporate reorganization to ensure that PTI can be distributed without duplicative taxation. Conversely, 961(c) Basis should be treated as 961(a) Basis (but only to the extent of the U.S. Shareholder's proportionate interest in the PTI) when shares in a lower tier CFC are distributed from an upper tier CFC to the U.S. Shareholder in a Section 332 liquidation to prevent double taxation of the PTI.

The critical issue in designing basis rules under such a modified "hybrid" regime is the extent to which Congress really intended to preserve the potential for duplication of earnings not merely with respect to other shareholders but also with respect to the previously-taxed U.S. Shareholder. Thus, if Congress intended to preserve the potential for duplication inherent in subchapter C and its basis rules, 961(c) Basis should not prevent taxation under subpart F of a distribution of the proceeds from a disposition of a lower tier CFC. Conversely, 961(a) Basis should be duplicated when shares in a directly owned CFC are contributed to an upper tier CFC, reducing E&P and the amount of a distribution of the proceeds that is taxed to other shareholders while potentially double taxing the U.S. Shareholder on some portion of the earnings. If one adopts that view, in other words, 961(a) Basis should not morph into 961(c) Basis upon contribution or vice versa in a Section 332 liquidation. While the statutory scheme for PTI might be understood to evidence such an intent, it is far from clear, and in any event there is absolutely no direct indication in the legislative history that Congress wished to preserve the potential for duplication of earnings and economic double taxation of PTI. If anything, the focus on avoiding double taxation in the legislative history is to the contrary.

Because of the disparate treatment of distributive and dispositive transactions, U.S. Shareholders can avoid duplication in any event through self-help by distributing the earnings up the chain prior to a disposition. Such a distribution may incur foreign taxes which in some cases may be adverse to a U.S. Shareholder, but to the extent they can credit the taxes against U.S. income taxes on other income may actually be disadvantageous to the fisc. Given the concern of Congress to avoid duplicative taxation of PTI and the lack of a policy justification for treating distributive and dispositive transactions involving PTI differently, it is doubtful Congress intended to force shareholders to engage in this kind of self-help as the price for avoiding duplication of earnings.

On the other hand, if Congress did not focus on the consequences of duplication for the subpart F regime and merely intended to ensure appropriate subchapter C taxation of shareholders other than the U.S. Shareholders, different basis rules are appropriate. In that case, Section 961(a) Basis should not merely be duplicated at the U.S. Shareholder level but treated as converted into Section 961(c) Basis (with no impact on the E&P pool from the perspective of other shareholders) in the hands of the upper tier CFC following a contribution. However, it should reduce E&P (and potential duplicative taxation of the earnings) from the perspective of the U.S. Shareholder.

Probably the most elegant way to achieve this result, in addition to a specific rule converting 961(a) Basis to 961(c) Basis following contribution (or vice versa in a Section 332 liquidation), would be to treat a disposition of a lower tier CFC in a manner analogous to Section 964(e). Under this approach 961(c) Basis would only be taken into account in measuring subpart F income of the U.S. Shareholder on disposition of the lower tier CFC, but an appropriate portion of the PTI E&P Subaccounts at the lower tier CFC would "tier up" to the upper tier CFC. However, this would reduce E&P of the lower tier CFC which would otherwise benefit a successor. A more limited approach would be to treat the PTI attribute as "tiering up" without otherwise adjusting the lower tier CFC's E&P account.

Treasury and the IRS have broad authority under Section 961 to determine basis adjustments (which should include the authority to determine the manner in which basis is duplicated and replicated in corporate reorganizations) in a manner that reasonably achieves the intended policy goals of subpart F. On balance, it makes little sense to think Congress intended Treasury and the IRS to exercise that authority to prevent inappropriate duplication of basis (e.g., by requiring the conversion of 961(a) to 961(c) Basis where appropriate) without having equal authority to ensure consistent non-duplication of earnings.

We note that the failure of Section 964(e) as drafted to cause PTI to tier up along with non-PTI upon the disposition of a lower tier CFC is not a self-evident result of the language of Section 964(e) but the indirect result of the mechanical approach by which Section 1248 avoids duplication of tax on PTI (by technically excluding it from the Section 1248 definition of E&P even though it is E&P for other purposes). However, that result could equally have been reached by treating all E&P as tiering up as a dividend but allowing the U.S. Shareholder to exclude the PTI component. Were that the Section 1248 mechanic, Section 964(e) would cause PTI to tier up along with other E&P. Consequently, we hesitate to read too much into Congress's failure to provide an explicit analog to Section 964(e) under the PTI regime.

We would also note the plain statutory language of Section 961(c), which states that the adjustment applies "only for purposes of determining the amount included under Section 951 in the gross income of such United States shareholder" with no mention that it is limited to recognition of subpart F income from the disposition of a CFC rather than a subsequent distribution of proceeds. Whether or not the latter was what Congress was focused on, we do not think the language in that regard is in any sense "clear."⁷⁸ Indeed, the example of a Section 332 liquidation which converts PTI into taxable income (see Example 15, above) aptly illustrates the absurd results of reading the language in Section 961(c) narrowly to reach a result that surely cannot have been intended by Congress. Consequently, given the general technical and policy complexity of the PTI rules, we believe it would be well within the interpretive regulatory authority of Treasury and the IRS to draft rules that create a PTI Account to prevent duplicative Section 951 inclusions while at the same time maintaining the E&P consequences for other shareholders. In exercising its authority, we would recommend that Treasury and the IRS take the approach of treating the PTI attributes of a lower tier CFC as "tiering up" without otherwise adjusting its E&P account.

We also do not believe that from a practical perspective this would create much risk to Treasury and the IRS of being "whipsawed" on the authority question in future litigation. No U.S. Shareholder should have reason to complain about the result. A possible complainant would be shareholders in an upper tier CFC to which a lower tier CFC was contributed who are U.S. persons but not U.S. Shareholders who could argue that by failing to reduce E&P by the contributed carry-over 961(a) Basis Treasury has adversely affected them and exceeded its authority. However, given the express authority of Treasury to determine 961 Basis "under regulations," the argument that the statute mandates duplication of 961(a) Basis (when it is silent on how adjustments should apply in reorganizations or how these basis adjustment carry over) seems highly unlikely to prevail.

A more plausible complainant would be a U.S. purchaser of a lower tier CFC who inherits the corporation with the E&P intact but without PTI attributes that would otherwise have been preserved. As a practical matter, such a shareholder in many cases would have had the option of eliminating the E&P by making a Section 338(g) election which it chose not to exercise, likely reducing its status as a sympathetic complainant. In any event, however, we do not think the argument should prevail. The language regarding "successors" in Section 959(a) and (b) speaks of the U.S. Shareholder or any other U.S. person who acquires "any portion of the interest of such United States person" in such CFC. There is nothing in this language that clearly or unambiguously treats as a successor a person that acquires an interest in a lower tier CFC from an upper tier CFC rather than acquiring the interest of the U.S. Shareholder itself in the upper tier CFC. Treasury also has express authority to determine the proof of the identity of the interest acquired by any successor. Consequently, such an argument also seems unlikely to prevail.

VII. Recommendations

In the case of a transfer of interests to which 961(a) Basis attaches to a lower tier CFC in which basis otherwise carries over, we believe the U.S. Shareholder should receive a substituted 961(a) Basis in the upper tier CFC shares received in the exchange but the basis of the upper tier CFC in the transferred lower tier CFC shares should be the 961(c) Basis the shares would have had if the subpart F income had arisen following the transfer.

Similarly, in the case of a transfer by a CFC of interests to which 961(c) Basis attaches to a U.S. Shareholder in a transaction in which basis otherwise carries over, we believe the U.S. Shareholder should receive a 961(a) Basis in the lower tier CFC shares received limited in amount to the 961(a) Basis the shares would have had if the subpart F income had arisen following the transfer.

Equally, we recommend that the final regulations eliminate the potential for inappropriate duplication of earnings subject to subpart F inclusion in the case of a disposition by an upper tier CFC of a lower tier CFC. By analogy to Section 964(e), this could take the form of a deemed distribution of a proportionate amount of the E&P (replicating the result a shareholder could have achieved by self-help anyway but avoiding the potential for foreign taxation of an actual distribution). We recommend that the final regulations provide for a tiering up of PTI Accounts with respect to the lower tier shares to the upper tier shares to match the duplicated PTI-related E&P without otherwise reducing E&P at the level of the lower tier CFC. That would preserve the potential for taxation of the purchaser on distributions out of historic E&P. It would increase the likelihood of authority challenges by successors, although for reasons discussed above we believe these are unlikely to prevail. However, if authority challenges are a concern, Treasury and the IRS could consider allowing an election under which the PTI does not tier up but remains with the lower tier corporation. We do not recommend this, because it is unclear to what extent it would preclude authority challenges by successors following subsequent re-transfers of the shares and because the resulting electivity may be revenue reducing. In cases where the U.S. Shareholder is subject to lower rates of tax than the successor (or has no need to distribute the proceeds of sale and actually incur duplicative subpart F taxation) presumably the election will be made. On the other hand, given the availability of Section 338(g) elections to many purchasers it is not clear how significant this revenue concern really is.

However, if Treasury and the IRS do not feel it is appropriate to allow PTI attributes to tier up in the case of a lower tier CFC disposition, we do not recommend piecemeal elimination of duplication of basis but not earnings. This makes little sense from a subpart F policy perspective and does not eliminate the risk of authority challenges. If Treasury and the IRS believe that Congress intended the subpart F regime to apply the constructs of subchapter C except where the statute is expressly to the contrary, then the subchapter C "chips" should be allowed to fall where they may.

We do not express a view as to whether it is appropriate to permit a U.S. Shareholder to share PTI attributes in shares it is deemed to own under Section 958(a), but actually owns indirectly, in a CFC with other directly owned shares in the CFC and vice versa.

We believe strong arguments can be made in favor of, and we generally support permitting sharing of PTI attributes within a consolidated group. However, to the extent this result is appropriate, it is not because it is appropriate based on subpart F policies considered in isolation, but rather because it is consistent with the "single entity" treatment of consolidated groups under the policies and approach adopted by the consolidated return regulations. Because "single entity" treatment does not apply to nonconsolidated affiliated U.S. corporations we do not support extending "sharing" to that context. Consideration should be given to including the sharing rules for PTI attributes of consolidated return members in a separate regulation under Section 1502 rather than including those rules in PTI regulations.

² REG-121509-00, 71 Fed. Reg. 51,155 (August 29, 2006), corrected 71 Fed. Reg. 71,116 (December 8, 2006). Except as otherwise noted, all "Section" and " § " references in this Report are to sections of the Internal Revenue Code of 1986, as amended (the "Code"), references to "Regulations" or "Reg." are to the Treasury Regulations promulgated thereunder, and references to "Prop. Reg." are to proposed Treasury Regulations.

³ For this purpose, "United States shareholders" are U.S. persons owning 10% or more of the voting stock of the CFC ("U.S. Shareholders"), and a CFC is any foreign corporation in which U.S. Shareholders own more than 50% of the stock by vote or value. §§ 951(b), 957(a).

⁴ § 951(a).

⁵ The rest of the Report refers to basis that arose under Section 961(a) as "961(a) Basis" and basis that arose under Section 961(c) as "961(c) Basis" and collectively to "961 Basis." It generally refers to "regular" basis as "SubC Basis" to distinguish it from 961 Basis.

⁶ Instead of a foreign person, F1 could also be a group of unrelated US persons who each own less than 10% of U-CFC.

⁷ There may be circumstances in which, for reasons of administrative simplicity, it may be reasonable for the rules to depart from the "pure" form of such a regime. For example, under Prop. Treas. Reg. § 1.959-3(e)(4), if a distribution to a taxpayer exceeds its own PTI account in the shares, it may be treated as a dividend only with respect to the non-PTI E&P of the corporation. In theory, this means a U.S. taxpayer could invest directly in shares of a CFC after it has earned subpart F income that was taxed to the original U.S. shareholder and receive a distribution of cash out of PTI E&P that is not a dividend (assuming no non-PTI E&P). The rule preventing treatment as a dividend is necessary to avoid depleting the corporate level PTI attributes that "belong" to the original shareholder and have the incidental effect of benefitting the new shareholder. However, developing rules that treat the distribution as a dividend, but then permit the corporate level PTI attributes to be replenished with future E&P and to reallocate foreign taxes on the PTI credited by the original shareholder may reasonably have been considered too complex to be worth the effort. We do not, however, view such departures from an intellectually pure form of the regime for reasons of administrative simplicity in narrow circumstances as inconsistent with the general approach described above.

⁸ § 959(a).

⁹ Prop. Treas. Reg. § 1.959-1(b)(5).

¹⁰See H.R. Rep. No. 87-1447, at 1332-1333 (1962).

¹¹See Sen. Rep. No. 87-1881, at 2439 (1962).

¹²See Sen. Rep. No. 87-1881, at 2452 (1962); H.R. Rep. No. 87-1447, at 1200 & 1336 (1962).

¹³See H.R. Rep. No.105-220, at 620-621 (1997) (Conf. Rep.).

¹⁴Id.

¹⁵ Importantly, for purposes of the discussion below, Section 1248 does not treat the constructive distribution as a distribution of all of the E&P attributable to the shares to the extent of the gain, but with amounts attributable to PTI then excluded as a PTI distribution. Instead, it defines the distributed E&P to exclude PTI and treats the entire amount as a dividend.

¹⁶ § 1248(a).

¹⁷ Treas. Reg. §§ 1.1248-2 and -3.

¹⁸ Treas. Reg. § 1.1248-2(b).

¹⁹See Treas. Reg. § 1.951-1(e).

²⁰ § 964(e)(2).

²¹See H.R. Rep. No.105-220, at 619-620 (1997) (Conf. Rep.).

²² 71 Fed. Reg. 51,115 (Aug. 29, 2006).

²³ Prop. Treas. Reg. § 1.959-3(b)(1) & (2).

²⁴ Prop. Treas. Reg. § 1.959-3(b)(2).

²⁵ Prop. Treas. Reg. §§ 1.959-1(c)(1), 1.959-4.

²⁶ Prop. Treas. Reg. § 1.959-3(e)(4).

²⁷ Prop. Treas. Reg. § 1.959-2(a)(1).

²⁸See Prop. Treas. Reg. § 1.959-2(a)(2)(Ex. 1). As discussed below, U-CFC's 961(c) Basis in L-CFC will be correspondingly reduced by $100.

²⁹ Prop. Treas. Reg. § 1.959-1(b)(12).

³⁰ Prop. Treas. Reg. § 1.959-3(e)(2)(iv).

³¹ Prop. Treas. Reg. § 1.959-1(d)(1).

³² Prop. Treas. Reg. § 1.959-3(f).

³³See Prop. Treas. Reg. § 1.959-3(f)(3)(Ex. 1). Corresponding adjustments to 961(a) Basis are made pursuant to Prop. Treas. Reg. § 1.961-1(b) and -2(a).

³⁴ Prop. Treas. Reg. § 1.959-3(f)(2).

³⁵ Prop. Treas. Reg. § 1.959-3(g).

³⁶See Prop. Treas. Reg. § 1.959-3(g)(4)(Ex. 1).

³⁷ Prop. Treas. Reg. § 1.959-3(g)(2).

³⁸ Prop. Treas. Reg. §§ 1.1502-32(b)(3)(ii)(D), -32(b)(3)(iii)(B).

³⁹ Prop. Treas. Reg. § 1.959-1(b)(5).

⁴⁰ Prop. Treas. Reg. § 1.959-3(h)(2).

⁴¹ Prop. Treas. Reg. § 1.959-3(h)(3).

⁴² Prop. Treas. Reg. § 1.959-3(h)(3)(ii). By cross-referring to the regulations under section 302, the Proposed Regulations suggest that, in a case in which the redeemed shareholder has no remaining shares but the redemption is treated as a dividend due to attribution, the remaining PTI would transfer to the CFC shares owned (under Section 958(a)) by the related person(s) whose ownership of CFC shares caused the redeemed shareholder to be treated as receiving a dividend distribution.

⁴³ Prop. Treas. Reg. § 1.959-3(h)(4).

⁴⁴See Prop. Treas. Reg. § 1.959-3(h)(4)(ii).

⁴⁵ Prop. Treas. Reg. § 1.959-3(e)(2). See generally Doernberg, Koenenn, Lowry and Teigen, "Ordering Rules Make Your Head Spin? Here's Some Aspirin," 2001 TNT 151-71 (Aug. 6, 2001).

⁴⁶ We assume, although the Proposed Regulations do not specify, that this is intended to mean the CFC's U.S. taxable year rather than any foreign taxable year.

⁴⁷ Prop. Treas. Reg. § 1.959-3(c).

⁴⁸ Prop. Treas. Reg. § 1.959-3(d).

⁴⁹ Prop. Treas. Reg. § 1.959-3(e)(5).

⁵⁰ Treas. Reg. § 1.367(b)-7(b)(2).

⁵¹See Treas. Reg. § 1.312-11.

⁵² Prop. Treas. Reg. § 1.961-1(b)&(c).

⁵³ Prop. Treas. Reg. § 1.961-2(a)&(b).

⁵⁴ Prop. Treas. Reg. § 1.961-3(a)(1).

⁵⁵ Prop. Treas. Reg. § 1.961-3(b)(1).

⁵⁶See Prop. Treas. Reg. § 1.961-3(b)(2).

⁵⁷ Prop. Treas. Reg. § 1.961-2(a)(1).

⁵⁸ Prop. Treas. Reg. § 1.959-1(b)(5).

⁵⁹ Treas. Reg. § 1.312-11(a), 1.381(a)-1(b)(2)(i).

⁶⁰ Treas. Reg. § 1.312-10.

⁶¹ Even in that case it is not identical. While the distribution and recontribution eliminates E&P and effectively converts it to basis at each CFC tier, if a lower tier CFC subsequently earned income, a distribution would be deemed to come first out of that E&P, not out of the previously distributed and recontributed amounts.

⁶² If USS is willing to entirely terminate its 70% interest in CFC for $100, presumably the CFC stock in aggregate is worth $142.85. Of this amount, $121.50 corresponds to the realized E&P and $21.35 represents unrealized appreciation in CFC's assets.

⁶³See Prop. Treas. Reg. § 1.959-3(h)(2). See also 1984 Conference Report. Compare Rev. Rul. 82-72, obsoleted by Rev. Rul. 95-71.

⁶⁴See Rev. Rul. 90-31, obsoleting Rev. Rul. 83-182, suspending Rev. Rul. 71-388.

⁶⁵ The example assumes Congress does not extend the look-through rule of Section 954(c)(6), with the result that this amount presumably would be taxable as a dividend. We note that the look-through rule was recently extended through December 31, 2014. H.R. 5771, 113^th Cong. § 135 (2014) (enacted).

⁶⁶ This corresponds to 70% of the unrealized appreciation in L-CFC assets.

⁶⁷ Section 312(n)(7) treats a redemption governed by Section 302(a) as reducing the distributing corporation E&P to the extent of the lesser of (1) the amount distributed and (2) the redeemed stock's "ratable share" of the E&P. In this case, L-CFC's E&P is reduced by 70% of the E&P, or by approximately $85.

⁶⁸ H.R. Conf. Rep. No. 105-220, July 31, 1997 at 620.

⁶⁹ As the "look-through" rule for inter-CFC dividends in Section 954(c)(6) was not adopted until 2005 (under the Tax Increase Prevention Reconciliation Act), this would have been the effect under the prevailing law unless the same-country exception applied.

⁷⁰See Prop. Reg § 1.959-3.

⁷¹See Prop. Reg § 1.961-1(c)(2).

⁷² Prop. Reg § 1.959-3(f)(1).

⁷³See 1.959-3(h)(4)(i) (which apparently allows a covered shareholder to treat itself as having PTI Accounts with respect to the share of the corporation from which it is deemed to receive a distribution out of the corporation's E&P PTI Subaccounts). It is not clear under the Proposed Regulations in this fact pattern how USP's 961(a) Basis and U-CFC's 961(c) Basis is adjusted, although presumably there should be an overall downward adjustment of $100.

⁷⁴ In fact, $20 of the E&P at L-CFC is non-PTI which would be taxed as a dividend to USS under Section 1248(a) upon a disposition of L-CFC.

⁷⁵ The same distribution in year 3 would result in a dividend to the extent of $50 under the nimble dividend rule, but in year 4 assume there is no current E&P and the C3 E&P Subaccount would be zero ($50 deficit offset by $50 of year 3 E&P)

⁷⁶ It is unclear whether this $50 basis reduction would reduce SubC Basis, 961(c) Basis or both pro rata, although USS would presumably prefer to reduce 961(c) Basis subject to the "Solely for Subpart F" limitation.

⁷⁷ For a discussion of some of these issues, see Bowers and Leyva, The Application of Sections 959 and 961 to Indirectly Owned CFCs, 34 Tax Mgt. Int'l J. 307 (2005).

⁷⁸See Chevron U.S.A., Inc. v. Natural Resources Defense Counsel, Inc., 467 U.S. 837 (1984).