Menu
Tax Notes logo

News Analysis: Big Decisions Loom as IRS Updates Transfer Pricing Rules

Posted on Feb. 25, 2019

Over a year after the Tax Cuts and Jobs Act enacted the first statutory change to the transfer pricing rules since 1986, efforts to update the regulations and align them with the statute remain underway.

After prioritizing guidance projects involving the TCJA’s more novel anti-profit-shifting provisions in 2018, the IRS and Treasury included the section 482 regulations in two of the high-priority projects listed in the 2018-2019 priority guidance plan. According to the guidance plan, the IRS is working on “regulations addressing the changes to [sections] 367(d) and 482” as part of its ongoing TCJA implementation effort. The regulations will presumably reflect the TCJA’s expanded definition of intangible property under section 367(d)(4) and amended section 482, which now explicitly authorizes valuation of intangible transfers on an aggregate basis or use of realistic alternatives to the transaction.

The guidance plan also includes “final regulations under [section] 482, including with respect to the treatment and allocation of risk” among its general guidance projects. The description of the project cites the temporary regulations (T.D. 9738), which expired in September 2018, on aggregation of transactions issued in 2015 under reg. section 1.482-1T, but does not specify whether the controversial provisions of those regulations will be reintroduced or how they relate to risk.

The guidance plan’s terse descriptions do not provide much indication of what changes are under consideration and IRS officials declined requests for comment. However, the nature of the TCJA’s amendments and the IRS’s historic struggles hint at some of the more likely changes. Whether more fundamental changes are forthcoming is unclear, but IRS officials have in the past pondered the need for a rewrite in response to the IRS’s poor litigation record and practitioners have suggested there is ample room for improvement.

Alignment Error

The TCJA either fundamentally changed the transfer pricing rules or made no substantive changes at all, depending on the hotly debated meaning of prior law. Although the TCJA’s section 482 amendment codified concepts that were part of the 1994 final section 482 regulations and have not faced serious challenges to their validity, the IRS has repeatedly failed to use aggregation or realistic alternatives to defend its intangible valuation methods in court. The TCJA simultaneously removed one of the main reasons courts have rejected such methods by explicitly including goodwill, going concern value, and workforce in place — each arguably covered by “other similar items” under prior law — in the definition of intangible property.

Treasury and IRS officials have said the Internal Revenue Code’s definition of intangible property never created an “artificial valuation limitation” that allows intangible assets not covered by that definition to be transferred free of charge. However, the definition of intangible property under former section 936(h)(3)(B) has plagued the IRS for years in its efforts to enforce the section 482 regulations in cases involving high-profit intangible transfers, including some of the highest-profile U.S. transfer pricing cases. (Prior analysis.)

The dispute over the interpretation of prior law continues in the government’s appeal of Amazon.com v. Commissioner, 148 T.C. No. 8 (2017), but the TCJA seemingly settled the issue by removing section 936(h)(3)(B) and replacing it with a definition under section 367(d)(4) that explicitly includes goodwill, going concern value, and workforce in place. However, the Tax Court’s analysis in Amazon and its predecessor, Veritas Software Corp. v. Commissioner, 133 T.C. 297 (2009), did not turn directly on the definition in section 936(h)(3)(B) itself. The court reasoned that the cost-sharing regulations then in force required a buy-in payment only for “intangibles,” which the pre-2009 regulations defined by cross-reference to reg. section 1.482-4(b), which in turn contains a definition taken almost verbatim from then-section 936(h)(3)(B).

Because the Amazon and Veritas decisions contain no analysis of the section 482 commensurate with income standard, which cross-references section 936(h)(3)(B) instead of reproducing it, the only definition directly relevant to the court’s assessment was the nearly identical definition that still appears in reg. section 1.482-4(b).

Under this interpretation, intangible transfers structured other than as cost-sharing arrangements could theoretically benefit from the same “artificial valuation limitation” until conforming changes are made to reg. section 1.482-4. The 2009 temporary (T.D. 9441) and 2011 final (T.D. 9568) cost-sharing regulations avoid the issue by introducing the concept of a “platform contribution,” defined as “any resource, capability, or right that a controlled participant has developed, maintained, or acquired externally to the intangible development activity (whether prior to or during the course of the [cost-sharing arrangement]) that is reasonably anticipated to contribute to developing cost-shared intangibles.” However, there is no automatic update mechanism for transactions subject to reg. section 1.482-4.

One likely change will therefore be to either cross-reference the new section 367(d)(4) or reproduce the updated definition. Although it bears on one of the most contentious issues in U.S. transfer pricing, officials who have signaled that this change will be forthcoming have characterized the update as a “housekeeping” exercise.

The other need for alignment concerns section 482’s newly added sentence on aggregation and realistic alternatives. The IRS allowed the September 2018 deadline to pass for extending temp. reg. section 1.482-1T, which attempted to shore up the government’s authority to aggregate transactions as necessary to reflect “all value provided between controlled taxpayers in a controlled transaction.”

Some practitioners believe the temporary regulations’ sweeping “all value” approach may be unnecessary and excessive even under the amended statutes. Citing synergies as an example, one practitioner suggested that the “all value” concept draws in value attributable to some things that the post-TCJA definition of intangibles still may not cover.

“One has to question how necessary the temporary [section] 482 regs are, because by definition now when you analyze a series of transactions, all of the value should be captured,” Thomas Vidano of EY said during a February 5 Practising Law Institute seminar. “But I think the question is how far can they go? How far is this ‘all value’ concept going to go within the temporary regulations?”

Government officials have been ambiguous on whether the updated regulations will ultimately retain the temporary regulations’ explicit assertions of broad aggregation authority or simply defer to the new statutes. The IRS did not immediately abandon its plan to finalize the temporary regulations and in mid-2018 officials still predicted final regulations before the September sunset deadline. However, after the deadline passed, officials said the project was delayed by the need to align the final regulations with the amended statute.

Good Housekeeping

As the IRS carries out its section 482 housekeeping project, it should consider other changes that would improve clarity and reduce compliance burdens, according to Guy Sanschagrin of WTP Advisors . The introduction of safe harbors and materiality thresholds could help reduce compliance costs for taxpayers while allowing the IRS to better allocate its enforcement resources, Sanschagrin told Tax Notes. As an example, Sanschagrin proposed a safe harbor for services that better aligns with the OECD transfer pricing guidelines’ simplified method for low-value-adding services.

“To help the IRS and taxpayers prioritize their efforts, I would encourage the IRS to provide a safe harbor markup rate for ‘routine’ and ‘ancillary’ services beyond the services cost method. This safe harbor would explicitly exclude high-value services, services that utilize intangible property, and services that are core to the business (e.g., software development in a software company),” Sanschagrin said. “A safe harbor markup rate would enable taxpayers to focus their analysis of the cost base — which typically represents the bulk of the risk associated with services transactions.”

The regulations would benefit from a general cleanup that clarifies when measures that arguably diverge from the traditional arm’s-length standard apply, Sanschagrin said.

“The guidance and regulations could benefit from harmonization of the arm’s-length standard with other aspects of the regulations, such as the commensurate with income standard and safe harbors,” Sanschagrin said. “The regulations and guidance could clarify that the arm’s-length standard is paramount — and that any additional standards or safe harbors are meant to support the application of the arm’s-length standard and assist taxpayers and the IRS in prioritizing their compliance efforts.”

Sanschagrin cautioned, however, that adding new language stating that targeted provisions are entirely consistent with the arm’s-length standard may create more confusion than clarity.

In addition to the organizational issues noted by Sanschagrin, the existing structure and layout of the regulations may contribute to confusion, including the reliance on chains of cross-references in defining the arm’s-length standard and other fundamental concepts. The definition of the arm’s-length standard incorporates the best method rule by reference, and the best method rule incorporates the subsection on comparability and the various sections on specific types of transactions.

The resulting confusion may contribute to the major interpretive differences between the IRS and the Tax Court, as cases decided against the IRS have placed relatively heavy emphasis on the brief description of the arm’s-length standard in reg. section 1.482-1(b) at the expense of the cross-referenced sections, which impose standards that seemingly rule out the comparable uncontrolled transaction method analyses accepted in Medtronic Inc. v. Commissioner, T.C. Memo. 2016-112, Amazon, and Veritas.

Other complications may arise because of the regulations’ outline, which alternates between sections devoted to general principles, sections on specific categories of transactions to which multiple transfer pricing methods may apply, and sections on specific transfer pricing methods that could apply to multiple types of transactions. This layout may contribute to the perception that transactions arguably not covered by any of the sections organized by transaction type — for example, transfers of assets that may not be “intangibles” subject to reg. section 1.482-4 or reg. section 1.482-7A — are not subject to the general transfer pricing rules under reg. section 1.482-1.

Risky Business

The priority guidance plan notes that temporary regulations were issued in 2015 and that the final regulations will address the “treatment and allocation of risk.” It does not indicate how far-reaching the potential changes to risk-related provisions may be or how they might relate to temporary regulations focused on aggregation of transactions. Unlike aggregation and intangible valuation methods, the transfer pricing treatment of risk was not directly addressed by the TCJA.

However, the reference to the “allocation of risk” does borrow a term that was fundamental to the OECD’s recent revisions to its transfer pricing guidance. According to the OECD’s 2015 release of its base erosion and profit-shifting report on actions 8-10, which has since been incorporated by the OECD transfer pricing guidelines, risk and the returns associated with bearing it should be reallocated if the entity that contractually bears the risk does not exercise control over the risk or does not have the financial capacity to bear it. Despite its role in developing and ultimately approving the OECD’s 2015 BEPS report, the United States has not yet taken any steps to adopt the risk allocation concepts that were central to the report’s recommendations.

The resistance is not surprising for a country that has generally sought to limit recharacterization and other concepts that resemble it. U.S. representatives to the BEPS project negotiations initially opposed OECD proposals on risk reallocation and recharacterization and later tried to narrow their scope. The IRS has generally sought to counter base-eroding intangible transfers with strong, economics-based ex ante pricing rules instead of substance requirements or antiabuse measures and this approach is built into the section 482 regulations.

In light of the IRS’s inability to persuade the Tax Court to apply these pricing rules, the success of this approach is questionable. The existing regulatory scheme has often worked against the IRS in major cases, as it did when the Tax Court cited the self-imposed restrictions on recharacterization under reg. section 1.482-1(f)(2)(ii)(A) while dismissing the government’s attempt to distinguish recharacterization from repricing.

Whether the guidance plan’s invocation of a key term of art in OECD guidance foreshadows a move to align with OECD standards is unclear. According to Reuven S. Avi-Yonah of the University of Michigan Law School, more fundamental changes should be considered that would simply disregard contractual risk allocations altogether. Avi-Yonah, a prominent critic of the arm’s-length standard, told Tax Notes that the internal allocation of risk within a multinational should carry no weight.

“I don’t think risk is a meaningful thing to allocate within multinationals. Most real risks are external and contractually allocating them to certain companies within a multinational invites profit shifting,” Avi-Yonah said.

Assuming the IRS has no intention of abandoning the arm’s-length standard, the most meaningful improvement would place greater emphasis on the profit-split method, according to Avi-Yonah. “I think adopting a formula to allocate the residual in applying the profit-split method would be the most significant improvement possible under the current [arm’s-length standard] based regime and would be consistent with the evolving view of the OECD as reflected in its recent policy note,” he said.

A greater emphasis on the profit-split method in the regulations may instead be a way to preserve and improve the arm’s-length standard in the post-BEPS environment, according to Thomas Zollo of KPMG LLP . Zollo told Tax Notes that the IRS, either through regulations or administrative practice, should recognize that times have changed since the 1994 regulations were issued. Although the BEPS project has effectively shut down cash boxes, the regulations still assume that the transferee is a tax haven entity with no functions other than providing cash, he said.

“Under the current regulations, the Service has attempted to relegate the profit split to the method of last resort,” Zollo said. “That should flip. The profit split is going to be the most often applicable approach, and if that’s the case, they ought to be giving some thought as to how they better define how you go about the profit split.”

The profit-split method, which under the U.S. regulations and OECD guidance is a “two-sided method,” is suitable when multiple parties contribute unique intangibles, generally calculate a routine return due to each party using a different transfer pricing method, and then divide the residual profit based on an allocation formula. In contrast, the IRS has generally favored valuation methods that, like the discounted cash flow methods used in Veritas and Amazon and the income method favored by the cost-sharing regulations, allocate all residual profit to the U.S. entity.

The flawed assumptions of this approach contributed to the IRS’s loss in Amazon and Veritas, according to Zollo. “When you have some intangibles or other capabilities on both sides of the transaction, whether it’s because the offshore participant is bringing its own intangibles, existing clientele, or complementary products to the transaction, the regulations don’t adequately provide guidance for rewarding the offshore participant for its contributions,” Zollo said. “They’re built on a paradigm that increasingly is not going to reflect the real facts.”

The downside of the profit-split method is its subjectivity, according to Zollo. Neither the U.S. regulations nor the OECD guidelines specify what profit should be split or what allocation factors are appropriate. If profit splits are the way of the future, the IRS would be wise to use this opportunity to provide more concrete guidance on how profit splits should be applied, Zollo said.

“Obviously it’s a big project to figure out how you come away with what you think would be workable profit-split methods for different fact scenarios, but that’s where I think it’d be worth investing more time than the regs have previously,” Zollo said.

Copy RID