Introduction

The Organisation for Economic Co-operation and Development (OECD) sees treaty shopping as an important source of Base Erosion and Profit Shifting (BEPS). In this context, it identified the prevention of treaty abuse as one of fifteen issues or “actions” in respect of which recommendations were to be formulated as part of its Action Plan on BEPS, released in July 2013 (Action 6).[1] Following a two-year consultation process between the OECD and G20, the OECD presented final reports with respect to its BEPS project on October 5, 2015. This article provides an overview of the consensus that was reached between the OECD and G20 with respect to Action 6.

Context

Building upon a preliminary discussion draft released in March 2014, the OECD published a package of initial proposals with respect to the prevention of treaty shopping and treaty abuse strategies in September 2014 (2014 Report).[2] The 2014 Report was promptly followed in November 2014 by a list of 20 issues that remained to be addressed, and which invited comments from members of the tax community. A response to these comments and ensuing proposals were provided in a discussion draft released in May 2015 (2015 Draft). Unsurprisingly, the Final Report on Action 6 (Final Report) has made certain superseding changes to the 2014 Report.[3] The Final Report is divided in three sections.

The Final Report

Section A

Section A of the Final Report (Section A) indicates that states should be given a degree of flexibility with respect to the prevention of treaty abuse. As was first agreed to in the 2014 Report, states are only expected to adhere to a “minimum standard” consisting of (1) making changes to the title and preamble of tax treaties as described below, and (2) adopting either (a) a principal purpose test (PPT) or (b) a limitation on benefits (LOB) provision, together with a rule aimed at “conduit arrangements”.

However, Section A also indicates that a state which is not concerned by the effect of treaty shopping on its own taxation rights will not be obliged to apply provisions such as the LOB or the PPT, as long as it agrees to include provisions that its treaty partner will be able to use such provisions. This is the first instance in which the OECD and G20 agreed to reduce the “minimal standard” to which states are expected to adhere in countering treaty abuse. The 2015 Draft did not make a recommendation to that effect.

Principal Purpose Test

The PPT will apply a general anti-avoidance rule that would deny treaty benefits when it is reasonable to conclude that one of the main purposes of arrangements or transactions is to secure such benefits, and when obtaining such benefits in the circumstances would be contrary to the object and purpose of the relevant provisions of the tax treaty.

Following the 2014 Report, it was suggested that states should consider establishing a form of administrative process (i.e. a mechanism along the lines of the GAAR committee) that would ensure that the PPT rule is only applied after approval at a senior level. Language to that effect was consequently proposed in the 2015 Draft, and now constitutes a formal recommendation of Section A. Similarly, revisions introduced by the 2015 Draft and providing for some form of discretionary relief under the PPT rule have been reproduced in Section A.

Limitation on Benefits

The comprehensive model LOB provision provided in the 2014 Report, along with the revisions to its text and commentary formulated in the 2015 Draft, has been reproduced in Section A. However, building upon the 2015 Draft recommendations, Section A now proposes that tax treaties may only contain a “skeleton” or “simplified” version of the LOB rule. An example of such abridged rule is provided in Section A, as an option for countries that have adopted a PPT and will consequently not need to rely heavily on an LOB. In all material respects, the simplified LOB of Section A is textually identical to the version proposed in the 2015 Draft.

However, the United States contemporaneously released an updated version of the LOB provision included in their model treaty, which is not expected to be finalized prior to the end of 2015. For that reason, it was decided that the Final Report’s LOB-related proposals and commentary will need to be further reviewed and finalized in 2016, in light of the comments received by the United States on their updated version.

Collective Investment Vehicles

The OECD defines Collective Investment Vehicles (CIV) as funds that are widely-held, hold a diversified portfolio of securities, and are subject to investor-protection regulation in the country in which they are established (i.e. mutual funds).[4] The 2014 Report indicated that further work was needed with respect to the policy considerations relevant to the treaty entitlement of CIVs and non-CIV funds. As a result of the follow-up work on these issues and of the comments received from stakeholders, it was agreed in the 2015 Draft that the LOB rule dealt with the application of the LOB to CIVs in a satisfactory way, and that there was therefore no need for additional changes. Accordingly, section A does not take the issue any further.

While many comments were also received as part of the 2015 Draft with respect to non-CIV funds (such as private equity funds[5], real estate investment trusts, and pension funds), Section A states that an examination of the issues related to the treaty entitlement of these types of funds will need to continue after September 2015, as a consensus was not reached on how they should be treated.

Conduit Financing Arrangements

Section A also modifies the anti-conduit provision recommended for adoption by the OECD, confirming the 2015 Draft proposals in this respect. Whereas the 2014 Report contained a proposed text for the rule, Section A now emulates the 2015 Draft and limits itself to providing a series of examples of situations that should or should not be caught by the rule, in order to guide states which may not accept the rule as proposed in the 2014 Report.

Section B

Section B of the Final Report (Section B) proposes revisions to the title and preamble of tax treaties which would clarify the intention of contracting states that tax treaties are not meant to be used to generate double non-taxation, including through treaty shopping strategies. The Section B revisions were not addressed by the 2015 Draft and are identical to those proposed in the 2014 Report.

Section C

Section C of the Final Report (Section C) modifies the introduction of the OECD Model Tax Convention[6], as did the 2014 Report, to include tax policy considerations which should help countries explain their decisions not to enter into certain tax treaties, and which would also be relevant for countries that need to consider whether they should modify or terminate a treaty previously concluded in the event that a change of circumstances raises BEPS concerns related to that treaty.

Section C also introduces new proposals seeking to restrict treaty benefits with respect to taxpayers that profit from certain preferential tax rules or with respect to certain drastic changes that could be made to a country’s domestic law after the conclusion of a treaty. These proposals are identical to those first released for comments as part of the 2015 Draft. They did not appear in the 2014 Report. Around the same time, however, the United States released new versions of similar proposals which, much like the new version of the LOB rule, are not expected to be finalized prior to the end of 2015. For that reason, it was decided that the new Section C proposals will also need to be reviewed in 2016, in light of the comments received by the United States on their proposals.

Conclusion

Now that the Final Report is released, focus will likely shift to the implementation of the treaty-related measures into domestic legislation. As part of the 2014 Federal Budget, the Department of Finance of Canada had announced a decision to adopt a domestic rule, rather than a treaty-based approach, that would have denied treaty benefits to a person, who, it was reasonable to conclude was engaging in treaty shopping.[7] Such adoption was later put on hold, the government deciding that it would “await further work by the OECD” before implementing the proposal set out in the 2014 Federal Budget.

Nearly 90 countries are currently working together on the development of a multilateral instrument capable of incorporating the recommendations of Action 6 into the existing network of bilateral treaties (Action 15). The instrument will be open for signature by all interested countries in 2016. It remains to be seen whether Canada and the countries typically implicated in treaty shopping strategies will sign any such instrument.