Action 6 of BEPS introduced the principal purpose test (PPT) as one of the Minimum Standards to be implemented by the countries/jurisdictions taking part in the BEPS Inclusive Framework (IF). The PPT has been also introduced in the Multilateral Instrument (MLI), in force since 1 July 2018. Even though the choice given to countries was either: (i) PPT (with or without detailed or simplified LOB), or (ii) detailed LOB with anti-conduit financing arrangements, the peer review report on Action 6 (March 2019) shows that countries are choosing to implement option (i) and that option (ii) has only been chosen until now by the United States (IRC § 1.881-3 and section 3.2.2. of the peer review report on Action 6).
Therefore, in principle, it can be argued that the PPT is one of the minimum international tax standards applicable by countries members of the BEPS Inclusive Framework and/or MLI (Cyprus is the only signatory of the MLI). However, the implementation of the PPT raises questions about the uniformity in the application of the standard and the role of different actors (tax administrations, tax advisors, business, business associations, judges and academia) in the interpretation of this standard.
The impact of BEPS Actions and MLI in the treaty network will be one of the topics discussed at the Congress of the International Fiscal Association in 2020, and it has been also the topic of several conferences around the world organized by tax advisors, academia, business associations, and/or international/regional organizations. In this context, the following paragraphs will focus on some of the main issues that may need to be addressed in the implementation of PPT to achieve a minimum standard that contributes to global tax governance.
- The PPT and the challenges
Main elements of PPT
The PPT aims to tackle treaty abuse including treaty shopping and it is based to some extent in the guiding principle of the 2003 OECD Commentary to art. 1 (para. 9.5). However, unlike the guiding principle that was not extensively followed by countries, the PPT as one of the BEPS 4 minimum standards is at the time of writing followed by 131 jurisdictions (130 from the IF and 1 Cyprus in the MLI).
In a nutshell, PPT aims to deny tax treaty benefits and it consists of three main elements:
- A benefit under a tax treaty includes a tax deduction, exemption, deferral or refund. In the OECD Model, the benefit under tax treaty is found in the provisions of art. 6 to 22, art. 23, and art. 24 of the OECD Model Tax Treaty. In addition, it can include tax sparing (para. 175 Commentary to art. 29 2017 OECD Model).
- Subjective element: Tax administrations need to reasonably conclude, having taken into account all relevant facts and circumstances, that obtaining the benefit was one of the principal purposes of any arrangement or transaction that resulted directly or indirectly in that benefit.
- Objective element: The taxpayer needs to establish that granting that benefit in these circumstances would be in accordance with the object and purpose of the relevant provisions of the Covered Tax Agreement.
Challenges to the implementation of PPT and the MLI
The literature highlighting the challenges in the implementation of the principal purpose test has been extensive. Therefore, to understand how the PPT works, it is necessary to read BEPS Action 6 report, scholarly articles, the commentary to art. 29 2017 OECD Model, terms of reference and peer review report of Action 6. This literature is more than 400 pages that should be read by tax administrations, tax advisers, businesses, judges, and scholars to have a general overview of the problems in the principal purpose test. The amount of literature shows the difficulties to understand this principal purpose test. In addition, some of the differences in implementation have been also highlighted by the CIAT in the BEPS Database.
One of the problems is the complex menu of options due to the introduction of the MLI. For the countries participating in the IF but not signatories of the MLI, the PPT will be introduced following the amendment of bilateral tax treaties. For countries which are signatories of the MLI, the choices are either: (i) to include the tax treaty as a covered tax agreement for the MLI; or (ii) to negotiate a tax treaty bilaterally. Some countries have decided to follow a mixed approach including both choices, while other countries have decided to introduce the PPT bilaterally — very few countries have decided to apply the PPT in the MLI to all its tax treaties.
Despite these choices, the discussion of the MLI in the Parliament/Congress for ratification purposes can also result in different choices than the ones made by the Government. This is the case of, for instance, the Netherlands. In spite of the choice of the Dutch Government for the application (opt in) of almost all provisions of the MLI, the Lower House of the Dutch Parliament decided to change the Government’s choice in the case of commissionaire arrangements in permanent establishments (PEs) available in the MLI by introducing a reservation to this provision. The main reason for this change was the uncertainty that the application of a commissionaire arrangement will bring for business. In this case, the Lower House decided that the Netherlands should opt out from this provision until there “is either sufficient clarity on profit allocation to agency PEs or that there is an effective dispute resolution mechanism in place with sufficient other MLI parties. If adequate progress is made, a legislative proposal to withdraw the Dutch reservation may be submitted by the end of 2020.”
Another problem are the mismatches between countries signing the MLI. The choice made by a country will also depend on the choice of the other treaty country, and if there is a mismatch, the result will be multiple mini-treaty negotiations. Furthermore, countries have decided to choose the PPT either with a simplified LOB or a detailed LOB, that also needs to be matched by the other treaty country. Following these choices, countries are having problems to manage these choices mainly due to capacity constraints and tax treaty policy choices, and in some cases countries signing the MLI are waiting for other countries’ choices before ratifying the MLI.
Until the time of writing (16 July 2019) only 29 of the 89 signatories countries have ratified the MLI, therefore only until all signatories have ratified the MLI, it cannot be clearly established whether the PPT will contribute to more convergence or if the result is more complexity in the international tax system mainly due to these mini-treaty negotiations.
Discretionary relief in the application of the PPT and the MLI
Another feature of the MLI that is important to mention is the choice made by countries to apply the discretionary relief (art. 7(4) MLI). This relief allows a tax administration to grant the tax benefit but subject to consultation (no approval is required) to the other tax administration. However, this consultation also creates more delay in the application of a tax treaty, as well as more complexity in the way in which discretionary relief will be applicable by the tax administration. Since it is based in a discretionary power, it is not clear how tax authorities will use this relief in practice, and if this relief will also depend on the general features of the tax system of the country (e.g. depending on the formalistic approach of the tax administration, for instance, in civil law countries such as Colombia or France, or in a more consensual approach, for instance, the “polder” model in the Netherlands that allows for open discussions between tax administrations, taxpayers and tax advisors) .
Interaction between domestic GAARs and PPT
Another problem that creates more complexity is the interaction between the PPT and tax treaty anti-abuse clauses. Examples of these anti-abuse clauses are the main purpose test, Limitation on Benefits, and treaty provisions that make the application of domestic anti-avoidance rules possible when a tax treaty is being abused (e.g. art. 27 Panama-the Netherlands bilateral tax treaty). In principle, since the PPT is regarded as an umbrella clause, it will apply even if the transaction has passed the LOB test, or any other test in the treaty. However, in the case of domestic anti-avoidance rules (GAARs) there is no clarity in the interaction between these GAARs and the PPT.
One issue is the threshold for the application of the principal purpose test, which has provided for an intermediate threshold test (i.e., if one of the main purposes is met). Since there are three different approaches — intermediate (one of the main purposes) versus narrow (exclusive/sole purpose) versus broad (one of the purposes) — available to countries and their courts, the analysis of the introduction of an intermediate approach for countries who apply a narrow approach (Belgium, Brazil, France, Luxembourg, Spain and Turkey) or broad approach (India, New Zealand) will need to be taken into account (2018 IFA General Report: Anti-avoidance measures of general nature and scope GAAR and other rules Vol. 103b). Two questions that can be raised: (i) are these countries going to change their approach due to the principal purpose test?; and, if not, (ii) how will the interpretation by courts take place? Will the approach be addressed separately (i.e. an intermediate approach for the PPT and a narrow or broad approach for GAARs)?
Another issue is that, unlike domestic GAARs, the PPT does not have a ‘genuine economic’ or ‘artificial’ test. Some countries apply GAARs taking into account the economic reality for the application and, in the case of genuine economic reality, the tax benefit will not be denied. Furthermore, following EU case law by the Court of Justice of the European Union (CJEU) originated by the Cadbury Schweppes case, GAARs can only be applied to “wholly artificial arrangements”, thus limiting the application of the GAAR in EU countries. The approach differences between PPTs and GAARs have also been highlighted in the 2018 EU IFA Congress Report, the EU Commission 2016 Recommendation and, more recently, in the CJEU Danish Beneficial Ownership cases (C-115/16, C-118/16, C-119/16, C-299/16, and C-116/16 and C-117/16) stating that the general EU principle to prevent abuse is still applicable, and that it should be interpreted and applied in accordance to EU law. These differences are not only of importance for the tax treaty concluded by EU countries, but also for countries outside the EU, mainly due to the general application of the EU principle to prevent abuse that provides for interpretation of the CJEU and also of domestic courts of the abuse including the PPT in light of the EU principle.
The Danish ownership cases have raised two questions: (i) how economic genuine activity should be interpreted?; and (ii) who is analyzing the compatibility of the EU abuse principles with the EU countries domestic GAARs and treaty anti-abuse rules including PPT? Is it the CJEU? The EU country domestic court? And if it is the CJEU, can the CJEU recharacterize the transaction or will it be up to the EU domestic court to do that? There is a precedent in the interpretation of a tax treaty provision by the CJEU in the Austria-Germany case which, even though it refers to tax treaty arbitration cases, can also open the door for other tax treaty interpretations by the CJEU in this case justified by the current application of the EU principle to prevent abuse by EU countries in light of CJEU case law.
- What does this say about global tax governance?
The description of the developments in the international tax (minimum) BEPS PPT standard shows that countries are making different choices in the implementation of BEPS, but also that there are issues that will need to be addressed by tax administrations, ministries of finance, law-makers, scholars, tax advisors and taxpayers. The aim should be to find out how the PPT affects legal certainty for businesses, how to enhance revenue mobilization and prevent treaty abuse while balancing the need of (mainly) developing countries to attract investment. The current ongoing process of ratification of the MLI shows that it is difficult to have a coordinated position regarding international tax standards. This brings more uncertainty, but also more compliance burdens for the tax administration and taxpayers in general.
In addition, some countries are also implementing other actions, e.g. BEPS Action 12 (Disclosure of Aggressive Tax Planning Arrangements). This Action is required for EU countries in light of the amendments to the Administrative Cooperation Directive (Council Directive 2018/822/EU of 25 May 2018, DAC 6). Furthermore, some countries are opting in or opting out of the MLI anti-abuse provision of articles 8 (for dividend transfer transactions) and 9 (capital gains from alienation of shares or interests of entities deriving their value principally from immovable property). The result is more complexity in the implementation of the BEPS standards and the PPT, since the PPT as general GAAR in the MLI will have to interact with article 8 and 9 of the MLI. In addition, the information provided to comply with DAC 6 (i.e. BEPS Action 12) can be used by tax administrations for applying the PPT. Therefore, the relationship between PPT and Action 12 and PPT and art. 8 and 9 of the MLI should be further object of analysis.
This shows how difficult is to reach multilateral solutions if countries are choosing their own solutions (e.g. the United States regarding the use of anti-conduit financing arrangement provisions and countries signatories of the MLI making different choices). These choices can also change mainly to the interaction between Government and law-maker (Parliament/Congress), for instance in the case of the Netherlands.
The broad scope of application of the PPT is also the result of the lack of reference to “genuine economic reality” or to “artificiality” as elements for a PPT analysis (except in the EU), and this also gives more freedom to tax administrations to act, which also causes even more uncertainty. Countries, including their tax administrations, may act differently and provide different interpretations of a transaction resulting in treaty abuse or not. This interpretation can also be challenged before a court, based on general principles of law such as good faith. Therefore, it remains to be seen how the application of the PPT by tax administrations will be followed in judicial practice, and if there will be differences between common law (using judicial precedent) or civil law (more formalistic, literal interpretation of the law) countries.
Irma Johanna Mosquera Valderrama is Associate Professor of Tax Law at the Faculty of Law of the Institute of Tax Law and Economics, University of Leiden.