From best endeavours to binding arbitration: Eliminating double taxation

When is the transfer pricing (TP) of a transaction considered final? The key feature of TP as a discipline is its subjective nature. This is recognised by the OECD's approval of a range of possible prices in any TP transaction. By definition, TP is a study of facts and analysis of comparable economic circumstances to assess arm's-length pricing. Businesses and their circumstances change over time, and in turn, TP guidance also changes, as seen by the many updates to TP guidelines that were introduced through the G20/OECD's BEPS process, which ultimately led to the modification of domestic TP laws and practices.

Transfer pricing needs to be right when a tax return is filed. A subsequent adjustment by a tax administration creates a new price (that is right at that time). The new post-audit TP adjustment can then be subject to a mutual agreement procedure (MAP), and as a result, a third TP outcome is agreed upon in many occasions.

The subjective nature of TP rules can often mean that no tax penalty is levied on a TP adjustment. Many TP regimes focus on the effort the taxpayer has made, or its behaviour, rather than the TP reported. This implicitly recognises the genuine difficulty in adopting appropriate transfer prices.

One may conclude that no transfer price is final until two competent authorities (CAs) have agreed together on that price, since there is no further arena in which that price can be changed (absent developments such as state aid). Alternatively, the available time for a tax administration audit may have passed, and therefore, the TP arrangement becomes final by default. It is thus no surprise that TP disputes are numerous and may take considerable time to resolve. Viewing the dispute in its full cycle (the time it takes to finalise an audit and resolve any double taxation), will render the true cost of a TP dispute apparent.

The OECD's MAP statistics remain the most empirical guide to the level of ongoing TP controversy. These statistics, however differently presented since records began in 2006, have consistently shown that the number of MAPs are increasing. Experience has reinforced the conclusion that the number of disputes involving TP (and the attribution of profits to permanent establishments) have increased.

It is important to note that the actual number of controversies may be much higher than the number of MAP cases reported. Ongoing 'risk assessment' interactions with a tax administration can resemble a full audit in everything but name. Furthermore, many full TP audits end without any adjustments, and are often just as burdensome and resource-intensive as those that do give rise to an adjustment. Even in situations where audits end in an adjustment, some cases are not taken forward into a MAP (sometimes by choice, but sometimes because of more formal factors, such as the taxpayer running out of time or being ineligible for some other reason).

Improvements in dispute resolution processes (essential in a MAP itself) are a part of the BEPS project, and a recognition (tacit or otherwise) that increased powers for tax administrations would mean more controversy. BEPS outcomes were designed to ensure that single taxation was allocated to the appropriate jurisdiction, not to increase double taxation. The Multilateral Convention to Implement Tax Treaty Related Measures to Prevent BEPS (BEPS multilateral instrument, or MLI) contains text designed to improve access, resolution and implementation of a MAP.

The MAP remains the sole path to eliminate double taxation caused by TP rules. That is universally accepted as the proper approach. There are few, if any instances, in which multinational enterprises (MNEs) can legitimately address double taxation caused by a tax administration adjustment. Therefore, a MAP process that works is crucial.

The three areas of concern identified by the OECD's 2005/2006 dispute resolution work that led to the manual for effective mutual agreement procedures (MEMAP) and the subsequent insertion of an arbitration clause into Article 25 of the Model Tax Convention (MTC) remain relevant.

These are:

1) Accessing a MAP;

2) Resolving a MAP; and

3) Implementing a MAP.

The consensus remains that the MAP works when available (availability depends on a number of factors, not all of them technical or legal), but it does not always work quickly enough.

The sheer number of disputes reported by the OECD has ramifications for the speed of resolution. Resolution times are not improving, while the changing nature of disputes is already impacting the resolution time frames for MAPs. The probability of a resolution based only on best endeavours has also been impacted. The latter underlines the importance of binding arbitration mechanisms.

Traditionally, MAP cases in which the only issue was price was relatively easy to resolve. This may no longer be the case when the amount at stake is objectively large. Following the introduction of Chapter 9 into the OECD's TP guidelines, there are now more binary adjustments based on one tax administration's view of what an MNE would have done (had its constituent parts been independent).

Adjustments that have arisen from re-characterisation are difficult to resolve as it can be difficult for CAs to find common ground (or a point on which to make a principled compromise). A different arm's-length price to the one in a tax return or imposed on audit can be agreed more easily than whether intellectual property would (or would not) have been sold by unrelated parties in similar circumstances, given a tax administration's view of what the realistically available option would have been.

Increased openness during the MAP process is welcomed and is paying dividends in terms of resolving complex double taxation issues. In most countries, it is now easier to determine the name and address of the relevant person (the CA) who should receive the MAP. For the country receiving the TP adjustment (the country that did not conduct the audit), questions around access are often narrower, and tend to focus on whether a case can be made in time and whether enough information has been made available to the CA. The main concerns around accessing the MAP continue to be in the country where the TP adjustment has been made.

Some tax administrations are open about having a policy that allows the audit to close with a lower adjustment if the taxpayer agrees not to open a subsequent MAP. The OECD may frown on this, but from a cost-saving perspective, one can see the point. However, MNEs should carefully weigh the consequences of settling for double taxation in this manner.

The biggest single issue to take into account is the interaction between the audit settlement stage and the impending (or desired) MAP. Strategic thinking in this area remains critical. For instance, it is vital that any adjustment to the price of a transaction with an affiliate (or denial of relief for tax purposes) is clearly agreed to be a TP adjustment. Adjustments denying deductibility under non-TP rules can result in a denial of entry into a MAP for taxpayers without considerable (and not always successful) effort.

Another barrier to MAP entry that often needs to be overcome occurs at the end of the audit. In any negotiated settlement, the taxpayer may feel persuaded to agree that no MAP case will be made. In effect, that would mean agreeing to double taxation.

All of the above practices that seek to deny access to a MAP are not in line with the minimum standards developed by the OECD. If relief is available "at the other end of the transaction" through a MAP, deciding to forgo a MAP should not be done lightly.

The OECD's statistics, as well as the MLI and BEPS minimum standards improvements, reveal some ground for optimism. For example, unilateral relief is granted in 20% of cases. This figure is not analysed further to distinguish between TP and non-TP cases, and it seems likely that a significant portion of these unilateral resolutions are for non-TP cases. Notwithstanding, the inclination of CAs to grant relief without the need to enter into bilateral negotiations is welcomed. After all, the notion is articulated in paragraph 2, Article 25 of the MTC.

Competent authority independence from the audit arm of the tax administration remains crucial in allowing for a principled resolution of MAPs. As the internal governance of TP audits by tax administrations increases, MAP agreements fleshed out between CAs are also coming under increased scrutiny to determine whether they should be ratified by the tax administration.

If a pause is reached after a potential agreement, that can be a good thing, as it can be used to inform or consult with the taxpayers involved. However, subjecting a MAP agreement to the same (or essentially similar) internal governance as TP audits may lead to an increased focus on the tax-raising (or tax-base defence) aspects of the tax adjustments involved in the MAP. This would be an unwelcome development.

The OECD was rightfully clear during the first round of dispute resolution improvement work when the arbitration clause was inserted into the MTC, noting that the CA's own performance should not be impacted by tax yield. This was to maintain that the MAP is about determining TP in terms of Article 9, rather than on narrower national self-interest. This philosophy is increasingly under threat.

A longstanding 'separation of the powers' concept for MAPs has helped to sustain principled outcomes in a MAP. But that concept is under threat in various ways, and the MAP outcomes may move away from treaty concepts and become mired in tax raising efforts. Irrespective of whether the CA is a function of the finance ministry or a government's treasury department, having independence from the audit/tax division is vital to the non-partisan conclusion of a MAP.

When the auditor or tax inspector effectively sits in the room with the CA during any negotiation in a MAP, a non-partisan conclusion becomes less likely. Influence over the MAP process by the audit arm of the tax authority often hinders the conclusion of a MAP.

However, influence can be exerted in other, less apparent ways. Subjecting MAP outcomes to the same, or similar governance as audit outcomes, or having substantially the same people within a tax administration carrying out such governance or internal audit, is essentially limiting the ability of the CAs to operate as impartial 'officers of the treaty'.

Tax administrations that impose adjustments during an audit that can be seen as the more extreme end of the arm's-length position may emerge from a subsequent MAP with a higher agreed adjustment. This places a commensurately higher burden on the CAs charged with eliminating double taxation, and may create expectations within the tax administration that a significant amount of tax remains to be collected.

Taking all this into account, it comes as no surprise that arbitration mechanisms are becoming increasingly important and, in theory, increasingly available. The groundwork was laid by the EU's Arbitration Convention in 1990. The OECD took the most effective parts of this document, and 17 years later introduced a far-reaching arbitration article into the MTC. Approximately a decade later, we have the arbitration potential in the Multilateral Instrument, and in the EU at least, in the Dispute Resolution Directive.

All these developments bode well for the potential availability of arbitration to resolve TP disputes between tax administrations. For example, the arbitration article has been incorporated regularly and consistently into tax treaties. However, there has been no great increase in TP cases that have actually gone to arbitration panels. In fact, such cases have been few and far between.

The availability of arbitration has traditionally been seen as a good thing for dispute resolution. But if this were universally true, there would have been an uplift in the numbers of cases being settled in proportion to the number of MAP cases being brought forward. However, this has not occurred. Given that the number of cases to proceed into formal arbitration have not increased, one conclusion is that cases are being kept out of the arbitration stage by a combination of tax administration or taxpayer action (or inaction).

However, it is likely that reality is more nuanced. For example, cases are being settled shortly before arbitration deadlines fall due. On the whole, arbitration is having its intended result: deterrence. The quid pro quo of having mandatory binding arbitration governing the eventual outcome of a MAP has been an increase in the barriers to entry of a MAP. Time limits, information requirements, and domestic legal proceedings must be taken into account if actual arbitration is going to be relied upon. It is more important than ever to take a strategic view of TP dispute resolution.

© 2019. For information, contact Deloitte Touche Tohmatsu Limited.This communication contains general information only, and none of Deloitte Touche Tohmatsu Limited, its member firms or their related entities (collectively, the "Deloitte network") is, by means of this communication, rendering professional advice or services. Before making any decision or taking any action that may affect your finances or your business, you should consult a qualified professional adviser. No entity in the Deloitte network shall be responsible for any loss whatsoever sustained by any person who relies on this communication.

Edward Morris

Partner Deloitte UK Tel: +44 (0) 207 007 6568 Mobile: +44 (0) 7788 366382 edmorris@deloitte.co.uk Edward Morris of Deloitte UK specialises in transfer pricing dispute prevention and resolution. He has been instrumental in setting up Deloitte's global TP controversy team, consisting of former tax administration personnel and in particular former government competent authorities. This network has a global reach and expertise and helps clients with mutual agreement procedures, advance pricing agreements and transfer pricing enquiry resolution. Previously, Edward worked for the HMRC as a tax inspector and as a UK competent authority, at the EU Commission and at the various OECD working parties concerned with tax treaties, transfer pricing and permanent establishments. Edward helps clients across all industry sectors deal with the increasingly complex world of tax disputes. Since joining Deloitte, he has appeared consistently in Euromoney's Transfer Pricing Expert Guide.