Special Measures: Within or Beyond the Arm's-Length Principle

Trust Bloomberg BNA's Premier International Tax offering for the news and guidance to navigate the complex tax treaty networks and business regulations.


By Gary D. Sprague, Esq.

Baker & McKenzie LLP, Palo Alto, CA

On December 19, 2014, the OECD published a discussion draft under Actions 8 - 10 of the BEPS Action Plan dealing with proposed revisions to Chapter 1 of the OECD Transfer Pricing Guidelines (TPG), including risk, recharacterization, and special measures. The question of what "special measures" might be introduced to the TPG has, to my mind, been one of the most interesting, and challenging, proposals in the BEPS project. With the publication of the discussion draft, business, civil society, and academics now have an opportunity to comment on what role, if any, "special measures" should play in the application of the arm's-length principle.

"Special measures" have been contemplated as part of the BEPS output from the beginning. Each of Actions 8, 9, and 10, dealing respectively with intangibles, risks and capital, and "other high-risk transactions," noted the need to adopt "transfer pricing rules or special measures" to deal with each of the specified BEPS challenges. The Action Plan itself noted that the special measures needed to address the "flaws" in the current transfer pricing system might be "either within or beyond the arm's length principle."

The suggestion that the TPG might contain operative rules "beyond the arm's length principle" broke new ground, and immediately became the focus of attention in discussions of the transfer pricing Actions.2 Understandably, OECD spokesmen were circumspect in their comments on what was meant by a transfer pricing rule "beyond" the arm's-length principle as Working Party 6 began work on the topic. In public comments, spokesmen stated that it was not important to define when a special measure might be within, and might be beyond, the arm's-length principle.  With the publication of this discussion draft, we have for the first time some insight into what is contemplated.

Briefly, the discussion draft describes five options for potential special measures. In each of the five cases, the option describes a set of circumstances which might trigger the application of the special measure, and the consequences of the special measure being applied.

The first option addresses "hard-to-value intangibles." The discussion draft justifies the need for a special measure "because of the potential for systematic mispricing in circumstances where no reliable comparables exist, where assumptions used in valuation are speculative, and where information asymmetries between taxpayers and tax administrations are acute." The trigger would be cases where the taxpayer fixes the price either as a lump sum or as a fixed royalty rate on the basis of projections without a contingent payment mechanism, and does not contemporaneously document the projections and make them available to the tax administration.  The consequence of the special measure would be that the tax administration could presume that a price adjustment would have been adopted between the parties, and rebase the calculations based on the actual outcome, imputing a contingent payment mechanism.

The second and third options deal with "inappropriate returns for providing capital." These measures "seek to address issues arising from the freedom multinational enterprise (MNE) groups have (except in certain regulated sectors) to control their structures, including the creation and capitalisation of companies." The trigger for Option 2 is a case where a capital-rich, asset-owning company is dependent on another group company to earn a return on the asset. The policy foundation of this special measure is that an independent investor, having the option of investing in either the capital-rich company or the company the former relies upon to earn a return on the asset, might choose to invest in the latter on the basis that it constitutes the more rational investment opportunity.  The consequence of the Option 2 special measure is that the capital is treated as being contributed to the latter company for the purpose of acquiring the asset, and no return would be allocated to the capital-rich, asset-owning company.

Option 3 would impose a thick capitalization rule on group companies based on a pre-determined capital ratio.  While the discussion draft provides few details on this option, the trigger presumably would be capitalization in excess of a set ratio relative to debt or perhaps to other operating measures of the entity, and the consequence would be deemed interest deductions at the level of the capital-rich company and deemed interest income at the level of the capital provider.

Option 4 is to apply in the case of a "minimal functional entity." The policy goal of this option is to address cases where one of the parties to a transaction, "especially transactions transferring key business risks or intangibles," has minimal functions. The discussion draft notes that the participation of the minimal functional entity "may also be the root cause of an arrangement lacking the fundamental economic attributes that normally underpin arrangements between unrelated parties," which is a reference to a principle developed elsewhere in the discussion draft for inclusion in Chapter 1 of the TPG to address the application of the arm's-length principle in these cases. Nevertheless, the draft suggests that "[i]t may prove simpler and more effective, therefore, in dealing with such cases to adopt a targeted special measure that focuses on a level of functionality that, where lacking, would cause the profits of that entity to be reallocated."

The proposed trigger for Option 4 would be circumstances where the minimal functional entity lacks the functional capacity to create value through exploiting its assets and managing its risks, and instead is reliant on arrangements with other group companies to do so, and is seen to make modest contributions to profits based on a variety of measures, including its functions, personnel, intercompany arrangements, and assets. The consequence would be that the entity's profits would be reallocated. The draft notes three alternatives to be considered: (1) apply a mandatory profit split with other group companies providing the functional capacity to exploit the affected company's assets and manage its risks; (2) allocate its profits to its parent (and further up the chain, if necessary, until reaching an entity which is not a minimal functional entity); and (3) allocate its profits to those companies which provide the functional capacity to exploit the assets and manage the risks of the minimal functional entity.

Option 5 deals with "ensuring appropriate taxation of excess returns" and has a connection to the work on CFC rules. The policy goal here is to address cases where income is shifted intra-group to locations where it is taxed at a low rate or not taxed at all. This option would be triggered when the entity earns "excess returns," which could be targeted on returns associated with intangibles and risk, and such returns are taxed in the CFC below a certain rate. The consequence would be that the excess profits would be taxed in the parent jurisdiction at that threshold rate (not the normal rate in the parent jurisdiction). If the parent jurisdiction did not have a CFC rule that caused this result, then a secondary rule would apply to allocate taxing jurisdiction over the entity's excess returns to other jurisdictions. While not stated in the discussion draft, those other jurisdictions presumably would be some or all of those jurisdictions in which associated enterprises are located that participate in the value chain organized through the CFC.

The discussion draft invites comments on the options, and certainly each of them requires considerable refinement before it could be implemented in practice.

Before any of these options are incorporated into the TPG, however, I think it is important for the healthy development of the international tax law to address the question of whether these "special measures" are within or beyond the arm's-length principle.  The discussion draft continues to assert that this question does not need to be answered, at least at the moment:

"Some of these measures could be seen as within the arm's length principle and others beyond. At this stage, it is not critical to determine whether a potential measure is on one side or the other of the boundary, but the aim is to consider the effectiveness of the measure."

Perhaps avoiding stating whether a proposal is within or beyond the arm's-length principle is useful at the stage of developing options, as delegates consider what sorts of rules might achieve the objectives they seek. But ultimately the distinction is an important one, and I think that eventually the OECD/G20 delegates should be clear whether an option indeed is or is not intended to be within the arm's-length principle. The difference could affect how the rule is designed, how it is interpreted, and how it is implemented as a matter of national law.

In some ways, a rule that is beyond the arm's-length principle is easier to administer. It is like a reverse safe harbor; if the taxpayer crosses the threshold, the consequences follow. There is no need to refer to comparables and no need to speculate about market behavior. The drafting imperatives in this case are that both the trigger and the consequences are expressed with precision. Since this rule presumably in many cases will have a cliff effect with a pretty long fall to the bottom for the taxpayer that finds itself on the wrong side of the precipice, fairness to taxpayers and the interest of reducing controversy means that the line on the cliff edge needs to be painted as clearly as possible.

On the other hand, a rule that is beyond the arm's-length standard poses some very interesting jurisprudential challenges. If the rule indeed is intended to be outside the arm's-length principle, then presumably it could not be applied by tax administrations as a mere interpretation of existing national legislation which requires intercompany dealings to be priced according to arm's-length principles. Absent implementing domestic legislation, it is hard to see how introducing the special measure to the TPG or even incorporating the rule into a treaty amendment could authorize taxation on that basis, as the treaty merely would then permit something that national law had not been amended to authorize.

In contrast, the jurisprudential challenges would seem to be significantly less severe if the rule is intended to be within the arm's-length principle, as a tax administration then presumably could enhance its regulatory or administrative interpretation of its statutory imposition of the arm's-length principle to include the new measures. In this case, however, it would seem that there must be some opportunity for the taxpayer to argue that its reported intercompany pricing indeed complied with the arm's-length principle, despite the possible application of the special measure. If the special measure is justified as being an appropriate method to implement the arm's-length principle, and is intended to lead to results that are in conformity with the arm's-length principle, then taxpayers should have at least the opportunity to show why their results are consistent with that principle without the application of the special measure.  At an extreme, one can anticipate taxpayer challenges to the special measure as a whole, if its terms create outcomes that are demonstrably outside arm's-length results.

The relationship between the special measure and the arm's-length principle also may affect the route to, or even availability of, double tax relief. If a special measure is within the arm's-length principle, then presumably Article 9(2) of the OECD Model Convention applies to require a corresponding adjustment, and the issue is eligible for MAP, including arbitration. If the special measure is outside the arm's-length principle, then in order to guarantee double taxation relief, Article 9(1) and 9(2) may need to be revised to authorize the special measure and require a corresponding adjustment.

Comments on the discussion draft most likely will focus on the terms of the proposed options, perhaps to the exclusion of this issue. But the OECD/G20 delegates will need to come to grips at some point with the question of the underlying authority to impose tax on the basis of a special measure. If the special measure is to be justified as within the arm's-length principle, then perhaps merely revising the TPG will be sufficient to implement the option.  But if the special measure is meant to operate independently of the arm's-length principle, something else will need to be done to give the special measure the force of law.

This commentary also will appear in the January 2015 issue of the  Tax Management International Journal. For more information, in the Tax Management Portfolios, see Maruca and Warner, 886 T.M., Transfer Pricing: The Code, the Regulations, and Selected Case Law, Culbertson, Durst, and Bailey, 894 T.M., Transfer Pricing: OECD Transfer Pricing Rules and Guidelinesand in Tax Practice Series, see ¶3600, Section 482 - Allocations of Income and Deductions Between Related Taxpayers.

  Sprague, "Special Measures" and the Arm's-Length Principle, 43 Tax Mgmt. Int'l J. 32 (Jan. 10, 2014).