AGENCY:

Internal Revenue Service (IRS), Treasury.

ACTION:

Final and temporary regulations.

SUMMARY:

This document contains final regulations that provide guidance to determine the amount of global intangible low-taxed income included in the gross income of certain United States shareholders of foreign corporations, including United States shareholders that are members of a consolidated group. This document also contains final regulations relating to the determination of a United States shareholder's pro rata share of a controlled foreign corporation's subpart F income included in the shareholder's gross income, as well as certain reporting requirements relating to inclusions of subpart F income and global intangible low-taxed income. Finally, this document contains final regulations relating to certain foreign tax credit provisions applicable to persons that directly or indirectly own stock in foreign corporations.

DATES:

Effective date: These regulations are effective on June 21, 2019.

Applicability dates: For dates of applicability, see §§ 1.78-1(c), 1.861-12(k), 1.951-1(i), 1.951A-7, 1.1502-51(g), 1.6038-2(m), and 1.6038-5(e).

FOR FURTHER INFORMATION CONTACT:

Concerning the regulations §§ 1.951-1, 1.951A-0 through 1.951A-7, 1.6038-2, and 1.6038-5, Jorge M. Oben at (202) 317-6934; concerning the regulations §§ 1.951A-1(e) and 1.951A-3(g), Jennifer N. Keeney at (202) 317-5045; concerning the regulations §§ 1.1502-12, 1.1502-32, and 1.1502-51, Katherine H. Zhang at (202) 317-6848 or Kevin M. Jacobs at (202) 317-5332; concerning the regulations §§ 1.78-1, 1.861-12, 1.861-12T, and 1.965-7, Karen J. Cate at (202) 317-6936 (not toll free numbers).

SUPPLEMENTARY INFORMATION:

Background

Section 951A was added to the Internal Revenue Code (the “Code”) ^[1] by the Tax Cuts and Jobs Act, Public Law 115-97, 131 Stat. 2054, 2208 (2017) (the “Act”), which was enacted on December 22, 2017. On October 10, 2018, the Department of the Treasury (“Treasury Department”) and the IRS published proposed regulations (REG-104390-18) under sections 951, 951A, 1502, and 6038 in the Federal Register (83 FR 51072) (the “proposed regulations”). A public hearing on the proposed regulations was held on February 13, 2019. The Treasury Department and the IRS also received written comments with respect to the proposed regulations.

In addition, on December 7, 2018, the Treasury Department and the IRS published proposed regulations (REG-105600-18) relating to foreign tax credits in the Federal Register (83 FR 63200) (“foreign tax credit proposed regulations”). A public hearing on these regulations was scheduled for March 14, 2019, but it was not held because there were no requests to speak. However, the Treasury Department and the IRS received written comments with respect to the foreign tax credit proposed regulations. Certain rules in the foreign tax credit proposed regulations are being finalized in this Treasury decision to ensure that the applicability dates of these rules coincide with the applicability dates of the statutory provisions to which they relate. See section 7805(b)(2). The rules being finalized relate to §§ 1.78-1, 1.861-12(c)(2), and 1.965-7(e). See part XI of the Summary of Comments and Explanation of Revisions section.

Comments outside the scope of this rulemaking are generally not addressed but may be considered in connection with future guidance projects. In this regard, the Treasury Department and the IRS expect that future guidance will address issues concerning the allocation and apportionment of expenses in order to determine a taxpayer's foreign tax credit limitation under section 904. All written comments received in response to the proposed regulations and the foreign tax credit proposed regulations are available at www.regulations.gov or upon request. Terms used but not defined in this preamble have the meaning provided in these final regulations.

Summary of Comments and Explanation of Revisions

I. Overview

The final regulations retain the basic approach and structure of the proposed regulations and foreign tax credit proposed regulations, with certain revisions. This Summary of Comments and Explanation of Revisions section discusses those revisions as well as comments received in response to the solicitation of comments in the notices of proposed rulemaking accompanying those regulations.

II. Comments and Revisions to Proposed § 1.951-1—Amounts Included in Gross Income of United States Shareholders

A. Hypothetical Distribution of Allocable E&P

A United States shareholder (“U.S. shareholder”) who owns stock of a foreign corporation on the last day of the foreign corporation's taxable year on which the foreign corporation is a controlled foreign corporation (“CFC”) includes in gross income its “pro rata share” of the CFC's subpart F income (as defined in section 952) for the taxable year. See section 951(a)(1) and § 1.951-1(a). In general, a U.S. shareholder's pro rata share of subpart F income is determined based on its proportionate share of a hypothetical distribution of all the current earnings and profits (“E&P” and “current E&P”) of the CFC. See section 951(a)(2)(A) and § 1.951-1(b)(1)(i) and (e)(1). A U.S. shareholder's pro rata share of tested income (as defined in section 951A(c)(2)(A) and § 1.951A-2(b)(1)), tested loss (as defined in section 951A(c)(2)(B)(i) and § 1.951A-2(b)(2)), qualified business asset investment (“QBAI”) (as defined in section 951A(d)(1) and § 1.951A-3(b)), tested interest expense (as defined in § 1.951A-4(b)(1)), and tested interest income (as defined in § 1.951A-4(b)(2)) (each a “tested item”) generally are also determined based on a hypothetical distribution of current E&P, with certain modifications to account for the differences between each tested item and subpart F income. See section 951A(e)(1) and § 1.951A-1(d).

For purposes of the hypothetical distribution, the proposed regulations define “current E&P” for a taxable year as the greater of (i) the E&P of the corporation for the taxable year determined under section 964, or (ii) the sum of the subpart F income (as determined under section 952, as increased under section 951A(c)(2)(B)(ii) and proposed § 1.951A-6(d)) and the tested income of the corporation for the taxable year. See proposed § 1.951-1(e)(1)(ii). One comment asserted that using the term “current earnings and profits” for this purpose is confusing because the definition differs significantly from the definition of “earnings and profits” provided in section 964(a), and therefore suggested using a different term for this purpose. In response to this comment, the final regulations replace the term “current earnings and profits” with “allocable earnings and profits” (“allocable E&P”).

B. Pro Rata Share Anti-Abuse Rule

The proposed regulations provide that any transaction or arrangement that is part of a plan a principal purpose of which is the avoidance of Federal income taxation, including, but not limited to, a transaction or arrangement to reduce a U.S. shareholder's pro rata share of the subpart F income of a CFC, which transaction or arrangement would otherwise avoid Federal income taxation, is disregarded in determining such U.S. shareholder's pro rata share of the subpart F income of the corporation (the “pro rata share anti-abuse rule”). See proposed § 1.951-1(e)(6). The pro rata share anti-abuse rule also applies in determining the pro rata share of each tested item of a CFC for purposes of determining a U.S. shareholder's global intangible low-taxed income (“GILTI”) inclusion amount under section 951A(a) and § 1.951A-1(b). See id.

Several comments suggested that the pro rata share anti-abuse rule is overbroad and could be interpreted to apply to nearly all transactions, arrangements, or tax elections that reduce the pro rata share amounts of a U.S. shareholder. In particular, comments noted that, under one interpretation of the rule, a U.S. shareholder that disposes of CFC stock could be required indefinitely to include its “pro rata share” of the CFC's subpart F income or tested items with respect to such stock. These comments recommended that the final regulations clarify the scope of the rule and, in particular, provide that the rule applies only to reallocate subpart F income and tested items of a CFC as of a hypothetical distribution date among persons that own, directly or indirectly, shares of the CFC on such date. According to these comments, the rule, as narrowed in this manner, could not apply to cause a U.S. person that disposes of stock of a CFC before a hypothetical distribution date to be treated as having a pro rata share of the CFC's subpart F income or tested items as of such date by reason of such stock.

The Treasury Department and the IRS agree that the scope of the pro rata share anti-abuse rule should be clarified. Accordingly, the final regulations clarify that the rule applies only to require appropriate adjustments to the allocation of allocable E&P that would be distributed in a hypothetical distribution with respect to any share outstanding as of the hypothetical distribution date. See § 1.951-1(e)(6). Thus, under the rule, if applicable, adjustments will be made solely to the allocation of allocable E&P in the hypothetical distribution between shareholders that own, directly or indirectly, stock of the CFC as of the relevant hypothetical distribution date. As clarified, the rule will not apply to adjust the allocable E&P allocated to a shareholder by reason of a transfer of CFC stock, except by reason of a change to the distribution rights with respect to stock in connection with such transfer (for example, an issuance of a new class of stock, including by recapitalization).

Other comments suggested that the final regulations limit the pro rata share anti-abuse rule to transactions or arrangements that lack economic substance or are artificial, or only to transactions or arrangements that result in non-economic allocations that shift subpart F income or tested items away from a U.S. shareholder. One comment suggested that the rule should apply only to enumerated transactions identified by the Treasury Department and the IRS as being abusive, and another comment suggested that the regulations should include examples illustrating transactions to which the pro rata share anti-abuse rule would or would not apply.

The Treasury Department and the IRS do not adopt these recommendations. Transactions that lack economic substance or are artificial would typically be disregarded under general tax principles, and non-economic allocations would generally be addressed through the facts and circumstances approach of § 1.951-1(e)(3) (as discussed in part II.C of this Summary of Comments and Explanation of Revisions section), such that limiting the pro rata share anti-abuse rule in the manner recommended could render it superfluous. Moreover, the concerns underlying the rule may arise in non-artificial transactions, or transactions with substance, that would be respected under general tax principles. In addition, attempting to specifically identify all the transactions covered by the rule or to specify such transactions by example would be impractical and inconsistent with one of the purposes underlying any anti-avoidance rule—that is, to deter the development and implementation of new transactions or arrangements intended to avoid the operative rule.

Another comment recommended an exception to the pro rata share anti-abuse rule for transactions entered into with unrelated parties and for transactions entered into with related parties located in the same country of tax residence as the relevant CFC. The comment also recommended a “small business” exception for U.S. shareholders with worldwide gross receipts under $25 million. The Treasury Department and the IRS have determined that the policy concerns underlying the rule can be implicated by transactions that involve unrelated parties, such as accommodation parties (for instance, a financial institution) that hold stock with certain distribution rights in order to reduce an unrelated U.S. shareholder's pro rata share of subpart F income or tested items. Further, these concerns can arise regardless of whether the parties involved are located in the same country of tax residence as the CFC. Finally, the Treasury Department and the IRS have concluded that the level of gross receipts of the shareholders is not relevant to, and therefore does not justify, an exception to the rule. Any administrative burden on small businesses would not stem from the rule itself but rather from engaging in a transaction a principal purpose of which is to avoid Federal income taxation. Accordingly, these recommendations are not adopted.

C. Facts and Circumstances Approach

Section 1.951-1(e)(3)(ii) of the existing regulations provides special rules applicable to CFCs with two or more classes of stock with discretionary distribution rights. Under these rules, the allocation of current E&P is primarily based on the relative fair market value of the stock with discretionary distribution rights. The preamble to the proposed regulations notes that this fair market value allocation method had been the basis of certain attempted avoidance structures. Accordingly, the proposed regulations adopt a facts and circumstances approach in allocating current E&P in a hypothetical distribution between multiple classes of stock, including stock with discretionary distribution rights. See proposed § 1.951-1(e)(3). The proposed regulations provide that, where appropriate, the relative fair market value of the stock may still be taken into account, but as one of several factors, none of which is dispositive. See id.

A comment asserted that the facts and circumstances approach set forth in the proposed regulations is a vague and subjective standard that would create uncertainty, while the fair market value approach in the existing regulations for stock with discretionary distribution rights is a long-standing and objective standard. The comment further noted that the preamble to the 2005 Treasury decision that adopts the fair market value approach specifically rejects the facts and circumstances approach, stating that “the interests of sound tax policy and administration are served by requiring the value-based allocation.” TD 9222, 70 FR 49864 (August 25, 2005). The comment recommended that the fair market value approach be retained in the final regulations, in lieu of the proposed facts and circumstances approach, for purposes of determining the pro rata share of subpart F income and tested items.

The Treasury Department and the IRS have determined, based on experience administering the fair market value approach, that a facts and circumstances approach, in which the fair market value of stock is relevant but not determinative, would be a more reliable method for determining a U.S. shareholder's pro rata share of subpart F income (and tested items) than the fair market value approach. While fair market value is easily determinable for publicly traded stock, determining the fair market value of privately-held stock is more difficult and typically requires a determination of the stock's rights to distributions of current and accumulated E&P and capital, as well as the voting rights with respect to such stock. In contrast, under section 951(a)(2) and § 1.951-1(b)(1), a shareholder's pro rata share of subpart F income is determined based solely on a hypothetical distribution of subpart F income for the taxable year. Furthermore, the amount of subpart F income treated as distributed in the hypothetical distribution is determined under § 1.951-1(e) based on a distribution of allocable E&P. Thus, the most relevant attribute of any share of CFC stock for purposes of the hypothetical distribution is its economic rights with respect to the allocable E&P of the CFC, which is generally determined by reference to its current E&P. Generally, a share's voting rights, rights to distributions of E&P accumulated before the current year, and rights to capital, all of which are also taken into account in determining fair market value, are not relevant to the hypothetical distribution of allocable E&P, and therefore a fair market value approach can distort the determination required under section 951(a)(2) and § 1.951-1(b)(1). A more flexible facts and circumstances approach that considers fair market value as a factor can also take into account other factors related to the expected distributions of allocable E&P with respect to such stock, without taking into account capital liquidation rights and other factors that are not relevant to the distribution of allocable E&P. Accordingly, the final regulations do not adopt this recommendation.

D. Modifications to Example 4

The proposed regulations provide that no amount of current E&P is distributed in the hypothetical distribution with respect to a particular class of stock to the extent that a distribution of such amount would constitute a redemption of stock (even if the redemption would be treated as a dividend under sections 301 and 302(d)), a distribution in liquidation, or a return of capital. See proposed § 1.951-1(e)(4)(i). The proposed regulations include an example to illustrate the application of this rule. See proposed § 1.951-1(e)(7)(v) Example 4. A comment asserted that proposed § 1.951-1(e)(4)(i) and the example illustrating the rule are confusing because, given the definition of current E&P in the proposed regulations, the hypothetical distribution would typically not give rise to a return of capital (other than through a redemption).

This rule is not intended to refer to the consequences of the hypothetical distribution itself (for example, the extent to which it could give rise to a return of capital), but rather is intended to provide that terms of the stock or related agreements and arrangements that could give rise to redemptions, liquidations, or returns of capital if actually exercised (or otherwise taken into account) are not taken into account for purposes of the hypothetical distribution. The final regulations and the related example are clarified to reflect this intent. See § 1.951-1(e)(4)(i) and § 1.951-1(e)(7)(v) Example 4. Similarly, the final regulations clarify that the facts and circumstances taken into account in determining the distribution rights of a class of stock do not include actual distributions (or any amount treated as a dividend) made during the taxable year that includes the hypothetical distribution date. See § 1.951-1(e)(3). Such distributions (or dividends) are not relevant in determining a class of stock's economic rights and interest in the allocable E&P (which are not reduced by actual distributions during the taxable year) as of the hypothetical distribution date.

E. Application of Section 951(a)(2)(B) to Subpart F Income and Tested Income in the Same Taxable Year

Under section 951(a)(2)(B), a U.S. shareholder's pro rata share of subpart F income with respect to stock for a taxable year (as determined under section 951(a)(2)(A)) is reduced by the amount of distributions received by any other person during the year as a dividend with respect to the stock, subject to a limitation based on the period of the taxable year in which the shareholder owned the stock within the meaning of section 958(a). Section 951A(e)(1) provides that the pro rata share of tested income, tested loss, and QBAI is determined under the rules of section 951(a)(2) in the same manner as such section applies to subpart F income. Accordingly, the proposed regulations provide that a U.S. shareholder's pro rata share of tested income is determined under section 951(a)(2) and § 1.951-1(b) and (e), generally substituting “tested income” for “subpart F income” each place it appears. See proposed § 1.951A-1(d)(2).

Because section 951(a)(2)(B) applies for purposes of determining the pro rata share of both subpart F income and tested income, the proposed regulations could be interpreted as permitting a dollar-for-dollar reduction under section 951(a)(2)(B) in both a U.S. shareholder's pro rata share of subpart F income and its pro rata share of tested income. The Treasury Department and the IRS have determined that this would be an inappropriate double benefit that is not contemplated under section 951(a)(2)(B) and section 951A(e)(1). Accordingly, the regulations under section 951(a)(2)(B) are revised to clarify that a dividend received during the taxable year by a person other than the U.S. shareholder reduces the U.S. shareholder's pro rata share of subpart F income and its pro rata share of tested income in the same proportion as its pro rata share of each amount bears to its aggregate pro rata share of both amounts. See § 1.951-1(b)(1)(ii).

The examples in § 1.951-1(b)(2) are modified solely to illustrate the application of the revised rule in § 1.951-1(b)(1) and to conform to the terminology in the final regulations. The Treasury Department and the IRS are studying the application of section 951(a)(2)(A) and (B) in certain cases that may lead to inappropriate results, for example, due to the concurrent application of the provisions. In addition, the Treasury Department and the IRS are studying the application of section 951(a)(2)(B) with respect to dividends paid to foreign persons, dividends that give rise to a deduction under section 245A(a), and dividends paid on stock after the disposition of such stock by a U.S. shareholder. Comments are requested in this regard.

F. Revisions to Cumulative Preferred Stock Rule

The proposed regulations provide a special rule applicable to preferred shares with accrued but unpaid dividends that do not compound annually at or above the applicable Federal rate (“AFR”) under section 1274(d)(1) (“cumulative preferred stock rule”). See proposed § 1.951-1(e)(4)(ii). If the cumulative preferred stock rule applies with respect to stock, the current E&P allocable to the stock may not exceed the amount of dividends actually paid during the taxable year with respect to the stock plus the present value of the unpaid current dividends with respect to the stock determined by using the AFR that applies on the date the stock is issued for the term from such issue date to the mandatory redemption date and assuming the dividends will be paid at the mandatory redemption date. See id.

A comment stated that it is unclear whether the applicability of the cumulative preferred stock rule is determined based on the AFR as of the issuance date or, alternatively, the AFR for the current year. The comment suggested that, because the amount of the preferred dividend determined under the cumulative preferred stock rule is based on the AFR as of the issue date, for consistency, the applicability of the rule should be determined by reference to the AFR as of the issue date as well. The Treasury Department and the IRS agree with this comment, and the final regulations are revised accordingly. See § 1.951-1(e)(4)(ii).

The proposed regulations provide that the amount of any arrearage on cumulative preferred stock is determined taking into account the time value of money principles in the cumulative preferred stock rule. See proposed § 1.951-1(e)(4)(iii). A comment recommended that the rule be clarified to reference the calculation of the present value of the unpaid current dividends described in the cumulative preferred stock rule. The Treasury Department and the IRS agree with this comment, and the final regulations are revised accordingly. See § 1.951-1(e)(4)(iii).

The proposed regulations contain a special rule for purposes of sections 951 through 964 to treat a controlled domestic partnership as a foreign partnership to determine stock ownership in a CFC by a U.S. person for purposes of section 958(a) if certain conditions are met. See proposed § 1.951-1(h). A comment suggested that because the proposed regulations define a “controlled domestic partnership” by reference to a specific U.S. shareholder, the rule could be read to apply only with respect to that shareholder but not with respect to other partners of the controlled domestic partnership, for which the partnership would therefore still be treated as domestic. The comment requested that the final regulations clarify that the treatment as a foreign partnership is with respect to all partners of the partnership. The rule, if applicable, is intended to treat a domestic partnership as a foreign partnership with respect to all its partners. The final regulations revise the definition of controlled domestic partnership to clarify the scope of the rule. See § 1.951-1(h)(2); see also § 1.965-1(e)(2). A change is also made to § 1.951-1(h) to conform to the change in the final regulations to the treatment of domestic partnerships for purposes of section 951A. See part VII.C of this Summary of Comments and Explanation of Revisions section.

Finally, certain regulations have been revised to reflect the repeal of section 954(f) (regarding foreign base company shipping income) and section 955 (regarding foreign investments in less developed countries). See Public Law 108-357, 415(a)(2) (2004) and Public Law 115-97, 14212(a) (2017). The Treasury Department and the IRS intend to revise other regulations to reflect the repeal of these provisions in future guidance projects.

III. Comments and Revisions to Proposed § 1.951A-1—General Provisions

A. CFC Inclusion Date

The proposed regulations provide that, for purposes of determining the GILTI inclusion amount of a U.S. shareholder for a U.S. shareholder inclusion year, the U.S. shareholder takes into account its pro rata share of a tested item with respect to a CFC for the U.S. shareholder inclusion year that includes a CFC inclusion date with respect to the CFC. See proposed § 1.951A-1(d)(1). Under the proposed regulations, the term “U.S. shareholder inclusion year” means a taxable year of a U.S. shareholder that includes a CFC inclusion date of a CFC of the U.S. shareholder, the term “CFC inclusion date” means the last day of a CFC inclusion year on which a foreign corporation is a CFC, and the term “CFC inclusion year” means any taxable year of a foreign corporation beginning after December 31, 2017, at any time during which the corporation is a CFC. See proposed § 1.951A-1(e)(1), (2) and (4).

Several comments noted that, under certain circumstances, the requirement that a U.S. shareholder take into account its pro rata share of a CFC's tested items for a U.S. shareholder inclusion year that includes a CFC inclusion date could have the effect of requiring a U.S. shareholder to take into account its pro rata share of the CFC's tested items for a U.S. shareholder inclusion year that does not include the last day of the CFC inclusion year. This could happen, for instance, if a U.S. person with a taxable year ending December 31, 2019, sells a wholly-owned foreign corporation with a taxable year ending November 30, 2020, to a foreign person on December 1, 2019 and, as a result of the sale, the foreign corporation ceases to be a CFC; in that case, under the proposed regulations, the CFC inclusion date with respect to the foreign corporation would be December 1, 2019, whereas the CFC inclusion year of the foreign corporation would not end until November 30, 2020. The comments raised several concerns, in particular, that the U.S. person in this example would be unable to determine its pro rata share of any tested item of the foreign corporation as of December 31, 2019, since the foreign corporation's tested items could not be determined until November 30, 2020. The comments also noted that the proposed regulations' definition of CFC inclusion date was inconsistent with section 951A(e)(1), which provides that the pro rata share of certain amounts is taken into account in the taxable year of the U.S. shareholder in which or with which the taxable year of the CFC ends. The comments recommended that the relevant definitions be revised to accord with section 951A(e)(1).

The Treasury Department and the IRS agree with these comments. Accordingly, the final regulations provide that a U.S. shareholder takes into account its pro rata share of a tested item of a CFC in the U.S. shareholder inclusion year that includes the last day of the CFC inclusion year. See § 1.951A-1(d)(1). However, consistent with sections 951(a)(2) and 951A(e)(1), a U.S. shareholder's pro rata share of each tested item of a CFC is still determined based on the section 958(a) stock owned by the shareholder on the last day of the CFC's taxable year on which it is a CFC (the “hypothetical distribution date”). See §§ 1.951-1(e)(1)(i) and 1.951A-1(f)(3). The term “hypothetical distribution date” in the final regulations has the same meaning as the term “CFC inclusion date” in the proposed regulations.

B. Pro Rata Share of Certain Tested Items

1. Pro Rata Share of QBAI

The proposed regulations provide that, in general, a U.S. shareholder's pro rata share of the QBAI of a tested income CFC is proportionate to the U.S. shareholder's pro rata share of the tested income of the tested income CFC for the CFC inclusion year. See proposed § 1.951A-1(d)(3)(i). However, the proposed regulations provide that, to the extent the amount of a tested income CFC's QBAI is greater than ten times its tested income for the year (that is, the point at which the shareholder's deemed tangible income return (“DTIR”) attributable to the QBAI would fully offset its pro rata share of the tested income CFC's tested income), the excess QBAI is allocated solely to common shares (and not to preferred shares) (the “excess QBAI rule”). See proposed § 1.951A-1(d)(3)(ii). The excess QBAI rule is intended to ensure that a shareholder cannot obtain an increase in its DTIR by reason of preferred stock that exceeds the increase in its aggregate pro rata share of tested income from the ownership of the stock. Without the excess QBAI rule, U.S. persons would be incentivized to acquire debt-like preferred stock of CFCs that have significant amounts of QBAI and minimal tested income in order to effectively exempt some or all of the U.S. person's pro rata shares of tested income from other CFCs from taxation under section 951A. The preamble to the proposed regulations requested comments on the approach in the proposed regulations, including the excess QBAI rule, for determining a U.S. shareholder's pro rata share of a CFC's QBAI.

The only comment received with respect to the QBAI allocation approach in the proposed regulations agreed that it was appropriate to limit the allocation of QBAI to a preferred shareholder, because the debt-like claim that a preferred shareholder has on a CFC should not entitle it to an amount of QBAI that could be used to effectively exempt tested income of the shareholder's other CFCs. The comment noted that, in cases where a CFC has minimal tested income and substantial QBAI, the approach in the proposed regulations could result in a common shareholder receiving a pro rata share of QBAI that is disproportionate to its pro rata share of tested income, but acknowledged that this effect would be reversed in future years when the CFC generates more tested income.

The Treasury Department and the IRS agree with the comment that the approach in the proposed regulations achieves the correct result over a multi-year period. Accordingly, the final regulations generally adopt the QBAI allocation rule of the proposed regulations, with certain modifications to the excess QBAI rule to better effectuate the purposes of the rule. Specifically, the final regulations provide that, in the case of a tested income CFC with tested income that is less than ten percent of its QBAI (the tested income CFC's “hypothetical tangible return”), a shareholder's pro rata share of QBAI is determined based on the shareholder's pro rata share of this hypothetical tangible return. See § 1.951A-1(d)(3)(ii)(A) and (C). A U.S. shareholder's pro rata share of the hypothetical tangible return is determined under the rules for determining the shareholder's pro rata share of tested income, for this purpose treating the hypothetical tangible return as tested income. See § 1.951A-1(d)(3)(ii)(B). In most cases, the excess QBAI rule in the final regulations will produce the same results as the excess QBAI rule in the proposed regulations. However, unlike the excess QBAI rule in the proposed regulations, the application of the excess QBAI rule in the final regulations is not limited to preferred stock.^[2] Further, with respect to common stock, by untethering the allocation of excess QBAI from the allocation of tested income, and instead applying a hypothetical distribution model to the excess QBAI, the rule ensures that the reduction under section 951(a)(2)(B) and § 1.951A-1(b)(1)(ii) to a U.S. shareholder's pro rata share of tested income does not result in an excessive reduction to the U.S. shareholder's pro rata share of QBAI. See § 1.951A-1(d)(3)(iii)(C) Example 3.

One comment recommended that the final regulations allocate QBAI to convertible preferred stock or participating preferred stock by bifurcating the stock into preferred stock (to the extent of the dividend and liquidation preference) and common stock (to the extent that the participation right is “in the money”), and then allocating QBAI to each component separately. This issue has been mooted because the determination of a U.S. shareholder's pro rata share of QBAI no longer depends on whether the stock owned by the shareholder is common or preferred. Accordingly, the final regulations do not adopt this recommendation.

Finally, for the avoidance of doubt, the final regulations clarify that the aggregate amount of any tested item (including QBAI) of a CFC for a CFC inclusion year allocated to the CFC's stock cannot exceed the amount of such tested item of the CFC for the CFC inclusion year. See § 1.951A-1(d)(1).

2. Pro Rata Share of Tested Loss

The proposed regulations provide that a CFC's tested loss is allocated based on a hypothetical distribution of an amount of current E&P equal to the amount of tested loss, except that, in general, tested loss is allocated only to common stock. See proposed § 1.951A-1(d)(4)(i)(C). The general rule that tested loss is allocated only to common stock is subject to two exceptions. First, the proposed regulations allocate tested loss to preferred shares to the extent the tested loss reduces the E&P accumulated since the issuance of those preferred shares to an amount below the amount necessary to satisfy any accrued but unpaid dividends with respect to such preferred shares. See proposed § 1.951A-1(d)(4)(ii). Second, when the common stock has no liquidation value, the proposed regulations allocate tested loss to classes of preferred stock with liquidation value in reverse order of priority. See proposed § 1.951A-1(d)(4)(iii). These two exceptions result in tested loss allocations corresponding to changes in the economic value of the CFC stock. The preamble to the proposed regulations requested comments on the proposed approach for determining a U.S. shareholder's pro rata share of a CFC's tested loss, including how (or whether) to allocate tested loss of a CFC when no class of CFC stock has positive liquidation value.

Comments were supportive of the approach taken in the proposed regulations to determine pro rata shares of tested loss because the approach avoids complexity, minimizes the potential for abusive allocations of tested loss, and is consistent with the economic reality that common stock generally bears the risk of loss before preferred stock. One comment that was supportive of the approach in the proposed regulations suggested a possible alternative approach of allocating tested loss to preferred shares to the extent the preferred shares were allocated subpart F income. However, the comment noted that the approach of the proposed regulations is simpler and that the suggested approach would require additional rules to ensure that corresponding allocations of tested income were made in future periods to the preferred shares to reflect an actual payment of a dividend to the preferred shares. The Treasury Department and the IRS agree with the comment that the approach for allocating tested loss in the proposed regulations is simpler and that the suggested approach would require adjustments to the pro rata share rules for tested income as well, resulting in more complex tracking of previous year pro rata allocations for CFCs and their shareholders to determine current year allocations. Accordingly, the suggestion is not adopted.

One comment recommended that if no class of stock has liquidation value, the tested loss should be allocated first to any shareholders that hold guaranteed debt of the CFC, and then to the most senior class of common stock, unless another class of stock will in fact bear the economic loss. The Treasury Department and the IRS have determined, based on experience with pro rata share rules in the subpart F context, that the facts and circumstances approach provides a flexible and appropriate allocation of tested loss, including in cases where no class of stock has liquidation value. Therefore, this comment is not adopted.

IV. Comments and Revisions to Proposed § 1.951A-2—Tested Income and Tested Loss

A. Determination of Gross Income and Allowable Deductions

For purposes of determining tested income or tested loss, gross tested income is reduced by deductions (including taxes) properly allocable to the gross tested income (or which would be properly allocable to gross tested income if there were such gross income) under rules similar to the rules of section 954(b)(5). See section 951A(c)(2)(A)(ii). The proposed regulations provide that, for purposes of determining tested income and tested loss, the gross income and allowable deductions of a CFC for a CFC inclusion year are determined under the rules of § 1.952-2 for determining the subpart F income of a CFC. See proposed § 1.951A-2(c)(2). Section 1.952-2 provides rules for determining gross income and taxable income of a foreign corporation. For this purpose, and subject to certain exceptions, these rules generally treat foreign corporations as domestic corporations. See § 1.952-2(a)(1) and (b)(1).

The preamble to the proposed regulations requested comments on the application of § 1.952-2 for purposes of determining subpart F income, tested income, and tested loss, including whether other approaches for determining tested income and tested loss, or whether additional modifications to § 1.952-2 for purposes of calculating tested income and tested loss, would be appropriate. Several comments were received in response to this request. The comments generally supported applying § 1.952-2 for purposes of determining tested income. However, a number of comments requested modifications to, or clarifications regarding, the application of § 1.952-2. Some comments suggested that § 1.952-2 be revised for purposes of determining tested income and tested loss to allow the use of net operating loss carryforwards under section 172 and net capital losses subject to limits under section 1212. Another comment requested that the Treasury Department and the IRS provide a list of specific deductions allowed to a CFC that would be disallowed to a domestic corporation, such as under section 162(m) or 280G. The same comment requested clarification that carryforwards of a CFC's disallowed interest deduction under section 163(j)(2) are not subject to any limitation or restrictions. Several comments suggested that section 245A should apply to determine a CFC's subpart F income and tested income and tested loss under § 1.952-2. There is also a concern that § 1.952-2 could be interpreted so expansively as to entitle a CFC to a deduction expressly limited to domestic corporations, such as a deduction under section 250.

The Treasury Department and the IRS intend to address issues related to the application of § 1.952-2, taking into account these comments, in connection with a future guidance project. This guidance is expected to clarify that, in general, any provision that is expressly limited in its application to domestic corporations, such as section 250, does not apply to CFCs by reason of § 1.952-2. The Treasury Department and the IRS continue to study whether, and to what extent, section 245A should apply to dividends received by a CFC and welcome comments on this subject.

Section 1.952-2(b)(2) provides that the taxable income of a CFC engaged in the business of reinsuring or issuing insurance or annuity contracts and which, if it were a domestic corporation engaged in such business, would be taxable as a life insurance company to which subchapter L applies, is generally determined by treating such corporation as a domestic corporation taxable under subchapter L and by applying the principles of §§ 1.953-4 and 1.953-5 for determining taxable income. These regulations, which were promulgated in 1964, have not been updated to reflect current sections 953(a), 953(b)(3), and 954(i). A comment requested that the final regulations confirm that the rules of current sections 953 and 954(i) apply in determining the tested income or tested loss of a CFC described in § 1.952-2(b)(2). The Treasury Department and the IRS agree that the tested income or tested loss of a CFC described in § 1.952-2(b)(2) is calculated in the same manner as its insurance income under sections 953 and 954(i), and the rule is revised accordingly. See § 1.951A-2(c)(2)(i). However, no inference is intended that a CFC may determine reserve amounts based on foreign statement reserves in the absence of a ruling request. See section 954(i)(4)(B)(ii). In this regard, the Treasury Department and the IRS intend to address, in separate guidance, the use of foreign statement reserves for purposes of measuring qualified insurance income under section 954(i).

B. Gross Income Excluded by Reason of Section 954(b)(4)

Section 951A(c)(2)(A)(i)(III) provides that gross tested income does not include any item of gross income excluded from foreign base company income (as defined in section 954) (“FBCI”) or insurance income (as defined in section 953) “by reason of section 954(b)(4)” (the “GILTI high tax exclusion”). The proposed regulations clarify that the GILTI high tax exclusion applies only to items of gross income that are excluded from FBCI or insurance income solely by reason of an election under section 954(b)(4) and § 1.954-1(d)(5). See proposed § 1.951A-2(c)(1)(iii). Thus, this exclusion does not apply to any item of gross income excluded from FBCI or insurance income by reason of an exception other than section 954(b)(4), regardless of the effective rate of foreign tax to which such item is subject.

One comment noted that this clarification is consistent with the language of the GILTI high tax exclusion, which is limited by its terms to income subject to the high tax exception of section 954(b)(4). Several comments, however, requested that the final regulations expand the GILTI high tax exclusion to exclude additional categories of high-taxed income. These comments asserted, based on the legislative history of the Act, that Congress intended that income of a CFC would be subject to tax under the GILTI regime only if it is subject to a low rate of foreign tax. Some of these comments suggested that the exclusion be expanded to apply to high-taxed income that would be FBCI or insurance income but for the application of one or more exceptions in section 954(c), (h), or (i). Others recommended that the final regulations apply the GILTI high tax exclusion to any item of gross income subject to a sufficiently high effective foreign tax rate, regardless of whether such income would be FBCI or insurance income but for an exception. Comments suggested that the Treasury Department and the IRS could exercise their authority under section 951A(f)(1)(B) to treat a GILTI inclusion as a subpart F inclusion that could potentially be excludible, on an elective basis, from FBCI (or insurance income) under section 954(b)(4).

Comments recommending an expansion of the GILTI high tax exclusion to any item of high-taxed income suggested various methods to determine the appropriate foreign tax rate for this purpose. One comment recommended the same threshold as used for the high tax exception for subpart F income under section 954(b)(4)—that is, a rate that is 90 percent of the maximum rate specified in section 11 (21 percent), or 18.9 percent. Another comment recommended a 13.125 percent rate, citing the conference report accompanying the Act that indicated that, in general, no residual U.S. tax would be owed on GILTI subject to a foreign tax rate greater than or equal to that rate. H.R. Rep. No. 115-466, at 627 (2017) (Conf. Rep.) (“Conference Report”).

Other comments suggested that even if the GILTI high tax exclusion is not expanded to take into account all high-taxed income, taxpayers should be permitted to elect to treat income that would otherwise be gross tested income as subpart F income in order to qualify for the exception under section 954(b)(4), for example, through a rebuttable presumption that all income (or alternatively, all high-taxed income) of a CFC is subpart F income. One comment asserted that such a rule would be consistent with taxpayers' historical ability to elect through the choice of transactional or operational structure to subject their CFC income to current taxation under subpart F. For example, the comment stated that a taxpayer could cause a CFC to make a loan to its U.S. shareholder, resulting in an inclusion under section 956, or could intentionally structure its operations in a manner that causes income to be characterized as FBCI. The comment also asserted that a rule that effectively permits a taxpayer to elect into subpart F income is consistent with the regulations under section 954, which permit an election to be made with respect to high-taxed income under section 954(b)(4) notwithstanding that that provision, similar to section 954(a) itself, is expressed as a mandatory rule. See § 1.954-1(d).

The final regulations do not adopt these comments. The Treasury Department and the IRS have declined to exercise regulatory authority under section 951A(f)(1)(B) because that authority relates to the treatment of a GILTI inclusion amount, rather than an item of gross tested income. A GILTI inclusion amount is determined based on a U.S. shareholder's pro rata share of all the tested items of one or more CFCs and, as a result, the determination of the extent to which foreign tax is imposed on any single item of net income for purposes of section 954(b)(4) cannot be made by reference to a GILTI inclusion amount. The final regulations also do not permit taxpayers to elect to treat income that would otherwise be gross tested income as subpart F income in order to qualify for the exception under section 954(b)(4). Unlike section 954(b)(4), nothing in section 954(a) or the legislative history suggests that taxpayers should be permitted to treat income that is not described in section 954(a), such as gross tested income, as FBCI through a rebuttable presumption or otherwise. In addition, this type of rebuttable presumption could give rise to significant administrability concerns. These concerns are discussed further in a notice of proposed rulemaking published in the same issue of the Federal Register addressing an election under section 954(b)(4) with respect to income that would otherwise qualify as tested income.

The Treasury Department and the IRS continue to believe that the GILTI high tax exclusion, as articulated in the proposed regulations, reflects a reasonable interpretation of section 951A(c)(2)(A)(i)(III) and section 954(b)(4), for the reasons stated in the notice of proposed rulemaking accompanying the proposed regulations. Accordingly, the final regulations retain the GILTI high tax exclusion without modification. See § 1.951A-2(c)(1)(iii). However, the Treasury Department and the IRS are studying, in light of the addition of section 951A by the Act, the appropriate circumstances under which taxpayers should be permitted to make an election under section 954(b)(4), with respect to income that would not be FBCI or insurance income, to exclude such income from gross tested income under the GILTI high tax exclusion using authority other than section 951A(f)(1)(B). In that regard, existing § 1.954-1(d)(1) does not provide the necessary framework for applying the exception under section 954(b)(4) to income that would be gross tested income, such as rules to determine the scope of an item of gross tested income to which the election applies and rules to determine the rate of foreign tax on such items. Therefore, the Treasury Department and the IRS are issuing a notice of proposed rulemaking published in the same issue of the Federal Register as these final regulations that will propose a framework under which taxpayers would be permitted to make an election under section 954(b)(4) with respect to income that would otherwise be gross tested income in order to exclude that income from gross tested income by reason of the GILTI high tax exclusion. However, until the regulations described in the separate notice of proposed rulemaking are effective, a taxpayer may not exclude any item of income from gross tested income under section 951A(c)(2)(A)(i)(III) unless the income would be FBCI or insurance income but for the application of section 954(b)(4) and § 1.954-1(d).

C. Gross Income Taken Into Account in Determining Subpart F Income

1. In General

Section 951A(c)(2)(A)(i)(II) provides that gross tested income is determined without regard to any gross income taken into account in determining the subpart F income of the corporation (the “subpart F exclusion”). Section 952(a) defines “subpart F income” as the sum of certain categories of income, including FBCI and insurance income.

Other than with respect to the coordination between the subpart F exclusion and section 952(c) (discussed in part IV.C.2 of this Summary of Comments and Explanation of Revisions section), the proposed regulations do not provide guidance on income that is “taken into account in determining the subpart F income” of a CFC within the meaning of the subpart F exclusion. In this regard, the final regulations provide rules for determining gross income included in FBCI and insurance company for purposes of the subpart F exclusion, including by reason of the application of the de minimis and full inclusion rules in section 954(b). See § 1.951A-2(c)(4)(ii)(A), (B), and § 1.951A-2(c)(4)(iii)(C); see also part IV.C.3 of this Summary of Comments and Explanation of Revisions section. The final regulations also clarify the circumstances in which the subpart F exclusion applies to less common items included in subpart F income under section 952(a)(3) through (5) (subpart F income resulting from participation in or cooperation with certain international boycotts, payments of illegal bribes, kickbacks, or other payments, or income derived from any country during which section 901(j) applies to that country). See § 1.951A-2(c)(4)(ii)(C) through (E).

2. Coordination With Section 952(c)

a. In General

The amount of subpart F income for a taxable year is subject to the E&P limitation and recapture provisions in section 952(c). Section 952(c)(1)(A) provides that a CFC's subpart F income for any taxable year cannot exceed its E&P for that year. See also § 1.952-1(c)(1). However, section 952(c)(2) provides that, to the extent subpart F income is reduced by reason of the E&P limitation in any taxable year, any excess of the E&P of the corporation for any subsequent taxable year over the subpart F income for that year is recharacterized as subpart F income. See also § 1.952-1(f)(1). An amount recaptured under section 952(c)(2) is treated as subpart F income in the same separate category (as defined in § 1.904-5(a)) as the subpart F income that was subject to the E&P limitation in a prior taxable year. See § 1.952-1(f)(2)(ii).

The Code does not provide a rule that explicitly coordinates the subpart F exclusion with section 952(c), which commenters identified as a source of confusion and potential inconsistency. In order to resolve this ambiguity, the proposed regulations set forth such a coordination rule by providing that the gross tested income and allowable deductions properly allocable to gross tested income are determined without regard to the application of section 952(c) (the “section 952(c) coordination rule”). See proposed § 1.951A-2(c)(4)(i). Thus, income that would be subpart F income but for the application of the E&P limitation in section 952(c)(1)(A) is excluded from gross tested income by reason of the subpart F exclusion. In addition, income that gives rise to E&P that results in subpart F recapture under section 952(c)(2) is not excluded from gross tested income by reason of the subpart F exclusion. In effect, the section 952(c) coordination rule treats an item of gross income as “taken into account” in determining subpart F income to the extent, and only to the extent, that the item would be included in subpart F income absent the application of section 952(c).

The proposed regulations include an example that illustrates this rule. See proposed § 1.951A-2(c)(4)(ii)(A). In the example, in Year 1, FS, a CFC wholly owned by a U.S. shareholder, has $100x of foreign base company sales income, a $100x loss in foreign oil and gas extraction income, and no E&P. In Year 2, FS has gross income of $100x that is not otherwise excluded from the definition of gross tested income in proposed § 1.951A-2(c)(1)(i) through (v), and no allowable deductions, and $100x of E&P. The example concludes that in Year 1 FS has no subpart F income because of the E&P limitation in section 952(c)(1)(A) and no gross tested income because gross tested income is determined without regard to section 952(c). In Year 2, the example concludes that, because FS's E&P ($100x) exceed its Year 2 subpart F income ($0), the subpart F income of Year 1 is recaptured in Year 2 under section 952(c)(2), and FS also has $100x of gross tested income in Year 2 because gross tested income is determined without regard to section 952(c).

One comment agreed that the section 952(c) coordination rule was an appropriate interpretation of the statute, noting that the rule preserves the ability for section 952(c)(2) to recapture subpart F income generated in prior years, while preventing recapture under section 952(c)(2) from permanently exempting gross tested income generated in subsequent years. However, several comments suggested that the section 952(c) coordination rule be withdrawn. These comments asserted that the section 952(c) coordination rule can lead to double taxation because the rule can result in the taxation of an aggregate amount of CFC income in excess of the net economic CFC income over a multi-year period. Some comments further suggested that the section 952(c) coordination rule is contrary to the language of the subpart F exclusion, on the grounds that any income of a CFC that generates E&P that are recharacterized as subpart F income by reason of the E&P recapture rule is “taken into account in determining the subpart F income” of the CFC and should therefore be excluded from gross tested income under the subpart F exclusion. Other comments recommended that the section 952(c) coordination rule be retained as it pertains to the E&P limitation rule under section 952(c)(1)(A), but be modified to exclude from its scope the E&P recapture rule of section 952(c)(2). Under that approach, both the subpart F income subject to E&P limitation in a prior year and gross income in a subsequent year that generates E&P giving rise to recapture of subpart F income would be excluded from gross tested income.

The Treasury Department and the IRS have determined that the section 952(c) coordination rule is consistent with the relevant statutory provisions and results in the appropriate amount of income that is subject to tax under sections 951 and 951A. Gross income that would be subpart F income during the current year but for the application section 952(c)(1)(A) is literally “taken into account” in determining subpart F income in that it potentially gives rise to future subpart F income by reason of section 952(c)(2). Furthermore, gross tested income is not subject to an E&P limitation analogous to the E&P limitation on subpart F income under section 952(c)(1)(A). In this regard, the determination of tested income under the GILTI regime is based on a taxable income concept, similar to the determination of income earned directly by a U.S. taxpayer, whereas the subpart F regime is rooted in a distributable dividend model, and thus predicated on the existence of E&P. Therefore, for example, a CFC may have $100x of gross tested income but no E&P in a taxable year (due, for instance, to a loss in foreign oil and gas extraction income), and the U.S. shareholder of the CFC (assuming no QBAI or other CFCs) will nonetheless have a $100x GILTI inclusion amount for the taxable year. This is the result under section 951A notwithstanding that the CFC in this case has no net economic income and no E&P for the year. If the same CFC for the same taxable year also has $100x of foreign base company sales income and $100x of E&P related to such income, in addition to the $100x GILTI inclusion amount, the CFC's U.S. shareholder would have a $100x subpart F inclusion. Under these facts, the U.S. shareholder is taxed on an aggregate amount of taxable income of the CFC ($200x) that exceeds the CFC's net economic income and E&P ($100x). In this example, the U.S. shareholder is not subject to tax twice with respect to a single item of income, but rather is subject to tax once with respect to each of two items—the CFC's subpart F income of $100x and the CFC's gross tested income of $100x. The section 952(c) coordination rule merely ensures that the same result obtains whether all items of income and loss arise in a single year (as in this example) or arise in different taxable years (as in the example in proposed § 1.951A-2(c)(4)(ii)(A)).

The Treasury Department and the IRS have also determined that it is not appropriate to exclude the E&P recapture rule from the scope of the section 952(c) coordination rule. Because section 951A contains no analog to the E&P limitation in section 952(c)(1)(A), it also contains no analog to the E&P recapture rule in section 952(c)(2). Without a GILTI recapture rule, the approach recommended by comments would effectively allow prior year losses in categories of income excluded from gross tested income (for example, subpart F income or foreign oil and gas extraction income) to permanently exempt gross tested income in subsequent years. For instance, if, in a taxable year, a CFC has $100x of foreign base company sales income, a $100x loss in foreign base company services income, and thus no subpart F income by reason of the E&P limitation of section 952(c)(1)(A), any gross tested income earned by the CFC in a subsequent year would recapture the foreign base company sales income from the previous year, and thus such gross income would never be subject to section 951A.

In excluding certain categories of income from gross tested income (namely, subpart F income, foreign oil and gas extraction income, and effectively connected income), Congress not only ensured that such income would not be subject to the GILTI regime, but also that losses with respect to such income would not be permitted to reduce income subject to the GILTI regime. Likewise, section 951A(c)(2)(B)(ii) provides that a loss in a category of income subject to the GILTI regime (that is, tested loss) cannot reduce the income subject to the subpart F regime by reason of the E&P limitation rule of section 952(c)(1)(A). See also § 1.951A-6(b) and part VIII.A of this Summary of Comments and Explanation of Revisions section. It is apparent, based on the purpose and structure of section 951A, that Congress intended for the GILTI and subpart F regimes to act as parallel, independent systems of taxation with respect to prescribed categories of CFC income, and losses with respect to one regime (or subject to neither regime) should not be permitted to permanently exempt the income subject to another regime. Therefore, an interpretation of section 952(c) that permits losses related to GILTI-exempt categories of income to reduce gross tested income would be contrary to the purpose and structure of section 951A.

A comment recommended, as an alternative to taking into account section 952(c)(2) recapture in determining gross tested income, that the recapture rules of section 952(c)(2) be modified so that E&P derived from gross tested income does not trigger recapture under section 952(c)(2). Although such amount would not be recaptured as subpart F income, the comment recommended that, in order to avoid double taxation of the same earnings, any recapture account should nonetheless be reduced by the amount treated as gross tested income. The Treasury Department and the IRS have determined that this recommendation is inconsistent with the language and purpose of section 952(c)(2). Section 952(c)(2) requires recapture in any taxable year in which E&P exceed subpart F income, and the recommendation would not result in recapture in these circumstances. Further, the purpose of section 952(c)(2) is to postpone the inclusion of subpart F income to a subsequent taxable year in which the CFC has sufficient E&P. The recommendation, by reducing a recapture account without recapture of subpart F income, would result in the permanent exemption of subpart F income. Finally, as illustrated in this part IV.C of the Summary of Comments and Explanation of Revisions section, the simultaneous recapture of subpart F income and the inclusion of gross tested income does not amount to double taxation of a single item of income, but rather the single taxation of each of two items of income. Accordingly, this recommendation is not adopted.

A comment recommended as another alternative that the section 952(c)(2) coordination rule not be applied with respect to recapture accounts that existed before the Act. The comment asserted that it would be inappropriate for income that triggers recapture under section 952(c)(2) based on pre-Act recapture account balances to also be treated as gross tested income because section 951A did not exist before 2018 and therefore no tested losses could have reduced subpart F income. The final regulations do not adopt this recommendation. Nothing in the statute or legislative history suggests that pre-Act recapture account balances should be treated differently than post-Act account balances. Further, there appears to be no stronger policy rationale for permitting losses that arose before the Act to permanently exempt gross tested income from taxation than for permitting GILTI-exempt losses that arise after the Act to do the same.

While the comments with respect to the section 952(c) coordination rule generally pertained to the application of the E&P limitation in section 952(c)(1)(A), the same issues as discussed in respect to section 952(c)(1)(A) arise with respect to application of the qualified deficit rule in section 952(c)(1)(B) and the chain deficit rule in section 952(c)(1)(C). Accordingly, the final regulations revise the section 952(c) coordination rule to apply also to disregard the effect of a qualified deficit or a chain deficit in determining gross tested income. See § 1.951A-2(c)(4)(ii).

One comment requested clarification that income subject to the high tax exception of section 954(b)(4) is not included in gross tested income even if such income would also be excluded from subpart F income by reason of section 952(c)(1)(A). The comment provided an example in which a CFC has $100x of foreign base company services income, a $100x loss in another category of subpart F income, no E&P, and thus no subpart F income by reason of the E&P limitation of section 952(c)(1)(A). According to the comment, if the election under section 954(b)(4) is made with respect to the foreign base company services income, one interpretation of the proposed regulations is that the $100x of foreign base company services income is not excluded from gross tested income by either the subpart F exclusion under section 951A(c)(2)(A)(i)(II) (because such income is not included in subpart F by reason of the high tax exception of section 954(b)(4)) or the GILTI high tax exclusion under section 951A(c)(2)(A)(i)(III) (because such income is not excluded from subpart F income “solely” by reason of the high tax exception of section 954(b)(4)). The Treasury Department and the IRS have determined that such clarification is unnecessary because an election under section 954(b)(4) cannot be made with respect to a net item eliminated by reason of section 952(c)(1)(A). Section 1.954-1(d)(4)(ii) provides that the net item of income to which the high tax exception of section 954(b)(4) applies is the subpart F income of a CFC determined after taking into account the earnings and profits limitation of section 952(c)(1)(A). Therefore, the net item of income that can be excluded under the high tax exception is determined after the application of section 952(c)(1)(A). Indeed, in the example presented by the comment, because the subpart F income of the CFC after application of the E&P limitation is zero, there is no net item of income for which an election under section 954(b)(4) and § 1.954-1(d)(5) can be made. Accordingly, the $100x of foreign base company services income is excluded from gross tested income solely by reason of the subpart F exclusion under section 951A(c)(2)(A)(i)(II).

b. Coordination With Qualified Deficit Rule in Section 952(c)(1)(B)

The qualified deficit rule in section 952(c)(1)(B) reduces a U.S. shareholder's subpart F inclusion attributable to a qualified activity (defined in section 952(c)(1)(B)(iii)) to the extent of that shareholder's pro rata share of any qualified deficit (defined in section 952(c)(1)(B)(ii)). A comment suggested that a tested loss could, in some cases, also give rise to a qualified deficit that could reduce subpart F income in a subsequent taxable year. The comment asserted that this could occur, for example, if certain deductions and losses that make up a qualified deficit are also properly allocable to gross tested income. Accordingly, the comment recommended that the final regulations deny a U.S. shareholder the ability to both reduce its net CFC tested income and increase a qualified deficit by reason of the same economic loss.

The Treasury Department and the IRS agree that the same deduction or loss should not result in a double benefit under section 951A and the qualified deficit rule, but have not identified a situation in which a single deduction or loss can both reduce tested income (or increase tested loss) and also give rise to or increase a qualified deficit. A deduction or loss that is properly allocable to gross tested income cannot also be attributable to a qualified activity that gives rise to subpart F income, and the same deduction cannot be taken into account more than once under sections 954(b)(5) and 951A(c)(2)(A)(ii). Nevertheless, for the avoidance of doubt, the final regulations provide that deductions that are allocated and apportioned to gross tested income are not attributable to a qualified activity and thus do not also increase or give rise to a qualified deficit. See § 1.951A-2(c)(3).

c. Coordination With Section 952(c)(1)(B)(vii)

Section 952(c)(1)(B)(vii)(I) contains an election to apply section 953(a) without regard to the same country exception in section 953(a)(1)(A). Comments requested that the section 952(c) coordination rule be modified to clarify that gross tested income is determined after giving effect to the election in section 952(c)(1)(B)(vii)(I). The rule in proposed § 1.951A-2(c)(4) was not intended to address the election in section 952(c)(1)(B)(vii)(I). Accordingly, the final regulations modify the section 952(c) coordination rule to apply only with respect to the E&P limitation rules of section 952(c)(1) (including the qualified deficit and chain deficit rules) and the E&P recapture rule of section 952(c)(2).

3. Coordination With De Minimis Rule, Full Inclusion Rule, and High Tax Exception

Section 954(a) provides that FBCI for a taxable year is equal to the sum of foreign personal holding company income (as determined under section 954(c)) (“FPHCI”), foreign base company sales income (as determined under section 954(d)) and foreign base company services income (as determined under section 954(e)). However, section 954(b)(3)(A) provides that if the sum of FBCI (determined without regard to allocable deductions) (“gross FBCI”) and gross insurance income for the taxable year is less than the lesser of five percent of gross income or $1,000,000, then no part of the gross income for the taxable year is treated as FBCI or insurance income (the “de minimis rule”). Conversely, section 954(b)(3)(B) provides that if the sum of gross FBCI and gross insurance income for the taxable year exceeds 70 percent of gross income, the entire gross income for the taxable year is treated as gross FBCI or gross insurance income, as appropriate (the “full inclusion rule”).

One comment requested that the de minimis and full inclusion rules be taken into account for purposes of determining “gross income taken into account” in determining subpart F income within the meaning of the subpart F exclusion. The comment asserted that such a rule would prevent double taxation because full inclusion subpart F income would be taxed solely under section 951 (and not section 951A), whereas de minimis subpart F income would be taxed solely under section 951A (and not section 951).

The Treasury Department and the IRS agree with this comment. Accordingly, subject to the application of the section 952(c) coordination rule, discussed in part IV.C.2 of this Summary of Comments and Explanation of Revisions section, the final regulations provide that the subpart F exclusion applies to gross income included in FBCI (adjusted net FBCI as defined in § 1.954-1(a)(5)) or insurance income (adjusted net insurance income as defined in § 1.954-1(a)(6)). See § 1.951A-2(c)(4)(i). Thus, for purposes of the subpart F exclusion, gross income taken into account in determining subpart F income does not include any item of gross income excluded from FBCI or insurance income under the de minimis rule or the high tax exception of section 954(b)(4), but generally does include any item of gross income included in FBCI or insurance income under the full inclusion rule. In addition, for purposes of the subpart F exclusion, gross income taken into account in determining subpart F income does not include gross income that qualifies for an exception to a category of FBCI described in section 954(a), including amounts excepted from the definition of FPHCI, such as rents and royalties derived from an active business under section 954(c)(2)(A) and § 1.954-2(b)(5) and (6) or active financing income under section 954(h).

Section 1.954-1(d)(6) provides that an item of gross income that is included in FBCI or insurance income under the full inclusion rule (“full inclusion FBCI”) is excluded from subpart F income if more than 90 percent of the gross FBCI and gross insurance income for the taxable year (determined without regard to the full inclusion rule) is attributable to net amounts excluded from subpart F income under the high tax exception of section 954(b)(4). The Treasury Department and the IRS have determined that it would be inappropriate for an item of gross income that would be included in gross tested income but for the full inclusion rule to be excluded from both gross tested income (by reason of the subpart F exclusion) and subpart F income (by reason of § 1.954-1(d)(6)). Accordingly, the final regulations provide that full inclusion FBCI excluded from subpart F income by reason of § 1.954-1(d)(6) is not excluded from gross tested income by reason of the subpart F exclusion. See § 1.951A-2(c)(4)(iii)(C). The final regulations further clarify that income excluded from subpart F income under § 1.954-1(d)(6) is also not excluded from gross tested income by reason of the GILTI high tax exclusion (discussed in part IV.B of this Summary of Comments and Explanation of Revisions section). See id. Accordingly, income excluded from subpart F income by reason of § 1.954-1(d)(6) is included in gross tested income.

D. Effect of Basis Adjustments Under Section 961(c)

Section 961(c) provides that, under regulations prescribed by the Secretary, if a U.S. shareholder is treated under section 958(a)(2) as owning stock of a CFC which is owned by another CFC, then adjustments similar to those provided under section 961(a) and (b) are made to the basis in such stock, and the basis in stock of any other CFC by reason of which the U.S. shareholder is considered under section 958(a)(2) as owning the stock. The provision further provides, however, that these adjustments are made only for the purposes of determining the amount included under section 951 in the gross income of such U.S. shareholder (or any successor U.S. shareholder). There are no regulations in effect under section 961(c).

Comments have questioned whether basis adjustments under section 961(c) should be taken into account for purposes of determining gross tested income of a CFC upon the CFC's disposition of stock of another CFC. One comment noted that, while section 951A(f)(1)(A) treats a GILTI inclusion in the same manner as a subpart F inclusion for purposes of basis adjustments under section 961, the resulting basis under section 961(c) only applies for purposes of determining amounts included in gross income under section 951. The comment recommended nonetheless that regulations provide that section 961(c) basis adjustments apply both for purposes of determining subpart F income and gross tested income to prevent certain items of income from being inappropriately taxed twice; the comment further noted, however, that unintentional non-taxation should also be avoided.

The interaction of basis adjustments under section 961(c) and section 951A will be further considered in connection with a guidance project addressing previously taxed E&P (“PTEP”) under sections 959 and 961. See Notice 2019-1, 2019-2 I.R.B. 275, section 3 (announcing an intention to address PTEP in forthcoming proposed regulations). The Treasury Department and the IRS are sensitive to the concern expressed in the comment but are also aware that taking into account section 961(c) basis adjustments for purposes of determining gross tested income could inappropriately reduce the amount of stock gain subject to tax. This may occur because, as was the case before the Act, section 961(c) adjustments are not taken into account for purposes of determining E&P, and thus a disposition of lower-tier CFC stock may generate E&P for the upper-tier CFC to the extent of the amount of the gain in the stock determined without regard to section 961(c). If the resulting E&P give rise to a dividend (including by reason of a disposition under section 1248) to a corporate U.S. shareholder, the dividend may result in an offsetting dividends received deduction. See sections 245A(a) and 1248(j). If section 245A(a) applies to the dividend, the taxable portion of any unrealized appreciation in the upper-tier CFC stock, to the extent attributable to unrealized appreciation in assets of the upper-tier CFC, would effectively be reduced in an amount equal to the dividend, either because of a dividend distribution that reduces the value in the upper-tier CFC stock without a corresponding basis reduction (section 961(d) applies only to the extent loss would otherwise be recognized) or by reason of a disposition to the extent the gain is recharacterized under section 1248(j) as a dividend for purposes of applying section 245A. Comments are requested on this issue, including the extent to which adjustments should be made to minimize the potential for the same item of income being subject to tax more than once and to minimize the inappropriate reduction of gain in CFC stock held by corporate U.S. shareholders.

E. Deduction or Loss Attributable to Disqualified Basis

1. In General

The proposed regulations include a rule that generally disallows, for purposes of calculating tested income or tested loss, any deduction or loss attributable to disqualified basis in depreciable or amortizable property (including, for example, intangible property) resulting from a disqualified transfer of the property. See proposed § 1.951A-2(c)(5). The relevant terms for purposes of applying the rule in proposed § 1.951A-2(c)(5) are defined by reference to certain provisions and terms in proposed § 1.951A-3(h)(2) (disregarding disqualified basis for purposes of determining QBAI), with certain modifications. See proposed § 1.951A-2(c)(5)(iii). In general, the term “disqualified basis” is defined as the excess of a property's adjusted basis immediately after a disqualified transfer, over the sum of the property's adjusted basis immediately before the disqualified transfer and the amount of gain recognized by the transferor in the disqualified transfer that is subject to tax as subpart F income or effectively connected income. See proposed § 1.951A-3(h)(2)(ii)(A) and (B). The term “disqualified transfer” is defined as a transfer of property by a transferor CFC during the transferor CFC's disqualified period to a related person in which gain was recognized, in whole or in part. See proposed § 1.951A-3(h)(2)(ii)(C). Finally, the term “disqualified period” is defined with respect to a transferor CFC as the period that begins on January 1, 2018, and ends as of the close of the transferor CFC's last taxable year that is not a CFC inclusion year. See proposed § 1.951A-3(h)(2)(ii)(D). Income generated by fiscal-year CFCs during the disqualified period is subject to neither the transition tax under section 965 nor the tax on GILTI under section 951A.

In response to comments, the Treasury Department and the IRS have revised these rules in a manner consistent with the purpose of the rule in the proposed regulations, as discussed in this part IV.E of the Summary of Comments and Explanation of Revisions section. Certain comments and revisions related to the determination of disqualified basis for purposes of both proposed §§ 1.951A-2(c)(5) and 1.951A-3(h)(2) are discussed in part IV.E.3 and 4 of this Summary of Comments and Explanation of Revisions section. For a discussion of additional comments and revisions related to the determination of disqualified basis for purposes of both proposed §§ 1.951A-2(c)(5) and 1.951A-3(h)(2), see part V.G of this Summary of Comments and Explanation of Revisions section.

2. Authority

Several comments recommended that the rule in proposed § 1.951A-2(c)(5) be withdrawn or substantially narrowed and re-proposed. Some of these comments recommended that the rule be revised to apply only to “non-economic” transactions or transactions engaged in with a tax-avoidance purpose, or that avoidance-type transactions be addressed through existing statutory or judicial doctrines. One comment recommended that the rule continue to be limited to transfers between related persons because third-party sales are fundamentally different from the “non-economic transactions” described in the legislative history. However, one comment opposed any additional limitations or weakening of the anti-abuse rules in the proposed regulations.

Several comments questioned the Treasury Department and the IRS's authority for issuing the rule. Many of these comments asserted that section 951A(d)(4), which provides authority to issue regulations that are “appropriate to prevent the avoidance of the purposes of this subsection,” does not authorize the Treasury Department and the IRS to promulgate rules that apply for any purpose other than for purposes of determining QBAI under section 951A(d). Also, two comments stated that the disallowance of deductions under proposed § 1.951A-2(c)(5) is contrary to, and therefore not authorized by, section 951A(c)(2)(A)(ii), which requires that the deductions of the CFC be allocated to gross tested income under rules similar to the rules of section 954(b)(5) for purposes of calculating tested income or tested loss.

In response to these comments, the Treasury Department and the IRS have revised the proposed rule in a manner that better reflects the source of its authority. Section 7805(a) provides that “the Secretary shall prescribe all needful rules and regulations for the enforcement of this title, including all rules and regulations as may be necessary by reason of any alteration of law in relation to internal revenue.” Section 951A(c)(2)(A) defines “tested income” by reference to certain items of gross income, reduced by “the deductions (including taxes) properly allocable to such gross income under rules similar to the rules of section 954(b)(5) (or to which such deductions would be allocable if there were such gross income).” Section 954(b)(5) provides that FPHCI, foreign base company sales income, and foreign base company services income are reduced, “under regulations prescribed by the Secretary,” by deductions “properly allocable” to such income. Similarly, section 882(c)(1)(A) provides that, for purposes of determining a foreign corporation's income which is effectively connected with the conduct of a trade or business within the United States (“effectively connected income”), “proper apportionment and allocation” of deductions of the foreign corporation “shall be determined as provided in regulations prescribed by the Secretary.” The rule, as revised in the final regulations, provides guidance for determining whether certain deductions or losses are “properly allocable” to gross tested income, subpart F income, or effectively connected income within the meaning of section 951A(c)(2)(A), section 954(b)(5), or section 882(c)(1)(A), respectively. See, for example, Redlark v. Commissioner, 141 F.3d 936, 940-41 (9th Cir. 1998) and Miller v. United States, 65 F.3d 687, 690 (8th Cir. 1995) (determining that the term “properly allocable” in section 163(e) is ambiguous and therefore there is an implicit legislative delegation of authority to the Commissioner to define the term).

The legislative history to the Act indicates that section 965 was intended as a transition measure to the new territorial tax system in which section 951A applies, and that Congress intended that all earnings of a CFC would be potentially subject to tax under either section 965 or section 951A. Conference Report, at 613 (“The [transition tax applies in] the last taxable year of a deferred foreign income corporation that begins before January 1, 2018, which is that foreign corporation's last taxable year before the transition to the new corporate tax regime elsewhere in the bill goes into effect.”). Because the final date for measuring the E&P of a CFC for purposes of section 965 is December 31, 2017 (the “final E&P measurement date”), and the effective date of section 951A is the first taxable year of a CFC beginning after December 31, 2017, all the earnings of a calendar year CFC are potentially subject to taxation under either section 965 or section 951A. However, a fiscal year CFC (for example, a CFC with a taxable year ending November 30) may have a gap between its final E&P measurement date under section 965 (December 31, 2017) and the date on which section 951A first applies with respect to its income (December 1, 2018, for a CFC with a taxable year ending November 30). Congress was aware that taxpayers could take advantage of this period to create “cost-free” basis in assets that could be used to reduce their U.S. tax liability in subsequent years, and expected the Treasury Department and the IRS to issue regulations to prevent this result. See Conference Report, at 645 (“The conferees intend that non-economic transactions intended to affect tax attributes of CFCs and their U.S. shareholders (including amounts of tested income and tested loss, tested foreign income taxes, net deemed tangible income return, and QBAI) to minimize tax under this provision be disregarded. For example, the conferees expect the Secretary to prescribe regulations to address transactions that occur after the measurement date of post-1986 earnings and profits under amended section 965, but before the first taxable year for which new section 951A applies, if such transactions are undertaken to increase a CFC's QBAI.”).

Consistent with the statute and the legislative history, the Treasury Department and the IRS have determined that a deduction or loss attributable to basis (disqualified basis) created by reason of a transfer from a CFC to a related CFC (a disqualified transfer) during the period between the final E&P measurement date and the effective date of section 951A (the disqualified period), to the extent no taxpayer included an amount in gross income by reason of such disqualified transfer, should not be permitted to reduce a taxpayer's U.S. income tax liability in subsequent years. Accordingly, the final regulations treat any deduction or loss attributable to disqualified basis as not “properly allocable” to gross tested income, subpart F income, or effectively connected income of the CFC (“residual CFC gross income”). See § 1.951A-2(c)(5)(i).

While the rules that allocate and apportion expenses generally depend on the factual relationship between the item of expense and the associated gross income, the relevant statutory language in sections 882(c)(1)(A), 951A(c)(2)(A)(ii), and 954(b)(5) does not constrain the Secretary from taking into account other considerations in determining whether it is “proper” for a certain item of expense to be allocated to, and therefore reduce, a particular item of gross income. Indeed, the Treasury Department and the IRS are not required to issue rules that mechanically allocate an item of expense to gross income to which such expense factually relates if taxable income would be distorted by reason of such allocation. In this regard, the Treasury Department and the IRS have determined that the rule in § 1.951A-2(c)(5) is necessary to ensure that transactions during the disqualified period, the income or earnings from which are not subject to tax, are not permitted to improperly reduce or eliminate a taxpayer's income that would be subject to tax after the disqualified period. This rule creates symmetry between the category of income generated by reason of a transfer during the disqualified period and the category of income to which any deduction or loss attributable to the resulting basis is allocated. That is, a disqualified transfer, by definition, generates residual CFC gross income (income that is not subpart F income, tested income, or effectively connected income), and the rule in § 1.951A-2(c)(5) allocates the deduction or loss attributable to the disqualified basis to the same category of income. In the case of a depreciable or amortizable asset with disqualified basis that is held until the end of its useful life, the aggregate amount of deduction or loss attributable to the disqualified basis allocated to residual CFC gross income under the rule will equal the amount of residual CFC gross income generated in the disqualified transfer.

The rule in proposed § 1.951A-2(c)(5) provides that any deduction or loss attributable to disqualified basis is disregarded for purposes of determining tested income or tested loss. In contrast, the rule in the final regulations allocates and apportions any such deduction or loss to gross income other than gross tested income, subpart F income, or effectively connected income. With respect to the determination of tested income or tested loss, whether an item of deduction or loss is disregarded (under the proposed regulations) or allocated to income other than gross tested income (under the final regulations) does not provide a different result. In either case, the deduction or loss is not permitted to reduce tested income or increase tested loss. However, by allocating an item of deduction or loss to residual CFC gross income, the rule in the final regulations ensures that any deduction or loss attributable to disqualified basis is also not taken into account for purposes of determining the CFC's subpart F income or effectively connected income. The broadening of the rule to allocate any deduction or loss attributable to disqualified basis away from subpart F income and effectively connected income is intended to ensure that taxpayers cannot simply circumvent the rule by converting their gross tested income into either subpart F income or effectively connected income, and thus be permitted to use the deduction or loss attributable to the disqualified basis against such income. The preamble to the proposed regulations evidenced an intention that taxpayers not be permitted to claim tax benefits with respect to cost-free disqualified basis, and the rule in the final regulations effectuates this intent by closing an obvious loophole. Furthermore, the rule ensures that the words “properly allocable” are interpreted consistently across provisions—sections 882(c)(1)(A), 951A(c)(2)(A)(ii), and 954(b)(5)—with respect to any deduction or loss attributable to disqualified basis.

The rule in proposed § 1.951A-2(c)(5) applies only to deductions or losses attributable to disqualified basis in “specified property,” which is defined as property that is of a type with respect to which a deduction is allowable under section 167 or 197. See proposed § 1.951A-2(c)(5)(ii). The Treasury Department and the IRS have concluded, however, that the rule should not be limited to specified property because deductions or losses attributable to disqualified basis in other property may also be used to inappropriately reduce a taxpayer's U.S. income tax liability. On the other hand, the Treasury Department and the IRS have concluded that it would be unduly burdensome to require CFCs to determine the disqualified basis in each item of inventory and that it is reasonable to expect that most inventory acquired during the disqualified period will be sold at a gain such that the disqualified basis in an item of inventory would rarely be relevant. Accordingly, the rule in the final regulations applies to deductions or losses attributable to disqualified basis in any property, other than property described in section 1221(a)(1), regardless of whether the property is of a type with respect to which a deduction is allowable under section 167 or 197. See §§ 1.951A-2(c)(5)(iii)(A) and 1.951A-3(h)(2)(ii).

One comment asserted that the use of the phrase “non-economic transactions” in the Conference Report means that the authority to draft anti-abuse rules pursuant to sections 7805 and 951A(d)(4) is limited to non-economic transactions, which necessitates a facts and circumstances test. The rule in § 1.951A-2(c)(5) is not premised upon facts and circumstances, such as a taxpayer's intent; rather, the rule is based on an interpretation of the term “properly allocable” in the context of a deduction or loss attributable to disqualified basis. Moreover, the rule applies only to a narrow subset of transactions—that is, transfers by fiscal year CFCs to related parties that occur between the final E&P measurement date under section 965 and the effective date of section 951A—and only has the effect of allocating a deduction or loss attributable to the cost-free basis created in such transaction to residual CFC gross income. The Treasury Department and the IRS have concluded that these narrowly circumscribed transactions will in almost all cases be motivated by tax avoidance rather than business exigencies, and that the allocation and apportionment of deduction or loss to residual CFC gross income is an appropriately tailored measure to address these transactions.

Based on the foregoing, the Treasury Department and the IRS have concluded that the rule in § 1.951A-2(c)(5), with the modifications discussed in this part IV.E of the Summary of Comments and Explanation of Revisions section, represents an appropriate exercise of its authority under sections 951A and 7805.

3. Effect of Disqualified Basis for Purposes of Determining Income or Gain

Some comments noted that the rule in proposed § 1.951A-2(c)(5) addresses only deductions or losses attributable to disqualified basis and does not address the effect of disqualified basis in determining a CFC's income or gain upon the disposition of property. For example, assume USP, a domestic corporation, wholly owns CFC1, which holds property with a fair market value of $100x and an adjusted basis of $80x, $70x of which is disqualified basis. CFC1 sells the property to an unrelated party in exchange for $100x of cash and, without regard to proposed § 1.951A-2(c)(5), recognizes $20x of gain. The comments asked whether, under the rule, the disqualified basis of $70x in the property is disregarded such that the sale results in $90x (rather than $20x) of gross tested income to CFC1.

The Treasury Department and the IRS have determined that the rule in § 1.951A-2(c)(5) should apply only for purposes of determining whether a deduction or loss is properly allocable to gross tested income, subpart F income, or effectively connected income. Thus, disqualified basis is not disregarded for purposes of determining income or gain recognized on the disposition of the property. However, because many taxpayers capitalize depreciation or amortization expense to other property, including inventory, and recover those costs through cost of goods sold or depreciation of the other property, the final regulations also provide that any depreciation, amortization, or cost recovery allowances attributable to disqualified basis is not properly allocable to property produced or acquired for resale under section 263, 263A, or 471. See § 1.951A-2(c)(5)(i). This rule ensures that depreciation or amortization expenses attributable to disqualified basis are not permitted to indirectly reduce taxable income through the depreciation expense of other property or from the disposition of inventory.

As discussed in part V.G of this Summary of Comments and Explanation of Revisions section, disqualified basis is generally reduced or eliminated to the extent that such basis reduces taxable income. Therefore, a sale of property with disqualified basis generally results in the elimination of the disqualified basis, because the basis is taken into account in determining the CFC's taxable income. As a result, absent a special provision, a CFC could “cleanse” the disqualified basis in property by selling the property to a related person after the disqualified period; the related person would have no disqualified basis in the property, and the selling CFC would recognize income only to the extent the amount realized exceeded its adjusted basis in the property (for this purpose, including its disqualified basis). To address this obvious loophole, the final regulations provide that, except to the extent that any loss recognized on the transfer of such property is treated as attributable to disqualified basis under § 1.951A-2(c)(5), or the basis is reduced or eliminated in a nonrecognition transaction within the meaning of section 7701(a)(45), a transfer of property with disqualified basis in the hands of a CFC to a related person does not reduce the disqualified basis in the hands of the transferee. See § 1.951A-3(h)(2)(ii)(B)(1)(ii). Thus, for example, if a CFC sells property with an adjusted basis of $80x and disqualified basis of $70x to a related person for $100x in a fully taxable exchange, the selling CFC would recognize $20x of gross income on the sale, which income may be included in gross tested income, and the disqualified basis in the property immediately after the transfer would remain $70x in the hands of the related person.

4. Concurrent Application of the Rule With Other Provisions

One comment asserted that if the Treasury Department and the IRS retain the rule in proposed § 1.951A-2(c)(5), then the disqualified transfer should be disregarded for all U.S. tax purposes, including for purposes of determining the gain or loss recognized by the transferor CFC by reason of the transfer and the tax attributes of the transferor CFC created by reason of the transfer. The comment expressed concern with potentially adverse consequences to the transferor CFC from the concurrent application of the rule and certain other provisions, such as incremental subpart F income generated by reason of the transfer, additional E&P that could dilute foreign tax credits with respect to a subpart F inclusion, and immediate U.S. taxation on any effectively connected income under section 882 from the transfer.

As discussed in part IV.E.2 of this Summary of Comments and Explanation of Revisions section, the rule in § 1.951A-2(c)(5) is intended to provide guidance on determining whether deductions of a CFC attributable to disqualified basis are properly allocable to gross tested income, subpart F income, and effectively connected income. The rule is not intended to disregard the transfer that created the disqualified basis in its entirety. Moreover, the Treasury Department and the IRS have determined that disregarding the transfer for all U.S. tax purposes is not appropriate because the property has in fact been transferred. In addition, disqualified basis in property does not include basis resulting from “qualified gain,” which is gain from the transfer included by the transferor CFC as effectively connected income or by a U.S. shareholder as its pro rata share of subpart F income. See § 1.951A-3(h)(2)(ii)(C)(3). Thus, the rule in § 1.951A-2(c)(5) does not apply to basis created in connection with amounts that are taxed under sections 882 and 951. Accordingly, this recommendation is not adopted.

Section 901(m) disallows certain foreign tax credits on foreign income not taken into account for U.S. tax purposes as a result of a “covered asset acquisition,” which includes an acquisition of assets for U.S. tax purposes that is treated as the acquisition of stock of a corporation (or is disregarded) for foreign tax purposes and an acquisition of an interest in a partnership which has an election in effect under section 754. See section 901(m)(2)(B) and (C). One comment noted that a disqualified transfer subject to the rule in proposed § 1.951A-2(c)(5) could also constitute a covered asset acquisition under section 901(m), such as the sale of an interest in a disregarded entity during the disqualified period. In such a case, according to the comment, a deduction or loss that is not taken into account for purposes of determining tested income or tested loss under the rule may nevertheless be taken into account for purposes of section 901(m) such that foreign tax credits under section 960 might be disallowed. The comment asserted that the concurrent application of the rule and section 901(m) could be unduly punitive to taxpayers that engaged in disqualified transfers that were also covered asset acquisitions and therefore recommended that a deduction or loss attributable to disqualified basis also be disregarded for purposes of section 901(m).

Disqualified basis could give rise to policy concerns under section 901(m) even when a deduction attributable to the disqualified basis is not taken into account in determining tested income or tested loss (or subpart F income or effectively connected income). For example, a deduction or loss attributable to the disqualified basis can reduce E&P for a taxable year, with the result that subpart F income for the taxable year may be limited under section 952(c)(1)(A). Indeed, proposed § 1.901(m)-5(b)(1) provides that basis differences must be taken into account under section 901(m) regardless of whether the deduction is deferred or disallowed for U.S. income tax purposes.

Based on the foregoing, the Treasury Department and the IRS have determined that it is not appropriate to disregard disqualified basis for purposes of section 901(m). However, in response to this comment, the final regulations permit taxpayers to make an election pursuant to which the adjusted basis in each property with disqualified basis held by a CFC or a partnership is reduced by the amount of the disqualified basis and the disqualified basis is eliminated. See § 1.951A-3(h)(2)(ii)(B)(3). This reduction in adjusted basis is for all purposes of the Code, including section 901(m). Thus, if an election is made, a disqualified transfer of property that is also a covered asset acquisition of a relevant foreign asset will result in neither disqualified basis in the property within the meaning of § 1.951A-3(h)(2)(ii) nor a basis difference with respect to the relevant foreign asset within the meaning of section 901(m)(3)(C). As a result, in the case of an election, the rule in § 1.951A-2(c)(5) and section 901(m) will not apply concurrently with respect to a disqualified transfer that is also a covered asset acquisition.

F. Other Comments and Revisions

1. Tested Loss Carryforward

In determining a U.S. shareholder's net CFC tested income for a taxable year, the U.S. shareholder's aggregate pro rata share of tested losses for the taxable year reduces the shareholder's aggregate pro rata share of tested income for the taxable year. See section 951A(c)(1). Comments recommended that the final regulations include a provision allowing a U.S. shareholder's aggregate pro rata share of tested losses in excess of the shareholder's aggregate pro rata share of tested income for the taxable year to be carried forward to offset the shareholder's net CFC tested income in subsequent years.

A GILTI inclusion amount is an annual calculation, and nothing in the statute or legislative history suggests that unused items, such as a U.S. shareholder's aggregate pro rata share of tested losses in excess of the shareholder's aggregate pro rata share of tested income for the taxable year, can or should be carried to another taxable year. Accordingly, this recommendation is not adopted.

2. Deemed Payments Under Section 367(d)

In general, section 367(d) provides that if a U.S. person transfers intangible property to a foreign corporation in an exchange described in section 351 or 361, the person is treated as having sold the property in exchange for payments contingent upon the productivity, use, or disposition of such property. The regulations under section 367(d) provide that the deemed payment may be treated as an expense (whether or not that amount is actually paid) of the transferee foreign corporation that is properly allocated and apportioned to gross income subject to subpart F under the provisions of §§ 1.954-1(c) and 1.861-8. See § 1.367(d)-1T(c)(2)(ii) and (e)(2)(ii).

In response to comments, the final regulations clarify that a deemed payment under section 367(d) is treated as an allowable deduction for purposes of determining tested income and tested loss. See § 1.951A-2(c)(2)(ii). Accordingly, consistent with the regulations under section 367(d), such deemed payments may be allocated and apportioned to gross tested income to the extent provided under § 1.951A-2(c)(3).

3. Compute Tested Income in the Same Manner as E&P

A comment requested that the final regulations provide that tested income and tested loss be determined under the principles of section 964, which provides rules for the calculation of E&P of foreign corporations. Another comment requested that the final regulations permit small CFCs to make an annual election to treat their tested income or tested loss for a CFC inclusion year to be equal to their E&P for such CFC inclusion year. Section 951A(c)(2) is clear that tested income or tested loss for a CFC inclusion year is computed by subtracting properly allocable deductions from gross tested income, and there is nothing in the statute or legislative history that indicates that tested income or tested loss should be limited by, or otherwise determined by reference to, E&P for such year. Accordingly, these recommendations are not adopted.

4. Effect of Losses in Other Categories of Income

The proposed regulations provide that allowable deductions are allocated and apportioned to gross tested income under the principles of section 954(b)(5) and § 1.954-1(c), by treating gross tested income within a single category (as defined in § 1.904-5(a)) as a single item of gross income, in addition to the items in § 1.954-1(c)(1)(iii). See proposed § 1.951A-2(c)(3). The final regulations clarify that losses in other categories of income (such as FBCI) cannot reduce gross tested income, and that tested losses cannot reduce other categories of income. See § 1.951A-2(c)(3).

V. Comments and Revisions to Proposed § 1.951A-3—Qualified Business Asset Investment

A. Inability of Tested Loss CFCs To Have QBAI

A U.S. shareholder's GILTI inclusion amount is equal to the excess of its net CFC tested income over its net DTIR for the taxable year. See section 951A(b)(1) and § 1.951A-1(c)(1). A U.S. shareholder's net DTIR is equal to 10 percent of its aggregate pro rata share of the QBAI of its CFCs. See section 951A(b)(2) and § 1.951A-1(c)(3). A CFC's QBAI is equal to its aggregate average adjusted basis in specified tangible property. See section 951A(1) and proposed § 1.951A-3(b). Specified tangible property is defined as tangible property used in the production of tested income. See section 951A(d)(2)(A) and proposed § 1.951A-3(c)(1). Consistent with the statute and the Conference Report, the proposed regulations clarify that tangible property of a tested loss CFC is not used in the production of tested income within the meaning of section 951A(d)(2)(A). See Conference Report, at 642, fn. 1536. In this regard, the proposed regulations provide that tangible property of a tested loss CFC is not specified tangible property and thus a tested loss CFC's QBAI is zero (the “tested loss QBAI exclusion”). See proposed § 1.951A-3(b), (c)(1), and (g)(1).

Comments recommended that the final regulations eliminate the tested loss QBAI exclusion, such that a tested loss CFC could have specified tangible property and therefore QBAI. One of the comments noted that the version of section 951A in the House bill defined specified tangible property as any tangible property to the extent such property is used in the production of tested income or tested loss. See H.R. 1, 115th Cong. § 4301(a) (2017). The comment posited that the text of the statute is ambiguous, the tested loss QBAI exclusion is otherwise inconsistent with section 951A, and the exclusion is not compelled by the statute. The comment also asserted that this rule may be easily avoided by combining a tested loss CFC with a tested income CFC (including through an election under § 301.7701-3 to change the classification of either entity for U.S. tax purposes) because there is no corollary to the tested loss QBAI exclusion for partnerships or disregarded entities.

The Treasury Department and the IRS reject this recommendation. The Senate amendment to the House bill struck the reference to “tested loss” in the definition of specified tangible property, and the Conference Report explains that the term “used in the production of tested income” means that “[s]pecified tangible property does not include property used in the production of a tested loss, so that a CFC that has a tested loss in a taxable year does not have QBAI for the taxable year.” See Conference Report, at 642, fn.1536. Thus, the statute, taking into account the footnote in the Conference Report, unambiguously provides that tested loss CFCs cannot have QBAI. Accordingly, the final regulations retain the tested loss QBAI exclusion. But cf. part VI.D of this Summary of Comments and Explanation of Revisions section regarding a reduction to tested interest expense of a CFC for a “tested loss QBAI amount,” a new component in computing specified interest expense.

One comment requested that, if the tested loss QBAI exclusion is retained, proposed § 1.951A-3(b) and (c) should be revised to clarify that the exclusion applies only for a CFC inclusion year with respect to which a CFC is a tested loss CFC. The final regulations do not revise these provisions because it is sufficiently clear that the tested loss QBAI exclusion rule applies only with respect to a CFC inclusion year of a CFC for which it is a tested loss CFC and that a CFC is a tested loss CFC only for a CFC inclusion year in which the CFC does not have tested income. See § 1.951A-2(b)(2).

B. Determination of Depreciable Property

Section 951A(d)(1)(B) provides that specified tangible property is taken into account in determining QBAI only if the property is of a type with respect to which a depreciation deduction is allowable under section 167. Similarly, the proposed regulations define “specified tangible property” as tangible property used in the production of tested income, and define “tangible property” as property for which the depreciation deduction provided by section 167(a) is eligible to be determined under section 168 (even if the CFC has elected not to apply section 168). See proposed § 1.951A-3(c)(1) and (2).

A comment recommended that, for purposes of determining QBAI, the final regulations take into account the entire adjusted basis in precious metals and other similar tangible property that are used in the production of tested income, even if only a portion of the adjusted basis in such property is depreciable in calculating regular taxable income. The comment suggested that if property is depreciable in part, then the entire asset is “of a type” with respect to which a deduction is allowable under section 167 within the meaning of section 951A(d)(1)(B).

In defining QBAI, section 951A(d) distinguishes between depreciable tangible property and non-depreciable tangible property, such as land. Section 951A(d) defines QBAI as specified tangible property “of a type” for which a deduction is allowable under section 167. The proposed and final regulations interpret the phrase “of a type” consistent with the interpretation of the phrase “of a character” with respect to section 168. See Rev. Rul. 2015-11, 2015-21 I.R.B. 975. See § 1.951A-3(c)(2) (defining tangible property as property for which the depreciation deduction provided by section 167(a) is eligible to be determined under section 168 (with certain exclusions)). The Treasury Department and the IRS determined that for consistency, the same standard for determining whether property is depreciable should apply for determining whether property qualifies as QBAI.

In Newark Morning Ledger Co. v. United States, 507 U.S. 546 (1993), the Supreme Court provided that “[w]hether or not . . . a tangible asset, is depreciable for Federal income tax purposes depends upon the determination that the asset is actually exhausting, and that such exhaustion is susceptible of measurement.” Newark Morning Ledger Co. v. United States at 566. Although unrecoverable commodities used in a business are depreciable, recoverable commodities used in a business are not depreciable because they do not suffer from exhaustion, wear and tear, or obsolescence over a determinable useful life. O'Shaughnessy v. Commissioner, 332 F.3d 1125 (8th Cir. 2003); Arkla, Inc. v. United States, 765 F.2d 487 (5th Cir. 1985). The recoverable quantity of a commodity used in the business suffers no change in its physical characteristics or value as a result of its use in the business. The comment seemed to imply that precious metals were a single unit of property that was partially depreciable and partially non-depreciable, rather than quantities of metal in separate categories of property, one of which is depreciable.

The Treasury Department and the IRS have determined that it would not be appropriate for purposes of determining a CFC's QBAI to take into account the CFC's entire adjusted basis in an asset that is only partially depreciable. Taking into account basis that is not subject to a depreciation allowance would overstate a CFC's QBAI. For example, in the case of precious metals that are partially depreciable, such as platinum used in a catalyst, a portion of the metal may be subject to exhaustion, wear and tear, or obsolescence during its useful life. The remainder of the metal is recoverable for reuse or sale. When initially purchased, the value and tax basis of the recoverable portion generally should reflect the forward price of such metal. The value and tax basis of the depreciable portion of the metal generally should reflect the net present value of the expected returns generated by the metal. QBAI is a proxy for the base upon which non-extraordinary, tangible returns should be calculated. See S. Comm. on the Budget, Reconciliation Recommendations Pursuant to H. Con. Res. 71, S. Print No. 115-20, at 371 (2017) (“Senate Explanation”) (The provision approximates . . . tangible income . . . as a 10-percent return on . . . the adjusted basis in tangible depreciable property.”). Therefore, only the depreciable portion of the precious metal, which is associated with the tangible returns, should be taken into account in this measurement. Given that liquid commodity markets exist for these precious metals, taxpayers could sell the future rights to the recoverable portion of the asset (thereby reducing their economic outlay and exposure with respect to the property). Cf. Guardian Industries v. Commissioner, 97 T.C. 308 (1991) (taxpayer regularly sold silver waste from photographic development process to refiners). Thus, the depreciable portion of the asset represents the taxpayer's economic investment in generating tangible returns. Accordingly, the comment is not adopted.

The comment also requested that in calculating the adjusted basis in precious metals for QBAI purposes, the final regulations provide that class lives applied to precious metals for purposes of the alternative depreciation system (“ADS”) are the same class lives determined under the principles of Rev. Rul. 2015-11, rather than the ADS class lives of the equipment to which the precious metals attach. This recommendation is not adopted because Rev. Rul. 2015-11 does not establish principles for determining class lives of the precious metals discussed therein, but rather addresses whether certain precious metals are depreciable under the facts and circumstances described in the ruling.

One comment requested that all expenditures paid or incurred with respect to the acquisition, exploration, and development of a mine or other natural deposit should be taken into account in determining QBAI. The comment stated that such exploration and development costs for mining operations are “of a type” for which depreciation is allowed, even though the costs are recovered through depletion rather than depreciation. The comment also recommended that the adjusted basis in a mine or other natural deposit included as QBAI should be determined using cost depletion, rather than percentage depletion.

Section 951A(d)(1)(B) limits property taken into account in determining QBAI to tangible property of a type with respect to which a deduction is allowable under section 167. Congress did not extend the definition of QBAI to property of a type with respect to which a deduction is allowed under section 611 (the allowance of deduction for depletion). Although the comment focused on the similarities between cost depletion and depreciation, there are also similarities between cost depletion of mineral properties and the acquisition cost of inventory. The inventory cost of a severed mineral includes the cost depletion attributable to the severed mineral. See section 263A and § 1.263A-1(e)(3)(ii)(J). In essence, the acquisition cost of the mineral property recovered through cost depletion is the inventory cost of the severed mineral, and QBAI does not include inventory. Accordingly, the recommendation is not adopted.

The proposed regulations define “tangible property” as property for which the depreciation deduction provided by section 167(a) is eligible to be determined under section 168 without regard to section 168(f)(1), (2), or (5) and the date placed in service. See proposed § 1.951A-3(c)(2). Section 168(k) increases the depreciation deduction allowed under section 167(a) with respect to qualified property, which includes tangible and certain intangible property. The final regulations revise the definition of tangible property in § 1.951A-3(c)(2) to exclude certain intangible property to which section 168(k) applies, namely, computer software, qualified film or television productions, and qualified live theatrical productions described in section 168(k)(2)(A).

C. Determination of Basis Under Alternative Depreciation System

For purposes of determining QBAI, the adjusted basis in specified tangible property is determined by using ADS under section 168(g), and by allocating the depreciation deduction with respect to such property for the CFC inclusion year ratably to each day during the period in the taxable year to which such depreciation relates. See section 951A(d)(3) ^[3] and § 1.951A-3(e)(1). ADS applies to determine the adjusted basis in property for purposes of determining QBAI regardless of whether the property was placed in service before the enactment of section 951A, or whether the basis in the property is determined under another depreciation method for other purposes of the Code. See section 951A(d)(3) and § 1.951A-3(e)(2). In addition, for purposes of determining income and E&P, a CFC is generally required to use ADS for depreciable property used predominantly outside the United States. See section 168(g) and §§ 1.952-2(c)(2)(ii) and (iv) and 1.964-1(a)(2). However, a CFC may instead use for this purpose a depreciation method used for its books of account regularly maintained for accounting to shareholders or a method conforming to United States generally accepted accounting principles (a “non-ADS depreciation method”) if the differences between ADS and the non-ADS depreciation method are immaterial. See §§ 1.952-2(c)(2)(ii) and (iv) and 1.964-1(a)(2).

A comment recommended that ADS not be required under section 951A(d) for specified tangible property placed in service before the enactment of section 951A. This comment asserted that section 951A(d)(3) does not compel the conclusion that ADS must be used for assets placed in service before the enactment of section 951A, and cited compliance concerns as a justification for not requiring the use of ADS with respect to such assets. Another comment recommended that the final regulations permit taxpayers to elect to compute the adjusted basis in all specified tangible property of a CFC—not just specified tangible property placed in service before the enactment of section 951A—under the method that the CFC uses to compute its tested income and tested loss, even if such method is not ADS.

Section 951A(d)(3) is clear that the adjusted basis in specified tangible property is determined using ADS under section 168(g), and therefore the final regulations do not adopt the recommendation to permit taxpayers an election to compute the adjusted basis in all specified tangible property under the CFC's non-ADS depreciation method. However, recognizing the potential burden of re-determining the basis under ADS of all specified tangible property held by a CFC placed in service before the enactment of section 951A, and given that a non-ADS depreciation method is permissible only when there are immaterial differences between ADS and such other method, the Treasury Department and the IRS have determined that a transition rule is warranted for CFCs that are not required to use ADS for purposes of computing income and E&P. Accordingly, the final regulations provide that a CFC that is not required to use ADS for purposes of computing income and E&P may elect, for purposes of calculating QBAI, to use its non-ADS depreciation method to determine the adjusted basis in specified tangible property placed in service before the first taxable year beginning after December 22, 2017, subject to a special rule related to salvage value. See § 1.951A-3(e)(3)(ii). The election also applies to the determination of a CFC's partner adjusted basis under § 1.951A-3(g)(3) in partnership specified tangible property placed in service before the CFC's first taxable year beginning after December 22, 2017. See id. This transition rule does not apply for purposes of determining the foreign-derived intangible income (“FDII”) of a domestic corporation. Cf. section 250(b)(2)(B) (in calculating deemed tangible income return for purposes of FDII, QBAI is generally determined under section 951A(d)).

A comment requested that the final regulations confirm that the use of ADS in determining the basis in specified tangible property, whether placed in service before or after the enactment of section 951A, for purposes of determining QBAI is not a change in method of accounting or, if it is a change in method, that global approval under section 446(e) be given for such a change. Another comment recommended that a CFC switching to ADS for property placed in service before the enactment of section 951A should not be required to file Form 3115 to request an accounting method change for depreciation, and that the cumulative adjustment should be taken into account for the adjusted basis in the specified tangible property as of the CFC's first day of the first year to which section 951A applies.

The determination of the adjusted basis in property under section 951A(d) is not a method of accounting subject to the consent requirement of section 446(e). As a result, a CFC does not need the Commissioner's consent to use ADS for purposes of determining its adjusted basis in specified tangible property in determining its QBAI. A CFC that uses ADS for purposes of determining QBAI should determine the correct basis in the property under ADS as of the CFC's first day of the first taxable year to which section 951A applies and apply section 951A(d)(3) accordingly. The final regulations also clarify that the adjusted basis in property is determined based on the cost capitalization methods of accounting used by the CFC for purposes of determining its tested income and tested loss. See § 1.951A-3(e)(1).

A change to ADS from another depreciation method for purposes of computing tested income or tested loss is a change in method of accounting subject to section 446(e). The Treasury Department and the IRS expect that many CFCs that are not already using ADS for purposes of computing income and E&P will change their method of accounting for depreciation to the straight-line method, the applicable recovery period, or the applicable convention under ADS to comply with § 1.952-2(c)(2)(iv) and § 1.964-1(c)(1)(iii)(c) and that most of such changes are already eligible for automatic consent under Rev. Proc. 2015-13, 2015-5 I.R.B. 419. The Treasury Department and the IRS intend to publish another revenue procedure further expanding the availability of automatic consent for depreciation changes and updating the terms and conditions in sections 7.07 and 7.09 of Rev. Proc. 2015-13 (related to the source, separate limitation classification, and character of section 481(a) adjustments) to take into account section 951A. After the change in accounting method, the basis in specified tangible property will be the correct basis for purposes of determining income, E&P, and QBAI.

The final regulations clarify the interaction between the daily proration of depreciation rule in section 951A(d)(3) and the applicable convention under ADS. Under section 951A(d)(3), the adjusted basis in property is determined by allocating the depreciation deduction with respect to property to each day during the period in the taxable year to which the depreciation relates. The half-year convention, mid-month convention, and mid-quarter convention in section 168(d) treat property as placed in service (or disposed of) for purposes of section 168 at the midpoint of the taxable year, month, or quarter, as applicable, irrespective of when the property was placed in service (or disposed of) during the taxable year. The final regulations clarify that the period in the CFC inclusion year to which such depreciation relates is determined without regard to the applicable convention under section 168(d). See § 1.951A-3(e)(1). Accordingly, in the year property is placed in service, the depreciation deduction allowed for the taxable year is prorated from the day the property is actually placed in service, and, in the year property is disposed of, the depreciation deduction allowed for the taxable year is prorated to the date of disposition. Allocating depreciation to each day during the period in which the property is used irrespective of the applicable convention ensures that the average of the aggregate adjusted basis as of the close of each quarter is properly adjusted to reflect the depreciation allowed for the taxable year.

The Treasury Department and the IRS continue to study issues related to the determination of QBAI for purposes of section 951A. In particular, the Treasury Department and the IRS are aware that a CFC that is a partner in a foreign partnership may have difficulty determining the basis in partnership property under ADS, particularly when the partnership is not controlled by U.S. persons. Comments are requested on methodologies for determining the basis in partnership property owned by a foreign partnership that is not controlled directly or indirectly by U.S. persons.

D. Dual Use Property

Section 951A(d)(2)(B) provides that if property is used both in the production of tested income and income that is not tested income, the property is specified tangible property in the same proportion that the gross income described in section 951A(c)(1)(A) produced with respect to such property bears to the total gross income produced with respect to such property. The proposed regulations provide that if tangible property is used in both the production of gross tested income and other income, the portion of the adjusted basis in the property treated as adjusted basis in specified tangible property is determined by multiplying the average of the adjusted basis in the property by the dual use ratio. See proposed § 1.951A-3(d)(1). If the property produces directly identifiable income for a CFC inclusion year, the dual use ratio is the ratio of the gross tested income produced by the property to the total amount of gross income produced by the property. See proposed § 1.951A-3(d)(2)(i). In all other cases, the dual use ratio is the ratio of the gross tested income of the tested income CFC to the total amount of gross income of the tested income CFC. See proposed § 1.951A-3(d)(2)(ii).

Under the proposed regulations, the dual use ratio requires a determination of whether and how much gross income is “directly identifiable” with particular specified tangible property. The Treasury Department and the IRS recognize that application of the directly identifiable standard could result in substantial uncertainty and controversy. In addition, the Treasury Department and the IRS have determined that the rules under section 861 for allocating a depreciation or amortization deduction attributable to property owned by a CFC to categories of income of the CFC represent a reliable and well-understood proxy for determining the type of income produced by the property, even in circumstances where there is no income that is “directly identifiable” with the property. Accordingly, the final regulations provide that the dual use ratio, with respect to tangible property for a CFC inclusion year, is the ratio calculated as the sum of the amount of the depreciation deduction with respect to the property for the CFC inclusion year that is allocated and apportioned to gross tested income for the CFC inclusion year under § 1.951A-2(c)(3) and the depreciation with respect to the property capitalized to inventory or other property held for sale, the gross income or loss from the sale of which is taken into account in determining tested income for the CFC inclusion year, divided by the sum of the total amount of the depreciation deduction with respect to the property for the CFC inclusion year and the total amount of depreciation with respect to the property capitalized to inventory or other property held for sale, the gross income or loss from the sale of which is taken into account for the CFC inclusion year. See § 1.951A-3(d)(3). The dual use ratio also applies with respect to partnership specified tangible property, except, for this purpose, determined by reference to a tested income CFC's distributive share of the amounts described in the preceding sentence. See § 1.951A-3(g)(3)(iii) and part V.E of this Summary of Comments and Explanation of Revisions section.

A comment recommended that the final regulations clarify, through additional examples, that the method for determining the dual use ratio with respect to specified tangible property does not change if (i) the dual use property becomes or ceases to be specified tangible property during the year, or (ii) the dual use property gives rise to increasing or decreasing gross tested income across quarters in a taxable year. The Treasury Department and the IRS have determined that additional examples are unnecessary. As the comment suggests, the dual use ratio is not determined on the basis of the type and amount of gross income produced by the property as of any particular quarter close, but rather is determined based on the type and the amount of gross income produced by the property for the entire taxable year. In this regard, there is no ambiguity in the language in the regulations, and therefore no need for additional clarification.

The rules in § 1.951A-3 do not apply in determining QBAI for purposes of computing the deduction of a domestic corporation under section 250 for its FDII. See proposed § 1.250(b)-2 (REG-104464-18, 84 FR 8188 (March 6, 2019)) for the QBAI rules related to the FDII deduction. However, it is anticipated that, except as indicated in part V.D of this Summary of Comments and Explanation of Revisions section with respect to the election to use a non-ADS depreciation method for assets placed in service before the enactment of section 951A, revisions similar to the revisions to proposed § 1.951A-3 discussed in parts V.B through E of this Summary of Comments and Explanation of Revisions section will be made to proposed § 1.250(b)-2.

E. Partnership QBAI

Section 951A(d)(3) ^[4] provides that, for purposes of calculating QBAI, if a CFC holds an interest in a partnership at the close of the CFC's taxable year, the CFC takes into account its distributive share of the aggregate of the partnership's adjusted basis in depreciable tangible property used in its trade or business that is used in the production of tested income (determined with respect to the CFC's distributive share of income with respect to such property). For this purpose, a CFC's distributive share of the adjusted basis in any property is the CFC's distributive share of income with respect to such property. See section 951A(d)(3) (flush language).

The proposed regulations implement the rule in section 951A(d)(3) by providing that, if a tested income CFC holds an interest in one or more partnerships as of the close of a CFC inclusion year, the QBAI of the tested income CFC for the CFC inclusion year is increased by the sum of the tested income CFC's partnership QBAI with respect to each partnership for the CFC inclusion year. See proposed § 1.951A-3(g)(1). A tested income CFC's partnership QBAI with respect to a partnership is the sum of the tested income CFC's share of the partnership's adjusted basis in partnership specified tangible property as of the close of a partnership taxable year that ends with or within a CFC inclusion year. See proposed § 1.951A-3(g)(2)(i). A tested income CFC's share of the partnership's adjusted basis in partnership specified tangible property is determined by multiplying the partnership's adjusted basis in the property by the tested income CFC's partnership QBAI ratio with respect to the property. See id. Similar to the rule for dual use property, under the proposed regulations, the tested income CFC's partnership QBAI ratio with respect to partnership specified tangible property depends on whether the property produces directly identifiable income. In the case of partnership specified tangible property that produces directly identifiable income for a partnership taxable year, a tested income CFC's partnership QBAI ratio with respect to the property is the tested income CFC's distributive share of the gross income produced by the property for the partnership taxable year that is included in the gross tested income of the tested income CFC for the CFC inclusion year to the total gross income produced by the property for the partnership taxable year. See proposed § 1.951A-3(g)(2)(ii)(A). In the case of partnership specified tangible property that does not produce directly identifiable income for a partnership taxable year, a tested income CFC's partnership QBAI ratio with respect to the property is the tested income CFC's distributive share of the gross income of the partnership for the partnership taxable year that is included in the gross tested income of the tested income CFC for the CFC inclusion year to the total amount of gross income of the partnership for the partnership taxable year. See proposed § 1.951A-3(g)(2)(ii)(B).

The partnership QBAI ratio in the proposed regulations is effectively an amalgamation of two ratios—a ratio that describes the portion of the partnership specified tangible property that is used in the production of gross tested income (that is, the dual use ratio) and a ratio that describes a tested income CFC's proportionate interest in all the income produced by the property. The final regulations disaggregate the partnership QBAI ratio into these two ratios—the dual use ratio (as defined in § 1.951A-3(d)(3)) and a new proportionate share ratio (as defined in § 1.951A-3(g)(4)(ii)). Accordingly, the final regulations provide that a tested income CFC's “partner adjusted basis” with respect to partnership specified tangible property—that is, the adjusted basis in partnership specified tangible property taken into account in determining the tested income CFC's partnership QBAI—is generally, in the case of partnership specified tangible property used in the production of only gross tested income (“sole use partnership property”), the tested income CFC's proportionate share of the partnership's adjusted basis in the property for the partnership taxable year. See § 1.951A-3(g)(3)(ii). A tested income CFC's partner adjusted basis with respect to partnership specified tangible property used in the production of gross tested income and gross income that is not gross tested income (“dual use partnership property”) is generally the tested income CFC's proportionate share of the partnership's adjusted basis in the property for the partnership taxable year, multiplied by the tested income CFC's dual use ratio with respect to the property (determined by reference to the tested income CFC's distributive share of amounts described in § 1.951A-3(d)(3)). See § 1.951A-3(g)(3)(iii). In either case, a tested income CFC's proportionate share of the partnership's adjusted basis in partnership specified tangible property is the partnership's adjusted basis in the property for the partnership taxable year multiplied by the tested income CFC's proportionate share ratio with respect to the property for the partnership taxable year.

As discussed in part V.D of this Summary of Comments and Explanation of Revisions section, a rule that determines adjusted basis in specified tangible property taken into account in determining QBAI by reference to the “directly identifiable income” attributable to such property would lead to substantial uncertainty and controversy, whereas the rules under section 861 for allocating and apportioning depreciation attributable to property owned by a CFC to categories of income represent a longstanding proxy for determining the types of income produced by the property. For this reason, the final regulations determine the dual use ratio by reference to the amount of depreciation deductions allocated to gross tested income under § 1.951A-2(c)(3). Similarly, the Treasury Department and the IRS have determined that calculating partnership QBAI by reference to the “directly identifiable income” produced by partnership specified tangible property would lead to substantial uncertainty and controversy, and that a partner's share of a depreciation deduction with respect to partnership specified tangible property is a reliable proxy for determining a CFC's distributive share of income with respect to such property. Accordingly, the final regulations determine the proportionate share ratio with respect to partnership specified tangible property also by reference to the depreciation with respect to the property, rather than the directly identifiable income attributable to the property or the gross income of the partner. See § 1.951A-3(g)(4)(ii).

A comment requested clarification that the partnership QBAI ratio in the proposed regulations, which references the amount of “gross income” produced by the property, is determined by reference to “gross taxable income,” rather than gross section 704(b) income. The comment also recommended that if the partnership QBAI ratio is determined by reference to a partnership's gross taxable income, that section 704(c) allocations (including items of income under the remedial method) be taken into account in determining the CFC's distributive share of the gross income produced by the property for the partnership taxable year. The specific comment regarding the calculation of gross income produced by property has been mooted by the change to determining the dual use and proportionate share ratios by reference to the depreciation with respect to the property. However, the comment remains relevant to the calculation of the depreciation with respect to property for purposes of determining the dual use ratio and proportionate share ratio.

For purposes of the proportionate share ratio, the final regulations do not adopt this recommendation. Section 704(b) income represents a partner's economic interest in the partnership and therefore more closely aligns with the economic production of income from partnership property that QBAI is intended to measure. Accordingly, the final regulations clarify that the proportionate share ratio is determined by reference to the amount of depreciation with respect to property (and a tested income CFC's distributive share of such amount) determined under section 704(b). See § 1.951A-3(g)(4)(i). Therefore, items determined under section 704(c) are not taken into account for purposes of determining a tested income CFC's partner adjusted basis in partnership specified tangible property held by a partnership and thus the tested income CFC's partnership QBAI with respect to the partnership. However, because the dual use ratio is determined by reference to the allocation and apportionment of depreciation deductions to gross tested income of a tested income CFC, and thus is based on a taxable income concept, items determined under section 704(c) are taken into account for purposes of determining the dual use ratio.

The proposed regulations provide that partnership QBAI is the sum of the tested income CFC's share of the partnership's adjusted basis in partnership specified tangible property. See proposed § 1.951A-3(g)(2)(i). A comment recommended that the final regulations clarify that the adjusted basis in partnership specified tangible property includes any basis adjustment under section 743(b). In response to this comment, the final regulations clarify that an adjustment under section 743(b) to the adjusted basis in partnership specified tangible property with respect to a tested income CFC is taken into account in determining the tested income CFC's partner adjusted basis in the partnership specified tangible property. See § 1.951A-3(g)(3) and (7). In addition, to ensure that the adjusted basis in property other than tangible property is not inappropriately shifted to tangible property for purposes of determining QBAI, the final regulations provide that basis adjustments to partnership specified tangible property under section 734(b) are taken into account only if they are basis adjustments under section 734(b)(1)(B) or 734(b)(2)(B) attributable to distributions of tangible property or basis adjustments under section 734(b)(1)(A) or 734(b)(2)(A) by reason of gain or loss recognized by a distributee partner under section 731(a). See § 1.951A-3(g)(6).

A comment also requested that the final regulations clarify that a CFC's QBAI is increased not only for partnership specified tangible property owned by partnerships in which the CFC is a direct partner, but also for lower-tier partnerships in which the CFC indirectly owns an interest through one or more upper-tier partnerships. The final regulations make this clarification. See § 1.951A-3(g)(1).

Finally, a comment suggested that, under section 951A(d)(3) and the proposed regulations, a disposition of a partnership interest by a tested income CFC could result in the CFC including its distributive share of partnership income in its gross tested income, but not taking into account any of the partnership's basis in partnership specified tangible property for purposes of calculating the CFC's QBAI. Under section 951A(d)(3) and proposed § 1.951A-3(g)(1), if a CFC holds an interest in a partnership at the close of the taxable year of the CFC, the CFC takes into account its share of a partnership's adjusted basis in certain tangible property for QBAI purposes. However, neither section 951A(d)(3) nor the proposed regulations have a rule that would allow a tested income CFC to increase its QBAI for its share of partnership QBAI if the tested income CFC owned the partnership interest for part of the year but not at the close of the CFC taxable year. However, a partner that disposes of its entire partnership interest before the close of the CFC taxable year could have a distributive share of partnership income if the partnership taxable year closes before the close of the CFC taxable year, including by reason of the disposition itself. See section 706(c)(2)(A) (taxable year of partnership closes with respect to partner whose entire interest terminates, including by reason of a disposition).

The Treasury Department and the IRS agree that a partner that has a distributive share of income from a partnership should also be permitted partnership QBAI with respect to the partnership. Therefore, the final regulations are revised to provide that a partner need only hold an interest in a partnership during the CFC inclusion year to have partnership QBAI with respect to the partnership. See § 1.951A-3(g)(1). The final regulations also provide that section 706(d) applies to determine a tested income CFC's partner adjusted basis in partnership specified tangible property owned by a partnership if there is a change in the tested income CFC's interest in the partnership during the CFC inclusion year. See § 1.951A-3(g)(3)(i).

F. Disregard of Basis in Specified Tangible Property Held Temporarily

Section 951A(d)(4) authorizes the issuance of regulations or other guidance that the Secretary determines are appropriate to prevent the avoidance of the purposes of section 951A(d), including regulations or other guidance which provide for the treatment of property that is transferred, or held, temporarily. The proposed regulations provide that if a tested income CFC (“acquiring CFC”) acquires specified tangible property with a principal purpose of reducing the GILTI inclusion amount of a U.S. shareholder for any U.S. shareholder inclusion year, and the tested income CFC holds the property temporarily but over at least the close of one quarter, the specified tangible property is disregarded in determining the acquiring CFC's average adjusted basis in specified tangible property for purposes of determining the acquiring CFC's QBAI for any CFC inclusion year during which the tested income CFC held the property (the “temporary ownership rule”). See proposed § 1.951A-3(h)(1). If an acquisition of specified tangible property would, but for the temporary ownership rule, reduce the GILTI inclusion amount of a U.S. shareholder, then the property is “per se” treated as temporarily held and acquired with a principal purpose of reducing the GILTI inclusion amount of a U.S. shareholder if the tested income CFC holds the property for less than a 12-month period that includes at least the close of one quarter during its taxable year (the “12-month per se rule”). See id. Therefore, the specified tangible property is disregarded under the proposed regulations for purposes of determining QBAI.

Although some comments supported the temporary ownership rule and, in particular, stated that the principal purpose standard was a reasonable interpretation of section 951A(d)(4), many comments asserted that it was overbroad. Comments expressed particular concern with the scope of the 12-month per se rule, noting for example that it could (i) apply to transactions not motivated by tax avoidance such as ordinary course transactions, (ii) require burdensome asset-level tracking of CFC property, and (iii) lead to uncertain return filing positions or financial accounting volatility if property acquired by a CFC has not yet been held for 12 months when a U.S. shareholder files its return or publishes a financial statement.

Comments suggested various ways to minimize the scope of the temporary ownership rule, including (i) eliminating the 12-month per se rule; (ii) converting the 12-month per se rule into a rebuttable presumption; (iii) providing an exception for property transferred among related CFCs owned by a U.S. shareholder when there is no decrease in that shareholder's GILTI inclusion amount (for this purpose, treating a consolidated group as a single entity); (iv) providing that, for purposes of applying the 12-month per se rule, a CFC's holding period in property received in a nonrecognition transaction include a tacked holding period under section 1223(2); (v) providing de minimis or ordinary course transaction exceptions; (vi) excepting acquisitions of property that result in effectively connected income or subpart F income to the transferor; (vii) tailoring the rule's application depending on whether property is acquired from or transferred to unrelated parties; and (viii) establishing a period of ownership that will not be considered temporary.

In response to the comments, the Treasury Department and the IRS have determined that it is appropriate to narrow the scope of the temporary ownership rule, and that the following changes strike the appropriate balance between mitigating the compliance burden and identifying transactions that have the potential to avoid the purposes of section 951A(d). First, the final regulations make certain technical changes that are intended to refine and clarify the application of the temporary ownership rule. For example, the rule applies, in part, based on a principal purpose of increasing the DTIR of a U.S. shareholder (“applicable U.S. shareholder”) and, for this purpose, certain related U.S. persons are treated as a single applicable U.S. shareholder. See § 1.951A-3(h)(1)(i) and (vi). Further, in response to comments, the final regulations clarify that property held temporarily over a quarter close is subject to the temporary ownership rule only if the holding of the property over the quarter close would, without regard to the temporary ownership rule, increase the DTIR of an applicable U.S. shareholder for its taxable year. See § 1.951A-3(h)(1)(i).

The final regulations also clarify that a CFC's holding period for purposes of this rule does not include the holding period for which the property was held by any other person under section 1223. See § 1.951A-3(h)(1)(v). The final regulations do not adopt the request to permit a tacking of holding periods for purpose of the temporary ownership rule, because temporary acquisitions of property through nonrecognition transactions, particularly between related parties, can artificially increase a U.S. shareholder's DTIR by, for instance, causing the property to be taken into account for an additional quarter close for purposes of calculating QBAI.

The final regulations also modify the 12-month per se rule to make it a presumption rather than a per se rule. Therefore, under the final regulations the temporary ownership rule is presumed to apply only if property is held for less than 12 months. See § 1.951A-3(h)(1)(iv)(A). This presumption may be rebutted if the facts and circumstances clearly establish that the subsequent transfer of the property was not contemplated when the property was acquired by the acquiring CFC and that a principal purpose of the acquisition of the property was not to increase the DTIR of the applicable U.S. shareholder. See id. As a result of this change, a taxpayer generally will know when it files its return whether the temporary ownership rule will apply. In order to rebut the presumption, a taxpayer must attach a statement to the Form 5471 filed with the taxpayer's return for the taxable year of the CFC in which the subsequent transfer occurs disclosing that it rebuts the presumption. See id. In response to a comment, the final regulations include a second presumption that generally provides that property is presumed not to be subject to the temporary ownership rule if held for more than 36 months. See § 1.951A-3(h)(1)(iv)(B).

The final regulations clarify that the adjusted basis in property may be disregarded under the rule for multiple quarter closes. See § 1.951A-3(h)(1)(ii). However, in the case that the temporary holding results in the property being taken into account for only one additional quarter close of a tested income CFC in determining the DTIR of a U.S. shareholder inclusion year, the adjusted basis in the property is disregarded under this rule only as of the first tested quarter close that follows the acquisition. See id.; see also § 1.951A-3(h)(1)(vii)(C) (Example 2) (disregarding the adjusted basis in specified tangible property for a single quarter due to differences in CFC taxable years). This rule ensures that the adjusted basis in property is not inappropriately disregarded in excess of the amount necessary to eliminate the increase in the DTIR of the applicable U.S. shareholder by reason of the temporary holding.

The final regulations also include a safe harbor for certain transfers involving CFCs. See § 1.951A-3(h)(1)(iii). Under the safe harbor, the holding of property as of a tested quarter close is not treated as increasing the DTIR if certain conditions are satisfied. In general, the safe harbor applies to transfers between CFCs that are owned in the same proportion by the U.S. shareholder, have the same taxable years, and are all tested income CFCs. The safe harbor is intended to exempt non-tax motivated transfers from the rule when the temporary holding of the property does not have the potential for increasing the DTIR of an applicable U.S. shareholder. The addition of the safe harbor responds to the comment requesting that the rule be tailored depending on whether the transfers involve related or unrelated parties.

In addition, in response to comments, the final regulations include four new examples to illustrate the application of the rule. See § 1.951A-3(h)(1)(vii). The examples identify a transaction that is not subject to the rule due to the application of the safe harbor, and three transactions that are subject to the rule, including transfers of property between CFCs that have different taxable years, and an acquisition of property by a tested income CFC from a tested loss CFC, which cannot have QBAI pursuant to § 1.951A-3(b) and (c)(1).

The final regulations do not adopt the comments requesting a de minimis or ordinary course transaction exception. The Treasury Department and the IRS have determined that these types of exceptions are unnecessary due to the narrowed and refined scope of the rule in the final regulations, including as a result of converting the 12-month per se rule into a rebuttable presumption, adding the safe harbor, and illustrating certain transactions that are targeted by the rule through new examples. Moreover, because the rule is limited to the temporary holding of depreciable property used in a CFC's trade or business (that is, specified tangible property), the Treasury Department and the IRS do not anticipate that many such assets will be acquired and disposed of in the “ordinary course” of a CFC's business, however that standard is defined.

Finally, the final regulations do not adopt the comment requesting an exception for acquisitions of property that result in effectively connected income or subpart F income to the transferor. The Treasury Department and the IRS have concluded that, unlike the rule that addresses disqualified basis in § 1.951A-2(c)(5) and § 1.951A-3(h)(2), the treatment of gain recognized by the transferor (if any) is not relevant for purposes of determining whether it is appropriate to take into account specified tangible property held temporarily for purposes of determining QBAI. Nothing in section 951A(d)(4) or the legislative history suggests that transfers of property that result in income or gain that is subject to U.S. tax should be exempt from the rule. Indeed, the policy concern underlying this rule—the temporary holding of specified tangible property with a principal purpose of increasing the DTIR of a U.S. shareholder—is present regardless of whether the basis in the specified tangible property reflects gain that is subject to U.S. tax.

G. Determination of Disqualified Basis

The determination of disqualified basis is relevant for purposes of both the rule in § 1.951A-2(c)(5) (allocating deductions attributable to disqualified basis to residual CFC gross income) and the rule in § 1.951A-3(h)(2) (disregarding disqualified basis for purposes of calculating QBAI). This part V.G of the Summary of Comments and Explanation of Revisions section describes comments and revisions related to the computation of disqualified basis both for purposes of § 1.951A-2(c)(5) and § 1.951A-3(h)(2). For other comments and revisions related to the computation of disqualified basis discussed in the context of the application of § 1.951A-2(c)(5), see part IV.E.3 and 4 of this Summary of Comments and Explanation of Revisions section.

As described in part IV.E.1 of this Summary of Comments and Explanation of Revisions section, the proposed regulations define “disqualified basis” in property as the excess of the property's adjusted basis immediately after a disqualified transfer, over the sum of the property's adjusted basis immediately before the disqualified transfer and the qualified gain amount with respect to the disqualified transfer. See proposed § 1.951A-3(h)(2)(ii)(A). In addition, the proposed regulations define “disqualified transfer” as a transfer of property by a transferor CFC during a transferor CFC's disqualified period to a related person in which gain was recognized, in whole or in part. See proposed § 1.951A-3(h)(2)(ii)(C). One comment recommended that the definition of disqualified transfer not be expanded to include transfers of property to unrelated persons. The final regulations do not modify the definition of disqualified transfer, and therefore the term continues to be limited to transfers of property by a CFC to a related person. See § 1.951A-3(h)(2)(ii)(C)(2).

A comment noted that the proposed regulations do not explain whether the computation of disqualified basis in property takes into account basis adjustments under section 743(b) or section 734(b) allocated to that property under section 755 during the disqualified period. The final regulations clarify that adjustments under sections 732(d), 734(b), and 743(b) can create, increase, or reduce disqualified basis in property. See § 1.951A-3(h)(2)(ii)(A) and (B).

The proposed regulations provide that disqualified basis may be reduced or eliminated through depreciation, amortization, sales or exchanges, section 362(e), and other methods. See proposed § 1.951A-3(h)(2)(ii)(A). The final regulations clarify the circumstances under which disqualified basis is reduced. Specifically, the final regulations provide that disqualified basis in property is reduced to the extent that a deduction or loss attributable to the disqualified basis in the property is taken into account in reducing gross income, including any deduction or loss allocated to residual CFC gross income by reason of the rule in § 1.951A-2(c)(5). See § 1.951A-3(h)(2)(ii)(B)(1)(i).

The proposed regulations provide that, if the adjusted basis in property with disqualified basis and adjusted basis other than disqualified basis is reduced or eliminated, then the disqualified basis in the property is reduced or eliminated in the same proportion that the disqualified basis bears to the total adjusted basis in the property. See proposed § 1.951A-3(h)(2)(ii)(A). The final regulations adopt this rule without substantial modification, except that the final regulations provide a special rule where a loss is recognized on a taxable sale or exchange. See §§ 1.951A-2(c)(5)(ii) and 1.951A-3(h)(2)(ii)(B)(1)(i). In the case of a loss recognized on a taxable sale or exchange of the property, the loss is treated as attributable to disqualified basis to the extent thereof. See id. Therefore, to the extent of the disqualified basis, the loss on the sale is allocated to residual CFC gross income and the disqualified basis in the property is reduced.

A comment noted that the proposed regulations do not specify when the proportion of the disqualified basis to the total adjusted basis in the property is determined for purposes of determining the reduction to disqualified basis. The comment recommended that the Treasury Department and the IRS clarify that this proportion is determined immediately after the disqualified transfer and does not change throughout the useful life of the property absent a subsequent disqualified transfer. The final regulations do not adopt this recommendation, because the proportion of disqualified basis to total adjusted basis in property can change by reason of one or more transactions subsequent to a disqualified transfer. For instance, a loss recognized on a taxable sale of property with disqualified basis and adjusted basis other than disqualified basis, which reduces disqualified basis to the extent of the loss under § 1.951A-3(h)(2)(ii)(B)(1)(i), will have the effect of decreasing the proportion of disqualified basis to total adjusted basis. See, generally, 1.951A-3(h)(2)(ii)(B) and this part V.G of the Summary of Comments and Explanation of Revisions for additional adjustments to disqualified basis.

A comment recommended that the Treasury Department and the IRS clarify that depreciation or amortization that is disregarded for purposes of determining tested income or tested loss under proposed § 1.951A-2(c)(5) nonetheless reduces the adjusted basis in the property. The final regulations do not disregard a deduction or loss attributable to disqualified basis, but rather allocate and apportion such deduction or loss to residual CFC gross income. Depreciation or amortization that is allocated and apportioned to residual CFC gross income continues to reduce the adjusted basis in the property in accordance with section 1016(a)(2). Accordingly, clarification that any depreciation or amortization attributable to disqualified basis in property reduces adjusted basis in the property is unnecessary.

Disqualified basis in property is generally an attribute specific to the property itself, rather than an attribute of a CFC or a U.S. shareholder with respect to the property. The final regulations, however, provide rules to treat basis in other property as disqualified basis if such basis was determined, in whole or in part, by reference to the basis in property with disqualified basis. See § 1.951A-3(h)(2)(ii)(B)(2). These rules are intended to prevent taxpayers from eliminating disqualified basis in nonrecognition transactions that would otherwise have the effect of granting taxpayers the benefit of the disqualified basis. This could occur, for example, if property with disqualified basis is transferred in a nonrecognition transaction, such as a like-kind exchange under section 1031, in exchange for other depreciable property. In that case, a portion of the basis in the newly acquired property is treated as disqualified basis. Also, disqualified basis may be duplicated through certain nonrecognition transactions. For example, if property with disqualified basis is transferred in a section 351 exchange, both the stock received by the transferor and the property received by the transferee will have disqualified basis, in each case determined by reference to the disqualified basis in the property in the hands of the transferor immediately before the transaction. See § 1.951A-3(h)(2)(ii)(B)(2)(ii). The final regulations also provide that basis arising from other transactions, such as distributions of property from a partnership to a partner, can create disqualified basis in property to the extent the transaction has the effect of shifting disqualified basis from one property to another. See § 1.951A-3(h)(2)(ii)(B)(2)(i). This might occur, for example, if low-basis property is distributed in liquidation of a high-basis partner under section 732(b) resulting in a decrease to disqualified basis in other partnership property under section 734(b)(2)(B). See § 1.951A-3(h)(2)(iii)(D) Example 4.

The final regulations also clarify how disqualified basis is disregarded under § 1.951A-3(h)(2)(i) in the case of dual use property and partnership specified tangible property for purposes of determining QBAI and partnership QBAI, respectively. The portion of the adjusted basis in dual use property with disqualified basis that is taken into account for determining QBAI is the average adjusted basis in the property, multiplied by the dual use ratio, and then reduced by the disqualified basis in the property. See § 1.951A-3(h)(2)(i)(B); see also § 1.951A-3(d)(4) Example. For purposes of determining partnership QBAI, a CFC's partner adjusted basis with respect to partnership specified tangible property with disqualified basis is first determined under the general rules of § 1.951A-3(g)(3)(i) and then reduced by the partner's share of the disqualified basis in the property. See § 1.951A-3(h)(2)(i)(C). In either case, the allocation and apportionment rules of § 1.951A-2(c)(5) are not taken into account for purposes of applying the dual use ratio and the proportionate share ratio to determine the amount of the adjusted basis in property that is reduced by the disqualified basis. See § 1.951A-3(h)(2)(i)(B) and (C).

The Treasury Department and the IRS request comments on the application of the rules that reduce or increase disqualified basis including, for example, how the rules should apply in an exchange under section 1031 where property with disqualified basis is exchanged for property with no disqualified basis.

VI. Comments and Revisions to Proposed § 1.951A-4—Tested Interest Expense and Tested Interest Income

A. Determination of Specified Interest Expense Under Netting Approach

Section 951A(b)(2)(B) reduces net DTIR of a U.S. shareholder by interest expense that reduces tested income (or increases tested loss) for the taxable year of the shareholder to the extent the interest income attributable to such expense is not taken into account in determining such shareholder's net CFC tested income. The proposed regulations adopt a netting approach to determine the amount of interest expense of a U.S. shareholder described in section 951A(b)(2)(B) (“specified interest expense”), defining such amount as the excess of such shareholder's pro rata share of “tested interest expense” of each CFC over its pro rata share of “tested interest income” of each CFC. See proposed § 1.951A-1(c)(3)(iii).

Several comments agreed with the adoption of the netting approach, principally on the grounds of administrability and policy. However, one comment noted that the netting approach for determining specified interest expense is potentially more favorable to taxpayers than permitted by the statute because it provides that specified interest expense is reduced by all interest income included in the tested income of the U.S. shareholder (subject to certain exceptions), even if earned from unrelated parties.

The Treasury Department and the IRS have determined that the netting approach appropriately balances administrability concerns with the purpose and language of section 951A(b)(2)(B). As discussed in the preamble to the proposed regulations, the netting approach avoids the complexity related to a tracing approach, under which a U.S. shareholder's pro rata share of each item of interest expense of a CFC would have to be matched to the shareholder's pro rata share of the interest income attributable to such interest expense received by a CFC. Furthermore, the amount of specified interest expense should, in most cases, be the same whether determined under a netting approach or under a tracing approach. In this regard, while the netting approach does not require a factual link between the interest income and interest expense, only interest income included in gross tested income, other than income included by reason of section 954(h) or (i) (that is, “qualified interest income”), is taken into account for this purpose. Because interest income is generally FPHCI under section 954(c)(1)(A) and qualified interest income is not taken into account under the netting approach, interest income taken into account under the netting approach is generally limited to interest income that is excluded from subpart F income by reason of section 954(c)(3) or (6). Furthermore, because the exceptions under section 954(c)(3) and (6) apply only to interest income paid or accrued by related party foreign corporations, both the interest income excluded by reason of section 954(c)(3) or (6) and the interest expense attributable to such interest income will generally be taken into account in determining the net CFC tested income of either the same U.S. shareholder or a related U.S. shareholder. Accordingly, the final regulations retain the netting approach for determining specified interest expense, with certain modifications described in part VI.B through D of this Summary of Comments and Explanation of Revisions section. See § 1.951A-1(c)(3)(iii).

B. Definition of Tested Interest Expense and Tested Interest Income

For purposes of determining specified interest expense, “tested interest expense” is defined in the proposed regulations as interest expense paid or accrued by a CFC that is taken into account in determining the tested income or tested loss of the CFC, reduced by the qualified interest expense of the CFC. See proposed § 1.951A-4(b)(1)(i). For this purpose, “interest expense” is defined as any expense or loss treated as interest expense under the Code or regulations, and any other expense or loss incurred in a transaction or series of integrated or related transactions in which the use of funds is secured for a period of time if such expense or loss is predominantly incurred in consideration of the time value of money. See proposed § 1.951A-4(b)(1)(ii). The proposed regulations include similar definitions for “tested interest income” and “interest income.” See proposed § 1.951A-4(b)(2)(i) and (ii).

One comment asserted that the concepts of “predominantly incurred in consideration of the time value of money” and “predominantly derived from consideration of the time value of money” are new and unclear, and lack analogies in other authorities. The comment also stated that this new standard is further complicated by references to “a transaction or series of integrated or related transactions.” Other comments asserted that creating a new standard for interest expense and interest income specifically for specified interest expense would result in additional confusion and complexity. Another comment questioned the inclusion of interest equivalents in the definition of interest in the proposed regulations and noted that, because the definition covers both interest income and interest expense, there is a particular risk of whipsaw to the government unless the authority for the regulations is clear. Some comments recommended that the final regulations replace the definitions of interest expense and interest income in the proposed regulations with references to interest expense or interest income under any provision of the Code or regulations, or as a consequence of issuing or holding an instrument that is treated as indebtedness for Federal income tax purposes, such as instruments characterized as indebtedness under judicial factors or administrative guidance, or payments “equivalent to interest.”

The Treasury Department and the IRS did not intend to create a new standard of interest solely for purposes of determining specified interest expense. In this regard, the reduction of net DTIR by specified interest expense under section 951A(b)(2)(B) and the limitation on business interest under section 163(j) are meant to achieve similar policy goals, namely preventing certain interest expense in excess of interest income from being taken into account in determining taxable income. Further, because the amount of interest expense subject to each of these provisions is determined, in part, by reference to interest income received, each of these provisions need clear and consistent definitions of both interest expense and interest income, including when and to what extent transactions that result in a financing from an economic perspective may be treated as generating interest expense and interest income. Finally, the relevant terms used in each provision—“interest expense” and “interest income” in section 951A(b)(2)(B) and “business interest” and “business interest income” in section 163(j)—do not differ meaningfully in their respective contexts and therefore do not necessitate different definitions. As a result of the foregoing, and in order to reduce administrative complexity, the Treasury Department and the IRS have determined that taxpayers and the government would benefit from the application of a single definition of interest for both section 951A(b)(2)(B) and section 163(j) (rather than the application of two partially overlapping, but ultimately different standards). Accordingly, the final regulations define “interest expense” and “interest income” by reference to the definition of interest expense and interest income under section 163(j). See § 1.951A-4(b)(1)(ii) and (2)(ii).

The regulations under section 163(j), when finalized, will address comments on the validity of the definition of interest expense and interest income that are used in those regulations. Because the final regulations adopt this definition for purposes of determining specified interest expense, the discussion in the regulations under section 163(j) will, by extension, address the validity of the definitions as used in these final regulations.

Finally, the definition of tested interest expense is revised in the final regulations to mean interest expense that is “allocated and apportioned to gross tested income” of a CFC under § 1.951A-2(c)(3). See § 1.951A-4(b)(1)(i). This revision does not reflect a substantive change to the definition in the proposed regulations—interest expense “taken into account in determining the tested income or tested loss”—but rather is intended to more clearly articulate that definition.

C. Determination of Qualified Interest Expense and Qualified Interest Income

The proposed regulations provide that, for purposes of determining the specified interest expense of a U.S. shareholder, the tested interest expense and tested interest income of a “qualified CFC” are reduced by its “qualified interest expense” and “qualified interest income,” respectively. See proposed § 1.951A-4(b)(1) and (2). The reduction for qualified interest expense and qualified interest income is intended to neutralize the effect of interest expense and interest income attributable to the active conduct of a financing or insurance business on a U.S. shareholder's net DTIR. For example, absent the rule for qualified interest expense, the third-party interest expense of a captive finance company—to the extent its interest expense exceeds its interest income—could inappropriately increase specified interest expense (and thus reduce the net DTIR) of its U.S. shareholder. Alternatively, under a netting approach to calculating specified interest expense, the third-party interest income of a captive finance company—to the extent its interest income exceeds interest expense—could inappropriately reduce the specified interest expense (and thus increase the net DTIR) of its U.S. shareholder.

For purposes of these rules, the proposed regulations define a “qualified CFC” as an eligible controlled foreign corporation (within the meaning of section 954(h)(2)) or a qualifying insurance company (within the meaning of section 953(e)(3)). See proposed § 1.951A-4(b)(1)(iv). Further, “qualified interest income” is defined as interest income included in the gross tested income of the qualified CFC that is excluded from FPHCI by reason of section 954(h) or (i). See proposed § 1.951A-4(b)(2)(iii). The proposed regulations define “qualified interest expense” as the portion of the interest expense of a qualified CFC, which portion is determined based on a two-step approach. First, a qualified CFC's interest expense is multiplied by a fraction, the numerator of which is the CFC's average basis in assets which give rise to income excluded from FPHCI by reason of section 954(h) or (i), and the denominator is the CFC's average basis in all its assets. See proposed § 1.951A-4(b)(1)(iii)(A). Second, the product of the first step is reduced by the interest income of the qualified CFC that is excluded from FPHCI by reason of section 954(c)(3) or (6). See proposed § 1.951A-4(b)(1)(iii)(B). This two-step approach effectively treats all interest expense of a qualified CFC as attributable ratably to the assets of the qualified CFC that give rise to income excluded from FPHCI by reason of section 954(h) and (i), but then traces such interest expense, after attribution to such assets, to any interest income received from related CFCs to the extent thereof.

A comment indicated that the two-step approach in the proposed regulations can understate the amount of qualified interest expense. Specifically, the comment noted that the proposed regulations include related party receivables in the denominator of the fraction under the first step, thus diluting the fraction and resulting in less qualified interest expense, and then interest income from such receivables further reduce qualified interest expense dollar-for-dollar under the second step. The comment recommended that, to avoid double counting, related party receivables should be excluded from the fraction in the first step.

The Treasury Department and the IRS agree with the comment that, under the two-step approach to the proposed regulations, related party receivables are effectively double-counted, and therefore the final regulations eliminate the second step reduction for interest income included in the gross tested income of a qualified CFC that is excluded from FPHCI by reason of section 954(c)(3) or (6). See § 1.951A-4(b)(1)(iii)(A). This revision ensures that a related party receivable is not double-counted in the determination of qualified interest expense, and thus qualified interest expense as calculated under the final regulations more accurately reflects the interest expense incurred to earn income earned from unrelated parties in an active financing or insurance business. Further, the Treasury Department and the IRS preferred the elimination of the second step reduction for resolving the double-counting issue, rather than the recommended alternative of excluding related party receivables from the fraction in the first step, because the elimination of an additional step substantially simplifies the calculation of qualified interest expense.

In addition, with regard to the effect of related party receivables on the computation of qualified interest expense, the final regulations clarify that a receivable that gives rise to income that is excludible from FPHCI by reason of section 954(c)(3) or (6) is excluded from the numerator of the fraction (that is, the receivable is not a “qualified asset” within the meaning of § 1.951A-4(b)(1)(iii)(B), a new term in the final regulations), notwithstanding that such receivable may also give rise to income excluded from FPHCI by reason of section 954(h) or (i). See § 1.951A-4(b)(1)(iii)(B)(2). Similarly, the final regulations clarify that interest income that is excludible from FPHCI by reason of section 954(c)(3) or (6) is excluded from qualified interest income, notwithstanding that such income may also be excluded from FPHCI by reason of section 954(h) or (i). See § 1.951A-4(b)(2)(iii)(B). These clarifications ensure that the computation of qualified interest income and qualified interest expense is determined by reference only to interest expense and interest income attributable to a CFC's active conduct of a financing or insurance business with unrelated persons.

A comment recommended that, for purposes of determining the amount of qualified interest expense of a CFC, instruments or obligations that give rise to interest income derived by active securities and derivatives dealers that is excluded from FPHCI under section 954(c)(2)(C) should also be included in the numerator for calculating qualified interest expense. The final regulations adopt this recommendation by including such instruments or obligations in the definition of qualified assets. See § 1.951A-4(b)(1)(iii)(B)(1). Similarly, interest income excluded from FPHCI under section 954(c)(2)(C) is included in the definition of qualified interest income. See § 1.951A-4(b)(2)(iii)(A).

A comment suggested that the benefit to some U.S. shareholders from the exclusion for qualified interest expense may not justify the difficulty and expense to determine the amount excluded. Therefore, the comment recommended that the final regulations provide taxpayers the ability to either establish the amount of their qualified interest expense or, alternatively, to assume that none of their interest expense constitutes qualified interest expense. The Treasury Department and the IRS agree that taxpayers should not be required to reduce their CFCs' tested interest expense by their CFCs' qualified interest expense if the taxpayer determines that the value of such reduction is outweighed by the cost of compliance. Accordingly, the final regulations provide that a CFC's qualified interest expense is taken into account only to the extent established by the CFC. See § 1.951A-4(b)(1)(iii)(A). Thus, if a CFC does not establish an amount of qualified interest expense, the taxpayer can assume that none of the CFC's interest expense is qualified interest expense. However, regardless of whether a CFC avails itself of the reduction for qualified interest expense, the exclusion for qualified interest income is mandatory. See § 1.951A-4(b)(2)(iii)(A).

A comment recommended an exception from the qualified interest rules for a CFC that is a qualified insurance company under section 954(i), or in the alternative, an exception from the qualified interest rules for any CFC that is part of a financial services group defined in section 904(d)(2)(C)(ii), with the result that all interest income and interest expense of such CFCs would be tested interest income and tested interest expense taken into account in determining a U.S. shareholder's specified interest expense. The comment speculated that the qualified interest rules may have been crafted to address a CFC involved in a financial services business that was not a member of a business group primarily engaged in a financial services business. The Treasury Department and the IRS decline to adopt this recommendation. The qualified interest rules are intended to neutralize the effect of an active finance business or an active business of a CFC on the specified interest expense (and thus net DTIR) of its U.S. shareholder, irrespective of whether the CFC is a member of a business group primarily engaged in such activities. In contrast, the recommended exception would permit interest income from an active finance business or active insurance business in excess of the associated interest expense to net against other interest expense in the computation of specified interest expense.

The same comment also explained that some foreign financial service groups borrow externally through a holding company to fund their qualifying insurance company subsidiaries that earn qualified interest income. The comment noted that the proposed regulations create a mismatch between the treatment of the interest income of the subsidiaries, which is qualified interest income of a qualified CFC and thus not taken into account in calculating specified interest expense, and the interest expense of the holding company, which is not qualified interest expense of a qualified CFC and thus is taken into account in calculating specified interest expense. To address this mismatch, the final regulations eliminate the term “qualified CFC.” Therefore, if a holding company that is not engaged in an active financing or insurance business borrows to fund the activities of subsidiaries that are engaged in an active financing or insurance business, the interest expense of the holding company may constitute qualified interest expense and thus be disregarded in determining specified interest expense. In this regard, the final regulations retain the rule that the adjusted basis in stock of a subsidiary is treated as basis in a qualified asset to the extent that the assets of the subsidiary are qualified assets. See § 1.951A-4(b)(1)(iii)(B)(3). In addition, the final regulations provide a new rule that treats a CFC that owns 25 percent or more of the capital or profits interest in a partnership as owning its attributable share of any property held by the partnership, as determined under the principles of § 1.956-4(b). See § 1.951A-4(b)(1)(iii)(B)(4). Therefore, under the final regulations, whether, and to what extent, the interest expense of a CFC is qualified interest expense depends entirely on the nature of the assets it holds directly and indirectly, and not on whether the CFC itself is engaged in an active financing or insurance business.

Finally, the definition of qualified interest expense in the proposed regulations includes a parenthetical that indicates that the fraction for determining qualified interest expense cannot exceed one. See proposed § 1.951A-4(b)(1)(iii). The Treasury Department and the IRS have determined that, because the numerator (average basis in qualified assets) is a subset of the denominator (average basis in all assets), this fraction can never exceed one, even without regard to the parenthetical. Therefore, the final regulations eliminate the parenthetical in the definition of qualified interest expense as surplusage. See § 1.951A-4(b)(1)(iii)(A).

D. Interest Expense Paid or Accrued by a Tested Loss CFC

Under the proposed regulations, tested interest expense includes interest expense paid or accrued by a tested loss CFC, notwithstanding that the proposed regulations provide that a tested loss CFC has no QBAI. See proposed § 1.951A-3(b) and § 1.951A-4(b)(1). As discussed in part V.A of this Summary of Comments and Explanation of Revisions section, the final regulations continue to provide that a tested loss CFC has no QBAI. See § 1.951A-3(b). Comments recommended that, if the rule excluding the QBAI of a tested loss CFC were retained, the final regulations should also exclude all interest expense of a tested loss CFC from the calculation of tested interest expense. Comments asserted that exempting interest expense of tested loss CFCs from the calculation of specified interest expense, in conjunction with the exclusion of the QBAI of tested loss CFCs, would produce appropriate results, though one comment acknowledged that such a rule might need to be accompanied by an anti-abuse rule. One comment asserted that excluding interest expense of a tested loss CFC would be appropriate under section 951A(b)(2)(B), because that subparagraph refers only to interest expense “taken into account under subsection (c)(2)(A)(ii),” which, according to the comment, describes only deductions taken into account in determining tested income. Another comment recommended that, rather than excluding all the interest expense of a tested loss CFC, the final regulations should exclude the interest expense incurred to fund acquisitions of tangible property held by the tested loss CFC. The comments suggested that including interest expense of a tested loss CFC (or incurred to acquire tangible property of the tested loss CFC), which reduces net DTIR of a U.S. shareholder, while excluding the QBAI of a tested loss CFC, which increases the net DTIR of a U.S. shareholder, results in unfair and asymmetrical treatment of tested loss CFCs.

The final regulations do not adopt the recommendation to exclude all interest expense of a tested loss CFC, because such exclusion would be inconsistent with the text of section 951A(d)(2)(A) and footnote 1563 of the Conference Report and could create an incentive to inappropriately shift interest expense to a tested loss CFC in order to avoid reducing a U.S. shareholder's net DTIR. The reference to section 951A(c)(2)(A)(ii) in section 951A(b)(2)(B) encompasses all deductions properly allocable to gross tested income, including deductions taken into account in determining tested loss. See section 951A(c)(2)(B)(i) (defining tested loss as the excess of deduction described in section 951A(c)(2)(A)(ii) over gross tested income described in section 951A(c)(2)(A)(i)).

However, in response to the comments, the final regulations reduce a tested loss CFC's tested interest expense by its tested loss QBAI amount, an amount equal to 10 percent of the QBAI that the tested loss CFC would have had if it were instead a tested income CFC. See § 1.951A-4(b)(1)(i) and (iv) and (c) Example 5. This rule has the effect of not taking into account the tested interest expense of a tested loss CFC to the extent that such tested interest expense is less than or equal to a notional 10 percent return on the tested loss CFC's tangible assets that are used in the production of gross tested income.

E. Interest Expense Paid or Accrued to a U.S. Shareholder

As discussed in part VI.A of this Summary of Comments and Explanation of Revisions section, the proposed regulations adopt a netting approach with the result that specified interest expense is the excess of a U.S. shareholder's pro rata share of tested interest expense of each CFC over its pro rata share of tested interest income of each CFC. See proposed § 1.951A-1(c)(3)(ii). Several comments recommended that the final regulations exclude interest expense paid by a CFC to a U.S. shareholder or a related U.S. person from the definition of tested interest expense. One comment recommended that this exclusion be applied to a payment of interest to any U.S. person, whereas two comments suggested that this exclusion also apply to interest expense to the extent the related interest income is subject to U.S. tax as effectively connected income or subpart F income. These comments asserted that interest expense should not generally increase specified interest expense to the extent that the related interest income is subject to U.S. tax at the regular statutory rate, at least in the hands of a U.S. shareholder or related person. According to these comments, excluding interest expense under these circumstance would be consistent with the policy of section 951A(b)(2)(B), which does not reduce a U.S. shareholder's net DTIR for a CFC's interest expense to the extent that the related income increases the U.S. shareholder's net CFC tested income.

The final regulations do not adopt these recommendations. Section 951A(b)(2)(B) generally reduces net DTIR of a U.S. shareholder by the full amount of its pro rata share of the interest expense of a CFC, but then provides a limited exception for the CFC's interest expense to the extent the related interest income is taken into account in determining the net CFC tested income of the U.S. shareholder. In effect, the rule generally reduces net DTIR of a U.S. shareholder by its pro rata share of the net external interest expense incurred by its CFCs. Thus, borrowing between commonly-owned CFCs generally does not reduce net DTIR, whereas external borrowing generally does. The statute does not provide a similar exception for any payment of interest to the extent the related interest income is subject to U.S. tax, nor is there any indication in the legislative history of the Act that Congress intended that the Treasury Department and the IRS should provide such an exception. Further, an exception for interest paid to U.S. persons could permit taxpayers to circumvent section 951A(b)(2)(B) by borrowing externally at the U.S. shareholder level and then on-lending the borrowed funds to CFCs. In this case, the borrowing by the U.S. shareholder would not reduce net DTIR, notwithstanding that the borrowing is factually traceable to the acquisition by the CFC of specified tangible property and net DTIR would have been reduced if instead the CFC had borrowed directly from the third party.

VII. Comments and Revisions to Proposed § 1.951A-5—Domestic Partnerships and Their Partners

A. Proposed Hybrid Approach

The proposed regulations provide that, in general, a domestic partnership that is a U.S. shareholder (“U.S. shareholder partnership”) of a CFC (“partnership CFC”) determines a GILTI inclusion amount, and partners of the partnership that are not also U.S. shareholders of the partnership CFC take into account their distributive share of the partnership's GILTI inclusion amount. See proposed § 1.951A-5(b). Partners that are U.S. shareholders of a partnership CFC (“U.S. shareholder partners”), however, do not take into account their distributive share of the partnership's GILTI inclusion amount to the extent determined by reference to the partnership CFC but instead are treated as proportionately owning the stock of the partnership CFC within the meaning of section 958(a) as if the domestic partnership were an aggregate of its partners. To accomplish this result, the proposed regulations, with respect to U.S. shareholder partners, treat the domestic partnership in the same manner as a foreign partnership, which is treated as an aggregate of its partners under section 958(a)(2). As a result, a U.S. shareholder partner determines its GILTI inclusion amount taking into account its pro rata share of any tested item of the partnership CFC. If the U.S. shareholder partnership holds other partnership CFCs in which the partner is not a U.S. shareholder, then a separate GILTI computation is made at the partnership level with respect to such partnership CFCs' tested items, and the partner includes its distributive share of this separately determined GILTI inclusion amount as well. See proposed § 1.951A-5(c). This hybrid approach (“proposed hybrid approach”) of treating a domestic partnership as an entity with respect to partners that are not U.S. shareholders, but as an aggregate of its partners with respect to partners that are U.S. shareholders, is intended to balance the policies underlying GILTI with the relevant statutory provisions. In particular, a domestic partnership is a U.S. person under sections 957(c) and 7701(a)(30) and thus a U.S. shareholder under section 951(b), which suggests that a domestic partnership should generally be treated as an entity for purposes of subpart F. On the other hand, if a domestic partnership were treated strictly as an entity for purposes of section 951A, a domestic partnership with a GILTI inclusion amount would be ineligible for foreign tax credits under section 960(d) or a deduction under section 250 with respect to its GILTI inclusion amount.

In the proposed regulations, the Treasury Department and the IRS rejected an approach that would treat a domestic partnership as an entity with respect to all its partners (“pure entity approach”) for purposes of section 951A, because treating a domestic partnership as the section 958(a) owner of stock in all cases would frustrate the GILTI framework by creating unintended planning opportunities for well-advised taxpayers and traps for the unwary. However, the Treasury Department and the IRS also did not adopt an approach that would treat a domestic partnership as an aggregate with respect to all its partners (“pure aggregate approach”) for purposes of GILTI, because such an approach would be inconsistent with the treatment of domestic partnerships as entities for purposes of subpart F.

B. Comments on Proposed Hybrid Approach

Two comments were received on the treatment of domestic partnerships and their partners under the proposed regulations. These comments raised concerns regarding the procedural and computational complexity of the proposed hybrid approach. The comments highlighted the difficulty that some partnerships would have in determining whether and to what extent its partners are U.S. shareholder partners of partnership CFCs in order to determine whether and with respect to which partnership CFCs to calculate a partnership-level GILTI inclusion amount for each of its partners. In this regard, a partner of a U.S. shareholder partnership may itself be a U.S. shareholder of one or more partnership CFCs, but not a U.S. shareholder of one or more others. According to the comments, the proposed hybrid approach also raises administrability concerns under the centralized partnership audit regime enacted by section 1101 of the Bipartisan Budget Act of 2015, Public Law 114-74 (BBA) as some determinations are made at the partnership level and others at the partner level.

The comments also raised concerns that the determination of a GILTI inclusion amount at the partnership level and the disparate treatment of U.S. shareholder partners and non-U.S. shareholder partners under the proposed hybrid approach leads to uncertainty regarding the application of sections 959 and 961 (regarding PTEP and corresponding basis adjustments) with respect to domestic partnerships and partnership CFCs, basis adjustments with respect to partnership interests and partnership CFCs, and capital accounts determined and maintained in accordance with § 1.704-1(b)(2). For instance, there are no rules in the proposed regulations regarding whether and to what extent a U.S. shareholder partner's capital account in a partnership is adjusted when the U.S. shareholder partner computes its GILTI inclusion amount based on its pro rata shares of tested items of partnership CFCs. The comments noted that if the capital account of a U.S. shareholder partner is not adjusted for its pro rata shares of tested items of a partnership CFC, then the economic arrangement between the U.S. shareholder partner and other partners could be distorted.

Neither comment recommended a pure entity approach as its primary recommendation. One comment supported a pure entity approach over the proposed hybrid approach, although it recommended a pure entity approach only if a pure aggregate approach were not adopted. Another comment recommended that the pure entity approach not be adopted in any case. Both comments noted that the pure entity approach would avoid the complexities inherent in the proposed hybrid approach and conform the treatment of domestic partnerships for GILTI purposes with the treatment under subpart F before the enactment of section 951A. However, the comments noted that a pure entity approach is inconsistent with the purpose of section 951A, which is to compute a single GILTI inclusion amount for a taxpayer by reference to the items of all the taxpayer's CFCs. The comments agreed that the preamble to the proposed regulations articulated valid policy reasons for rejecting the pure entity approach, namely, that such approach presents both an inappropriate planning opportunity as well as a trap for the unwary.

Both comments primarily recommended a pure aggregate approach. Under a pure aggregate approach, a domestic partnership would not have a GILTI inclusion amount, and thus no partner of the partnership would have a distributive share of such amount. Rather, for purposes of determining the partner's GILTI inclusion amount, a partner would be treated as owning directly the stock of CFCs owned by a domestic partnership for purposes of determining its own GILTI inclusion amount. Thus, under a pure aggregate approach, unlike under the proposed hybrid approach or a pure entity approach, a partner that is not a U.S. shareholder of a partnership CFC would not have a pro rata share of the partnership CFC's tested items or a distributive share of a GILTI inclusion amount of the partnership. According to comments, a pure aggregate approach would reduce complexities inherent in the proposed hybrid approach in terms of administration and compliance. A pure aggregate approach would also avoid the disparate and arbitrary effects of a pure entity approach, under which a U.S. shareholder's GILTI inclusion amount may vary significantly depending on whether it owns CFCs through a domestic partnership as opposed to directly or through a foreign partnership. The comments acknowledged that while domestic partnerships have historically been treated as entities for purposes of subpart F, the enactment of section 951A and its reliance on shareholder-level calculations justifies a reconsideration of this approach.

One comment recommended that the pure aggregate approach apply also to the determination of whether a foreign corporation owned by a domestic partnership is a CFC. Under this approach, a domestic partnership would also be treated as a foreign partnership for purposes of determining whether a domestic partnership is a U.S. shareholder of a foreign corporation and therefore whether the foreign corporation is owned in the aggregate more than 50 percent (by voting power or value) by U.S. shareholders. The same comment suggested that if this approach were not adopted, the final regulations should either adopt the proposed hybrid approach or an aggregate approach that would require even non-U.S. shareholder partners to take into account their pro rata shares of tested items of CFCs owned by a domestic partnership. This approach, in contrast to the pure entity approach and the proposed hybrid approach, would permit a partner that is not a U.S. shareholder with respect to a partnership CFC to nonetheless aggregate its pro rata shares of the tested items of such partnership CFC with its pro rata shares of the tested items of any non-partnership CFCs with respect to which the partner is a U.S. shareholder for purposes of determining a single GILTI inclusion amount for the partner.

The other comment recommended that if the pure aggregate approach or the pure entity approach were not adopted, the final regulations adopt an approach under which a domestic partnership would be treated as an entity for purposes of determining its GILTI inclusion amount and each partner's distributive share of such amount, but then each partner's overall GILTI inclusion amount would be adjusted by its separately-computed GILTI inclusion amount with respect to non-partnership CFCs of the partner. This adjustment would be positive to the extent of the partner's net CFC tested income with respect to CFCs owned outside a domestic partnership, but it could be negative if the partner had a “net CFC tested loss” (that is, aggregate pro rata shares of tested loss in excess of aggregate pro rata share of tested income) with respect to such CFCs.

C. Adoption of Aggregate Treatment for Purposes of Determining GILTI Inclusion Amounts

After consideration of the comments received, the Treasury Department and the IRS have decided not to adopt the proposed hybrid approach in the final regulations. Instead, the final regulations adopt an approach that treats a domestic partnership as an aggregate for purposes of determining the level (that is, partnership or partner) at which a GILTI inclusion amount is calculated and taken into gross income. Specifically, the final regulations provide that, in general, for purposes of section 951A and the section 951A regulations, and for purposes of any other provision that applies by reference to section 951A or the section 951A regulations (for instance, sections 959, 960, and 961), a domestic partnership is not treated as owning stock of a foreign corporation within the meaning of section 958(a). See § 1.951A-1(e)(1). Rather, the partners of a domestic partnership are treated as owning proportionately the stock of CFCs owned by the partnership in the same manner as if the partnership were a foreign partnership under section 958(a)(2). See id. Because a domestic partnership is not treated as owning section 958(a) stock for purposes of section 951A, a domestic partnership does not have a GILTI inclusion amount and thus no partner of the partnership has a distributive share of a GILTI inclusion amount. Furthermore, because only a U.S. shareholder can have a pro rata share of a tested item of a CFC under section 951A(e)(1) and § 1.951A-1(d), a partner that is not a U.S. shareholder of a CFC owned by the partnership does not have a pro rata share of any tested item of the CFC. For the reasons discussed in this part VII.C of the Summary of Comments and Explanation of Revisions section, the Treasury Department and the IRS have determined that this approach best reconciles the relevant statutory provisions, the policies underlying GILTI, and the administrative and compliance concerns raised by the comments.

Since the enactment of subpart F, domestic partnerships have generally been treated as entities, rather than as aggregates of their partners, for purposes of determining whether a foreign corporation is a CFC. See § 1.701-2(f) Example 3 (concluding that a domestic partnership that wholly owns a foreign corporation is treated as an entity and the U.S. shareholder of the foreign corporation, and that the foreign corporation is a CFC for section 904 purposes). In addition, domestic partnerships have generally been treated as entities for purposes of determining the U.S. shareholder that has the subpart F inclusion with respect to such foreign corporation. But cf. §§ 1.951-1(h) and 1.965-1(e) (treating certain domestic partnerships owned by CFCs as foreign partnerships for purposes of determining the U.S. shareholder that has a subpart F inclusion with respect to CFCs owned by such domestic partnerships).

The GILTI rules employ the basic subpart F architecture in several regards, such as for purposes of determining a U.S. shareholder's pro rata share of tested items. See section 951A(e)(1). Nevertheless, there is no indication that Congress intended to incorporate the historical treatment of domestic partnerships under subpart F into the GILTI regime, particularly given that respecting a domestic partnership as the owner under section 958(a) of the stock of a CFC for purposes of GILTI would frustrate the statutory framework. In addition, no provision in the Code prescribes the treatment of domestic partnerships for purposes of section 958(a) in determining GILTI.

Given the silence in the statute with respect to the treatment of domestic partnerships for purposes of GILTI, the Act's legislative history, and the overall significance of the GILTI regime with respect to the taxation of CFC earnings after the Act, the Treasury Department and the IRS have determined that it is an appropriate occasion to reexamine whether a domestic partnership should be treated as an entity or an aggregate in determining the owners of section 958(a) stock for purposes of sections 951 and 951A. The 1954 legislative history makes clear that this determination should be based on the policies of the provision at issue. See H.R. Rep. No. 83-2543, at 59 (1954) (Conf. Rep.). In this regard, the Act fundamentally changed the policies relating to the taxation of CFC earnings relative to those in 1962. Moreover, an aggregate approach applies if it is appropriate to carry out the purpose of a provision of the Code, unless an entity approach is specifically prescribed and clearly contemplated by the relevant statute. Cf. § 1.701-2(e).

As discussed in the preamble to the proposed regulations, an aggregate approach to domestic partnerships furthers the purposes of the GILTI regime. It is consistent with the general intent of the GILTI regime to determine tax liability at the U.S. shareholder level on an aggregate basis rather than on a CFC-by-CFC basis. See Senate Explanation at 371 (“The committee believes that calculating GILTI on an aggregate basis, instead of on a CFC-by-CFC basis, reflects the interconnected nature of a U.S. corporation's global operations and is a more accurate way of determining a U.S. corporation's global intangible income.”); see also House Ways and Means Committee, 115th Cong., Rep. on H.R. 1, H.R. Rep. No. 115-409, at 389 (Comm. Print 2017) (“[I]n making this measurement, the Committee recognizes the integrated nature of modern supply chains and believes it is more appropriate to look at a multinational enterprise's foreign operations on an aggregate basis, rather than by entity or by country.”). A pure entity approach undermines this overall framework in two ways. First, under a pure entity approach, well-advised taxpayers might avail themselves of domestic partnerships to segregate tested items in a manner that is inconsistent with the overall framework of section 951A. In this regard, taxpayers generally would lower their tax liability by separating through one or more domestic partnerships their CFCs with high-taxed tested income and tested interest expense from their CFCs with low-taxed tested income, QBAI, and tested losses. Second, a pure entity approach would represent a trap for an unwary taxpayer by, for example, preventing the use of the tested losses of CFCs directly held by a taxpayer to offset the tested income of CFCs held by the taxpayer through one or more domestic partnerships. This result would not occur if the domestic partnership were treated as an aggregate of its partners. In this regard, the proposal to “adjust” a partner's distributive shares of its domestic partnerships' GILTI inclusion amount by the partner's net CFC tested income and the net CFC tested loss calculated with respect to the partner's CFCs held outside the partnership would not fully address these concerns. That is, the partner would be permitted the full benefit of its aggregate pro rata share of tested losses with respect to CFCs outside the partnership, but the specified interest expense with respect to CFCs outside the partnership would be effectively segregated from the QBAI of CFCs inside the partnership (and therefore would not reduce the partner's net DTIR), and vice versa.

In addition, an aggregate approach with respect to section 958(a) furthers the policies of other provisions related to section 951A. The legislative history makes clear that Congress intended for a domestic corporate partner of a domestic partnership to obtain the benefit of a foreign tax credit under section 960(d) and a deduction under section 250 with respect to a GILTI inclusion amount. See Conference Report, at 623, fn. 1517. However, only domestic corporations (not domestic partnerships) are eligible for a foreign tax credit under section 960(d) or a deduction under section 250. Moreover, absent treating a domestic partnership as an aggregate for purposes of section 951A, a domestic corporate partner's inclusion percentage under section 960(d)(2) is determined without regard to any CFC owned by the partnership because such partner has no pro rata share of the tested income of such CFC. See section 960(d)(2)(B) (the denominator of the inclusion percentage of a domestic corporation is the corporation's aggregate pro rata share of tested income amount under section 951A(c)(1)(A)). Therefore, a strict entity approach to section 960(d) might suggest that domestic corporate partners of a domestic partnership are ineligible for foreign tax credits with respect to a GILTI inclusion amount of the partnership. On the other hand, an aggregate approach to domestic partnerships furthers Congressional policy by treating domestic corporate partners as owning (within the meaning of section 958(a)) stock of CFCs owned by domestic partnerships and thus determining the domestic corporate partner's GILTI inclusion amount by reference to CFCs owned by the domestic partnership.

The final regulations treat a domestic partnership as an aggregate of its partners in determining section 958(a) stock ownership by providing that, for purposes of section 951A and the section 951A regulations, a domestic partnership is treated in the same manner as a foreign partnership. See § 1.951A-1(e)(1). For purposes of subpart F, a foreign partnership is explicitly treated as an aggregate of its partners, and rules regarding aggregation of foreign partnerships are relatively well-developed and understood. See section 958(a)(2). Therefore, rather than developing a new standard for the treatment of domestic partnerships as an aggregate, the Treasury Department and the IRS have determined that it would be simpler and more administrable to incorporate the aggregate approach by reference to the rules related to foreign partnerships under section 958(a)(2).

The final regulations do not adopt the recommendation to extend the treatment of a domestic partnership as an aggregate of its partners to the determination of U.S. shareholder and CFC status. The Treasury Department and the IRS have determined that an approach that treats a domestic partnership as an aggregate of its partners for purposes of determining CFC status would not be consistent with the relevant statutory provisions. A domestic partnership is a U.S. person under section 957(c) and section 7701(a)(30) and, therefore, can be a U.S. shareholder under section 951(b). Indeed, when subpart F was enacted in 1962, the legislative history indicated that domestic partnerships generally should be treated as U.S. shareholders. See S. Rep. No. 1881, 87th Cong., 2d Sess. 80 n.1 (1962) (“U.S. shareholders are defined in the bill as `U.S. persons' with 10-percent stockholding. U.S. persons, in general, are U.S. citizens and residents and domestic corporations, partnerships and estates or trusts.”). Furthermore, sections 958(b) and 318(a)(3) treat a partnership (including a domestic partnership) as owning the stock of its partners for purposes of determining whether the foreign corporation is owned more than 50 percent by U.S. shareholders, which suggests that partnerships are treated as entities for purposes of determining ownership under section 958(b). See also sections 958(b) and 318(a)(2) (treating stock owned by a partnership, domestic or foreign, as owned proportionately by its partners).

The final regulations also do not extend aggregate treatment to the determination of the controlling domestic shareholders (as defined in § 1.964-1(c)(5)) of a CFC for purposes of any election made under the section 951A regulations. See § 1.951A-3(e)(3)(ii) (election to use a non-ADS depreciation method for pre-enactment property) and § 1.951A-3(h)(2)(ii)(B)(3) (election to eliminate disqualified basis). As a result, a domestic partnership that satisfies the ownership requirements of § 1.964-1(c)(5) with respect to a CFC, and not its partners, is treated as the controlling domestic shareholder of the CFC and the partnership files the relevant elections with respect to the CFC. The treatment of a domestic partnership as the controlling domestic shareholder reduces the number of persons that need to comply with the rules of § 1.964-1(c)(3), and ensures that any election with respect to a CFC that could affect the tax consequences of a U.S. person that is a partner of a domestic partnership is made by such partnership. Accordingly, the final regulations provide that the aggregation rule for domestic partnerships does not apply for purposes of determining whether a U.S. person is a U.S. shareholder, whether a U.S. shareholder is a controlling domestic shareholder (as defined in § 1.964-1(c)(5)), or whether a foreign corporation is a CFC. See § 1.951A-1(e)(2).

The treatment of domestic partnerships as foreign partnerships in the final regulations is solely for purposes of section 951A and the section 951A regulations and for purposes of any other provision that applies by reference to a GILTI inclusion amount (such as sections 959 and 961). The rule does not affect the determination of ownership under section 958(a) for any other provision of the Code (such as section 1248(a)), nor does it change whether such partner has a distributive share of a domestic partnership's subpart F inclusion under section 951(a). However, the Treasury Department and the IRS are proposing in a notice of proposed rulemaking published in the same issue of the Federal Register as these final regulations to apply a similar aggregate treatment to domestic partnerships for purposes of section 951.

Under section 1373(a), an S corporation is treated as a partnership and its shareholders as partners for purposes of subpart F, including section 951A. Therefore, for purposes of determining a GILTI inclusion amount of a shareholder of an S corporation, under § 1.951A-1(e), the S corporation is not treated as owning stock of a foreign corporation within the meaning of section 958(a) but instead is treated in the same manner as a foreign partnership. The Treasury Department and the IRS are studying the application of section 1373(a) with respect to section 951A, as well as the broader implications of treating S corporations as partnerships for purposes of subpart F. Comments are requested in this regard.

Conforming changes are also made to other aspects of the final regulations to account for the aggregate treatment of domestic partnerships under § 1.951A-1(e). For instance, the proposed regulations provide that, for purposes of determining whether a U.S. shareholder has a pro rata share of an accrual for purposes of sections 163(e)(3)(B)(i) and 267(a)(3)(B), a domestic partnership's pro rata share of the accrual is taken into account only to the extent that U.S. persons include in gross income a distributive share of the domestic partnership's GILTI inclusion amount. See proposed § 1.951A-5(c)(2). This rule is no longer necessary under the final regulations because a domestic partnership does not have a GILTI inclusion amount, and partners that are U.S. shareholders have their own pro rata shares of the accrual. Therefore, this rule is eliminated in the final regulations. See § 1.951A-5(c). In addition, the partnership blocker rule is modified such that it no longer applies for purposes of section 951A. See § 1.951-1(h)(1). It is no longer necessary to apply the rule for purposes of section 951A because, for such purposes, a domestic partnership is not treated as owning stock of a foreign corporation within the meaning of section 958(a).

VIII. Comments and Revisions to Proposed § 1.951A-6—Treatment of GILTI Inclusion Amount and Adjustments to E&P and Basis Related to Tested Loss CFCs

A. Increase of E&P by Tested Losses for Purposes of Section 952(c)(1)(A)

Section 951A(c)(2)(B)(ii) provides that section 952(c)(1)(A) is applied by increasing the E&P of a tested loss CFC by the amount of its tested loss. See also proposed § 1.951A-6(d). Comments asserted that proposed § 1.951A-6(d) has the effect of increasing E&P by a tested loss even if, and to the extent, the tested loss does not provide a benefit to a U.S. shareholder because its aggregate pro rata share of tested losses exceeds its aggregate pro rata share of tested income. These comments argued that this result is not appropriate because, based on the heading of section 951A(c)(2)(B)(ii) (“Coordination with subpart F to deny double benefit of losses”), the provision is limited to denying a double benefit from a tested loss (that is, a reduction in both net CFC tested income and subpart F income), and that there can be no such double benefit to the extent that the tested loss does not reduce a U.S. shareholder's net CFC tested income. These comments recommended that proposed § 1.951A-6(d) be modified such that it applies only to a tested loss to the extent the tested loss is “used” within the meaning of proposed § 1.951A-6(e).

The final regulations do not adopt this recommendation. Section 951A(c)(2)(B)(ii), by its terms, increases E&P for purposes of section 952(c)(1)(A) by the amount of any tested loss. There is no indication in the provision or legislative history that limiting the application of section 951A(c)(2)(B)(ii) to a tested loss that reduces net CFC tested income would be appropriate, and the heading of the provision has no legal effect. See section 7806(b). Accordingly, the rule is adopted without modification in § 1.951A-6(b).

B. Treating GILTI Inclusion Amounts as Subpart F Inclusions for Purposes of the Personal Holding Company Rules

A comment requested clarification regarding the treatment of a GILTI inclusion amount for purposes of the personal holding company rules in sections 541 through 547. Section 541(a) imposes a 20-percent tax on the undistributed personal holding company income of a personal holding company. Section 542(a) defines a “personal holding company” as a corporation if at least 60 percent of its adjusted ordinary gross income for the taxable year is personal holding company income and certain ownership requirements are satisfied. Section 543(a) defines “personal holding company income” by reference to certain categories of passive income, including dividends. However, for this purpose, dividends received by a U.S. shareholder from a CFC are excluded from the definition of personal holding company income. See section 543(a)(1)(C). The comment noted that the existing regulations under section 951 provide that for purposes of determining whether a corporate U.S. shareholder is a personal holding company, the character of a subpart F inclusion of such domestic corporation is determined as if the amount that results in the subpart F inclusion were realized directly by the corporation from the source from which it is realized by the CFC. See § 1.951-1(a)(3).

The Treasury Department and the IRS have determined that it would be inappropriate to treat any portion of a GILTI inclusion amount as personal holding company income. A GILTI inclusion amount is determined by reference to income that would have been taxed, if at all, as dividends from CFCs before the enactment of section 951A, which are specifically excluded from the definition of personal holding company income under section 543(a)(1)(C). Further, there is no indication in the legislative history that Congress intended through the enactment of section 951A to substantially change the types of income that would be taken into account in determining personal holding company status. Accordingly, the final regulations clarify that in determining whether a corporate U.S. shareholder is a personal holding company, a GILTI inclusion amount is not treated as personal holding company income (as defined in section 543(a)). See § 1.951A-5(d).

C. Adjustments to Basis Related to Net Used Tested Loss

To eliminate the potential for the duplicative use of a loss, the proposed regulations set forth rules providing for downward adjustments to the adjusted basis in stock of a tested loss CFC to the extent its tested loss was used to offset tested income of another CFC. See proposed § 1.951A-6(e). These adjustments are generally made at the time of a direct or indirect disposition of stock of the tested loss CFC. See proposed § 1.951A-6(e)(1). Comments raised many significant issues with respect to these rules.

The Treasury Department and the IRS remain concerned that, absent basis adjustments, a tested loss can result in the creation of uneconomic or duplicative loss, but have determined that the rules in the proposed regulations related to basis adjustments should not be adopted in these final regulations. Instead, the rules related to basis adjustments, including the comments received with respect to such rules, will be considered in a separate project. Accordingly, the final regulations reserve on the rules related to adjustments to stock of tested loss CFCs. See § 1.951A-6(c). Any rules issued under § 1.951A-6(c) will apply only with respect to tested losses incurred in taxable years of CFCs and their U.S. shareholders ending after the date of publication of any future guidance.

For a discussion of corresponding rules for basis adjustments within a consolidated group, as provided for in proposed §§ 1.1502-13, 1.1502-32, and 1.1502-51, see part IX.C of this Summary of Comments and Explanation of Revisions section.

IX. Comments and Revisions to Proposed §§ 1.1502-13, 1.1502-32, and 1.1502-51—Consolidated Section 951A

A. Calculation of GILTI Inclusion Amount

Section 1502 provides that consolidated return regulations will be promulgated to clearly reflect the income tax liability of a consolidated group and each member of the consolidated group (a “member”). However, in the context of section 951A, clear reflection of the GILTI inclusion amounts of both individual members and the consolidated group as a whole is not feasible. Section 951A requires a U.S. shareholder-level calculation, where, for example, the shareholder's pro rata share of the tested income of one CFC may be offset by its pro rata share of the tested loss or QBAI of another CFC, to produce a smaller GILTI inclusion amount. Accordingly, calculating a member's GILTI inclusion amount on a completely separate-entity basis, solely based on its pro rata share of the items of its CFCs, would clearly reflect the income tax liability of the member. However, such an approach would mean that the consolidated group's GILTI inclusion amount would vary depending on which members own each CFC, particularly in cases in which the CFCs held by some members produce tested income, but the CFCs held by other members produce tested loss. This variability undermines the clear reflection of the income tax liability of the consolidated group as a whole. The Treasury Department and the IRS determined in the proposed regulations that members' GILTI inclusion amounts should be determined in a manner that clearly reflects the income tax liability of the consolidated group and that creates consistent results regardless of which member of a consolidated group owns the stock of the CFCs (“single-entity treatment”). This approach removes incentives for inappropriate planning and also eliminates traps for the unwary.

The proposed regulations accomplish these goals by providing that the GILTI inclusion amount of a member is determined pursuant to a multi-step process. As in the case of a non-member, the GILTI inclusion amount of a member equals the excess (if any) of the member's net CFC tested income over the member's net DTIR for the taxable year. See proposed § 1.951A-1(c)(1) and proposed § 1.1502-51(b). For purposes of determining a member's net CFC tested income, a member's aggregate pro rata share of tested income is determined on a separate-entity basis by aggregating its pro rata share of the tested income of each of its CFCs. See proposed § 1.1502-51(e)(1) and (12). However, a member's aggregate pro rata share of tested loss and its net DTIR for the taxable year is calculated in three steps—first, each member's pro rata share of each tested item other than tested income is determined on a separate-entity basis by reference to its pro rata share of each CFC; second, each member's pro rata share of each tested item other than tested income is aggregated into a consolidated sum; and third, each member is then allocated a portion of the consolidated sum of each such tested item based on its relative amount of tested income (the “aggregation approach”). See proposed § 1.1502-51(e)(2), (3), (4), (5), (7), and (10). The aggregation approach has the effect of determining the aggregate amount of GILTI inclusion amounts of members on a single-entity basis, but then determining each member's share of the consolidated group's aggregate GILTI inclusion amount based on its relative pro rata share of tested income as determined on a separate-entity basis.

The Treasury Department and the IRS received several comments addressing the calculation of a member's GILTI inclusion amount. These comments generally supported single-entity treatment, but they expressed concern about the lack of clear reflection of income at the member level. The concern arises from the movement of the economic benefit (in the GILTI computation) of one member's pro rata share of a tested loss with respect to stock held by the member to other members, including those not holding such stock. The comments considered whether alternative methods could be used that both provide for single-entity treatment and minimize uneconomic results to members. In particular, the comments raised the possibility that the tested loss of a CFC should first offset the tested income of a CFC owned by the same member (the “priority allocation approach”).

One comment evaluated the merits of the priority allocation approach versus the aggregation approach. The comment identified the tension in the section 951A context between clearly reflecting income tax liability at the consolidated group level and doing so at the member level, and it considered possible ways to alleviate this conflict. The comment ultimately endorsed maintaining the approach in proposed § 1.1502-51, due to the additional rules and complexities required to rationalize the priority allocation approach.

Two of the comments proposed similar methods for determining a member's GILTI inclusion amount. One of these comments suggested calculating the consolidated group's GILTI inclusion amount as if members holding CFC stock were divisions of a single corporation, then allocating the resulting consolidated group amount among members based on each member's net CFC tested income. For this purpose, net CFC tested income is calculated in a manner consistent with the priority allocation approach, by allowing the member's tested losses to be used first to offset the same member's tested income. The other comment suggested calculating and allocating the consolidated group's GILTI inclusion amount in the same manner, but would extend application of this method to foreign tax credits with respect to tested income. This second comment proposed using the aggregation approach to determine the amount of such credits available to the consolidated group (and the identity of the CFCs to whom the credits are attributable), but allocating certain basis adjustments in member stock related to such credits under the priority allocation approach. As an alternative, the second comment would base the allocations on the relative amounts of foreign tax credits paid by each member's CFCs.

The Treasury Department and the IRS decline to adopt these comments because they do not produce reasonable results that are consistent with single-entity treatment. In particular, the first of these comments does not provide for single-entity treatment when foreign tax credits are taken into account, instead allowing for wide variation in the availability of foreign tax credits depending on which member of a consolidated group owns the stock of the CFCs. The variation arises because a corporate U.S. shareholder is deemed to pay a portion of the foreign income taxes paid or accrued by its CFCs based on the shareholder's GILTI inclusion amount. See section 960(d). A priority allocation approach, like the separate entity calculations discussed in a preceding paragraph, would change members' GILTI inclusion amounts based on which member owns the stock of the CFCs. By extension, a priority allocation approach would also change the amount of foreign tax credits that are available to the consolidated group based on which member owns the stock of the CFCs. This disparity would allow for tax planning to maximize the availability of foreign tax credits with respect to tested income.

The second of these comments contains proposals that contravene longstanding foreign tax credit principles, by divorcing a member's income inclusion from the member's deemed payments of foreign tax. Absent a GILTI inclusion amount and ownership of a CFC that has paid or accrued foreign taxes on tested income, a U.S. shareholder can claim no foreign tax credits with respect to tested income. And yet under the proposed method, a consolidated group's foreign tax credits may reflect foreign taxes paid or accrued by CFCs of members that have no GILTI inclusion amount. For these reasons, the Treasury Department and the IRS do not adopt this method.

Based on the foregoing, the Treasury Department and the IRS continue to believe that the aggregation approach balances, to the greatest extent possible, the clear reflection of the income tax liability under section 951A of a consolidated group with reasonable results to its individual members. Accordingly, the final regulations generally adopt the aggregation approach from the proposed regulations without substantial changes.

B. Applicability Date for Consolidated Groups

For a discussion of the applicability date for § 1.1502-51, see part XI.A of this Summary of Comments and Explanation of Revisions section.

C. Basis Adjustments to Member Stock

The proposed regulations contain special rules, applicable to consolidated groups, that reflect the downward basis adjustments set forth in proposed § 1.951A-6(e) with respect to the stock of tested loss CFCs. See proposed §§ 1.1502-32(b)(3)(ii)(E) and (b)(3)(iii)(C), and 1.1502-51(c) and (d). As discussed above in part VIII.C of this Summary of Comments and Explanation of Revisions section, the Treasury Department and the IRS have determined that the rules related to basis adjustments for tested loss CFCs should not be adopted in these final regulations and will instead be considered in a separate project. Correspondingly, the special rules for consolidated groups that reflect such rules are likewise reserved. See §§ 1.1502-32(b)(3)(ii)(E) and (b)(3)(iii)(C), and 1.1502-51(c) and (d). These special rules, along with related comments, will be considered in the same project as the rules related to basis adjustments for tested loss CFCs and will apply only to taxable years of U.S. shareholders that are members of a consolidated group ending after the date of publication of the final rules.

D. Portion of Proposed Regulations not Being Finalized

The proposed regulations would treat a member as receiving tax-exempt income immediately before another member recognizes income, gain, deduction, or loss with respect to a share of the first member's stock (the “F adjustment”). See proposed § 1.1502-32(b)(3)(ii)(F). The amount of the tax-exempt income would be determined based in part on the aggregate tested income and aggregate tested losses of the member's CFCs in prior taxable years.

The Treasury Department and the IRS have become aware of serious flaws with the F adjustment. Examples of the problems include unintended and duplicative tax benefits, distortive effects, and possible avoidance of Code provisions and regulations. Therefore, the Treasury Department and the IRS have decided not to finalize the F adjustment. As a result, taxpayers may not rely on the F adjustment. The Treasury Department and the IRS continue to study a number of issues regarding consolidated stock basis in this area.

X. Comments and Revisions to Proposed §§ 1.78-1, 1.861-12(c)(2), and 1.965-7(e) of the Foreign Tax Credit Proposed Regulations

A. Special Applicability Date Under Section 78

The foreign tax credit proposed regulations revise § 1.78-1 to reflect the amendments to section 78 made by the Act, as well as make conforming changes to reflect pre-Act statutory amendments. In addition, the foreign tax credit proposed regulations provide that amounts treated as dividends under section 78 (“section 78 dividends”) that relate to taxable years of foreign corporations that begin before January 1, 2018 (as well as section 78 dividends that relate to later taxable years), are not treated as dividends for purposes of section 245A.

Comments questioned whether the Treasury Department and the IRS have authority to treat section 78 dividends relating to taxable years of foreign corporations beginning before January 1, 2018, as ineligible for the dividends-received deduction under section 245A, which generally applies to certain dividends paid after December 31, 2017. Although some comments acknowledged that allowing a dividends-received deduction for section 78 dividends would provide taxpayers with a double benefit that clearly was not intended by Congress, the comments claimed that the statutory language directly provides for the dividends-received deduction, and therefore the rule applying proposed § 1.78-1(c) to taxable years beginning before January 1, 2018, should be eliminated.

The Treasury Department and the IRS have determined that sections 7805(a), 7805(b)(2), and 245A(g) provide ample authority for the rule and therefore finalize the proposed applicability date without change. Section 7805(a) provides that the Treasury Department and the IRS shall prescribe all needful rules and regulations for the enforcement of title 26, including all rules and regulations as may be necessary by reason of any alteration of law in relation to internal revenue. The enactment of the Act and the addition of section 245A necessitated regulations to ensure that section 78 continues to serve its intended purpose. The purpose of the section 78 dividend is to ensure that a U.S. shareholder cannot effectively both deduct and credit the foreign taxes paid by a foreign subsidiary that are deemed paid by the U.S. shareholder. See Elizabeth A. Owens & Gerald T. Ball, The Indirect Credit § 2.2B1a n.54 (1975); Stanley Surrey, “Current Issues in the Taxation of Corporate Foreign Investment,” 56 Columbia Law Rev. 815, 828 (June 1956) (describing the “mathematical quirk” that necessitated enactment of section 78). Allowing a dividends-received deduction for a section 78 dividend would undermine the purpose of the section 78 dividend because taxpayers would effectively be allowed both a credit and deduction for the same foreign tax. For this reason, section 78 (as revised by the Act) provides that a section 78 dividend is not eligible for a dividends-received deduction under section 245A.

As noted in the preamble to the foreign tax credit proposed regulations, the special applicability date rule under § 1.78-1(c) is necessary to ensure that this principle is consistently applied with respect to a CFC that uses a fiscal year beginning in 2017 as its U.S. taxable year (a “fiscal year CFC”) in order to prevent the arbitrary disparate treatment of similarly situated taxpayers. Otherwise, a U.S. shareholder of a fiscal year CFC would effectively be able to take both a credit and a deduction for foreign taxes by claiming a section 245A deduction with respect to its section 78 dividend. In contrast, section 78 (as revised by the Act) would apply correctly to a U.S. shareholder of a CFC using the calendar year as its U.S. taxable year that was also subject to section 245A.

The special applicability date is also consistent with the grant of authority under section 245A(g) to provide rules as may be necessary or appropriate to carry out the provisions of section 245A. Section 245A was intended to provide for tax-exempt treatment of certain E&P earned through foreign subsidiaries as part of a new participation exemption system. See Conference Report, at 470 (2017) (section 245A “allows an exemption for certain foreign income”). Notably, the amount of a dividend eligible for a dividends-received deduction under section 245A is determined based on the amount of a foreign corporation's “undistributed foreign earnings.” It would be incompatible with the purpose of section 245A to exempt income arising by reason of a section 78 dividend, which is not paid out of a foreign corporation's undistributed foreign earnings but instead represents earnings that could not be distributed since they were used to pay foreign tax.

B. Application of Basis Adjustment for Purposes of Characterizing Certain Stock

Proposed § 1.861-12(c)(2) clarifies certain rules for adjusting the stock basis in a 10 percent owned corporation, including that the adjustment to basis for E&P includes PTEP. Proposed § 1.861-12(c)(2)(i)(B)(2). Additionally, in order to account for the application of section 965(b)(4)(A) and (B), relating to the treatment of reduced E&P of a deferred foreign income corporation and increased E&P of an E&P deficit foreign corporation, proposed § 1.861-12(c)(2)(i)(B)(1)(ii) provides that, for purposes of § 1.861-12(c)(2), a taxpayer determines the basis in the stock of a specified foreign corporation as if it had made the election under § 1.965-2(f)(2), even if the taxpayer did not in fact make the election. However, the taxpayer does not include the amount by which basis with respect to a deferred foreign income corporation is increased under § 1.965-2(f)(2)(ii)(A), because the amount of that increase would be reversed if the increase were by operation of section 961. After issuance of the foreign tax credit proposed regulations, final regulations issued under section 965 (TD 9864, 84 FR 1838 (February 5, 2019)) altered the election under § 1.965-2(f)(2) to allow taxpayers to limit the reduction in basis with respect to an E&P deficit foreign corporation under the election to the amount of the taxpayer's basis in the respective share of stock of the relevant foreign corporation.

One comment requested a special rule with respect to the adjustment to basis for E&P to account for the increase to E&P of an E&P deficit foreign corporation under section 965(b)(4)(B). Alternatively, the comment requested that the adjustment for E&P not include PTEP. However, proposed § 1.861-12(c)(2)(i)(B)(1)(ii) already accounts for the increase in E&P of an E&P deficit foreign corporation under section 965(b)(4)(B) by providing for an equivalent reduction in the adjusted basis of the foreign corporation. Accordingly, the recommendation is not adopted.

Another comment requested that the rule in proposed § 1.861-12(c)(2)(i)(B)(1)(ii) be revised in light of the changes to § 1.965-2(f)(2) to similarly provide that any reductions in basis be limited to the amount of the taxpayer's basis in the 10 percent owned corporation. This comment noted that in the absence of such a rule, the application of proposed § 1.861-12(c)(2)(i)(B)(1)(ii) could reduce the adjusted basis of the stock below zero, which would be inappropriate for purposes of applying the expense allocation rules. The Treasury Department and the IRS agree that, for purposes of applying the expense allocation rules, a taxpayer should not have an adjusted basis below zero in the stock of a 10 percent owned corporation. However, rather than limit the reduction in stock basis to the amount of the taxpayer's basis in the 10 percent owned corporation, the final regulations provide that § 1.861-12(c)(2)(i)(B)(1)(ii) may cause the taxpayer's adjusted basis in the stock of the corporation to be negative, as long as the adjustment for E&P provided for in § 1.861-12(c)(2)(i)(A) increases the taxpayer's adjusted basis to zero or an amount above zero. If the taxpayer's adjusted basis in the 10 percent owned corporation is still below zero after application of § 1.861-12(c)(2)(i)(A)(1) and (2), then for purposes of § 1.861-12, the taxpayer's adjusted basis in the 10 percent owned corporation is zero for the taxable year. Section 1.861-12(c)(2)(i)(A)(3); see also § 1.861-12(c)(2)(i)(C)(3) (Example 3) and (4) (Example 4). The Treasury Department and the IRS have determined that allowing the adjusted basis in stock to be negative before the application of the adjustment for E&P most accurately reflects the value of the stock in the 10 percent owned corporation.

Additionally, these final regulations modify proposed § 1.861-12(c)(2)(i)(B)(1)(ii) to make clear that the adjustment in § 1.861-12(c)(2)(i)(B)(1)(ii) may cause a taxpayer's adjusted basis in stock in the 10 percent owned corporation to be negative, and to account for the changes made to § 1.965-2(f)(2). Specifically, § 1.861-12(c)(2)(i)(B)(1)(ii) now provides that the taxpayer first adjusts its basis in the 10 percent owned corporation as if it did not make the election in § 1.965-2(f)(2)(i) and then, if applicable, adjusts the basis in the 10 percent owned corporation by the amount described in § 1.965-2(f)(2)(ii)(B)(1). These changes are not intended to alter the outcome of the application of the rule to the taxpayer's adjusted basis in the stock of the 10 percent owned corporation as compared to the rule articulated in the foreign tax credit proposed regulations; rather, the changes are intended to make the rule more straightforward for taxpayers to apply and to clarify any ambiguities about the application of the rule where the adjustment exceeded the taxpayer's adjusted basis in the stock. See § 1.861-12(c)(2)(i)(C)(1) (Example 1) and (2) (Example 2).

C. Effect of Section 965(n) Election

Under section 965(n), a taxpayer may elect to exclude the amount of section 965(a) inclusions (reduced by section 965(c) deductions) and associated section 78 dividends in determining the amount of the net operating loss carryover or carryback that is deductible in the taxable year of the inclusions. Section 1.965-7(e)(1), as added by TD 9846, 84 FR 1838 (February 5, 2019), provides that, if the taxpayer makes a section 965(n) election, the taxpayer does not take into account the amount of the section 965(a) inclusions (reduced by section 965(c) deductions) and associated section 78 dividends in determining the amount of the net operating loss for the taxable year.

Proposed § 1.965-7(e)(1)(i), included in the foreign tax credit proposed regulations, provides that the amount by which the section 965(n) election creates or increases the net operating loss for the taxable year is the “deferred amount.” Proposed § 1.965-7(e)(1)(iv)(B) provides ordering rules to coordinate the election's effect on section 172 with the computation of the foreign tax credit limitations under section 904. The foreign tax credit proposed regulations provide that the deferred amount comprises a ratable portion of the deductions (other than the section 965(c) deduction) allocated and apportioned to each statutory and residual grouping for section 904 purposes.

Before the issuance of the foreign tax credit proposed regulations, the Treasury Department and the IRS were aware that some taxpayers were taking the position that the source and separate category of the deferred amount consisted solely of deductions allocated and apportioned to the section 965(a) inclusion. Under this approach, the deferred amount would likely consist primarily of deductions allocated and apportioned to foreign source general category income because that is the likely source and separate category of the section 965(a) inclusion; as a result, the electing taxpayer would generally have a greater amount of foreign source general category income and thus be able to credit more foreign taxes paid or accrued with respect to general category income (relative to the result under the foreign tax credit proposed regulations).

After publication of the foreign tax credit proposed regulations, a comment recommended not finalizing the proposed ordering rules because taxpayers did not have a chance to consider those ordering rules before deciding to make an election under section 965(n). The comment also argued that the foreign tax credit proposed regulations are inconsistent with the statutory language in section 965(n), and with existing rules on the allocation and apportionment of expenses under section 904, to the extent they defer deductions that would be taken against income other than the section 965(a) inclusion. In addition, the comment stated that the foreign tax credit proposed regulations are inconsistent with the operation of section 965 and section 904 to the extent they treat the section 965(a) inclusion net of the section 965(c) deduction, rather than the section 965(a) inclusion without reduction for the section 965(c) deduction, as the gross income in the statutory grouping for section 904 purposes. The comment also suggested that the exclusion of the section 965(c) deductions from the deferred amount was inappropriate. The comment further stated that, if the regulations are finalized as proposed, taxpayers should be allowed to revoke the section 965(n) election. Finally, the comment recommended that proposed § 1.965-7(e)(1)(iv)(B) be revised to refer to allocation of all deductions (other than the net operating loss carryover or carryback to that year that is not allowed by reason of the section 965(n) election), rather than refer solely to allocation of deductions that would have been allowed for the year but for the section 965(n) election.

The final regulations include the ordering rules from the foreign tax credit proposed regulations, with some modifications to take into account the comments. In general, the Treasury Department and the IRS have determined that these rules are consistent with sections 965(n) and 904. Section 965(n) does not modify the generally applicable rules concerning the allocation and apportionment of expenses for section 904 purposes, nor does it provide an ordering rule for determining which deductions create or increase the amount of a current year net operating loss by reason of the section 965(n) election. Section 965(n) applies solely to determine the amount of the net operating loss for the election year and the amount of net operating loss carryover or carryback to that year. It does not require or permit the reallocation of deductions that are allocated and apportioned to the separate category containing the section 965(a) inclusion and associated section 78 dividends, regardless of whether any deductions are deferred by reason of the section 965(n) election. For example, if a taxpayer with only U.S. source and general category income has U.S. source taxable income exceeding the amount of deductions allocated and apportioned to foreign source general category income that includes a section 965(a) inclusion and associated section 78 dividends, a section 965(n) election would not result in a deferred amount and would not affect the calculation of the taxpayer's foreign tax credit limitation. Similarly, a taxpayer with U.S. source income in excess of its net operating loss carryover would have no basis to prevent general category income that includes a section 965(a) inclusion from being reduced by a general category section 172 deduction. A pro rata convention for determining the source and separate category of the deferred amount is more neutral and more consistent with the operation of the expense allocation rules in the absence of a deferred amount than a rule stacking the deferred amount first out of deductions that would reduce the section 965(a) inclusion and associated section 78 dividends. Therefore, the final regulations include the proposed rules applying the existing rules on the allocation and apportionment of expenses for purposes of section 904, and determining the source and separate category of the deferred amount on a pro rata basis. However, in response to the comment regarding the exclusion of the section 965(c) deductions from the deferred amount, the Treasury Department and the IRS agree that section 965(n) does not provide that the deferred amount includes or excludes specific deductions for purposes of section 904. Therefore, the final regulations include the section 965(c) deduction in determining the source and separate category of the deferred amount. See § 1.965-7(e)(1)(iv)(B)(2).

Separately, the Treasury Department and the IRS have determined that nothing in proposed § 1.965-7(e)(1)(iv)(B)(2) suggests that the allocation and apportionment of expenses is based on the section 965(a) inclusion net of the section 965(c) deduction, as opposed to the section 965(a) inclusion not reduced by the section 965(c) deduction. All expenses are allocated and apportioned according to the regulations under §§ 1.861-8 through 1.861-17. See proposed § 1.965-7(e)(1)(iv)(B)(1). The section 965(c) deduction is definitely related to the section 965(a) inclusion. See § 1.861-8(b). Other deductions are allocated and apportioned according to the regulations under §§ 1.861-8 through 1.861-17. For example, a deduction that is not definitely related to any gross income must be ratably apportioned between the statutory grouping of gross income and the residual grouping. The gross income utilized for such ratable apportionment is not reduced by the section 965(c) deduction. See § 1.861-8(c)(3).

The final regulations also adopt the comment's alternative suggestion to allow taxpayers a limited period to revoke a prior election under section 965(n) in order to account for the fact that the foreign tax credit proposed regulations were issued after some taxpayers were required to make the election under section 965(n). See § 1.965-7(e)(2)(ii)(B). For administrability reasons, in order to minimize the number of amended returns that a taxpayer may need to file in connection with section 965, the deadline for a revocation is based on the extended due dates for the taxpayer's returns. In addition, in response to the comment's request for clarification, proposed § 1.965-7(e)(1)(iv)(B)(1) is revised in the final regulation to clarify that it refers to all deductions (other than the net operating loss carryover or carryback to that year that is not allowed by reason of the section 965(n) election).

Another comment requested guidance providing that a taxpayer that had made a timely election under section 965(n) be treated as having made a timely election under section 965(h). Under section 965(h), a taxpayer may elect to pay its section 965(h) net tax liability in eight installments. Section 965(h)(5) provides that the election must be made no later than the due date for the tax return for the inclusion year and in the manner prescribed by the Secretary. Section 1.965-7(b)(2)(ii) provides that relief is not available under § 301.9100-2 or § 301.9100-3 to file a late election. The comment explained that, as a result of the ordering rules in the foreign tax credit proposed regulations, some taxpayers will have a section 965(h) net tax liability in excess of amounts paid with respect to the tax year ending December 31, 2017. Those taxpayers did not make a timely election under section 965(h) because they may have determined that they did not have a section 965(h) net tax liability in excess of amounts paid because they calculated their section 904 foreign tax credit limitation in the inclusion year without allocating or apportioning any expenses to reduce the amount described in § 1.965-7(e)(1)(ii), which is inconsistent with the rules in the foreign tax credit proposed regulations.

The final regulations do not adopt this recommendation. The statute requires that the election must be made not later than the due date for the tax return for the inclusion year. See section 965(h)(5); see also TD 9846, 84 FR 1838, 1868 (February 5, 2019) (denying a similar request to permit late elections under section 965). Moreover, regulations deeming an election to be made by default would not be appropriate, because the statute requires an affirmative election. Cf. 83 FR 39514, 39533-39534 (August 9, 2018) (denying a similar request to provide for default section 965(h) elections). For these reasons, these regulations do not treat a taxpayer that has made a timely election under section 965(n) as having made a timely election under section 965(h).

Finally, the final regulations include two new examples to illustrate the application of § 1.965-7(e)(1). See § 1.965-7(e)(3).

Consistent with § 1.965-9, the final regulations in § 1.965-7(e) apply to the last taxable year of a foreign corporation that begins before January 1, 2018, and with respect to a U.S. person, beginning the taxable year in which or with which such taxable year of the foreign corporation ends.

XI. Comments and Revisions Regarding Applicability Dates

A. Proposed Regulations

The proposed regulations provide that § 1.951-1(e), other than paragraph (e)(1)(ii)(B) (regarding the determination of allocable E&P), applies to taxable years of U.S. shareholders ending on or after October 3, 2018. Comments requested certain changes and guidance related to the applicability date of proposed § 1.951-1(e)(6), the substance of which is discussed more fully in part II.B of this Summary of Comments and Explanation of Revisions section. Comments recommended that the pro rata share anti-abuse rule in proposed § 1.951-1(e)(6) not be applied to transactions or arrangements entered into before the general applicability date of § 1.951-1(e). Under this recommendation, transactions or arrangements entered into before the general applicability date of § 1.951-1(e)(6), regardless of whether they would be subject to the pro rata share anti-abuse rule, would be given effect for purposes of determining a U.S. shareholder's pro rata share of subpart F income and tested items for taxable years ending after the general applicability date. The Treasury Department and the IRS do not adopt this recommendation because it would have the effect of grandfathering existing transactions or arrangements entered into with a principal purpose of avoiding Federal income taxation.

A comment also recommended that taxpayers be permitted, but not required, to apply the facts and circumstances method under § 1.951-1(e)(3), the substance of which is discussed more fully in part II.C of this Summary of Comments and Explanation of Revisions section, to taxable years ending on or after December 31, 2017, and before October 3, 2018. The comment stated that, under section 965, a U.S. shareholder with a taxable year ending on December 31 may be required to determine its pro rata share of the increase to subpart F income of its foreign subsidiaries in both its 2017 taxable year with respect to foreign subsidiaries with a taxable year ending December 31, and its 2018 taxable year with respect to foreign subsidiaries with a taxable year ending November 30. Accordingly, given the applicability date in the proposed regulations, for purposes of determining such U.S. shareholder's inclusion under section 965, the U.S. shareholder could be required to apply, with respect to its calendar year foreign subsidiaries, the fair market value method under the existing regulations for classes of stock with discretionary distribution rights, but then apply, with respect to its fiscal year foreign subsidiaries, the facts and circumstances method for stock with the same characteristics. The comment suggested that allowing U.S. shareholders to rely on the facts and circumstances method for taxable years ending on or after December 31, 2017, and before October 3, 2018, would enable taxpayers to apply a uniform method for allocating the section 965(a) earnings amounts of all relevant foreign subsidiaries among or between U.S. shareholders, would provide more certainty, would be less administratively burdensome, and would not result in improper allocations of subpart F income because the method is consistent with each shareholder's economic rights and interests.

The Treasury Department and the IRS have determined that it would be inappropriate to permit U.S. shareholders the ability to choose whether to rely on the new allocation rules under § 1.951-1(e)(3) for taxable years of foreign corporations that end within the U.S. shareholder's taxable year ending before October 3, 2018, the general applicability date of § 1.951-1(e). See § 1.951-1(i). Rather than simplifying the process of determining their pro rata shares with respect to their calendar year foreign subsidiaries, the proposal would incentivize taxpayers to invest additional time and resources to determine their U.S. tax liability under both sets of pro rata share rules in order to determine the rules that result in the least amount of U.S. tax liability. In addition, because most tax returns of U.S. shareholders that include income from a foreign subsidiary with a taxable year ending on December 31, 2017, by reason of section 965 have already been filed, the proposal would increase the number of amended returns filed for those taxable years, thus creating additional compliance burdens for taxpayers and administrative costs for the government. Accordingly, the final regulations do not adopt this proposal.

There were no comments related to the applicability dates of other provisions of the proposed regulations. The final regulations adopt the applicability dates of the proposed regulations without substantial changes. Therefore, consistent with the applicability date of section 951A, §§ 1.951A-1 through 1.951A-6, including §§ 1.951A-2(c)(5) and -3(h)(2), apply to taxable years of foreign corporations beginning after December 31, 2017, and to taxable years of U.S. shareholders in which or with which such taxable years of foreign corporations end. The applicability dates with respect to the rules in § 1.951-1 are as follows. Paragraphs (a), (b)(1)(ii), (b)(2), (e)(1)(ii)(B), and (g)(1) apply to taxable years of foreign corporations beginning after December 31, 2017, and to taxable years of U.S. shareholders in which or with which such taxable years of foreign corporations end. Paragraph (e), except for paragraph (e)(1)(ii)(B), applies to taxable years of U.S. shareholders ending on or after October 3, 2018. Paragraph (h) applies to taxable years of domestic partnerships ending on or after May 14, 2010. Sections 1.6038-2(a) and § 1.6038-5 apply to taxable years of foreign corporations beginning on or after October 3, 2018.

These final regulations modify applicability dates in the proposed regulations related to consolidated groups. Proposed § 1.1502-51 applies to taxable years of foreign corporations beginning after December 31, 2017, and to taxable years of U.S. shareholders in which or with which such taxable years of foreign corporations end. The Treasury Department and the IRS have determined that for U.S. shareholders that are members of a consolidated group, the applicability date for § 1.1502-51 should be postponed to taxable years of such members for which the due date (without extensions) of the consolidated return is after the date on which these final regulations are published in the Federal Register. However, the final regulations provide that a consolidated group may apply the rules of § 1.1502-51 in their entirety to all of its members for all taxable years described in § 1.951A-7. See § 1.1502-51(g).

B. Foreign Tax Credit Proposed Regulations

No significant changes were made to the applicability dates of the portions of the final regulations that relate to rules that were in the foreign tax credit proposed regulations. Under § 1.965-9(a), the provisions of § 1.965-7 contained in this final regulation apply beginning the last taxable year of a foreign corporation that begins before January 1, 2018, and with respect to a United States person, beginning the taxable year in which or with which such taxable year of the foreign corporation ends. In general, § 1.78-1 applies to taxable years of foreign corporations that begin after December 31, 2017, and to taxable years of U.S. shareholders in which or with which such taxable years of foreign corporations end, and § 1.861-12(c) applies to taxable years that both begin after December 31, 2017, and end on or after December 4, 2018.

A special applicability date was provided in proposed § 1.861-12(k) in order to apply § 1.861-12(c)(2)(i)(B)(1)(ii) to the last taxable year of a foreign corporation beginning before January 1, 2018, since there may be an inclusion under section 965 for that taxable year. In the final regulations, this special applicability date is extended to § 1.861-12(c)(2)(i)(A) to accommodate the changes that were made to that rule to further implement the rule in § 1.861-12(c)(2)(i)(B)(1)(ii). A special applicability date is provided in § 1.78-1(c) in order to apply the second sentence of § 1.78-1(a) to section 78 dividends received after December 31, 2017, with respect to a taxable year of a foreign corporation beginning before January 1, 2018. See part X.A of this Summary of Comments and Explanation of Revisions section regarding comments received about the special applicability date in § 1.78-1(c).

XII. Comment Regarding Special Analyses

One comment asserted that in issuing the proposed regulations, the Treasury Department and the IRS did not comply with the Regulatory Flexibility Act (“RFA”) due to the number of small business entities impacted. The comment also stated that the Treasury Department and the IRS did not comply with the Paperwork Reduction Act (“PRA”) when they authorized the collection of information. Lastly, the comment claimed that the Treasury Department and the IRS did not comply with Executive Orders 12866 and 13563, as well as the Memorandum of Understanding, Review of Tax Regulations under Executive Order 12866, when they issued the proposed regulations.

The Treasury Department and the IRS complied with the applicable requirements under the RFA, the PRA, and Executive Orders 12866 and 13563 when issuing the proposed regulations. See 83 FR 51072, 51084 Special Analyses section. The comment's assertion regarding the number of small business entities impacted by the proposed regulations is addressed in part III of the Special Analyses section.

Special Analyses

I. Regulatory Planning and Review—Economic Analysis

Executive Orders 13563 and 12866 direct agencies to assess costs and benefits of available regulatory alternatives and, if regulation is necessary, to select regulatory approaches that maximize net benefits (including potential economic, environmental, public health and safety effects, distributive impacts, and equity). Executive Order 13563 emphasizes the importance of quantifying both costs and benefits, of reducing costs, of harmonizing rules, and of promoting flexibility.

These final regulations have been designated as subject to review under Executive Order 12866 pursuant to the Memorandum of Agreement (April 11, 2018) between the Treasury Department and the Office of Management and Budget (OMB) regarding review of tax regulations. OMB has designated this final regulation as economically significant under section 1(c) of the Memorandum of Agreement. Accordingly, the final regulations have been reviewed by OMB's Office of Information and Regulatory Affairs. For purposes of E.O. 13771 this rule is regulatory. For more detail on the economic analysis, please refer to the following analysis.

A. Need for the Final Regulations

The final regulations are needed to address remaining open questions regarding the application of section 951A and comments received on the proposed regulations. In addition, certain rules in the foreign tax credit proposed regulations need to be finalized to ensure that the applicability dates of these rules coincide with the applicability dates of the statutory provisions to which they relate.

B. Background

The Tax Cuts and Jobs Act (the Act) established a system under which certain earnings of a foreign corporation can be repatriated to a corporate U.S. shareholder without U.S. tax. See section 14101(a) of the Act and section 245A. However, Congress recognized that, without any base protection measures, this system, known as a participation exemption system, could incentivize taxpayers to allocate income—in particular, mobile income from intangible property—that would otherwise be subject to the full U.S. corporate tax rate to controlled foreign corporations (CFCs) operating in low- or zero-tax jurisdictions. See Senate Explanation at 365. Therefore, Congress enacted section 951A in order to subject intangible income earned by a CFC to U.S. tax on a current basis, similar to the treatment of a CFC's subpart F income under section 951(a)(1)(A). However, in order to not harm the competitive position of U.S. corporations relative to their foreign peers, the global intangible low-taxed income (GILTI) of a corporate U.S. shareholder is taxed at a reduced rate by reason of the deduction under section 250 (with the resulting U.S. tax further reduced by a portion of foreign tax credits under section 960(d)). Id. Also, due to the administrative difficulty in identifying income attributable to intangible assets, intangible income (and thus GILTI) is determined for purposes of section 951A based on a formulaic approach. Intangible income for this purpose is generally all net income (other than certain excluded items) less a 10-percent return (“normal return”) on certain tangible assets (“qualified business asset investment” or “QBAI”). Id. at 366.

The final regulations address open questions regarding the application of section 951A and comments received on the proposed regulations. In addition, certain rules in the foreign tax credit proposed regulations are being finalized in this Treasury decision to ensure that the applicability dates of these rules coincide with the applicability dates of the statutory provisions to which they relate. The final regulations retain the basic approach and structure of the proposed regulations and foreign tax credit proposed regulations, with certain revisions.

The final regulations relating to GILTI provide general rules and definitions, guidance on the computation of a GILTI inclusion amount, rules regarding the interaction of certain aspects of section 951A with other provisions, guidance for consolidated groups and their members and partnerships and their partners, information reporting requirements, and rules to prevent the avoidance of GILTI. The regulations under sections 78, 861, and 965 finalize certain discrete provisions included in the foreign tax credit proposed regulations that relate to section 965.

C. Economic Analysis

1. Baseline

The Treasury Department and the IRS have assessed the economic effects of the final regulations relative to a no-action baseline reflecting anticipated Federal income tax-related behavior in the absence of these final regulations.

2. Summary of Economic Effects

To assess the economic effects of these final regulations, the Treasury Department and the IRS considered economic effects arising from three sorts of provisions of these final regulations. These are (i) effects arising from provisions that provide enhanced certainty and clarity; (ii) effects arising from provisions to prevent tax-avoidance behavior; and (iii) effects arising from other provisions.

These final regulations provide certainty and clarity to taxpayers regarding terms and calculations they are required to apply under the statute. Because a tax had not been imposed on GILTI before the enactment of section 951A and because the statute is silent on certain aspects of definitions and calculations, taxpayers can particularly benefit from enhanced specificity regarding the relevant terms and necessary calculations they are required to apply under the statute. In the absence of this enhanced specificity, similarly situated taxpayers might interpret the statutory rules of section 951A differently, potentially resulting in inefficient patterns of economic activity or litigation in the event that a taxpayer's interpretation of the statute differs from that of the IRS. For example, different taxpayers might pursue income-generating activities based on different assumptions about whether that income will be counted as GILTI, and some taxpayers may forego specific investments that other taxpayers deem worthwhile based on different interpretations of the tax consequences alone. If the foregone activities would have been more profitable than those that were undertaken, U.S. economic performance would be negatively affected. The guidance provided in these regulations helps to ensure that taxpayers face more uniform incentives when making economic decisions, thereby improving U.S. economic performance. This guidance also helps to ensure that taxpayers make tax-related decisions under interpretations that are more consistent with the intent and purpose of the statute.

The Treasury Department and the IRS have not undertaken quantitative estimates of these effects. Any such quantitative estimates would be highly uncertain because the mix of interpretations that taxpayers might have pursued in the absence of this guidance and the mix of economic behaviors stemming from those interpretations are not readily known. More importantly, the relationship between a taxpayer's interpretation absent this guidance and the taxpayer's GILTI inclusion under the final regulations, a difference that is key to understanding the economic effects of the final regulations, is also not readily known.

For example, the final regulations include provisions to address the treatment of domestic partnerships and partners for purposes of section 951A and the section 951A regulations. Part I.C.3.a.i of this Special Analyses section lays out some of the possible interpretations that taxpayers might have adopted in calculating their GILTI inclusion with respect to CFCs owned by a domestic partnership in the absence of specific guidance. Because GILTI and the GILTI partnership provisions are new and because taxpayers' ownership shares of CFCs both through and separate from domestic partnerships are not readily available, the Treasury Department and the IRS cannot readily predict the difference in taxpayers' marginal GILTI inclusion between any given interpretation under the baseline and the final regulation. Thus it is not feasible for the Treasury Department and the IRS to quantify with any reasonable precision the difference in economic activity that might be undertaken by those taxpayers based on those marginal GILTI inclusions.^[5] As data become available, the Treasury Department and the IRS will observe and monitor partner GILTI inclusions resulting from the statute and these supporting regulations.

With these considerations in mind, part I.C.3.a.i of this Special Analyses section explains the rationale behind the final regulations' approach to the treatment of partnerships and provides a qualitative assessment of the alternatives considered.

The final regulations also include provisions designed to curtail improper tax avoidance behavior. In the absence of these provisions, taxpayers could potentially reduce their GILTI by holding specified tangible property over an additional quarter close. See part I.C.3.d.i of this Special Analyses section. This activity is economically inefficient to the extent that the taxpayer acquires the property or holds property longer than the taxpayer would have held it in the absence of this tax-avoidance opportunity. The cost of this inefficiency (relative to the final regulations, which reduce the incentives for such behavior) is roughly proportional to the amount of specified tangible property held longer than optimal, multiplied by the length of the extra holding period, multiplied by the difference between the use value of this property to the taxpayer and its alternative use. The benefit of the final regulations is the reduction in this inefficiency.

The Treasury Department and the IRS have not undertaken a quantitative estimate of this benefit but expect it to be small because the difference between the use value to the taxpayer of property held for tax avoidance purposes and its alternative use is not likely to be large.^[6] The Treasury Department and the IRS do not have readily available data on the amount of specified tangible property that might otherwise be used for tax avoidance purposes, the taxpayers who might hold this property, or the value differential of the property that would be held for tax avoidance purposes.

While it is not currently feasible for the Treasury Department and the IRS to quantify these effects, part I.C.3.c.i of these Special Analyses explains the rationale behind the final regulations' approach to the temporary holding of specified tangible property and provides a qualitative assessment of the alternatives considered.

This economic analysis further considered the economic effects of all other provisions in the final regulations. For example, the statute dictates that, for the purpose of calculating QBAI, taxpayers should depreciate assets placed in service before the enactment of section 951A using the alternative depreciation system (ADS) but grants authority to the Secretary under 951A(d)(4) to issue regulations to prevent the avoidance of the purposes of section 951A(d). By providing taxpayers an alternative to ADS, the final regulations reduce taxpayers' compliance burden and, by effecting changes in QBAI, change some taxpayers' marginal GILTI inclusion, an effect that may result in changes in economic activity and the location of such activity. Furthermore, the final regulations determine partnership QBAI by reference to the depreciation deductions generated by partnership specified tangible property because a CFC partner's share of these depreciation deductions can be used as a reliable proxy for determining a CFC's distributive share of tested income produced with respect to such property. The use of the proxy simplifies, and reduces the uncertainty in the computation for taxpayers, thereby reducing taxpayer burden relative to the baseline.

The netting approach for specified interest expense adopted in these final regulations also reduces uncertainty and the complexity involved in characterizing income and matching expense to income which would be required under a tracing approach. Therefore, the netting approach simplifies the taxpayers' computations and reduces their compliance costs.

With respect to partially depreciable assets, such as platinum catalysts, the final regulations treat a portion of the adjusted basis of the asset as giving rise to QBAI, rather than the asset's entire adjusted basis. The Treasury Department and the IRS determined that applying the same standard for determining whether property qualifies as QBAI and whether the property is depreciable is simpler for tax administration and compliance purposes than having two standards. Moreover, since QBAI generally is determined for purposes of FDII under section 951A(d), it is expected that the final rule will incentivize the use of partially depreciable assets within the United States versus without relative to an alternative of treating the entire adjusted basis of the asset as QBAI.

Because GILTI is new and because tax filings do not report taxpayers' accounting methods for assets placed in service before the enactment of section 951A, the Treasury Department and the IRS do not have readily available data to project which taxpayers are affected by these regulations or to project their marginal GILTI inclusion for current income-generating activities. Thus it is not currently feasible for the Treasury Department and the IRS to estimate the economic effects of the final regulations relative to the baseline.

With these considerations in mind, part I.C.3 of these Special Analyses explains the rationale behind the final regulations and provides a qualitative assessment of the alternatives considered.

3. Economic Effects of Provisions Substantially Revised From the Proposed Regulations

a. Treatment of Domestic Partnerships Under Section 951A

i. Background and Alternatives Considered

Section 951A does not contain any specific rules on the treatment of a domestic partnership and their partners that directly or indirectly own stock of CFCs. The proposed regulations contain a rule that requires a domestic partnership that is a U.S. shareholder of a CFC to determine its GILTI inclusion amount. The proposed regulations then provide that partners of the partnership that are not separately U.S. shareholders of the CFC take into account their distributive share of the partnership's GILTI inclusion amount. In contrast, partners that are U.S. shareholders of the CFC are required to take into account their proportionate share of the partnership's pro rata share of tested items of the CFC for purposes of determining the U.S. shareholder's own GILTI inclusion amount. The proposed regulations thus adopt a hybrid approach under which the domestic partnership is treated as an entity with respect to partners that are not themselves U.S. shareholders of a CFC but as an aggregate with respect to partners that are themselves U.S. shareholders of the CFC. While the hybrid approach is consistent with the framework of section 951A, a number of comments pointed to administrative and procedural complexities with the approach of the proposed regulations. Thus the Treasury Department and the IRS re-evaluated this approach for the final regulations.

The Treasury Department and the IRS considered a number of alternatives for addressing the treatment of domestic partnerships in the final regulations. These alternatives were: (i) The hybrid approach set forth in the proposed regulations; (ii) an approach under which the domestic partnership would be treated as an entity for all purposes of section 951A; and (iii) an approach under which a domestic partnership would be treated as an entity for purposes of determining whether any U.S. person is a U.S. shareholder and any foreign corporation is a CFC, but as an aggregate for purposes of determining whether, and to what extent, any U.S. person has a GILTI inclusion. A fourth option, to apply a pure aggregate approach under which a domestic partnership would be treated as an aggregate of all of its partners for all purposes of section 951A, was rejected because the Treasury Department and the IRS determined that it is inconsistent with other sections of the Code.

The first option was to finalize the hybrid approach set forth in the proposed regulations. While the hybrid approach is consistent with the framework of section 951A, a number of comments pointed to administrative and procedural complexities with the approach of the proposed regulations, including coordination with partners' capital accounts and basis adjustments with respect to partnership interests and CFCs. In particular, comments noted the uncertainty under the hybrid approach whether, and to what extent, a U.S. shareholder partner's pro rata share of tested income or tested loss of a partnership CFC should increase or decrease the partner's capital account with respect to the partnership or its basis in the partnership interest. Comments also noted that the hybrid approach can result in varied GILTI computations for partners depending on whether the partner is a U.S. shareholder of a CFC owned by a domestic partnership. Finally, comments noted that the hybrid approach would result in disparate treatment between partners that own stock in a CFC through a domestic partnership and partners that own stock in a CFC through a foreign partnership. These latter outcomes have clearly detrimental economic effects because they do not treat similar taxpayers in a similar fashion.

The second option was to adopt a pure entity approach, meaning that the domestic partnership would determine its own GILTI inclusion amount and each partner would take into account its distributive share of the partnership's GILTI inclusion amount. This approach is consistent with the historical treatment of domestic partnerships for purposes of subpart F. However, this approach is inconsistent with the policies underlying the GILTI provisions and interrelated rules, such as the deduction under section 250 and certain foreign tax credits for GILTI that are determined at the partner level (rather than the partnership level). Further, under this approach, many taxpayers would be compelled to reorganize their ownership structure—for instance, by eliminating their ownership of CFCs through domestic partnerships—to obtain full aggregation of tested items of their CFCs as envisioned by Congress. Yet other taxpayers would be incentivized to reorganize in an attempt to avoid full aggregation so as to reduce their inclusion below an amount that accurately reflects their GILTI. For instance, taxpayers could separate tested items that generally decrease a U.S. shareholder's GILTI (for example, qualified business asset investment) from certain tested items that reduce the benefit of such tested items (for example, specified interest expense), thus minimizing the U.S. shareholder's aggregate GILTI inclusion amount. Potentially reorganizing to realize a specific GILTI treatment suggests that tax instead of market signals are determining business structures. This can lead to higher compliance costs and inappropriate investment.

The third option, which is adopted in the final regulations, is to apply an approach that treats a domestic partnership as an entity for purposes of determining whether any U.S. person is a U.S. shareholder and whether any foreign corporation is a CFC, but treats a domestic partnership as an aggregate for purposes of determining whether, and to what extent, a partner of a domestic partnership has a GILTI inclusion. Such an approach is consistent with the framework of section 951A and gives effect to the relevant statutory language that treats a domestic partnership as a U.S. shareholder and as owning stock for purposes of determining U.S. shareholder and CFC status. Moreover, this approach eliminates the administrative complexity identified by comments with respect to the hybrid approach in the proposed regulations by calculating a U.S. shareholder partner's GILTI inclusion amount solely at the partner level.

The final regulations treat a domestic partnership as an aggregate by providing that, in general, for purposes of section 951A and the section 951A regulations, a domestic partnership is treated in the same manner as a foreign partnership. The final regulations employ the existing framework for foreign partnerships (which are generally treated as an aggregate of their partners for purposes of subpart F), rather than creating new aggregation rules specifically for the treatment of domestic partnerships, because such framework is relatively well-developed and understood. Using the same treatment for domestic and foreign partnerships is more likely to result in market forces determining organization form instead of tax law. In addition, by eliminating the complexity and traps for the unwary associated with the hybrid and entity approaches, respectively, the chosen approach reduces compliance costs relative to the alternatives.

ii. Affected Taxpayers

The Treasury Department and the IRS estimate that there were approximately 7,000 U.S. partnerships with CFCs that e-filed at least one Form 5471 as Category 4 or 5 filers in 2015 and 2016.^[7] The identified partnerships had approximately 2 million partners, as indicated by the number of Schedules K-1 filed by the partnerships. This number includes both domestic and foreign partners, so it substantially overstates the number of partners that would actually be affected by the final regulations by including foreign partners.^[8] The final regulations affect domestic partners that are U.S. shareholders of a CFC owned by the domestic partnership because such partners will determine their GILTI inclusion amount by reference to their pro rata shares of tested items of CFCs owned by the partnership. Domestic partners that are not U.S. shareholders of a CFC owned by the domestic partnership will neither determine their own GILTI inclusion amount by reference to their pro rata shares of tested items of CFCs owned by the partnership nor include in their income a distributive share of the partnership's GILTI inclusion amount. This latter group is likely to be a substantial portion of domestic partners given the high number of partners per partnership and have lower compliance costs as a result of the final regulations. Because it is not possible to readily identify these types of partners based on available data, this number is an upper bound of partners who would have been affected by this rule had this rule been in effect in 2015 or 2016.

b. Rule for Transfers During the Disqualified Period

i. Background and Alternatives Considered

The proposed regulations include a rule in § 1.951A-2(c)(5) to address transactions intended to reduce a GILTI inclusion amount as a result of a stepped-up basis in CFC assets attributable to related party transfers that occur during the disqualified period. The disqualified period of a CFC is the period between December 31, 2017, which is the last earnings and profits (E&P) measurement date under section 965, and the beginning of the CFC's first taxable year that begins after December 31, 2017, which is the first taxable year with respect to which section 951A is effective. A taxpayer that caused a CFC to sell its assets to a related party during the disqualified period would not be subject to taxation on the income or earnings from such sales under either section 965 (because it was after the final E&P measurement date) or section 951A (because it was before its effective date). However, absent a special rule, in subsequent years, the transaction would reduce a U.S. shareholder's GILTI, by either reducing the transferee CFC's tested income (or increase its tested loss) through the depreciation or amortization attributable to the “cost-free” basis (disqualified basis) in assets created by reason of such related party transfer. Accordingly, the rule in the proposed regulations prevents the benefits of the disqualified basis by disallowing any deduction or loss attributable to the disqualified basis for purposes of determining tested income or tested loss.

Because the rule in proposed § 1.951A-2(c)(5) only disallows the stepped-up basis created by reason of a disqualified transfer for purposes of determining a CFC's tested income and tested loss, under the proposed regulations, a taxpayer would have to keep track of both a CFC's disqualified basis in an asset for purposes of section 951A and the CFC's adjusted basis in the asset for all other purposes of the Code. In addition, if the disqualified basis was not allowed for purposes of determining tested income and tested loss, a comment noted that it would be unfair for the basis to still be taken into account for purposes of section 901(m), which disallows foreign tax credits for foreign income not subject to U.S. tax by reason of certain basis differences that arise by reason of covered asset acquisitions. A transfer subject to the rule (a disqualified transfer) can also be a covered asset acquisition, and therefore section 901(m) and proposed § 1.951A-2(c)(5) could apply concurrently by reason of the same transaction.

The Treasury Department and the IRS considered three options to address the treatment of disqualified basis. These options were: (i) Adopt the proposed regulations without change; (ii) revise the regulations to provide that disqualified basis is also not taken into account for purposes of certain other provisions (in addition to section 951A) to ensure that the rule only prevents the GILTI benefits that taxpayers were trying to achieve; or (iii) allow taxpayers to make an election that would disregard the disqualified basis for all purposes of the Code.

The first option was to finalize without change the rule contained in the proposed regulations. On the one hand, this approach could be viewed as simple and targeted, because this rule would only disregard disqualified basis for purposes of determining GILTI, and the transactions subject to the rule were primarily intended to reduce GILTI. On the other hand, this rule could be considered unfair in certain cases because the concurrent application of both the rule and section 901(m), without a means for avoiding such concurrent application, could be viewed as unduly punitive to taxpayers that engaged in such transactions. In addition, this option would require taxpayers to track and maintain separate bases in the property for purposes of GILTI and all other purposes of the Code.

The second option was to not take into account disqualified basis for certain other provisions (in addition to section 951A) to ensure that the rule only prevented the GILTI benefits that taxpayers were trying to achieve. Such an approach would result in additional and considerable complexity because numerous other provisions would have to be considered. In addition, simply not taking into account the basis for purposes of these other provisions may not alone provide appropriate results, without taking into account the policies underlying the specific provisions. Such particular policy considerations could require additional special and detailed rules or modifications to the general disallowance rules. In addition, it would be difficult to assess the effect that the disqualified basis would have on other provisions of the Code, or how it could affect different taxpayers with different tax postures.

The third option, which is adopted in the final regulations, is to allow taxpayers to make an election that eliminates disqualified basis in property by reducing a commensurate amount of adjusted basis in the property for all purposes of the Code. Although this option was not as targeted as the second option, it was the simplest of the three options because it results in the property only having a single tax basis for all purposes of the Code such that different bases need not be tracked for different purposes. In addition, it does not result in additional complex rules, as would be required in the second option, because it simply applies for all purposes; once the basis is reduced, the Code simply applies to the property as if the basis were never stepped up. Finally, this approach permits taxpayers to decide whether the benefit of the additional adjusted basis associated with the disqualified basis outweighs the cost of complexity in applying the rule or, alternatively, whether the value of simplicity outweighs the benefit of the additional adjusted basis. By allowing this flexibility and adopting a single adjusted basis for all purposes of the Code, the adopted approach reduces complexity and compliance costs, relative to both alternatives considered.

ii. Affected Taxpayers

The final regulations apply to any deduction or loss attributable to disqualified basis. Disqualified basis is created by reason of a disqualified transfer, which is defined as a transfer of property by a fiscal year CFC during the disqualified period to a related person in which gain was recognized, in whole or in part. A fiscal year CFC's disqualified period is the period that begins on January 1, 2018, and ends as of the close of the CFC's last taxable year that is not a CFC inclusion year. The taxpayer affected is a U.S. shareholder of any CFC that holds property with disqualified basis. In general these final regulations affect U.S. shareholders with at least one fiscal year CFC that has at least one other CFC where the fiscal-year CFC has property with unrealized gains that can be transferred during the disqualified period.

The Treasury Department and the IRS do not have data identifying CFCs that engaged in transactions with related CFCs during the period after December 31, 2017 but before the effective date of section 951A. As an upper-bound estimate, there are approximately 3,000 U.S. shareholders of fiscal year CFCs with at least one related CFC that could potentially engage in a transaction.^[9] This is an overestimate since only those fiscal year CFCs with unrealized gains could take advantage of this disqualified period. The Treasury Department does not have data readily available to estimate these unrealized gains.

c. Transition Rule To Determine Normal Return Using the Alternative Depreciation System

i. Background and Alternatives Considered

A U.S. shareholder's GILTI inclusion amount is based on a formulaic approach under which a 10-percent return attributed to certain tangible assets (QBAI) is computed and then each dollar of certain income above such “normal return” is effectively treated as intangible income. Under the statute, QBAI is measured by determining the adjusted basis in certain tangible property using the alternative depreciation system (ADS). Section 951A(d)(4) directs the Secretary to issue regulations or other guidance that is appropriate to prevent the avoidance of the purposes of section 951A(d), including with respect to the treatment of temporarily held or transferred property.

The proposed regulations require the adjusted basis of all specified tangible property to be determined using ADS under section 168(g) for purposes of determining the QBAI of a CFC. In general, the Code requires that tangible property used by a CFC outside the United States must be depreciated using ADS. Accordingly, in most instances, the depreciation method required under the proposed regulations will correspond to the CFC's depreciation method used for computing income. However, under existing regulations under section 952, a CFC may compute its income and E&P using the depreciation method used in keeping its accounting books and records or a method conforming to United States generally accepted accounting principles (“non-ADS depreciation method”) if the differences between ADS and the non-ADS depreciation method are immaterial. In the case of a CFC that is permitted to use a non-ADS depreciation method, the proposed regulations would nonetheless require the CFC to determine its adjusted basis in its assets for purposes of calculating QBAI based on ADS. In particular, with respect to assets placed in service before the enactment of section 951A, the proposed regulations would require the CFC to determine the date the assets were placed in service, the ADS class life, and other information about the asset to correctly apply ADS as if the asset had been depreciated using ADS since the date the asset was placed in service. Several comments noted that this requirement could be onerous for specified tangible property acquired before the enactment of section 951A and requested relief from this requirement for such property.

Although section 951A(d)(3) specifically requires use of ADS to determine the adjusted basis in specified tangible property, section 951A(d)(4) authorizes the Secretary to issue regulations that are appropriate for purposes of determining QBAI. Thus, the Treasury Department and the IRS considered three options to address the use of ADS for specified tangible property placed in service prior to the enactment of section 951A. These options were: (i) Require the use of ADS for all property placed in service before the enactment of section 951A, consistent with the proposed regulations; (ii) require ADS for determining the adjusted basis of specified tangible property, but on a “cut-off basis”; or (iii) allow the CFC to continue using its non-ADS depreciation method for property placed in service prior to the enactment of section 951A, and to include a special rule that requires depreciation of the “salvage value.” These options apply only where the CFC is not required to use ADS to compute its income under § 1.952-2 or E&P under § 1.964-1 with respect to such property.

The first option considered was to require the use of ADS for all property placed in service before the enactment of section 951A, consistent with the proposed regulations. However, Treasury and the IRS recognize that re-determining the adjusted basis in assets using a new depreciation method could be a difficult, uncertain, and time-consuming process for CFCs that have numerous items of specified tangible property acquired before the enactment of section 951A, in part, because the CFCs may not have kept the records necessary to make the determinations. Notably as described above, CFCs are permitted to compute their income and E&P using their non-ADS depreciation method for specified tangible property used outside the United States when the differences between the non-ADS depreciation method and ADS are immaterial. Therefore, the Treasury Department and the IRS determined that some relief from the administrative burden of re-determining the adjusted basis of each property placed in service before December 22, 2017, should be available to CFCs that are not required to use ADS for computing income and E&P. Such relief will alleviate this administrative burden, but will not impact taxpayer incentives or cost of capital, because it pertains only to property already placed in service.

The second option considered seeks to relieve burden by requiring ADS for determining the adjusted basis in specified tangible property, but on a “cut-off basis.” Under this option, the CFC would apply ADS to the adjusted basis determined using its non-ADS depreciation method as of the beginning of the first taxable year subject to section 951A. This option eliminates the need to re-determine the adjusted basis in the property as if ADS had been used since the property was placed in service. This approach could be implemented by applying ADS for the remaining ADS class life of the property or by treating the property as newly placed in service and applying the full ADS class life to the property. Each of those options would still require the CFC to determine when the property was placed in service and its ADS class life. In addition, applying ADS for the remaining ADS class life of the property would also require special rules for situations in which the property would have been fully depreciated under ADS before the first taxable year subject to section 951A, and applying ADS to the property based on the full ADS class life of the property would extend the period that the property is taken into account in the computation of QBAI. The Treasury Department and the IRS concluded that applying ADS on a cut-off basis under either approach did not significantly reduce the administrative burden of computing QBAI with respect to property placed in service prior to the enactment of section 951A.

The third option considered was to allow the CFC to elect to use its non-ADS depreciation method for property acquired prior to the enactment of section 951A, and to include a special rule that requires depreciation of the “salvage value” (in other words, the portion of the basis of property that would not be fully depreciated under the non-ADS depreciation method). The special rule is required because otherwise the salvage value would be included in the CFC's QBAI until the CFC disposed of the asset. This option was the least administratively burdensome, and the least likely to result in controversy between taxpayers and the IRS. It reduces compliance costs relative to the two alternatives by eliminating the need to redetermine the adjusted basis, class life and date placed in service of property for which good records may not exist. As noted above, it does not impact taxpayers' incentives or cost of capital, because it applies to property already placed in service. Further, because relief is provided in instances in which the difference between ADS and a non-ADS depreciation method is immaterial, it is likely to result in only minimal differences in depreciation deductions and QBAI.^[10] Small changes in the QBAI have an even more muted impact on the determination of GILTI, because net DTIR, a component of the GILTI calculation, is only 10 percent of QBAI. Therefore, the impact of using a non-ADS depreciation method versus ADS for property placed in service before the enactment of section 951A is minimal. Accordingly, this is the option adopted in the final regulations.

ii. Affected Taxpayers

The population of taxpayers potentially affected by this aspect of these final regulations are the U.S. shareholders of CFCs that are not required to use ADS when computing E&P, subpart F income, and tested income or tested loss, because the differences in the tax liability of such U.S. shareholders resulting from the use of the CFCs' non-ADS depreciation method are immaterial relative to the use of ADS. Only those taxpayers whose CFCs use a non-ADS depreciation method for property placed in service before December 22, 2017 instead of ADS when computing E&P would be affected by these final regulations.

The Treasury Department and the IRS have previously projected that between 25,000 and 35,000 direct shareholders of CFCs would be potentially subject to GILTI and thus could be affected by this rule. This is an upper-bound estimate of taxpayers affected because it is not limited to those with CFCs that are permitted to use a non-ADS depreciation method with respect to property placed in service before the enactment of section 951A. Precise identification of these taxpayers is not possible from readily available data because taxpayers do not report on Form 5471 what depreciation method they used in computing E&P.

d. Anti-Abuse Rule for Specified Tangible Property Held Temporarily

i. Background and Alternatives Considered

The proposed regulations include an anti-abuse rule to address property that is held temporarily over the quarter close of a CFC with a principal purpose of reducing the GILTI inclusion amount of a U.S. shareholder of the CFC. In the absence of an anti-abuse rule, taxpayers could reduce their GILTI inclusion by having a CFC temporarily hold property over an additional quarter close in order to artificially increase the U.S. shareholder's “normal return” on tangible assets. The anti-abuse rule for temporarily held property in the proposed regulations included a “per se” rule, which deemed property to be held temporarily and acquired with a principal purpose of reducing a GILTI inclusion amount if held by the CFC for less than a 12-month period. Comments asserted that the anti-abuse rule was overbroad. In particular, comments expressed concerns that the 12-month per se rule could affect transactions not motivated by tax avoidance, such as ordinary course transactions, and create burdens resulting from having to track how long the specified tangible property is held.

The Treasury Department and the IRS considered four options to address these concerns. These options were: (i) Adopt the proposed regulations without change; (ii) shorten the per se rule; (iii) eliminate the per se rule and rely on a principal purpose rule; or (iv) convert the per se rule into a rebuttable presumption, add a safe harbor, and clarify the scope of the rule.

The first option was to finalize without change the rule contained in the proposed regulations. This approach is a simple and administrable rule for the IRS and taxpayers because it would not consider the taxpayer's motivation for holding property for less than 12 months; however, it would not address the concern raised by comments that the rule can potentially apply to transactions that were not tax motivated and could therefore lead to a reduction in otherwise economically valuable transactions.

The second option was to shorten the 12-month per se rule to, for example, six months. While this option could significantly reduce the number of transactions subject to the rule relative to the first option, and would be administrable for the IRS and taxpayers (because a taxpayer's motivation for holding the property would not be relevant), it could still apply to transactions that were not tax-motivated. In addition, it could increase the burden on IRS to enforce compliance because it would require additional resources to assert the rule for property held longer than six months, even though the property may still be held temporarily for tax-motivated reasons.

The third option was to eliminate the per se rule and rely on a principal purpose rule. The rule would disregard the adjusted basis in property for purposes of computing QBAI if the property is held temporarily and is acquired with a principal purpose of reducing a GILTI inclusion amount. While this option would have the benefit of being flexible and, therefore, in theory could apply only to temporary holdings that were intended to reduce a U.S. shareholder's GILTI inclusion amount, it could create uncertainty for both taxpayers and the IRS. This uncertainty would result, in part, from the need to determine the taxpayer's principal purposes for each relevant acquisition and not having general guidelines for when property is considered to be held temporarily. It would also increase administrative and compliance costs for the IRS and taxpayers because there could be more disputes over the taxpayer's principal purpose and when a property is held temporarily.

The fourth option that was considered involved several components. First, this option would convert the per se rule to a rebuttable presumption. Under this rule, property would be presumed to be temporarily held and acquired with a principal purpose of reducing a GILTI inclusion amount if the property is held for less than twelve months. However, the presumption could be rebutted if, in general, the facts and circumstances clearly establish that the subsequent transfer of the property by the CFC was not contemplated when the property was acquired and that a principal purpose of the acquisition of the property was not to increase the normal return of a U.S. shareholder. This option also would add a second presumption that generally provides that property is presumed to not be subject to the rule if held for more than 36 months. In addition, this option would include a “safe harbor” that generally applies to transfers between CFCs that are owned in the same proportion by U.S. shareholders, have the same taxable years, and are all tested income CFCs. Finally, this option would include examples to indicate types of transactions that are, and are not, subject to the rule.

This fourth option more accurately identifies cases of potential abuse in comparison to the proposed regulations and the other options discussed in this part I.C.3.d.i of the Special Analyses section. Because it more accurately identifies cases of potential abuse, it yields more efficient outcomes because it does not penalize taxpayers with a legitimate business purpose for temporarily holding tangible property. This option provides flexibility to taxpayers holding property less than 12 months to either accept the presumption (and thus disregard the basis of the property under the anti-abuse rule) or, if appropriate, to choose to rebut the presumption by filing the appropriate statement. Taxpayers will have the flexibility to make the choice that appropriately balances the compliance costs related to rebutting the presumption with the tax cost of not rebutting the presumption depending on their particular circumstances. This option also relieves taxpayers of the burden of monitoring assets that are held more than 36 months, relative to the other options. In addition, the safe harbor would provide additional certainty to both taxpayers and the IRS, and eliminate any resulting compliance and administrative costs, because these transactions, which generally do not give rise to avoidance concerns, would be entirely excluded from the application of the rule. Although the compliance costs associated with a rebuttal based on facts and circumstances will likely be higher than under the first and second alternatives, those alternatives do not provide taxpayers with an opportunity to demonstrate the economic substance of the transaction, and the electivity of the rebuttal leaves taxpayers no worse off than under the first and second options. It is not clear whether the adopted approach has higher or lower compliance costs than the third approach, but Treasury and IRS determined the adopted approach to be superior for the reasons discussed above.

The Treasury and the IRS determined that these changes strike an appropriate balance between (i) mitigating compliance burdens relative to the proposed regulations and providing certainty and flexibility to taxpayers and (ii) identifying transactions that have the potential for abuse. Thus, this is the approach adopted in the final regulations.

ii. Affected Taxpayers

In principle, this aspect of the final regulations could apply to any tested income CFC that purchases tangible property and holds it temporarily. Therefore, this aspect of the regulations could affect any of the 25,000-35,000 persons with a potential GILTI inclusion and should be treated as an upper-bound estimate. In practice, however, it would only apply to U.S. shareholders of CFC that temporarily hold tangible property for tax minimization purposes, which would only be a small subset of sophisticated tax planners. The Treasury Department and the IRS do not have readily available data to enable estimating how many taxpayers could minimize tax in this way, nor which taxpayers would likely undertake such behavior in the absence of these regulations.

e. Application of Basis Adjustment for Purposes of Characterizing Certain Stock

i. Background and Alternatives Considered

Under the Code, certain expenses, including interest, must be allocated based on the adjusted basis of the assets held by the taxpayer. For purposes of allocating expenses to stock of certain foreign corporations held directly by a taxpayer, section 864(e)(4) generally requires that a taxpayer adjust the adjusted basis of the stock by the aggregate amount of E&P of the foreign corporation and its subsidiaries. The combination of the adjusted basis of the stock of the foreign corporation and the increase or decrease (if the foreign corporation and its subsidiaries have a deficit in E&P) in that amount by the E&P of the foreign corporation approximate the value of the stock of the foreign corporation for purposes of the expense allocation rules. See Joint Committee on Tax'n, General Explanation of the Tax Reform Act of 1986 (Pub. L. 99-514) (May 4, 1987), JCS-10-87, at p. 946 (noting that “the failure to consider earnings and profits caused significant distortion” for purposes of expense allocation rules because the value of the earnings and profits is reflected in the fair market value of the stock).

Under section 965(b)(4)(B), if a taxpayer used a deficit in E&P to offset its inclusion under section 965(a), the deficit is eliminated by increasing the E&P of the foreign corporation with the deficit. However, because there is no offsetting reduction to the basis of the stock of the foreign corporation, the adjusted basis of that foreign corporation for purposes of section 864(e)(4) is increased as a result of the application of section 965(b)(4)(B), even though there has been no economic change to the value of the foreign corporation. Under final regulations under section 965, in general, a taxpayer may elect to reduce the basis in the stock of the foreign corporation, on a share by share basis, by the amount of the increase to the E&P of the foreign corporation under section 965(b)(4)(B). See § 1.965-2(f)(2)(i). However, the election does not cause the taxpayer's basis to be reduced below zero, even if the amount of the increase to the E&P of the foreign corporation under section 965(b)(4)(B) exceeds the taxpayer's basis in the stock.

The foreign tax credit proposed regulations provide that, for purposes of determining the adjusted basis of the stock of the foreign corporation under section 864(e)(4), a taxpayer should determine its adjusted basis in the stock of the foreign corporation as if the taxpayer had made in the election in § 1.965-2(f)(2)(i). See proposed § 1.861-12(c)(2)(i)(B)(1)(ii). After this adjustment, the taxpayer then follows the existing rule under section 864(e)(4) to increase or decrease the adjusted basis in the stock by the E&P of the foreign corporation and its subsidiaries.

A comment requested that the foreign tax credit proposed regulations be amended to make clear that, for purposes of section 864(e)(4), that the reduction in basis under proposed § 1.861-12(c)(2)(i)(B)(1)(ii) does not cause the taxpayer's basis in the stock in the foreign corporation to be less than zero. This could happen, for example, where the increase in the foreign corporation's E&P under section 965(b)(4)(B) exceeded the taxpayer's adjusted basis in the stock of that foreign corporation.

The Treasury Department and the IRS agreed that, for purposes of applying the expense allocation rules, a taxpayer should not have an adjusted basis below zero in the stock of a foreign corporation. When the adjusted basis of an asset is zero, no expenses are allocated to that asset and thus allowing a negative adjusted basis would serve no purpose for the expense allocation rules. However, because the adjustment to the stock of the foreign corporation in this case is two steps—the adjusted basis is reduced to account for the application of section 965(b)(4)(B) and then increased or decreased by the amount of E&P of the foreign corporation and its subsidiaries—the adjusted basis could be less than zero after the initial adjustment but still be positive after the second adjustment is taken into account. Accordingly, the Treasury Department and the IRS considered two options to address the concern expressed by the comment. These options were: (i) Adopt the foreign tax credit proposed regulations described above with a statement that the reduction in basis is limited to the taxpayer's adjusted basis in the stock of the foreign corporation; or (ii) allow a taxpayer's adjusted basis in the stock of the foreign corporation to be reduced below zero as a result of the adjustment for section 965(b)(4)(B) as long as the adjustment for E&P provided in section 864(e)(4) increased the adjusted basis of the foreign corporation to or above zero.

The first option was to adopt the proposed regulations with a statement that the reduction in basis is limited to the taxpayer's adjusted basis in the stock of the foreign corporation. On one hand, this would address the concerns that the adjustment could cause a taxpayer's adjusted basis in the stock of the foreign corporation to be less than zero for purposes of the expense allocation rules. On the other hand, this would perpetuate some of the distortion created by the application of section 965(b)(4)(B). That is, because the increase in the E&P of the foreign corporation would exceed the downward adjustment in the basis of the foreign corporation, the adjusted basis in the stock of the foreign corporation would still be higher for purposes of section 864(e)(4) than if section 965(b)(4)(B) had not applied.

The second option was to provide that the taxpayer's adjusted basis in the stock of the foreign corporation may be reduced below zero as a result of the adjustment for section 965(b)(4)(B) as long as the adjustment for E&P provided in section 864(e)(4) increased the adjusted basis of the foreign corporation to or above zero. This option fully addresses the non-economic increase to the E&P of the foreign corporation under section 965(b)(4)(B) because the adjusted basis of the foreign corporation is reduced by the full amount of the increase. However, it also still ensures that, for expense allocation purposes, the adjusted basis of the stock of the foreign corporation will not be below zero, after accounting for the E&P adjustment in section 864(e)(4). The Treasury Department and the IRS selected this option for the final regulations because it addressed the concerns regarding negative adjusted basis while most accurately reflecting the value of the stock in the foreign corporation for purposes of the expense allocation rules, and did not increase compliance costs relative to the alternatives.

ii. Affected Taxpayers

The taxpayers potentially affected by this aspect of the final regulations are those taxpayers that own at least 10 percent of a foreign corporation that had its E&P increased under section 965(b)(4)(B). The Treasury Department and the IRS have not estimated how many taxpayers are likely to be affected by these regulations because this level of detail regarding taxpayer filings under section 965 is not readily available. However, 100,000 taxpayers were estimated to pay the section 965 one-time tax. This is an upper-bound estimate of affected taxpayers since only those with an E&P adjustment under section 965(b)(4)(B) would be affected. Information on those taxpayers is not readily available to the Treasury Department and the IRS.

II. Paperwork Reduction Act

In response to comments addressing the notices of proposed rulemaking preceding the final regulations, the Treasury Department and the IRS have added new collections of information with respect to section 951A and revised a collection of information with respect to section 965(n).

The new collections of information in these regulations with respect to section 951A are in § 1.951A-3(e)(3)(ii), (h)(1)(iv)(A), and (h)(2)(ii)(B)(3). The revised collection of information with respect to the election under section 965(n) is in § 1.965-7(e)(2)(ii)(B).

The collection of information in § 1.951A-3(e)(3)(ii) is an election that the controlling domestic shareholders of a CFC may make in order for the CFC to continue to use its book depreciation method (rather than converting to ADS) for purposes of determining the adjusted basis in specified tangible property placed in service before its first taxable year beginning after December 22, 2017 if certain conditions are met. This election is made by controlling domestic shareholders by attaching a statement meeting the requirements of § 1.964-1(c)(3)(ii) with their income tax returns following the notice requirements of § 1.964-1(c)(3)(iii). This election, if made by a CFC, simplifies the calculation of the QBAI for the CFC attributable to property placed in service before December 22, 2017, which, and in turn, simplifies the calculation of the DTIR of the CFC's U.S. shareholders attributable to such property. For purposes of the Paperwork Reduction Act of 1995 (44 U.S.C. 3507(d)) (“PRA”), the reporting burden associated with § 1.951A-3(e)(3)(ii) will be reflected in the PRA submission associated with the Form 990 series, Form 1120 series, Form 1040 series, Form 1041 series, and Form 1065 series (see chart at the end of this part II of this Special Analyses section for the status of the PRA submissions for these forms).

The collection of information in § 1.951A-3(h)(1)(iv)(A) is a statement that a U.S. shareholder must attach to a Form 5471 with respect to a CFC in order to rebut the presumption that a transfer of specified tangible property held by the CFC for less than 12 months was held temporarily with a principal purpose of increasing the DTIR of the U.S. shareholder. The information included in the statement is required in order for the IRS to be aware if the taxpayer takes the position that the temporary ownership rule of § 1.951A-3(h)(1) does not apply. Without this statement, there is a presumption that such property is held temporarily with a principal purpose of increasing DTIR of a U.S. shareholder and a portion of the basis in the property may be disregarded for purposes of calculating QBAI of the CFC that holds the property temporarily. The statement indicates that the U.S. shareholder should be allowed the benefit of basis that would otherwise be disregarded for purposes of calculating QBAI. For purposes of the PRA, the reporting burden associated with § 1.951A-3(h)(1)(iv)(A) will be reflected in the PRA submission associated with Form 5471, “Information Return of U.S. Persons With Respect to Certain Foreign Corporations” (OMB control number 1545-0123).

The collection of information in § 1.951A-3(h)(2)(ii)(B)(3) is an election to disregard disqualified basis, which is certain basis that was created by reason of a disqualified transfer during the disqualified period of a transferor CFC, as those terms are defined in § 1.951A-3(h)(2)(ii)(C). This election would simplify recordkeeping with respect to the property because a separate record of the disqualified basis and total adjusted basis in the property would not have to be tracked. For purposes of determining disqualified basis, a disqualified transfer includes both a direct transfer during the disqualified period by one CFC to a related person, and also an indirect transfer of property owned by a partnership through, for example, a transfer by a CFC to a related person of an interest in the partnership, for which a section 754 election is in effect. Therefore, disqualified basis may exist in both property held by a CFC and property held by a partnership. Accordingly, there are two methods for making this election based upon whether the property with disqualified basis is held directly by a CFC or indirectly through a partnership in which the CFC is a partner. With respect to property held directly by the CFC, this election is made by controlling domestic shareholders of the CFC by attaching a statement meeting the requirements of § 1.964-1(c)(3)(ii) with their income tax returns following the notice requirements of § 1.964-1(c)(3)(iii). See § 1.951A-3(h)(2)(ii)(B)(3)(ii). With respect to property held in a partnership in which the CFC is a partner, this election is made by the partnership by filing a statement as described in § 1.754-1(b)(1) attached to the partnership return. See § 1.951A-3(h)(2)(ii)(B)(3)(iii). For purposes of the PRA, the reporting burden associated with § 1.951A-3(h)(2)(ii)(B)(3)(ii) will be reflected in the PRA submission associated with the Form 990 series, Form 1120 series, Form 1040 series, Form 1041 series, and Form 1065 series (see chart at the end of this part II of the Special Analysis section for the status of the PRA submissions for these forms). For purposes of the PRA, the reporting burden associated with § 1.951A-3(h)(2)(ii)(B)(3)(iii) will be reflected in the PRA submission associated with Form 1065 (see chart at the end of this part II of the Special Analysis section for the status of the PRA submissions for this form).

The collection of information in § 1.965-7(e)(2)(ii)(B) requires a taxpayer revoking a section 965(n) election to attach a statement to that effect to an amended income tax return. The information is required in order for the IRS to be aware if a taxpayer revokes an election. The Treasury Department and the IRS have determined that the reporting burden associated with § 1.965-7(e)(2)(ii)(B) to revoke a section 965(n) election is reflected in the reporting burden associated with making the election. For purposes of the PRA, the reporting burden associated with § 1.965-7(e)(2)(ii)(B) will be reflected in the PRA submission associated with TD 9846, 84 FR 1838 (February 5, 2019) (OMB control number 1545-2280).

The estimates for the number of impacted filers with respect to the collections of information described in this part II of the Special Analysis section are based on filers of income tax returns with a Form 5471 attached because only filers that are U.S. shareholders of CFCs or that have at least a 10 percent ownership in a foreign corporation would be subject to the information collection requirements. The IRS estimates the number of affected filers to be the following:

The current status of the PRA submissions related to the tax forms that will be revised as a result of the information collections in the section 951A regulations is provided in the accompanying table. As described above, the reporting burdens associated with the information collections in the regulations are included in the aggregated burden estimates for OMB control numbers 1545-0123 (which represents a total estimated burden time for all forms and schedules for corporations of 3.157 billion hours and total estimated monetized costs of $58.148 billion ($2017)), 1545-0074 (which represents a total estimated burden time, including all other related forms and schedules for individuals, of 1.784 billion hours and total estimated monetized costs of $31.764 billion ($2017)), 1545-0092 (which represents a total estimated burden time, including all other related forms and schedules for trusts and estates, of 307,844,800 hours and total estimated monetized costs of $9.950 billion ($2016)), and 1545-0047 (which represents a total estimated burden time, including all other related forms and schedules for tax-exempt organizations, of 50.450 million hours and total estimated monetized costs of $1,297,300,000 ($2017). The overall burden estimates provided for the OMB control numbers below are aggregate amounts that relate to the entire package of forms associated with the applicable OMB control number and will in the future include, but not isolate, the estimated burden of the tax forms that will be revised as a result of the information collections in the regulations. These numbers are therefore unrelated to the future calculations needed to assess the burden imposed by the regulations. These burdens have been reported for other regulations related to the taxation of cross-border income and the Treasury Department and the IRS urge readers to recognize that these numbers are duplicates and to guard against overcounting the burden that international tax provisions imposed prior to the Act. No burden estimates specific to the forms affected by the regulations are currently available. The Treasury Department and the IRS have not estimated the burden, including that of any new information collections, related to the requirements under the regulations. The Treasury Department and the IRS estimate PRA burdens on a taxpayer-type basis rather than a provision-specific basis. Those estimates would capture both changes made by the Act and those that arise out of discretionary authority exercised in the final regulations.

The Treasury Department and the IRS request comments on all aspects of information collection burdens related to the final regulations, including estimates for how much time it would take to comply with the paperwork burdens described above for each relevant form and ways for the IRS to minimize the paperwork burden. Proposed revisions (if any) to these forms that reflect the information collections contained in these final regulations will be made available for public comment at https://apps.irs.gov/​app/​picklist/​list/​draftTaxForms.html and will not be finalized until after these forms have been approved by OMB under the PRA.

In 2018, the IRS released and invited comments on drafts of the above forms in order to give members of the public advance notice and an opportunity to submit comments. The IRS received no comments on the portions of the forms that relate to section 951A during the comment period. Consequently, the IRS made the forms available in late 2018 and early 2019 for use by the public. The IRS is contemplating making additional changes to forms in order to implement these final regulations.

III. Regulatory Flexibility Act

It is hereby certified that this final regulation will not have a significant economic impact on a substantial number of small entities within the meaning of section 601(6) of the Regulatory Flexibility Act (5 U.S.C. chapter 6).

Sections 951 and 951A generally affect U.S. shareholders of CFCs. Section 965 generally affects U.S. taxpayers who are at least 10-percent shareholders of a foreign corporation. The reporting burdens in § 1.951A-3(e)(3)(ii), (h)(1)(iv)(A), and (h)(2)(ii)(B)(3), and § 1.965-7(e)(2)(ii)(B) generally affect U.S. taxpayers that elect to make or revoke certain elections or rebut a presumption. In general, foreign corporations are not considered small entities. Nor are U.S. taxpayers considered small entities to the extent the taxpayers are natural persons or entities other than small entities. For purposes of the PRA, the Treasury Department and the IRS estimate that there are 25,000-35,000 respondents of all sizes that are likely to file Form 5471. Only a small proportion of these filers are likely to be small business entities. This estimate was used in the proposed regulations (REG-104390-18), and comments were requested on the number of small entities that are likely to be impacted by the section 951A regulations.

Examining the gross receipts of the e-filed Forms 5471 that is the basis of the 25,000-35,000 respondent estimates, the Treasury Department and the IRS have determined that the tax revenue from section 951A estimated by the Joint Committee on Taxation for businesses of all sizes is less than 0.3 percent of gross receipts as shown in the table below. Based on data for 2015 and 2016, total gross receipts for all businesses with gross receipts under $25 million is $60 billion while those over $25 million is $49.1 trillion. Given that tax on GILTI inclusion amounts is correlated with gross receipts, this results in businesses with less than $25 million in gross receipts accounting for approximately 0.01 percent of the tax revenue. Data are not readily available to determine the sectoral breakdown of these entities. Based on this analysis, smaller businesses are not significantly impacted by these final regulations.

Although the Treasury Department and the IRS received one comment asserting that a substantial number of small entities would be affected by the proposed regulations, that comment was principally concerned with U.S. citizens living abroad that owned foreign corporations directly or indirectly through other foreign entities. U.S. citizens living abroad are not small business entities; thus, no small entity is affected in this scenario.

Specifically, the small business entities that are subject to the requirements of § 1.951A-3(e)(3)(ii), (h)(1)(iv)(A), and (h)(2)(ii)(B)(3) of the final regulations are domestic small entities that are U.S. shareholders of one or more CFCs. The data to assess the number of small entities potentially affected by § 1.951A-3(e)(3)(ii), (h)(1)(iv)(A), and (h)(2)(ii)(B)(3) are not readily available. However, businesses that are U.S. shareholders of CFCs are generally not small businesses because the ownership of sufficient stock of a CFC in order to be a U.S. shareholder generally entails significant resources and investment. Therefore, the Treasury Department and the IRS have determined that a substantial number of domestic small business entities will not be subject to § 1.951A-3(e)(3)(ii), (h)(1)(iv)(A), and (h)(2)(ii)(B)(3). Moreover, as discussed above, smaller businesses are not significantly impacted by the final regulations. Consequently, the Treasury Department and the IRS have determined that § 1.951A-3(e)(3)(ii), (h)(1)(iv)(A), and (h)(2)(ii)(B)(3) will not have a significant economic impact on a substantial number of small entities. Accordingly, it is hereby certified that the collection of information requirements of § 1.951A-3(e)(3)(ii), (h)(1)(iv)(A), and (h)(2)(ii)(B)(3) will not have a significant economic impact on a substantial number of small entities.

With respect to § 1.965-7(e)(2)(ii)(B) regarding the revocation of the election under section 965(n), the Treasury Department and the IRS have determined that § 1.965-7(e)(2)(ii)(B) will not have a significant economic impact on a substantial number of small entities for the reasons described in part III of the Special Analyses section in TD 9864, 84 FR 1838 (February 5, 2019). Accordingly, it is hereby certified that the collection of information requirements of § 1.965-7(e)(2)(ii)(B) will not have a significant economic impact on a substantial number of small entities.

Pursuant to section 7805(f), the proposed regulations preceding these final regulations (REG-104390-18 and REG-105600-18) were submitted to the Chief Counsel for Advocacy of the Small Business Administration for comment on their impact on small business.

IV. Unfunded Mandates Reform Act

Section 202 of the Unfunded Mandates Reform Act of 1995 requires that agencies assess anticipated costs and benefits and take certain other actions before issuing a final rule that includes any Federal mandate that may result in expenditures in any one year by a state, local, or tribal government, in the aggregate, or by the private sector, of $100 million in 1995 dollars, updated annually for inflation. In 2019, that threshold is approximately $154 million. These regulations do not include any Federal mandate that may result in expenditures by state, local, or tribal governments, or by the private sector in excess of that threshold.

V. Executive Order 13132: Federalism

Executive Order 13132 (entitled “Federalism”) prohibits an agency from publishing any rule that has federalism implications if the rule either imposes substantial, direct compliance costs on state and local governments, and is not required by statute, or preempts state law, unless the agency meets the consultation and funding requirements of section 6 of the Executive Order. These regulations do not have federalism implications and do not impose substantial direct compliance costs on state and local governments or preempt state law within the meaning of the Executive Order.

VI. Congressional Review Act

The Administrator of the Office of Information and Regulatory Affairs of the OMB has determined that this Treasury decision is a major rule for purposes of the Congressional Review Act (5 U.S.C. 801 et seq.) (“CRA”). Under section 801(3) of the CRA, a major rule takes effect 60 days after the rule is published in the Federal Register. Notwithstanding this requirement, section 808(2) of the CRA allows agencies to dispense with the requirements of section 801 of the CRA when the agency for good cause finds that such procedure would be impracticable, unnecessary, or contrary to the public interest and that the rule shall take effect at such time as the agency promulgating the rule determines.

Pursuant to section 808(2) of the CRA, the Treasury Department and the IRS find, for good cause, that a 60-day delay in the effective date is unnecessary and contrary to the public interest. The statutory provisions to which these rules relate were enacted on December 22, 2017 and apply to taxable years of foreign corporations and to the taxable years of United States persons in which or with which such taxable years of foreign corporations end. In certain cases, these taxable years have already ended. This means that the statutory provisions are currently effective, and taxpayers may be subject to Federal income tax liability for their 2017 or 2018 taxable years reflecting these provisions. In certain cases, taxpayers may be required to file returns reflecting this Federal income liability during the 60-day period that begins after this rule is published in the Federal Register.

These final regulations provide crucial guidance for taxpayers on how to apply the relevant statutory rules, compute their tax liability and accurately file their Federal income tax returns. These final regulations resolve statutory ambiguity, prevent abuse and grant taxpayer relief that would not be available based solely on the statute. Because taxpayers must already comply with the statute, a 60-day delay in the effective date of the final regulations is unnecessary and contrary to the public interest. A delay would place certain taxpayers in the unusual position of having to determine whether to file tax returns during the pre-effective date period based on final regulations that are not yet effective. If taxpayers chose not to follow the final regulations and did not amend their returns after the regulations became effective, it would place significant strain on the IRS to ensure that taxpayers correctly calculated their tax liabilities. For example, in cases where taxpayers and their CFCs have engaged in disqualified transfers or other abusive transactions, a delayed effective date may hamper the IRS' ability to detect such transactions. Moreover, a delayed effective date could create uncertainty and possible restatements with respect to financial statement audits. Therefore, the rules in this Treasury decision are effective on the date of publication in the Federal Register and apply in certain cases to taxable years of foreign corporations and United States persons beginning before such date.

The foregoing good cause statement only applies to the 60-day delayed effective date provision of section 801(3) of the CRA and is permitted under section 808(2) of the CRA. The Treasury Department and the IRS hereby comply with all aspects of the CRA and the Administrative Procedure Act (5 U.S.C. 551 et seq.).

Drafting Information

The principal authors of the regulations are Jorge M. Oben, Michael A. Kaercher, and Karen Cate of the Office of Associate Chief Counsel (International), Jennifer N. Keeney of the Office of the Associate Chief Counsel (Passthroughs and Special Industries), and Katherine H. Zhang and Kevin M. Jacobs of the Office of Associate Chief Counsel (Corporate). However, other personnel from the Treasury Department and the IRS participated in the development of the regulations.

Effect on Other Documents

The following publications are obsolete as of June 21, 2019:

Notice 2009-7 (2009-3 I.R.B. 312).

Notice 2010-41 (2010-22 I.R.B. 715).

Statement of Availability of IRS Documents

IRS Revenue Procedures, Revenue Rulings, notices, and other guidance cited in this document are published in the Internal Revenue Bulletin (or Cumulative Bulletin) and are available from the Superintendent of Documents, U.S. Government Publishing Office, Washington, DC 20402, or by visiting the IRS website at http://www.irs.gov.

List of Subjects in 26 CFR Part 1

• Income taxes
• Reporting and recordkeeping requirements

Adoption of Amendments to the Regulations

Accordingly, 26 CFR part 1 is amended as follows:

PART 1—INCOME TAXES

Paragraph 1. The authority citation for part 1 is amended by adding entries for §§ 1.78-1, 1.861-12, 1.951-1, 1.951A-2, 1.951A-3, 1.951A-5, 1.1502-51, 1.6038-5 in numerical order to read in part as follows:

Authority: 26 U.S.C. 7805 * * *

Section 1.78-1 also issued under 26 U.S.C. 245A(g).

Section 1.861-12 also issued under 26 U.S.C. 864(e)(7).

Section 1.951-1 also issued under 26 U.S.C. 7701(a).

Section 1.951A-2 also issued under 26 U.S.C. 882(c)(1)(A) and 954(b)(5).

Section 1.951A-3 also issued under 26 U.S.C. 951A(d)(4).

Section 1.951A-5 also issued under 26 U.S.C. 951A(f)(1)(B).

Section 1.1502-51 also issued under 26 U.S.C. 1502.

Section 1.6038-5 also issued under 26 U.S.C. 6038.

Par. 2. Section 1.78-1 is revised to read as follows:

Par. 3. Section 1.861-12 is amended by revising paragraph (c)(2) and adding paragraph (k) to read as follows.

Par. 4. Section 1.861-12T is amended by revising paragraph (c)(2)(i) to read as follows:

Par. 5. Section 1.951-1 is amended by:

1. Revising paragraph (a) introductory text.

2. Revising paragraphs (b)(1)(ii), (b)(2), (c), (e), and (g)(1).

3. Adding paragraphs (h) and (i).

The revisions and addition read as follows:

Par. 6. Sections 1.951A-0 through 1.951A-7 are added to read as follows:

Par. 7. Section 1.965-7 is amended by:

1. Revising the last sentence of paragraph (e)(1)(i).

2. Adding three sentences at the end of paragraph (e)(1)(i).

3. Adding paragraph (e)(1)(iv).

4. Revising paragraph (e)(2)(ii).

5. Adding paragraph (e)(3).

The revisions and additions read as follows:

Par. 8. Section 1.1502-12 is amended by adding paragraph (s) to read as follows:

Par. 9. Section 1.1502-32 is amended by adding and reserving paragraphs (b)(3)(ii)(E) and (b)(3)(iii)(C).

Par. 10. Section 1.1502-51 is added to read as follows:

Par. 11. Section 1.6038-2 is amended by revising the section heading, the introductory text of paragraph (a), and paragraph (m) to read as follows:

Par. 12. Section 1.6038-5 is added to read as follows:

Footnotes