The OECD recently released what it calls a "public discussion draft
" in connection with its work on the multilateral instrument (MLI), and seeks public input until June 30. As I explained in a post
a few months ago, the MLI is be used to 'modify' all existing tax treaties in force among signatory countries to conform to BEPS standards and recommendations. However, the released document is not actually a discussion draft of the MLI--there are no terms to be reviewed. The drafting committee, which currently includes 96 members (OECD members and "BEPS Associates"), only met for the first time two weeks ago so this is decidedly not a draft of substantive provisions to be debated in the public discourse. No, that would be chaos and contrary to the plan:
"the draft text of the multilateral instrument is the subject of intergovernmental discussions in a confidential setting."
Instead it is in effect a crowdsourced, and very carefully framed, issue-spotting exercise. The document consists of three pages: page one is the BEPS narrative (why the OECD undertook this project and what has happened so far). Page two describes what BEPS changes will be covered in the MLI once drafted. Page three lays out three "technical issues" the OECD faces in drafting the MLI, and finally gives the call for input. The discussion is very brief and in OECD-passive-speak so it's almost comical to summarize but here are the three issues, as I understand them:
- the MLI must be able to modify existing treaties, and this will be done with "compatibility clauses."
- the MLI will be broadly worded so will require commentary and maybe explanatory notes for consistent interpretation
- the MLI will be in French and English but will interpret thousands of treaties written in different languages.
These are all very interesting international law issues and not technical tax law issues at all. They are also clear expressions of intent to cement the OECD's role as the principal curator of international tax norms. This is especially clear in respect of point 2, because consistent interpretation means that the OECD commentary and "guidance" must harden ("ossify") into more persuasive and ultimately binding legal authority. For a discussion of what that implies for the rule of law, see this old thing I wrote a long time ago
Point 1 raises the issue that seems to me most difficult in terms of the transition to complete OECD domination of global tax policy: I am still not sure how the MLI is supposed to work on top of a network of individualized and distinct bilateral agreements among sovereign nations. The OECD says "If undertaken on a treaty-by-treaty basis, the sheer number of treaties in effect would make such a process very lengthy." Indeed it would but as a matter of law in many countries, revising an existing international agreement requires another international agreement that is ratified in the same manner as the original, which appears to require the signatories to come to a meeting of the minds as to the terms that govern their unique relationship. The OECD says that distinguished experts have carefully considered the public international law questions at hand. But I haven't seen any study and I don't quite understand how you get a coherent international tax law regime in anything like a "quick" process. The OECD's implied answer in point 1 only raises another question for me: what is a compatibility clause? Is this a well-understood mechanism in play in other areas of international law? Can I get a precedent somewhere to anticipate where we are going with this?
Further, is the MLI going to be a matchmaking exercise in practice? If country A agrees to revisions 1 through 6 as to countries B and C, but only revision 5 as to country D, and country B agrees to revisions 1-3 for countries A and D but only 5 and 6 for C, and countries C and D agree in principle but never ratify anything, then what, exactly, are the agreements between and among these countries?
I am also not sure what the agreement matrix looks like when there are multiple standards for several of the BEPS items. Notably the "prevent treaty shopping" minimum standard provides multiple choices for defending treaties against "abuse": a principal purpose test, a limitation on benefits provision, an anti-conduit provision, or some combination. May each of countries A, B, C, and D choose a different combination vis a vis each of the others? It is difficult to see convergence. At the panel I attended in Montreal a couple of weeks ago this was a topic of vigorous discussion. The more I think about this, the increasingly uncertain I become regarding how this is going to work out in practice.
Any thoughts on these observations are welcome as I develop my thinking on these issues. And if you are submitting comments, please note:
Comments and input should be submitted by 30 June 2016 at the latest, and should be sent by email
to email@example.com in Word format (in order to facilitate their distribution to government
officials). Please note that all comments received will be made publicly available. ... Persons and organisations who submit comments on this document are invited to indicate whether they wish to speak in support of their comments at a public consultation meeting that is scheduled to be held in Paris at the OECD Conference Centre on 7 July 2016 beginning at 10.00 am.