Oxford University Centre for Business Taxation
THE ARM’S LENGTH PRINCIPLE (“ALP”) – IS IT A PRINCIPLE AND IS IT ARM’S LENGTH?
What is the ALP?
The definitive statement of the ALP is set out in Article 9 (1) of the OECD Model Double Tax Treaty. Broadly, this tells us that where “conditions are made or imposed” between associated enterprises in their “commercial or financial relations” and these conditions both differ from those which would be made between independent enterprises and also lead to a shortfall in profits in one of those associated enterprises, the amount of that shortfall may be taxed. The provision is unchanged since its first introduction into the OECD Model in 1963.
Is it a principle?
In order to amount to a principle, the ALP would need to rank as some sort of proposition that serves as the foundation for a set of rules or which has numerous applications. In fairness, this is clearly the point of the ALP. However, there are two immediate caveats.
The first relates to the terms of Article 9 (1). There remains considerable uncertainty as to what is included in the various terms used in the Article (e.g. “conditions”, “commercial or financial arrangements”, etc.) and on top of that some knotty treaty issues as to how those terms are to be interpreted in the context of applicable tax treaties.
The second is that bandying about references to the ALP as some sacred principle does not tell us much about what it actually means or requires in any specific situation. That might seem an odd statement to make but is true because of the enormous evolution of the ALP over the period of its existence. The solitary (and very short) paragraph in the original 1963 Commentary to the OECD Model (which noted the ALP “seems to call for very little comment”) has now given way to literally hundreds of pages of OECD guidance on what the ALP does mean and require. This means that what the ALP is, and requires, changes considerably over time, which hardly supports any claim for the ALP as some kind of immutable principle.
Take the recent BEPS changes to the ALP. A taxpayer whose transfer pricing arrangements, (including approaches to compliance, applications of transfer pricing methods, documentation, etc.) were in full compliance with the ALP before BEPS would find that position inadequate in a large number of ways when judged by the “post BEPS” ALP, despite zero changes in the wording of the ALP, as enshrined in Article 9 (1) itself.
The point is even clearer in the context of the PE attribution rules in Article 7 which are often described as being based on the ALP. However, if the position before and after development of the “authorised OECD approach” (the “AOA”) is considered, the result is a huge difference in what is required under Article 7.
So, yes, the ALP is a principle of sorts but it is vague and uncertain and used in such a wide variety of contexts that it will virtually always need some extra explanation to clarify exactly what is being said. Reliance on the authority of “the ALP” alone is not obviously more precise than claims relating to the “right” amount of tax.
Is it arm’s length?
As to whether the ALP really is “arm’s length”, there are many questions as to what that concept can mean when applied to transactions within a MNC group which are not seen between independent parties – even though the conventional wisdom of the OECD’s Transfer Pricing Guidelines is that there is no problem in applying the ALP in such circumstances.
But a more immediate question relates to the BEPS project and whether, in its enthusiasm to stamp out all possible means of profit shifting and base erosion, the OECD has pressed the ALP into action so that it is, post BEPS, now to be applied in a way that actually undermines or contradicts the “principle” that it is supposed to enshrine. Such a conclusion is very hard to resist.
Take the question of how risk is dealt with by the transfer pricing rules in light of the BEPS changes. Understandably, there were in the BEPS project significant concerns about MNC groups stripping risk from group entities in high tax states and shifting it into group entities in low tax states. In response, the output from BEPS on transfer pricing (now incorporated into UK law by the adoption of the latest version of the OECD’s Transfer Pricing Guidelines) bolstered the general barrage of measures against this sort of planning by providing a specific response. Under that response, risk is not regarded as borne by a group entity (meaning the relevant taxpayer is denied a reward for risk) unless the risk is controlled by the entity concerned.
However, as a number of commentators have already pointed out, there are many cases where risks are clearly borne by third parties even though they exercise no meaningful control over those risks, for example where a specialist contractor manages those risks. This suggests a basic disconnect with the “principle” in the name of anti-avoidance. Further, given the highly complex interaction between the transfer pricing rules and the PE profit attribution rules post BEPS, this risk point is now also made relevant in the context of the PE rules. The position seems hardly sustainable. The expectation must be that, sooner or later, the status of this new “control” requirement will be litigated in the courts and found wanting by the arm’s length standard.
Perhaps the take-home point is the principle that, if not the OECD, someone will be looking after the essential arm’s length requirement of the ALP.
Some relevant CBT Publications
Transfer Pricing and the Arm’s Length Principle After BEPS, Oxford University Press 2017, Richard Collier and Joe Andrus
“OECD Discussion Draft: additional guidance on the attribution of profits to permanent establishments”, British Tax Review 2017, number 5 at p. 509, Joe Andrus and Richard Collier
“Section 75: transfer pricing: application of OECD principles”, British Tax Review 2016, number 5, at p. 590, Richard Collier
“Are we heading for a corporation tax fit for the 21st century?”, Fiscal Studies 2014, 35.4, 449–475, Michael Devereux and John Vella
“The arm’s length principle and distortions to multinational firm organisation”, Journal of International Economics, 2013, 89, 432-440, Christian Keuschnigg and Michael Devereux
Back to top of article