To the dismay of many practitioners and business trade groups, the foreign tax credit regulations issued in January (T.D. 9959) finalized what the 2020 proposed regulations (REG-101657-20) called the “jurisdictional nexus requirement” with only minor changes. Rebranded the “attribution requirement” by the final regulations, the controversial provisions prevent the creditability of foreign tax measures that do not conform to what Treasury and the IRS consider to be the international norms applicable to the taxing jurisdiction. As is evident from its initial name and the explanation provided in the preamble to the final regulations, the primary targets of the attribution requirement that now appears in reg. section 1.901-2(b)(5) were extraterritorial taxes such as digital services taxes and diverted profits taxes that flout traditional taxable nexus standards.
“Recently, many foreign jurisdictions have disregarded international taxing norms to claim additional tax revenue, resulting in the adoption of novel extraterritorial taxes that diverge in significant respects from U.S. tax rules and traditional norms of international taxing jurisdiction. These extraterritorial assertions of taxing authority often target digital services, where countries seeking additional revenue have chosen to abandon international norms to assert taxing rights over digital service providers,” the preamble to the final FTC regulations says. Noting that the digital economy did not exist when the FTC was first introduced, it states that the attribution requirement is “necessary and appropriate to more clearly delineate the circumstances in which a tax does not qualify as an income tax in the U.S. sense due to the foreign jurisdiction’s unreasonable assertion of jurisdictional taxing authority.”
Reg. section 1.901-2(b)(5)(i) establishes the attribution requirement’s conditions for a foreign tax imposed on nonresidents of the jurisdiction that imposes the tax, most of which deal with the taxable nexus issues that represent the clear focus of the regulatory project. However, reg. section 1.901-2(b)(5)(ii), which addresses foreign taxes imposed on residents of the same foreign jurisdiction, sets conditions that go considerably beyond nexus standards. As part of an attempt to enforce “jurisdictional taxing norms” more broadly, reg. section 1.901-2(b)(5)(ii) essentially conditions the foreign tax’s creditability on the foreign jurisdiction’s adoption of the arm’s-length principle as the basis for its transfer pricing regime. Under that section, a foreign jurisdiction’s tax on resident taxpayers must determine any allocation income, gain, deductions, or losses among associated enterprises “under arm’s length principles, without taking into account as a significant factor the location of customers, users, or any other similar destination-based criterion.”
Reg. section 1.901-2(b)(5)(ii) consists of two seemingly independent conditions: First, the foreign tax’s allocation mechanism must reflect arm’s-length principles and, second, it must not treat the “location of customers, users, or any other similar destination-based criterion” as a significant factor. A literal reading of the text suggests that failure to satisfy either condition constitutes sufficient grounds for disqualification, which would exclude any foreign tax that deviates from arm’s-length principles, even if the specific deviation has nothing to do with any prohibited destination-based criteria. It would also disqualify foreign taxes that apply arm’s-length principles but would do so in a manner that treats the location of customers, users, or other similar destination-based criteria as significant factors.
Failure to satisfy these conditions does not merely lead to the denial of a credit for the difference between the foreign taxes actually paid and the foreign taxes that would have been paid if the foreign tax law had complied with reg. section 1.901-2(b)(5)(ii)’s requirements. As the preamble to the regulations explains, conformity with “established international jurisdictional norms, reflected in the Internal Revenue Code and related guidance, for allocating profit between associated enterprises” goes to the heart of whether a foreign tax constitutes “an income tax in the U.S. sense.” Accordingly, a foreign tax imposed on residents of that jurisdiction is not a “foreign income tax” and cannot be credited at all, and there is little reason to expect Treasury and the IRS to offer any relief in the foreseeable future.
Although the breadth of these conditions and the drastic consequences for failing to satisfy them are striking to say the least, the new FTC regulations contain some mitigating aspects. One is the use of “foreign tax law,” as defined by reg. section 1.901-2(g)(4), instead of the specific manner in which the foreign law is applied in particular cases as the basis for assessment. If only the terms of “statutes, regulations, case law, and administrative rulings or other official pronouncements, as modified by an applicable income tax treaty” are considered, the behavior of an unscrupulous tax administration that routinely misapplies compliant national transfer pricing laws cannot, by itself, disqualify that country’s foreign tax. The relevant legislation or regulations in most countries at least nominally endorse the arm’s-length principle in general, and few (if any) such general legislative or regulatory endorsements explicitly mandate the consideration of any of reg. section 1.901-2(b)(5)(ii)’s prohibited destination-based criteria.
Another possible mitigating feature is reg. section 1.901-2(b)(5)(ii)’s use of the plural term “arm’s length principles,” along with language in the regulatory preamble noting that there is no requirement for “strict conformity” between U.S. and foreign tax law, which suggests that there is at least some flexibility in assessing the consistency of the foreign jurisdiction’s transfer pricing rules. Reg. section 1.901-2(a)(1)(iii) also contains a general treaty coordination rule, which deems any foreign levy covered by the double taxation article of one of the United States’ bilateral income tax treaties to be a qualifying foreign income tax.
None of these mitigating provisions are entirely reassuring. The treaty coordination rule is clearly of no use when the specific jurisdiction concerned is not part of the United States’
68-country treaty network, which excludes some significant U.S. trading partners such as Brazil, Chile, Malaysia, Singapore, Taiwan, and Vietnam. A much larger problem is that the treaty coordination rule appears to be of no use at all if the application of a foreign jurisdiction’s non-arm’s-length transfer pricing rules increases the tax liability of a subsidiary resident in that jurisdiction. As has been noted by trade groups, the regulatory text of the coordination rule covers only taxes paid by U.S. residents and, in limited cases, taxes paid in a treaty jurisdiction by a controlled foreign corporation resident in a
With no protection from the treaty coordination rule, a foreign tax on resident taxpayers faces a wide range of hazards that could jeopardize its creditability. Regarding reg. section 1.901-2(b)(5)(ii)’s required use of arm’s-length principles, there is no guidance on the extent of the flexibility contemplated by the regulations. Despite the emergence of major interpretive differences between the United States and other jurisdictions on transfer pricing issues after the base erosion and profit-shifting project, nothing in reg. section 1.901-2(b)(5)(ii), the examples, or the regulatory preamble helps taxpayers locate the demarcating line between a tolerable level of deviation and a failure to follow arm’s-length principles.
The lack of specificity is especially problematic considering the importance and extent of some of these disagreements. Many relate to critical concepts and principles, including the standards for reallocating risk; the significance of DEMPE (development, enhancement, maintenance, protection, and exploitation) functions for intangibles; the application of the profit-split method; and the use of transfer pricing principles to characterize financial instruments as debt versus equity. In some cases, the interpretations favored by non-U.S. jurisdictions are not only plainly inconsistent with the United States’ section 482 regulations, but are likely at odds with the OECD transfer pricing guidelines as well. Although such disparate interpretations often manifest themselves through inconsistent tax administration practices rather than inconsistent legal provisions, they can also appear in one or more of the sources of foreign tax law cited in reg. section 1.901-2(g)(4).
It’s also easy to imagine how an initially creditable foreign income tax could become non-creditable over time as tax administrations’ practices gradually work their way into recognized sources of foreign law. Even if a country’s statutory or regulatory law formally endorses the arm’s-length principle, a renegade tax administration intent on misapplying it may eventually attempt to legitimize its approach by formalizing it in “administrative rulings or other official pronouncements.” Such rulings or pronouncements could also expressly endorse the kinds of destination-based criteria that fall within the scope of reg. section 1.901-2(b)(5)(ii)’s second disqualifying condition.
The inclusion of case law in reg. section 1.901-2(g)(4)’s list of legal sources presents a similar risk. If a tax administration were to successfully persuade a national court to accept a transfer pricing adjustment based on non-arm’s-length principles or prohibited destination-based criteria, the foreign tax may be disqualified based on case law. In the EU, a judgment by the Court of Justice of the European Union endorsing the European Commission’s highly unorthodox interpretation of the arm’s-length principle could have the same result. (Prior coverage: Tax Notes Int’l, Nov. 21, 2016, p. 739.) The risk is not a trivial one: As clearly attested by U.S. section 482 case law, courts often struggle to correctly interpret transfer pricing legislation and regulations.
Disregarding Destination-Based Criteria
Another problem with reg. section 1.901-2(b)(5)(ii) is the awkward juxtaposition of one condition that deals with the features of the foreign jurisdiction’s transfer pricing regime with another that focuses on taxable nexus standards. Denying creditability for taxes that use the location of customers or users as a nexus standard is consistent with the stated goal of Treasury and the IRS, which was to ensure that the United States’ FTC regime does not subsidize foreign jurisdictions’ extraterritorial tax policies. However, the text of the provision appends this prohibition on the consideration of destination-based criteria to the condition concerning the foreign jurisdiction’s transfer pricing regime. As noted earlier, under reg. section 1.901-2(b)(5)(ii), the foreign jurisdiction’s transfer pricing rules must require that any allocation be “determined under arm’s length principles, without taking into account as a significant factor the location of customers, users, or any other similar destination-based criterion.”
This language is problematic because once a sufficient taxable nexus has been established, it may be perfectly appropriate to consider “destination-based” criteria when applying the arm’s-length principle. As explicitly recognized in paragraph 1.164 of the OECD transfer pricing guidelines, local market features “may affect the arm’s length price with respect to transactions between associated enterprises.”
“The comparability and functional analysis conducted in connection with a particular matter may suggest that the relevant characteristics of the geographic market in which products are manufactured or sold, the purchasing power and product preferences of households in that market, whether the market is expanding or contracting, the degree of competition in the market and other similar factors affect prices and margins that can be realized in the market,” the OECD guidelines say. “Similarly, the comparability and functional analysis conducted in connection with a particular matter may suggest that the relative availability of local country infrastructure, the relative availability of a pool of trained or educated workers, proximity to profitable markets, and similar features in a geographic market where business operations occur create market advantages or disadvantages that should be taken into account.”
The United States’ own transfer pricing regulations take a consistent approach in reg. section 1.482-1(d)(4)(ii), which acknowledges the potential importance of differences in geographic markets. However, under the text of reg. section 1.901-2(b)(5)(ii), the kinds of local market features that are almost universally accepted as relevant in the application of the arm’s-length principle — including by the OECD guidelines and the section 482 regulations — could be considered “similar destination-based” criteria that disqualify a foreign levy as a creditable foreign income tax. This may not be the intended meaning of the phrase “similar destination-based criterion,” but none of the regulations’ examples help to clarify the intended meaning.
A Questionable Target
The regulations’ clumsy incorporation of a nexus-related provision into its standard for an acceptable transfer pricing regime raises the broader question of why it was considered necessary or desirable to make the adoption of arm’s-length principles part of the new attribution requirement in the first place. The United States’ strong opposition to DSTs, diverted profits taxes, and other measures that disregard traditional nexus standards is well-established, and the sense of urgency on the part of the IRS and Treasury to ensure that the FTC regime does not subsidize such measures is understandable. But the arm’s-length principle specifies how profit should be allocated among associated enterprises, not whether those enterprises have a taxable nexus in any particular jurisdiction. Why was it necessary to draw the foreign jurisdiction’s transfer pricing system into the discussion when the failure to follow established nexus standards was the primary concern?
The preamble to the proposed regulations suggests that concerns about other countries’ adoption of formulary apportionment, which could theoretically sweep away the limits imposed by an independent taxable nexus standard, may have led Treasury and the IRS to introduce the conditions now contained in reg. section 1.901-2(b)(5)(ii). “Foreign transfer pricing rules that allocate profits by taking into account as a significant factor the location of customers, users, or any other similar destination-based criterion will not satisfy the jurisdictional nexus requirement,” the proposed regulations’ preamble says. “Comments are requested on whether special rules are needed to address foreign transfer pricing rules that allocate profits to a resident on a formulary basis (rather than on the basis of arm’s length prices), such as through the use of fixed margins in a manner that is not consistent with arm’s length principles.”
If the specter of foreign jurisdictions’ adoption of formulary apportionment was the motivation, reg. section 1.901-2(b)(5)(ii) represents a drastic solution to an almost entirely hypothetical problem. Countries participating in the inclusive framework on BEPS may have reached consensus on a major departure from established nexus and profit allocation principles in the form of pillar 1, but few (if any) are advocating a global system of formulary apportionment.
Reg. section 1.901-2(b)(5)(ii) is also disproportionate as a response to “the use of fixed margins in a manner that is not consistent with arm’s length principles,” a practice followed only by Brazil. There is nothing new about Brazil’s use of fixed-margin methods instead of the arm’s-length principle, and the country is currently in the process of reforming its system to better conform to OECD-approved transfer pricing principles. It is unclear why this was considered the opportune moment to target an unorthodox transfer pricing regime that a single country has followed for decades, just as that country is working on aligning its rules with “jurisdictional taxing norms.”
In its explanation of the new measures, the preamble to the final regulations claims that it is appropriate to include the attribution requirement in the definition of what constitutes “an income tax in the U.S. sense.” The implication is that an income tax isn’t really an income tax — at least not “in the U.S. sense” — unless its mechanism for allocating income among associated enterprises follows arm’s-length principles. But Treasury and the IRS’s claim that transfer pricing principles drawn from their own regulations, not from statutory law, are a necessary part of what makes the U.S. federal income tax an “income tax in the U.S. sense” is hard to take seriously. Accepting this view would mean that any tax regime that relies on formulary apportionment — even if it is universally understood and referred to as an income tax — is in fact something other than an income tax.
Income taxes imposed by U.S. states are an obvious example. In its response to the proposed regulations, the United States Council for International Business noted the bizarre implication that U.S. state income taxes are not income taxes. The preamble to the final regulations brushes this objection aside, noting that U.S. state law does not control the meaning of federal law.
But this is beside the point, which is that singling out one particular mechanism for allocating income among related parties, and abruptly claiming in the FTC context that this particular mechanism is in fact a required element of the definition of an income tax, is novel, extreme, and clearly at odds with the standard usage of the term. It is highly doubtful that, in the 109 years since the 16th Amendment’s ratification, a U.S. federal tax law practitioner has ever heard a reference to a state’s income tax and responded with a look of confusion or an admonishment that there’s actually no such thing.