Tax Notes logo

Shaking the CPM’s Persistent ‘Method of Last Resort’ Label

Posted on Nov. 14, 2022

This is the first article in a three-part series.

The comparable profits method is still often cast as uniquely conceptually flawed among transfer pricing methods and only suitable for use as a last resort, but this characterization often draws heavily on far-fetched statutory interpretations.

Despite the IRS’s victory in securing judicial acceptance of the CPM in Coca-Cola Co. v. Commissioner, 155 T.C. 145 (2020), the method still has not quite shaken its method-of-last-resort status in the eyes of some Tax Court judges, taxpayer expert witnesses, private practitioners, and commentators.

The CPM, described in reg. section 1.482-5, is a profit-based transfer pricing method often favored by the IRS as a means of limiting the taxable income shifted by transfers of U.S.-developed intangibles to a tax-favored offshore affiliate. By design, the method limits the returns available to the tested party — the offshore transferee in the typical scenario — to a fixed return in the form of operating profit as a share of sales, costs, or assets. Any residual profit (or loss) is allocated to the other controlled party to the transaction by default.

Despite its widespread use, the method has been controversial since its inception. A relatively new innovation in the long and eventful history of section 482 (and its predecessor, section 45), the CPM was not formally specified by the section 482 regulations as a potential method until the 1994 regulatory overhaul. Many multinational taxpayers and their representatives understandably dislike a method that drains away most of the benefit of their intellectual property tax planning schemes. This motivation, along with the theoretical objections of many observers, has ensured that efforts to delegitimize the method continue almost 30 years after its introduction.

The anti-CPM campaign was on display in the Medtronic II trial, when Medtronic expert witness Glenn Hubbard testified about the general superiority of the comparable uncontrolled transaction method over the CPM. (Prior coverage: Tax Notes Int’l, June 28, 2021, p. 1862.) The effort was evidently successful, at least for now, in persuading Judge Kathleen Kerrigan to reject the IRS’s attempt to apply the CPM using Medtronic’s offshore manufacturing subsidiary as the tested party. Kerrigan cites Hubbard at fawning length in her August opinion in Medtronic Inc. v. Commissioner, T.C. Memo. 2022-84. (Prior coverage: Tax Notes Int’l, Aug. 22, 2022, p. 893; prior analysis: Tax Notes Int’l, Oct. 3, 2022, p. 7.)

The anti-CPM effort is also visible in the disdainful commentary of some transfer pricing economists. (Prior analysis: Tax Notes Int’l, Dec. 6, 2021, p. 1129.) And it is evident in taxpayers’ explicit claims that the CPM, as a profit-based method, is an inherently disfavored method of last resort appropriate only when a traditional transactional method like the CUT method or the comparable uncontrolled price method cannot be applied. (Prior analysis: Tax Notes Int’l, Dec. 7, 2020, p. 1279.)

The attacks on the CPM fall into three distinct categories: statutory arguments, regulatory arguments, and theoretical economics arguments. This article focuses on statutory arguments; I will address the remaining two in subsequent articles.

Textualist Corrective

The statutory arguments for favoring traditional transactional methods over the CPM, like the regulatory arguments against the method, are flawed, have been flawed for as long the CPM has been a specified transfer pricing method, and will continue to be flawed for the foreseeable future. Regardless, they persist. Even after the 1986 statutory amendment and the 1994 regulatory overhaul, many still believe (or at least argue) that the statutory and regulatory framework favors other transfer pricing methods over the CPM. (Prior analysis: Tax Notes Int’l, June 27, 2022, p. 1609.)

This view often draws on some highly unorthodox statutory arguments. Some anti-CPM proponents evidently believe that section 482 — perhaps unlike any other statute ever enacted — has an inherent and immutable essence that stands apart from, and takes precedence over, the statutory text. (Prior analysis: Tax Notes Int’l, Aug. 15, 2022, p. 776.) This statutory essence is of course the arm’s-length standard, and it is so firmly fixed as the foundation of section 482 that it renders the law impervious to acts of Congress or regulatory action. (Prior analysis: Tax Notes Int’l, June 11, 2018, p. 1249.)

As is clear from the regulations, in particular reg. section 1.482-1(b)(1), Treasury has elected of its own volition to embrace the arm’s-length standard in principle. However, the detractors of the CPM and other inconvenient methodological rules often allege that the more specific and substantive provisions typically found outside reg. section 1.482-1, the IRS’s interpretation of the regulations in litigation, or both are often unfaithful to the inherently transactional arm’s-length standard. (Prior analysis: Tax Notes Int’l, Nov. 26, 2018, p. 872.)

The statute’s extratextual mandate, according to this view, inherently and inalterably favors transactional methods like the CUP method or the CUT method over profit-based methods like the CPM. Any express statements to the contrary that appear scattered throughout numerous regulatory provisions, most of which have never been subject to any Administrative Procedure Act or Chevron challenge, evidently must give way to this inherently tipped methodological scale. (Prior analysis: Tax Notes Int’l, Aug. 29, 2022, p. 993.)

Adherents of this and similar views also draw on language carefully extracted from introductory sections of the regulations that, read in isolation, could suggest regulatory submission to section 482’s unspoken statutory imperative. (Prior analysis: Tax Notes Int’l, Apr. 20, 2020, p. 277.) For example, reg. section 1.482-1(b)(1) says that a controlled transaction meets the arm’s-length standard if the result is consistent with what would have resulted if uncontrolled taxpayers had engaged in “the same transaction under the same circumstances” or, if no “identical transactions” can be located, “comparable transactions under comparable circumstances.” (Emphasis added.) The same section also says that “the standard to be applied in every case is that of a taxpayer dealing at arm’s length with an uncontrolled taxpayer,” and it is undeniable that parties dealing at arm’s length rarely mirror the mechanics of the CPM by awarding their counterparties a guaranteed return on sales, costs, or assets.

Overinterpreted statements by Treasury and the IRS in nonbinding guidance, such as those in a 1988 white paper (Notice 88-123, 1988-2 C.B. 458) and in the preamble to the 1994 final regulations, are often pressed into the service of anti-CPM arguments as well. In particular, the 1994 preamble’s reference to the CPM as “a method of last resort” was seized upon in Coca-Cola in an unsuccessful attempt to discredit the IRS’s selection of the CPM.

It’s worth taking a step back to appreciate the unusual nature of these arguments, which show strikingly little regard for the statutory text. The judicial philosophy of textualism, which has been marshaled in support of many highly debatable court decisions over the years, has earned its controversial status. But a measured dose of textualism may be a necessary corrective when, as is the case for section 482, the text of a statute and its construction long ago parted ways.

It would be hard to conceive of a more radically anti-textualist approach than one that recognizes the authority of some binding extratextual statutory essence that controls the meaning of section 482. Such an approach rests on some of the “false notions” identified by former Supreme Court Justice Antonin Scalia and academic Bryan A. Garner in their influential 2012 book on textual construction, Reading Law: The Interpretation of Legal Texts. These include the “false notion that the spirit of a statute should prevail over its letter” and “the false notion that the quest in statutory interpretation is to do justice” — where justice is presumably equated with the arm’s-length standard.

But one need not be a textualist fundamentalist or a Scalia acolyte to believe that a statute’s meaning comes principally from its text, not from its purported essence. And there is nothing in the text of section 482 that requires Treasury’s fealty to the arm’s-length standard, much less any anti-CPM interpretation thereof, as the one true basis for exercising its statutory authority to clearly reflect the income of controlled taxpayers. The relevant text comprises the first sentence of section 482, which accounted for the entire statute before its 1986 amendment:

In any case of two or more organizations, trades, or businesses (whether or not incorporated, whether or not organized in the United States, and whether or not affiliated) owned or controlled directly or indirectly by the same interests, the Secretary may distribute, apportion, or allocate gross income, deductions, credits, or allowances between or among such organizations, trades, or businesses, if he determines that such distribution, apportionment, or allocation is necessary in order to prevent evasion of taxes or clearly to reflect the income of any of such organizations, trades, or businesses.

Nothing in this text remotely supports the proposition that Treasury has a statutory obligation to irrevocably equate the phrase “clearly to reflect the income” with application of the arm’s-length standard. And contrary to popular belief, no binding case law precedent limits Treasury’s regulatory authority to implementation of the arm’s-length standard either.

The Supreme Court never pegged the meaning of section 482 to the arm’s-length standard, and even if it did, an opinion issued half a century ago cannot dictate the meaning of a 1986 statutory amendment or predetermine the amendment’s effects on Treasury’s regulatory authority. (Prior analysis: Tax Notes Int’l, Feb. 7, 2022, p. 646.) Neither can the Tax Court’s blocked income precedent, which generally misinterpreted Commissioner v. First Security Bank of Utah, 405 U.S. 394 (1972), as foisting the arm’s-length standard onto the pre-1986 version of the statutory text. If no statutory text or binding case law precedent compels Treasury to accept any interpretation of the arm’s-length standard, then there is certainly none that requires acquiescence to the antiquated transactional interpretation associated with attempts to discredit the CPM.

The only other possible external constraint on Treasury’s regulatory authority under section 482 comes from U.S. bilateral treaty commitments. Whether the articles of the United States’ tax treaties that correspond to article 9 of the 2016 U.S. model tax convention require adherence to the arm’s-length principle is debatable in light of the article 1 savings clause. (Prior analysis: Tax Notes Int’l, Oct. 10, 2022, p. 139.) Whether profit-based methods like the CPM (or its OECD equivalent, the transactional net margin method) are inherently inferior to traditional transactional methods under the arm’s-length principle as endorsed by article 9 is clear. The OECD transfer pricing guidelines, which serve as an authoritative extended commentary on article 9, have not endorsed any hierarchy of methods since the 2010 version of the guidelines adopted a “most appropriate method” rule similar to reg. section 1.482-1(c)’s best method rule.

No Regulatory Acquiescence

If Treasury’s section 482 regulations aren’t bound by any higher legal authority to favor transactional methods over the CPM, they do not voluntarily acquiesce to that approach either. In some cases, the regulations even expressly contradict flawed interpretations of the arm’s-length standard that CPM critics often read into section 482 and use to degrade the standing of the CPM. One example appears in reg. section 1.482-1(f)(2)(v)(A), which specifies that the best method under the regulations need not correspond to any method that uncontrolled parties might use to price uncontrolled transactions. Whether parties transacting at arm’s length would mimic the CPM by guaranteeing each other a fixed return is therefore beside the point.

To the extent that any introductory language contained in reg. section 1.482-1(b)(1) suggests a preference for transactional methods, that suggestion is expressly and thoroughly dispelled by reg. section 1.482-1(b)(2) (arm’s-length methods), reg. section 1.482-1(c) (best method rule), and the cross-referenced reg. sections on specific categories of transaction and individual methods. By the general/specific canon of construction (generalia specialibus non derogant) endorsed by Scalia and Garner, the specific rules detailed throughout the rest of the section 482 regulations should carry greater weight than the general introductory language found almost entirely near the beginning of reg. section 1.482-1.

The transaction-specific sections cross-referenced by reg. section 1.482-1(b)(2), and favored by the general/specific canon, include the sections on tangible property transfers, intangible property transfers, and services — which all identify the CPM described in reg. section 1.482-5 as a potential method. It is therefore the specific provisions of the section 482 regulations themselves, and not any imaginary statutory constraints, that determine the reliability of the CPM. And nothing in reg. section 1.482-5 or in the reg sections that incorporate it by reference lend support to the idea that the CPM is inherently disfavored.

If the criteria set out in reg. section 1.482-5(c) (comparability and reliability considerations) and any specific CPM reliability factors relevant for a particular category of transaction weigh in favor of the CPM, and no other method satisfies the reliability conditions set by the relevant regulatory provision, then the law compels selection of the CPM. Nothing in the statute, regulations, or case law supports a different conclusion.

Copy RID