Christos Theophilou of Taxand Cyprus compares the principal purpose test in the OECD’s 2017 Model Tax Convention with the EU’s general antiabuse rule.
In order to counter transactions whose main purpose is to grant inappropriate tax benefits, many countries introduced into their tax codes certain specific antiavoidance rules or targeted antiavoidance rules. Because these rules proved to be inadequate, both the OECD and the EU introduced two general antiabuse rules of their own.
The first one, established in October 2015, is the principal purpose test, a tax treaty GAAR introduced by the OECD under Action 6 of the BEPS project. The second one, approved by the EU Council in July 2016, is the Antitax Avoidance Directive, which includes a GAAR in Article 6.
The aim of this insight is to discuss the overlap between the PPT and EU GAAR and analyze their similarities and differences.
“… benefits of a double taxation convention should not be available where a main purpose for entering into certain transactions or arrangements was to secure a more favourable tax position and obtaining that more favourable treatment in these circumstances would be contrary to the object and purpose of the relevant provisions.”
In this context, the OECD Model Tax Convention (2017), in its commentary on Article 1, Paragraph 61, notes that the PPT codifies the guiding principle. Consequently, depending on the legal status of the OECD commentaries, it could be said that many states have already applied the PPT. (For more about the legal status of the OECD commentaries, see my article “Using the OECD Model Commentaries and Other Tools to Interpret Tax Treaties.”)
Likewise, at least for some EU member states, the ATAD is not new. Because the ATAD provides for a minimum standard within the EU, various member states, such as Germany, Ireland, the Netherlands, and Sweden, did not need to amend their domestic law as their domestic GAARs already followed a stringent approach. Notably, unlike the ATAD, which is not adopted in the same manner across all EU member states, the PPT adopted by various countries largely follows the 2017 OECD Model Tax Convention.
The table below depicts the wording of the PPT, which is divided, for the purposes of this comparison, into three parts, and the EU GAAR, which is comprised of three paragraphs. (For more information, see my previous article.)
To begin with, as the EU GAAR is applicable only within the EU, the PPT has a broader coverage because it is included in both the 2017 OECD Model Tax Convention and as a minimum standard of the 2016 Multilateral Convention to Implement Tax Treaty Related Measures to Prevent Base Erosion and Profit Shifting. As of June 28, 2022, 99 jurisdictions have signed the MLI, and another three have expressed their intent to sign.
Second, the scope of the EU GAAR is restricted to corporate tax, while the PPT is broader in scope as it applies to all OECD Model Tax Convention income taxes (articles 6 to 21) and capital tax (article 22). Thus, for example, the PPT applies to individuals, while the EU GAAR does not. Another example is withholding taxes on passive income. Because the term “corporate tax” in article 1 and the term “corporate tax liability” in article 6 of the ATAD are not defined, it is therefore left to domestic law to interpret what taxes are covered. Notably, withholding taxes on passive income would typically be considered a corporate tax liability within the EU member sates, yet this ambiguity does not arise under the PPT.
Third, the first part of the PPT in sum explains that the PPT applies irrespective of any other tax treaty SAAR. Tax authorities would normally apply tax treaty SAARs first. For example, the beneficial ownership test under the passive income articles, namely Article 10 (dividends), Article 11 (interest) and Article 12 (royalties), or the rent-a-star provision under Article 17 (entertainers and sportsperson). If such SAARs would not achieve the desired result, then the PPT will usually apply. Similarly, in practice, GAARs also apply after SAARs; therefore, it is expected that the EU GAAR would normally apply after any other domestic SAAR.
Fourth, the second part of the PPT, i.e., the “purpose test,” is broadly the same as the EU GAAR. From the PPT perspective, the triggering provision is “one of the principal purposes,” whereas the EU GAAR is “main purpose or one of the main purposes.”
Nevertheless, it can be argued that the EU GAAR, which uses a slightly different wording, should be interpreted in a more restrictive way as compared to the PPT. Moreover, the PPT applies to “any arrangement or transaction,” and according to paragraph 177 of the Commentary on Article 29 of the 2017 OECD Model Tax Convention, it should be interpreted broadly. On the other hand, the EU GAAR applies only to “arrangements” and should be construed in accordance with the EU’s fundamental freedoms, specifically the freedom of establishment. As a result, the EU GAAR is more restrictive than the PPT.
Fifth, under the third part of the PPT, when the burden of proof shifts from the tax authorities to the taxpayer, the latter must provide conclusive evidence that such “benefit” was “in accordance with the object and purpose” of the relevant tax treaty. Equally, the EU ATAD states that a “tax advantage” should be in accordance with “the object or purpose of the applicable tax law.”
Typically, the term “benefit” (under the PPT) and “advantage” (under the EU ATAD) would be interpreted broadly and in a comparable way. However, the EU GAAR provides an additional test that is not included in the wording of the PPT.
In particular, to fall under the scope of the EU GAAR, the “arrangements” should be “not genuine.” The term “not genuine” is clarified as arrangements that “are not put into place for valid commercial reasons which reflect economic reality.” Thus, this so-called “economic substance test” would explicitly prevent the EU GAAR from applying if the test is not satisfied. In contrast, from the PPT perspective, even though the economic substance test is not explicitly included, the commentary on Article 29 of the 2017 OECD Model Tax Convention provides several examples that resemble the economic substance test being in accordance with the object and purpose of the relevant tax treaty.
Sixth, and finally, where a taxpayer enters into abusive arrangements the tax authorities, in applying the PPT, will deny any tax treaty benefit or protection claimed by such taxpayer. As a result, they will apply the relevant domestic tax law without any tax treaty restrictions. Similarly, when calculating the tax due under the EU ATAD, any abusive arrangements will be ignored and domestic law applied. However, it can be argued that tax treaties are part of domestic law, and thus a taxpayer would normally have access to that country’s tax treaty network provided that it meets all the conditions under the relevant tax treaty.
In light of the above, it can be seen that the purpose of the two GAARs, put simply, is to work as a safety net when other provisions, such as SAARS or TAARS, are circumvented. Consequently, taxpayers within the EU should take into consideration both GAARs, particularly regarding passive income structures such as financing, holding and licensing, because they present difficulties with regard to meeting the economic substance test.
For example, a simple holding structure where a parent company’s purpose is limited to holding one subsidiary would be unlikely to meet the requirements of the economic substance test. That test would more likely be satisfied by a complex holding structure, where an ultimate parent company’s purpose is to hold and manage numerous subsidiaries.
The current international tax environment is certainly becoming more demanding for multinational enterprises, not only in terms of meeting economic substance tests but also in terms of meeting business and commercial conditions. Structures such as holding, financing, and licensing special purpose vehicles need to be revisited to ensure that they comply with the new requirements.
This article does not necessarily reflect the opinion of The Bureau of National Affairs, Inc., the publisher of Bloomberg Law and Bloomberg Tax, or its owners.