ARTICLE
8 October 1999

Implications for UK Tax Rules on Transfer Pricing of The National Westminster Case

United Kingdom

The US tax case of National Westminster Bank plc -v- United States was the summer's most discussed victory for a taxpayer. It concerned the question of whether or not the terms of the US/UK double tax treaty should override domestic US tax regulations. The case involved the taxation of NatWest's US branch. The US branch was not a separate legal entity from the UK headquarters. In order to carry out its business in the US it obtained funds from its UK headquarters, which were shown as a loan in its branch accounts. The branch accounts in turn also showed interest deductions on this loan, which were used in order to reduce the branch's profits for US tax purposes. The IRS argued that the amount of interest expense that could be allocatable to the branch should be limited by their own regulation 1.882-5. NatWest argued that this regulation contravened the terms of the US/UK double tax treaty, which provided that such a branch should be treated on a stand alone basis.

In order to support this argument NatWest relied upon the commentary to the OECD model treaty. This commentary stated that although loans from a headquarters to a branch should generally be disregarded when preparing the branch's accounts for tax purposes, a different treatment should apply in the case of the branch of a bank, as the payment of interest and the making of loans would be fundamental to that branch's business.

Although the judge found in favour of NatWest, many US commentators have argued that the commentary to the OECD model treaty should not have been conclusive, as the wording of the US/UK double tax treaty did not perfectly coincide with the relevant wording in the OECD model treaty. Specifically, the US/UK double tax treaty has wording which provides that a reasonable allocation of interest should be made to such a branch. This wording is absent from the OECD model treaty. It could be argued that regulation 1.882-5 provides a coherent methodology as to how such a reasonable allocation of interest expense could be made.

For UK tax purposes the point which is of particular interest is that the judge accepted the OECD commentary despite the fact that the language of the model treaty was not the same as the language of the US/UK double taxation convention. It has already been established in a UK tax case (Sun Life Assurance Company of Canada -v- Pearson [1984] STC 461 at 511B) that the commentaries to the OECD model "can and indeed must be referred to as a guide to the interpretation of the treaty". What is less clear, however, is the extent to which the following is true:

  1. Can model treaty commentary only be used in order to aid the construction of terminology contained in specific treaties which exactly mirrors the relevant words of the model treaty?
  2. Even if the relevant language of the treaty exactly mirrors the relevant language in the model convention, can the language of the model convention only be used as guidance if it can be shown that the intention behind the use of that language in the treaty concerned exactly mirrored the intention behind the use of that language in the model convention?
  3. Where the language of the relevant treaty does not exactly mirror the language in the model convention, can one infer a common intention behind the model convention and the model treaty that will allow the commentary to the model convention to be used in order to construe the language of the relevant treaty?
  4. Is it only possible to use the commentary of the model convention when the treaty concerned postdated the model convention, and therefore the commentary itself? Is it actually possible to use commentary which postdates the signing of the relevant treaty, as it could not in those circumstances be inferred that those that signed the treaty could possibly have had the intention that went behind a commentary that was only published on a later date?
  5. Given the general development by the UK courts in cases such as McGuckian of an increasingly purposive approach to the construction of UK tax statutes, should one similarly apply such a purposive approach to the construction of treaty language? Would such purposive approach be particularly difficult to apply when the commentary language was published at a time later than the signing of the treaty concerned?

Under the new UK transfer pricing rules the OECD guidelines can be used in order to construe UK rules, even where the OECD rules postdate the UK rules. However, the new OECD guidelines will first have to be designated as transfer pricing guidelines by the Treasury before they can be used by the courts as an aid to construction. It may be doubted, however, to what extent such a designation will actually follow a considered review of such guidelines, as this may well be merely a matter of formal designation in order to avoid any ambiguity on the question of their identification, rather than a question that a serious decision will have to be taken as to whether or not to allow such guidelines to be utilised for this purpose. The degree of elasticity that may be allowed, therefore, in the act of interpretation of the UK statutes is entirely open to argument before the UK courts. Furthermore, as it is ultimately a question of interpretation of existing UK law, rather than a formal delegation of the power to legislate in transfer pricing questions to some other international body - an act which would be contrary to the UK constitution, other than to the extent which it has already been achieved by the UK's accession to the European Community - it follows that there must come an awkward point at which this elasticity can be stretched to the point of snapping. UK legislation can be construed in a flexible manner, perhaps, but it will not be possible to make UK rules say something directly contrary to what they actually say, if there is a direct contradiction between the UK rules and the OECD rules, and the OECD commentary based upon those rules.

In this respect, the position is not entirely dissimilar to the manner in which the European Convention of Human Rights has now been assimilated into UK law. Under the Human Rights Act 1998 UK legislation must be construed to the extent to which it is possible to do so in a manner that is consistent with the European Convention on Human Rights. If it is not possible to construe the UK legislation in this manner, however, then the UK legislation must be applied in any immediate case in its current form. The court will in such a case make an order to the effect that UK legislation must be amended at the first opportunity in order to make it consistent with the European Convention on Human Rights, but this will not alter the decision in the case that is actually before the court in such an instance.

Were the UK courts to have to decide a case which was identical in its facts to that of the National Westminster case, only with regard to the UK branch of a US bank, they would seek to apply a transfer pricing methodology based upon the fiction of the separate entity, at an arms length price. The general tax problems presented by this area have been discussed by the OECD, notably in their 1998 publication "The Taxation of Global Trading of Financial Instruments". In their Tax Bulletin for December 1998 the Inland Revenue published a short article on this subject, in which they gave a brief summary of the difficulties attaching to the preparation of branch accounts for tax purposes when the activities concerned are not easily allocatable to any particular branch in any particular jurisdiction. Broadly speaking, of course, the issues concerned relate to capital adequacy, exposure to risk, and the value of the time of the personnel involved in the transactions that give rise to the taxable events, but having described the general issues it remains the case that the application for these principles to the particular circumstances will be debated on a case by case basis. Such cases therefore tend never to come before the courts, but instead to be settled by a lengthy process of negotiation between the Inland Revenue and the taxpayer concerned. For these purposes the OECD guidelines become useful rules which can be employed by either party in order to give some shape to the negotiations that take place. It is precisely because of the uncertainty attached to these negotiations that the Inland Revenue have introduced their new rules regarding advance pricing agreements, guidelines for which have recently been published. It remains to be seen to what extent such agreements will actually be entered into. It would appear that the guidelines were published in a form that was virtually unchanged from their initial draft version, because few people were prepared or interested in commenting on the draft form that was published for the purposes of consultation. From a taxpayer's point of view, the entering into of such an agreement would clearly be a risky matter, given the possibility that large amounts of material would have to be supplied to the Inland Revenue before any binding agreement had been reached. Having provided this information, the taxpayer may feel at a disadvantage to the Inland Revenue in subsequent negotiations, as it will have shown its hand in its entirety from the outset.

Through the use of guidelines and model conventions drafted by international bodies as aids to construction of domestic legislation, there has evolved a consequential flexibility of approach that in turn has led to the concept of any form of legislative certainty in this area becoming very weak indeed. Instead, an essentially commercial negotiation is now entered into in an uncertain context regarding the guidelines that provide the parameters within which such negotiation must be take place.

For further information please contact George Hardy, 2 Park Lane, Leeds LS3 1ES, Tel: +44 113 284 7000

This article was first published in the Autumn 1999 issue of Hammond Suddards' Tax Insight Newsletter.

The information and opinions contained in this article are provided by Hammond Suddards. They should not be applied to any particular set of facts without appropriate legal or other professional advice.

Mondaq uses cookies on this website. By using our website you agree to our use of cookies as set out in our Privacy Policy.

Learn More