Multilateral or bilateral Implementation of BEPS Treaty-related measures? Swiss-UK and UK-Uzbekistan Protocol show the way

Kluwer International Tax Blog
February 21, 2018

Please refer tot his post as: , ‘Multilateral or bilateral Implementation of BEPS Treaty-related measures? Swiss-UK and UK-Uzbekistan Protocol show the way’, Kluwer International Tax Blog, February 21 2018,

Non-inclusion of the Swiss-United Kingdom income tax treaty in the list of Covered Tax Agreements of both the states on signing the BEPS MLI on 7 June 2017 surprised some observers. This evident gap between two OECD member countries was quickly filled by a protocol to the Swiss-United Kingdom treaty concluded on 30 November 2017.

Swiss-UK Protocol

The protocol is a model of brevity and clarity. It reflects the extent to which the two states agree to gives effect to the BEPS treaty related measures.  The minimum standard in the Action 6 Final Report is given effect by adopting the preamble now found in the 2017 OECD Model. The Principal Purpose Test (PPT) is introduced as a new article 27A which reproduces the new article 29(9) in the 2017 OECD Model. The traditional approach in UK treaties for decades has been to include a purpose based anti-abuse rule in the dividend, interest, royalties and other income articles. The use of this provision has not however been consistently applied and varies from treaty to treaty. In order to tidy up the anti-abuse rules, these specific anti-abuse rules are deleted from articles 11(7) (Interest), 12(5) (Royalties) and 21(4) (Other Income) and the “conduit arrangement” concept is eliminated from the treaty. The result is a single anti-abuse provision applicable to all articles.

The Action 14 Final report on dispute resolution minimum standard is also given effect by amending the mutual agreement procedure in article 24(1) to permit a case to be presented to the competent authorities of either state. Resolution of disputes is further facilitated by the introduction of article 9(2) of the OECD Model to address corresponding adjustments. This replaces the rather unusual language to similar effect and aligns this fully with the OECD language.

A single optional provision that parallels the BEPS MLI relates to elimination of double taxation derived from article 23A of the OECD Model. A new article 22(7) removes the Swiss obligation to exempt income from Swiss tax where the UK applies the treaty to exempt that income (or reduces the rate of tax in the case of dividends).

UK-Uzbekistan Protocol

During the same period, the UK also negotiated a protocol to the UK-Uzbekistan tax treaty which was concluded on 24 January 2018. Uzbekistan is not a party to the MLI. Nonetheless, the BEPS minimum standards are reflected in the treaty: Action 6 provisions comprise a revised preamble and the PPT with discretionary relief introduced in article 23. Surprisingly, the specific purpose tests for interest (article 11(9)), royalties (article 12(7)) and other income (article 21(3)) are retained.

Other aspects of the BEPS Actions are included in the protocol. Action 2 proposals in relation to transparent entities appear in article 1 revised in line with the 2017 OECD Model and article 24 which dealt with partnerships is deleted. The place of effective management is displaced by competent authority agreement to determine the residence of persons other than individuals in article 4(3).

The Action 14 minimum standard is included in revisions to permit a case for mutual agreement to be present in either state and for any resulting resolution to be implemented notwithstanding domestic time limits. Article 26 is also amended to provide for arbitration of unresolved disputes (although dual residence of persons other than individuals is not arbitrable).

Other non-BEPS related updating to the UK-Uzbekistan treaty include the post 2010 business profits article and the inclusion of independent personal services within the meaning of enterprise and business, although, bizarrely, the independent personal services article is retained. Exchange of information in modernised and assistance in tax collection introduced by a new article 27A.

Both treaties reflect a common intention of the parties to only adopt limited parts of the BEPS treaty related actions. The protocols raise no novel or special issues of interpretation. The MLI, by comparison is complex, difficult to understand, and most of its terms are not relevant to these treaty relationships.

Where does this leave the MLI?

The original purpose of Action 15 was the need for a swift implementation of the treaty-related BEPS measures. It was thought that bilateral amendment might be very lengthy. See OECD , Action Plan on Base Erosion and Profit Shifting (2013) Action 15. While it is too early to make a final judgement on whether the MLI will achieve that objective, it is clear that the objective can be met by simpler protocols. In addition, these may require less effort and more certainty. The MLI has spawned a huge amount of activity including translating it into Arabic, Dutch, German, Greek, Italian, Russian, Spanish and Swedish. A huge effort is underway to assist states with their reservations and to encourage those who did not sign, to do so. An application toolkit and database have had to be developed. A depository for ratifications must be established and administered.

The amendments to the OECD Model were in place when the Final Reports to the BEPS Actions were published in October 2015. A revised Model could have been agreed shortly thereafter, but instead took until December 2017 to be approved. Has the absorption with the MLI itself been a distraction?

Perhaps the most useful innovation is not the instrument itself but the process of reaching agreement. The formation of the ad hoc Group with the participation of so many states provided the first opportunity for states to evaluate their own positions on the treaty-related actions. The speed-dating sessions where treaty negotiators from participating states gathered for multiple bilateral negotiations on the various measures and options in the MLI provided the key to rapid implementation of treaty changes. An alternative output from those sessions could easily have been protocols to individual treaties. The templates for such legal instruments are well established and call for no innovation. Adding the BEPS related provisions would be just  a cut and paste exercise. There is also no reason why the signing ceremony could not involve the same people gathering to execute a number of agreements together.

The same observation may be made about ratification by contracting states. The presentation of a number of protocols to a legislature for ratification simultaneously is no more onerous or difficult than presenting the MLI plus the state’s List of Reservations and Notifications at the Time of Signature.