There is no doubt that 7 June 2017 is a day that will be a milestone in the history of international tax law. Signature of the Multilateral Convention to Implement Tax Treaty Related Measures to Prevent Base Erosion and Profit Shifting (“MLI”) at a high-profile ceremony at the OECD in Paris by 68 ministers and other high-level representatives is the culmination of four year’s work on the BEPS project. The next step is the implementation of the Convention.
The MLI is an innovative, legally complex instrument. Its interpretation and application raises challenging questions for lawyers and others called upon to consider its effect. Although there are many examples of multilateral treaties that interact with bilateral treaties, most tend to either supersede bilateral treaties entirely or modify them substantially.
In the case of the MLI, it is intended to make significant and detailed changes to at least 10 of the articles of double tax treaties that follow the OECD Model, add at least one new article, as well as change or insert a preamble. Unlike most other multilateral treaties that establish a single set of rules for all contracting states as set out in the treaty (possibly preserving some aspects of existing bilateral treaties), the MLI sets out only to change existing bilateral treaties without creating a unified legal regime. Even in relation to the minimum standards that signatories are required to adopt in preventing abuse (articles 6 and 7) and in resolving tax treaty disputes (article 16), signatories are offered a wide range of choice, including retaining existing bilateral treaty provisions that are compatible with the corresponding MLI provisions.
To this mix must be added a choice by signatories of the particular treaties to which the MLI is to apply (Covered Tax Agreements). Unless two states both include a particular treaty in their respective lists of CTAs, that treaty will not be modified by the MLI. All the choices made by signatories so far, are notified as contemplated by articles 28 and 29 to the OECD in a standard format. States may change their position between the time of signature and deposit of instrument of ratification (article 29(3)).
There is, as a result, no core content at the time of signature. In order to determine whether a specific provision in a particular bilateral treaty between two states has come within the ambit of the MLI it is necessary to check the statements of position of both states to see if there is a match on that provision. The OECD estimates that more than 1,100 bilateral tax treaties are covered by the positions stated by the initial signatories with more than 40 choices that signatories may make in respect of each CTA.
Agreement by computer matching
In order to work out who has agreed what with whom, the OECD has commissioned software that will determine compatibility between the positions of signatories. The resort to information technology in order for a state to know what it has agreed to (and its tax administrators and taxpayers to know what rules govern) rather than by simply reading the legal instruments is certainly ground-breaking. It is true that the effect on an individual provision in a specific treaty can be worked out by reading the positions of the two signatories. In most cases, this is impractical and a technological solution seems inevitable.
This raises fundamental questions about the nature of law and law making. Treaties, like other contracts require agreement to come into existence. What is the minimum content of the subject matter needed for the necessary consensus ad idem between the parties? Do domestic legislators really know what they are agreeing to when ratifying the MLI?
Some have compared the process to computer dating (combined with speed-dating sessions organised by the OECD where treaty negotiators meet many of their counterparts for short periods in rapid succession). Online dating apps however only affect introductions and it is up to the parties what to do with the introduction. Here speed-dating provides the oportunity for possible agreeement and the computer matching is to enable everyone to identify how signatories have modified their existing treaty relationships.
Complexity created by humans that is best addressed through information technology makes one wonder if simpler solutions are available. Answering this question and assessing the suitability of the instrument for its intended purpose, is also difficult without recourse to the software under development. For example, how have states exercised the choices on offer? If many states have only adopted the minimum standard, (and selected the PPT) would it not be simpler to have a multilateral agreement on treaty abuse which just adopts the PPT and conditional on minimum standards of dispute resolution. A stand-alone multilateral agreement on dispute resolution (with optional arbitration) would be similar to the EU Arbitration Convention. If many states have relied heavily on compatibility of existing treaty provisions with the MLI (as the UK has), what additional legal protection have states obtained against treaty-related base erosion and profit shifting?
The MLI has an important political objective in finishing off the 15 BEPS Actions with a legally binding instrument to which many states are a party, and which has the prevention of base erosion and profit shifting as its stated purpose. The high-profile signing ceremony attended by political leaders and senior officials, signals that significance. That success cannot be doubted.
Like all innovation, legal innovation brings opportunites for new solutions. It also carries risks of creating new problems. The MLI is intended to reset the substantive international tax rules but has also thrown down a challenge to established ways of treaty making. Welcome to the 21st Century.
See my other posts on the MLI: Multilateral Negotiation of Bilateral Treaties;