INTRODUCTION
Globalisation has brought a scenario of significant interaction and interdependence between states and peoples. This framework of proximity between countries, however, creates opportunities for access, by abusive schemes, to the network of conventions on double taxation of a state by non-resident economic agents. This abusive mechanism is called treaty shopping, and it plays a critical role in the erosion of a nations’ tax base, being
indispensable its confrontation. In this regard, there is the development of new tools to address abusive schemes, such as the PPT rule, which plays the role of a General Anti- Avoidance Rule – or GAAR –, universally preventing abusive tax planning. This legal construction was recommended by Action 06 of the BEPS project and expressed in Article 7 (1) of the Multilateral Instrument – or MLI. Also, a replica of the PPT rule can be found in the 2017 update to the OECD Model Tax Convention, with very slight changes to address the nature of bilateral treaties (ELLIFFE, 2019, p. 6).
In relation to the Brazilian international tax policy, it is noteworthy the inception of the PPT rule into the Tax Agreement signed by Brazil with Argentina (Decree no 9482/2018, article XXVII, item 1). This new trend for the Brazilian international practice can be seen in the amendment signed but not yet into force that has incepted that GAAR into the Brazil- Sweden Double Tax Convention (Message from the President to the Congress no 43/2020, article XXVI-A, item 9). Furthermore, there are newly signed but not yet ratified Treaties with Singapore (Message from the President to the Congress no 796/2018, article 28, item 8), UAE (Message from the President to the Congress no 394/2019, article 29, item 4), Switzerland (Message from the President to the Congress no 242/2019, article 27, item 1) and Uruguay (Message from the President to the Congress no 44/2020, article 29, item 9) that have also adopted the PPT rule. Accordingly, a detailed study of such anti-avoidance rule, and its real applicability, is appropriate.
The objective of the study is limited to a critical analysis of the PPT rule, observing both conceptual and practical issues inherent to its core elements. In this sense, the article presents five parts. After this introduction, Part two discusses the treaty shopping and the fundamental aspects of the PPT rule, granting their definitions. Next, Part three analyses the issues that derives from the description of the PPT rule, whereas Part four examines the practical challenges that are intrinsic to the PPT rule. Part five finishes the paper by summarising the main arguments put forward in the article and drawing from them the conclusions arising from the questions examined in the study.
TREATY SHOPPING AND THE STRUCTURAL FEATURES OF THE PPT RULE The taxpayer may design tax schemes that provide greater tax efficiency, including taking advantage of the mismatches between one country domestic legal system and its tax
treaties. However, such tax planning cannot take place by the use of artificial conceptions,
in which the reduction of the tax burden is made regarding an abusive scheme, i.e., a mechanism that acts against the essence of the legal provisions provided for in the States’ internal laws or international tax treaties (COURINHA, 2015a, p. 293). In this regard, treaty shopping is the leading exponent of abusive practice of tax conventions.
To understand the phenomenon of treaty shopping, it is worth mentioning that only the residents of a particular country may benefit from the network of conventions on double taxation signed by that State (OECD, 2015, p. 17). Thus, the treaty shopping arises from a tax
structure organised in such a way as to enable a taxpayer to benefit from the Conventions of a State of which he is not a resident. And this phenomenon occurs, in short, when a non- resident person accesses the network of conventions of a certain State through the establishment or acquisition of a company or the relocation of his residence to that nation (COURINHA, 2015a, pp. 291-313).
Several proposals have already addressed ways to combat treaty shopping, being the usual solution the use of jurisprudential guidelines of the national courts, through the principle of substance over form, or restrictions provided for in the domestic law or the Tax Agreements signed by the State. The OECD itself, in the Commentary on Article 1 of the OECD MC, has set a guiding principle in addressing abusive practices, which calls for the denial of benefits to a tax scheme which has deprived the essence behind that benefit (OECD, 2017, C(1)-61).
However, in the context of cross-border transactions, abusive schemes, taking advantage of the lack of harmony between tax systems, always find a way to evade antiabuse rules. Along these lines, the G20 and the OECD jointly developed the BEPS project, an international soft law initiative (COURINHA, 2015b, p. 294) composed of 15 measures, arranged in individual and interdependent actions, which aim to cope with abusive or aggressive tax planning1. The strategy would work through the incorporation of antiabuse instruments into the internal legal order of States, as well as to the international tax treaties agreed upon by nations.
Regarding the abuse of conventions, the BEPS Project2 made in Action 06 three recommendations: (i) the inclusion, in the title and preamble of the Double Taxation Treaties, of the express caveat that the respective Contracting State will not favour evasive or avoidance practices, such as treaty shopping, whereas this country will commit itself to tackle the erosion of the tax base; (ii) the setting of a Specific Anti-Avoidance Rule – or SAAR – in the form of Limitations-on-Benefits rules – or LOB; and (iii) the elaboration of a General Anti-Avoidance Rule – or GAAR –, in the figure of the Principal Purpose Test rule
– or PPT rule (LANG, 2014, p. 655; CHAND, 2015, p. 485).
In the wake of the BEPS project comes the Multilateral Instrument, developed by the OECD, which is the primary vehicle for the implementation of antiabuse measures under international treaties. The MLI, signed by 94 jurisdictions as of May 2020 (OECD, 2019, pp.
The concepts are not interchangeable. In abusive tax planning, the taxpayer acts in opposition to the domestic rules of a State by obtaining improper advantages derived from artificial schemes (DOURADO, 2015, pp. 44-45). On the other hand, in aggressive tax planning, even if the artificiality of the structure is not detected, it should still be repressed if the incompatibilities between national orders are taken advantage of, thus acting as an “umbrella concept”, as is a vague definition that encompasses situations of avoidance, evasion and even lawful tax planning (DOURADO, 2015, p. 48; DOURADO, 2016a, pp. 287-288; DOURADO, 2016b, p. 441; MENESES, 2009, p. 442; BELLINGWOUT, 2015, p. 3).
The BEPS plan can be summed up as an initiative to combat tax planning that relies on asymmetries between the various tax systems to artificially reduce the tax base or profit shifting to low or zero tax jurisdictions where there is little or no economic activity performed.
1-3), acts as an amendment to existing Tax Treaties. Thus, instead of countries renegotiating all the Tax Treaties of which they are a party – a lengthy and inefficient procedure to combat abusive schemes – the MLI automatically transposes antiabuse rules into the Tax Agreements. Even so, the OECD has safeguarded the possibility for states to make comments and caveats at the time of signing the Multilateral Instrument.
It is also relevant to mention that the MLI has established that all signatory countries must adhere to a minimum threshold. In doing so, the OECD has classified the PPT rule as the leading standard for preventing abusive schemes since that GAAR has all the minimum requirements by itself. In this sense, the MLI also has created the opt-out system concerning anti-avoidance clauses, and, if MLI signatory states wish to apply a different rule than the standard, they must expressly select which norm will affect their tax treaties (OECD, 2016, pp. 3-4; CHRISTIANS; SHAY, 2017, 44-45; BAKER, 2016, pp. 683-689). Finally, the
PPT rule can be found in article 7 (1) of the MLI, as listed below:
“Notwithstanding any provisions of a Covered Tax Agreement, a benefit under the Covered Tax Agreement shall not be granted in respect of an item of income or capital if it is reasonable to conclude, having regard to all relevant facts and circumstances, that obtaining that benefit was one of the principal purposes of any arrangement or transaction that resulted directly or indirectly in that benefit, unless it is established 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.”
The PPT rule, as set out in the Multilateral Instrument, will deny a benefit stipulated in a Tax Agreement when there is the cumulative combination of three factors (OECD, 2015, pp. 18-19), namely: (i) the benefit under consideration falls within the scope of Article 7 (1) of the MLI – this is the scope of application of the rule; (ii) one of the taxpayer’s principal purpose when designing its tax structure is to obtain such a benefit – this is the subjective element of the PPT rule; and (iii) the access of that benefit is contrary to the design of the Tax Treaty
– this is the objective element.
In this manner, the PPT rule follows the same trail as the guiding principle set out in the Commentaries on Article 1 of the OECD MC (OECD, 2017, C (1)-61; OECD, 2015, p. 55). For this principle, the benefits of a Tax Agreement should not be granted in the presence of abusive tax planning, i.e. the artificial tax scheme designed to reduce the tax burden in breach of the object and purpose of the rules set out in the Treaty (KOK, 2016, p. 407; CHAND, 2015, p. 485). Below, the study highlights the nuances of the structural features of Article 7 (1) of the Multilateral Instrument.
Scope of the PPT Rule
The scope of this GAAR is at the beginning of MLI Article 7 (1):
“Notwithstanding any provisions of a Covered Tax Agreement, a benefit under the Covered Tax Agreement shall not be granted in respect of an item of income or capital […]”
As it is clear from the provision, the PPT rule shall apply regardless of the scope of the other sections provided for in a given Tax Agreement. This position adopted by the OECD will have an impact on the compatibility of the provisions of this GAAR with other regulations, especially those of a unique nature. Accordingly, the scope of the PPT rule is limited to the Tax Agreement, which will be modified by MLI. Consequently, tax benefits granted under national tax law cannot be denied under the PPT rule. It is equally inconceivable that such norm could withhold benefits provided for in other treaties concluded by one of the Contracting States or by entirely different States, even if they have the same personal or substantial scope (LANG, 2014, p. 656-657).
In another turn, what are the benefits that can be denied concerning article 7 (1) of the MLI? The same thought, as presented above, should be adopted. As the PPT rule only concerns the Tax Treaties modified by the Multilateral Instrument, the benefits to be refused must be set out in that Tax Agreement. If the benefit is derived from national law; or European law; or another Tax Convention, there is no scope by the PPT rule, as it only acts within the parameters of a particular treaty (OECD, 2015, p. 56; KOK, 2016, p. 408).
In a sum, the benefits that may suffer from the hindrance of the PPT rule, according to the OECD, are those established at articles 6 to 24 of the OECD MC, as articles 1 to 2 are mainly focused on defining concepts to be applied throughout the Tax Agreement, and as articles 3 to 5 provide for benefits that may only be exploited by abusive practice if combined with other provisions of the Tax Treaty (OECD, 2015, p. 56; KOK, 2016, p. 408). Conversely, the rule proposed in Article 7 (1) of the MLI can hardly affect article 28 of the OECD MC, which refers to the fiscal privileges of members of diplomatic missions or consular posts, as these benefits derive from rights established in specific treaties and international customs, not the Tax Treaty itself (LANG, 2014, pp. 656-657).
Subjective Element: Principal Purpose
The subjective element of the principal purpose test can be taken from the following excerpt from Article 7 (1) of the Multilateral Instrument:
“[…] a benefit under the Covered Tax Agreement shall not be granted […] if it is reasonable to conclude, having regard to all relevant facts and circumstances, that obtaining that benefit was one of the principal purposes of any arrangement or transaction that resulted directly or indirectly in that benefit […].”
Thus, to apply the PPT rule, the taxpayer’s real intentions must be analysed, so that if one
of the main reasons for a particular tax design is to acquire a benefit, this one should be
denied (CHAND, 2015, p. 485). The difficulty of this criterion is, therefore, apparent, as it is difficult – if not impossible – to prove an agent’s real intent. Moreover, the PPT rule does not need concrete evidence for its application, but only if it proves to be reasonable to conclude that the taxpayer wished, when designing the tax scheme, to acquire a benefit. On the one hand, such a provision assigns the responsibility for tax authorities to investigate and conclude that the structure was indeed abusive. On the other, the absence of the need for evidence puts the taxpayer at a disadvantage, as it may have a benefit denied without, however, the previous gathering of conclusive evidence proving the perpetrated abuse (LANG, 2014, pp. 658-659).
Objective Element: Purpose of the Tax Treaty
It is also underlined by Article 7 (1) of the Multilateral Instrument that:
“[…] a benefit […] shall not be granted […] if it is reasonable to conclude […] that obtaining that benefit was one of the principal purposes […], unless it is established 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.”
It is not enough to attest the presence of the subjective element, that is, the intention of a taxpayer to set up a structure that has as one of the primary purposes the obtaining of a benefit provided for in a Tax Treaty, being indispensable also to verify the objective criterion of the PPT rule. In this context, two points should be highlighted: (i) whereas the subjective aspect of the GAAR observes the taxpayer’s intentions, the objective element analyses the purposes of the Tax Treaty; and (ii) while the subjective feature requires only a reasonableness test, by which it is sufficient for the tax authority to understand the potential for the abuse to exist, the objective criterion requires that the tax structure under consideration be in compliance with the provisions of the Convention.
In short, the objective element of the PPT rule commands that obtaining a specific benefit through a particular tax planning is appropriate under the established rationale of the Tax Treaty (LANG, 2014, p. 660; CHAND, 2015, p. 485). In this point, it is essential to inquire about the meaning of the term “established” since it can be imagined that the granting of a benefit must be expressly set in the text of the Treaty. However, stating that the tax authority should only rely on a purely literal interpretation, the objective element will be reduced to insignificance, for in only rare cases will this criterion be met. The present paper does not follow this trail of thought, as it is inconceivable to expect that an international treaty will be so attentive to all the concrete cases in which the benefit may be granted. Also, the PPT rule is a GAAR, being by nature abstract in its definition (EY ABOGADOS, 2015, p. 18). In conclusion, the objective element shall be met when the granting of the benefit respects the purposes of a Tax Agreement without the need for an express provision. Thus, it is prized for a logical-systematic and not merely literal interpretation.
Nonetheless, it remains to be investigated what are the scope and purpose of the provisions of the Tax Agreement. In this sense, as these are not expressly written in the body of the Treaty, it is necessary to study the Treaty in its entirety to extract the meaning of its text. In this task, not only the rules laid down in the Agreement, but also the title and preamble of the Treaty must be observed. In this light, it should be noted that BEPS Action 06 has adopted as one of the minimum approaches to tackling treaty shopping the express caveat that states will not create opportunities for evasive or avoidance practices (OECD, 2015, p. 10). So, this measure entails the modification of the titles and preambles of the Tax Agreements. Finally, it is worth considering the Comments on the OECD MC articles, given that, although not binding, they serve as an interpretation of the ratio legis of the provisions and, consequently, of the Convention itself (KOK, 2016, p. 409).
CONCEPTUAL ISSUES OF THE PPT RULE
The Irrelevance of the Subjective Element
In relation to the subjective test, the present study argues that it is riddled with ineffectiveness, which culminates in the very insignificance of this core aspect of the PPT rule. Initially, as is apparent from the subjective element of the GAAR, access to the benefit of a Tax Agreement, signed by a particular State, should be denied if a particular tax planning has as one of its main objectives the attainment of such benefit. So, as an example, an enterprise may prove that a particular intra-corporate decision has multiple goals, such as to further its competitiveness in the market, but if the acquisition of a tax benefit is within the range of motivations, then such benefit shall be denied (OECD, 2015, p. 58).
This legal design abandons the conceptual construction of abusive tax planning that has been built in recent years, especially in the European Court of Justice. In the Halifax case (C-255/02), that Court applied an antiabuse rule because it found that the “essential aim”, above all others, was to obtain a tax advantage – and thus falling in the category of abusive tax planning making it necessary to refuse the acquisition of a benefit. This logical construction is not present at the PPT rule, as it only envisages that one of the main objectives is to obtain the benefit, hence considering such tax planning as abusive (LANG, 2014, pp. 659-660).
As a criticism, the ease for characterising a tax planning as abusive entails circumstances in which a sound corporate decision, conveyed with sufficient business purpose or economic substance, may be disqualified from their right to obtain a specific tax benefit by this GAAR. It should be remembered that the PPT rule focuses on the objectives of the tax scheme, but not its substance. In short, even though an enterprise’s choice contains economic significance, in respect of the substance over form principle, it may still be affected by the PPT rule, provided that one of its central purposes is merely to lessen its tax burden (CHAND, 2015, p. 487). In this trail of thought, rather than the multiple purpose
criterion, it would have been more appropriate to apply “the essential aim” criterion – at this point, with an ultimate sense of purpose and superior to all others – already employed by international jurisprudence, e.g., the emblematic Halifax case, but also in recent years: Terracult case (C-835/18, opinion of advocate general); KrakVet Marek Batko (C-276/18, opinion of advocate general); Kuršu zeme (C-273/18, Judgment); T Danmark (C-116/16, Judgment); N Luxembourg 1 (C-115/16, Judgment); Skarpa Travel (C-422/17, Judgment). Therefore, only actual abusive tax planning would have its benefits disregarded.
Notwithstanding the harshness imposed by the subjective element, which may even affect non-artificial tax schemes, it cannot be forgotten that the criteria of a multiplicity of purposes should be envisaged in line with the objective element. Indeed, the PPT rule itself provides a solution to the difficulties presented by the subjective element, as the objective aspect overcomes the conceptual flaws of its counterpart by establishing that the benefit will be granted to tax planning that, despite having as one of their primary purposes the acquisition of the tax advantage, respect the purposes of the Tax Agreement. Ultimately, it is not the subjective factor that will determine if a tax planning is to be denied of its benefits, but rather if the tax planning has held in high regard the intent of the Tax Treaty
more of this will be presented later.
It is noteworthy that, given the diversity of purposes of a given tax planning, the subjective element has high conceptual elasticity. In other words, the PPT rule, besides being severe due to the subjective element, is generic and can, through countless interpretations, be applied to a myriad of circumstances. The PPT rule, therefore, makes it possible to impose a severe punishment – the loss of the tax benefit – without, however, explicitly presenting the cases to which it relates. This terminological breadth potentially violates the principle of legal certainty, as the PPT rule does not provide a clear, precise and predictable rule, and this vagueness has serious consequences, since the rule in question does not allow taxpayers to realise in advance whether access to the benefit provided for in the Tax Agreement will be denied or not (CUNHA, 2016, pp. 186-187; TEIXEIRA, 2016, pp. 58-59; MARTINS, 2007, pp. 35- 36). On this point, however, it should not be forgotten that a GAAR must inevitably contain a more abstract definition. If not, it would no longer be a general rule, rather becoming a specific anti-avoidance rule.
In the same sense follows the cost-benefit critique of this GAAR. It can be argued that the inherent conceptual inaccuracy of such a GAAR enables the tax authority to outweigh any benefits that may accrue from the tax planning (ROSENBLATT; SANTOS, 2018, p. 250). The present paper does not follow this trend, as it acknowledges that the current specific clauses are not sufficient to combat the erosion of the tax base, as aggressive tax schemes either find the legal loopholes or, despite being in line with the literal meaning of the law, they violate the essence of the norm, recognising that it is imperative, for the maintenance of the current tax system, that a GAAR along the lines of the PPT rule should be established on the international stage.
Another criticism refers to the very terminology used by the Multilateral Convention since it exposes a semantic inconsistency. The PPT rule is based on the fact that a scheme can have multiple “principal” purposes. Now the very multiplicity of objectives disqualifies the motive as being “principal”. It makes no sense to say that at the top of the hierarchical chain of purpose of a tax scheme is a plurality of objectives – a pyramid has only one summit. It makes more sense to return to the conceptual construction already established in the legal literature and jurisprudence of “the essential aim” criterion, in which the abuse of the fiscal scheme is analysed given its primary purpose, rather than a multiplicity of objectives.
In relation to this criticism, despite the terminological inaccuracy, the rationale of Article 7 (1) of the MLI is rather clear, since it is based on the understanding that fiscal planning can have multiple purposes – fiscal or extra fiscal – which may be at different levels of relevance to the assessed structure, and the use of obtaining a benefit should be at a higher level to validate the limitation-on-benefits, not justifying the punishment when it is found in a secondary layer of intention. Therefore, it is noted that the Multilateral Instrument divides the taxpayer’s intentions into two categories: primary and secondary value (CUNHA, 2016, pp. 187-188). This distinction causes uncertainties, as it becomes challenging to determine the order of importance between the taxpayer’s intentions so that the tax authorities should advise the abuse on a case-by-case basis; and even if the taxpayer proves that in the upper tier of motives it was not intended to obtain an individual tax benefit, the tax authority, based on Article 7 (1) of the MLI, may conclude that the tax scheme is still abusive – which excludes the taxpayer’s true intentions, privileging the tax authorities’ discretion.
Another topic to consider is the absence in the Multilateral Instrument of the method that should be employed to determine if the attainment of a benefit was unlawful. In other words, the scope of the analysis to be promoted by the tax authority is inaccurate, so it is not known whether the taxpayer’s tax position will be assessed only at the domestic level or whether the taxpayer’s overall situation will be observed (LANG, 2014, p. 657). As the MLI has no answers, a logical and systematic interpretation of the international tax system is necessary. That is, the present study understands that the tax situation of the individual should be determined as a whole, not only regarding a particular jurisdiction. This reasoning is based on the fact that the benefit analysed is the result of the application of international rules, so that its actual occurrence should also be investigated in a cross- border context. After all, only a global analysis can maintain the coherence of the international tax regime. For example, whereas in State “A” the tax burden might have decreased – which would be considered a benefit – in State “B” a heavier tax imposition possibly took place. If the tax authority of State “A” analyses the tax context of the taxpayer only at a national level, it will be allowed to apply the GAAR, if the subjective and objective elements are met since the tax position of the taxpayer has been improved.
However, this stance will be an aggravation to the taxpayer as the advantage acquired in State “A”, which would oppose the loss suffered in State “B”, will be eliminated, thus resulting in double taxation.
Further examination of the subjective element also unveils that, under the expression “reasonable to conclude”, the real intentions of the taxpayer will be disregarded without the need for conclusive proof, since all the Tax Authority requires is a reasonable conclusion that points out the intents behind the taxpayers’ planning to access a benefit (OECD, 2015, pp. 57-58; LANG, 2014, pp. 658-659). This paper understands that the lack of evidence will give rise to arbitrariness by the Tax Administration. It is noteworthy that the Final Report of Action 06 of the BEPS Plan established that the Tax Authority will not be able to act on mere conclusions in ascertaining the taxpayer’s purposes (WEBER, 2017, pp. 49-50). In this sense, the tax authorities will only conclude that the achievement of a tax benefit was one of the taxpayer’s objectives if the assembled tax structure is only explained based on obtaining such benefit (OECD, 2015, p. 58). However, this guideline promoted by Action 06 will not avoid arbitrariness on the part of the tax authorities, because the PPT rule works with the understanding that there may be multiple purposes in the organisation of a tax planning, which makes it challenging to find that obtaining a tax benefit is not part of the intention of the structure; and the existence of an evidential set that supports the conclusion of the Fiscal Authority is necessary. Thus, in practice, the PPT rule will find no limits on the reasonableness test provided for in Action 06 of the BEPS project (KOK, 2016, p. 408). Actually, in the end, such criterion will increase the discretionary margin of the tax authority, which will only produce evidence that allows insight as to the existence of abuse (CUNHA, 2016, p. 188; CHAND, 2015, p. 487-488).
Finally, it is worth noting that this pattern to produce evidence is an innovation promoted by the Multilateral Instrument, as the OECD, in its Model Convention, adopts another type of approach. Indeed, the already established position is guided by the need to compile sound proof of the existence of abuse, as it can be seen from Comment No. 80 to Article 1 of the 2017 OECD MC (OECD, 2017, C(1)-80):
“Whilst these rules do not conflict with tax conventions, there is agreement that member countries should carefully observe the specific obligations enshrined in tax treaties to relieve double taxation as long as there is no clear evidence that the treaties are being abused.” – the remark was made by the author
This understanding is more appropriate in the interests of the tax authority and the taxpayer, as it does not require definitive evidence, but at the same time does not make room for a mere illation, as the evidence of abuse must be crystal clear.
Since it is difficult to demonstrate the taxpayer’s real intentions, as well as to provide reasoning based on hard evidence, regarding the fact that the tax authority’s reasonable
conclusion of the practice of avoidance is sufficient, the practical consequence of all the arguments presented above will be the ulterior indifference to the reasons that permeate the tax planning. Irrespective of the reasons exposed to the tax authorities by the taxpayer in formulating a particular tax design, if the tax authority concludes by a reasonableness judgment that there has been a higher purpose of obtaining a benefit, it may deny its attainment, provided there is also the presence of the objective element of the PPT rule. Rather than the Tax Authority, it is the taxpayer who bears the probative burden. It is not enough that the tax planning under analysis has several extra-fiscal purposes. If it has a tax purpose, the taxpayer will be required to prove that the benefit accrued meets the objective aspect of the PPT rule. Otherwise, the tax authorities will deny the benefit, based on its conclusions, without substantial evidence.
It is noted repeatedly that the will of the taxpayer is often pushed to a secondary role. Based on what has already been explained throughout this study, the PPT rule assumes that there may be multiple primary purposes for tax planning, all potentially at the same hierarchical level, without legal distinction between primary and secondary intentions, so that, although the taxpayer has as an overarching motive of his tax scheme an extra-fiscal purpose, the tax authorities may still understand that there was abuse. In this case-by-case verification process, the tax authority need not even provide crystal clear evidence of the taxpayer’s avoidance intention, let alone definitive evidence, being a reasonable conclusion a sufficient criterion. Therefore, it is perceived that the important thing is not the taxpayer’s will (LANG, 2014, p. 661), but the tax authority’s judgment of reasonableness, which may frame a particular tax structure as being abusive, but without solid proof to justify this decision. This will potentially cause great injustice, as a tax planning may be endowed with economic background, in compliance with the principle of substance over form, and yet be deprived of obtaining the tax benefit stipulated in a Tax Agreement.
In the end, though, there is a logical inconsistency within the PPT rule. This GAAR does not analyse, in its subjective element, the substance of the structure, but rather the primary purpose of the tax scheme, unless the objective element, which observes the purpose of the Tax Treaty itself, allows such tax planning. Nonetheless, it should be highlighted that one of the purposes of the treaties is to facilitate cross-border transactions by minimising the tax burden and thus allowing for a more dynamic trade relationship between states. The taxpayer is encouraged to design its business in a way that improves tax efficiency. The Tax Treaty aims to release taxpayers of its burdens by granting them tax relief and hence improving their stance in the international market. Therefore, there is a paradox within the PPT rule. When the taxpayer pursues the reduction of his expenses, through the use of a tax benefit, which is prohibited by the subjective element, it is actually acting according to the purposes of the Tax Treaty, which is allowed by the objective factor. Since the objective test was designed to be an exception to the subjective feature, it seems, therefore, that the first completely nullifies the
importance of the second. It is not precisely a setback since it was stated before that the subjective test is, in fact, too severe, vague, redundant and arbitrary.
Moreover, as it was seen, the subjective element ends up reduced to the discretion of the Tax Authority, as the taxpayer’s intention is relegated to a secondary role. Thus, the only real limit to the tax authorities’ application of the PPT rule will be the objective element, namely the verification that obtaining the respective tax benefit complies with the purpose of the Tax Agreement. From this reasoning, two consequences can be unveiled: (i) if the tax scheme obeys the contours for achieving the benefit established in the Convention, then there will be no impact on the objective element, so the PPT rule cannot apply; and (ii) if, however, it is not in line, the tax authorities may freely analyse the existence of the subjective element and, if it considers it reasonable to suppose that the tax scheme sought to enjoy the benefit as one of the main reasons, without the need to fully produce sound shreds of evidence in this regard, the tax authorities may apply the PPT rule, denying the benefit. This paper follows the thesis that the objective element will be the crucial component in the application of the PPT rule since the taxpayer’s purposes are not relevant in the finding of abuse so that the real limit to the denial of benefit is parity with the object and purpose of the tax treaty.3
The Redundancy of the Objective Element
As the subjective element is flawed in many respects, the objective component is relegated to the real role of limiting the benefits provided for in the Tax Agreements. Thus, this feature must be thoroughly studied for possible weaknesses. In this sense, the objective aspect analyses the object and purpose of the provisions provided for in the Tax Treaty to exceptionally rule out the application of the PPT rule and thereby prohibit the denial of a particular benefit. However, this logic is already reflected in Article 31 (1) of the Vienna Convention, which provides that the interpretation of the rules laid down in the international treaties must respect the object and purpose of such Conventions.4 Is the objective element of the PPT rule, therefore, redundant?
The objective test may only be regarded as a guideline for interpretation of the PPT rule, as expressed in the Vienna Convention. In other words, the objective element would be a mere interpretative instrument, in the sense that the norms provided for in the international treaty need to be glimpsed in the light of its object and purpose. This logic is found in the Action Final Report 06 itself, which illustrates the application of the PPT rule through examples, which would be a demonstration of how the interpretation of the
Reinout Kok (2016, p. 407) asserts that the subjective element is the primary test since if the purpose of the scheme was to achieve an individual benefit expressed in the Treaty, the PPT rule applies. Only at a second time, and if it is compatible with the objectives of the tax treaty, is the objective element observed.
Article 31 (1) of the Vienna Convention: “A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given
to the terms of the treaty in their context and in the light of its object and purpose.”
GAAR should be performed (LANG, 2014, p. 661). Through this reasoning, however, the PPT rule itself would become unnecessary, since the tax advantage would of course not be granted if it were contrary to the purposes of the Treaty, in accordance with Article 31 (1) of the Vienna Convention.
The present study does not follow this point of view. Bearing in mind that the Vienna Convention is merely an interpretative instrument, there is no ground to speak of denying tax benefits in the name of combating the abuse of the Conventions without the primary setting of an antiabuse rule. That is, Article 31 (1) of the Vienna Convention does not provide for an antiabuse rule, but only a means of interpreting the provisions of a Convention (KOK, 2016. p. 408; RANZ, 2017, p. 200). Therefore, the claim that the objective element is redundant does not sustain itself, as it is necessary to fix an anti-avoidance rule to cope effectively with abusive practices.
The Real Effectiveness of the PPT Rule
From all the above, the most logical conclusion is that the taxpayer is at a severe disadvantage vis-à-vis the Tax Authority. However, this reasoning may, in practice, not be correct. Contrary to what it is supposed, the PPT rule could be extremely ineffective, if not completely null, and this is due to the wording of Article 7 (1) of the Multilateral Instrument, which provides that the benefit will be denied “unless it is established 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”.
The fulfilment of the object and purpose of the Tax Agreement represents the major deadlock in the effectiveness of the PPT rule. As already highlighted before, for compliance with the objective element, it is necessary to employ the logical-systematic interpretation. Thus, the purpose of the Convention stems from the general reading of the Convention’s articles, title and preamble, and this is the crux of the matter: abusive tax planning, if well- crafted, generally fit within the legal purposes of the Tax Agreement (CHAND, 2015, p. 487). Indeed, tax avoidance typically exploits the rules set out in Articles 6 to 24 of the OECD Model Convention, so that the wording of the provisions is employed to elicit the avoidance effect. In other words, large abusive schemes use the letter of the law to generate tax avoidance. In this sense, the legal purpose of the articles is respected.
However, the basis of the PPT rule is not the legal purpose of the provisions, but the systematic-teleological purpose of the provisions contained in the Tax Agreement. The objective element of the PPT rule observes not the wording of the law, but the essence of the norm. This is the challenge that hinders the widespread effectiveness of the PPT rule and raises questions: What are the goals and well-defined boundaries between the legal and the essential purpose of the Convention’s provisions? How to say that a specific tax planning respects the word of law, but nonetheless disregards it as to its rationale? The
line is thin so that the analysis of the PPT rule will eventually take place on uncertain grounds. It should be expected several conflicts at the Courts, severing the legal certainty.
PRACTICAL ISSUES OF THE PPT RULE
Tax Rights Allocation Between Jurisdictions
In the light of all the above considerations, it remains to be seen the legal consequences of applying the PPT rule. As it was stated before, the principal purpose test was designed to deny a tax benefit if the requirements were met. Nevertheless, it is not clear which legal provision will govern the tax scheme, since the former, which provided for the benefit, was dismissed. For addressing this question, an example is presented: Company “X”, resident in State “A”, invests in Company “Y”, its subsidiary, resident in State “B”, through a financial instrument characterised as debt, as opposed to equity, so that payments made by company “Y” will be classified as interest, not as dividends. By setting up this tax scheme, company “X” avoids taxation at 5% or 15% rates of the State of Source, as provided in article 10 of OECD MC for income configured as dividends. If this structure is riddled with artificiality, it should be considered abusive and, therefore, the tax benefit obtained in the scheme – that is, the tax difference that would occur if the income were considered “dividend” and not “interest” – should be denied by the PPT rule. However, as the legal consequences of applying this GAAR are not known, there are three ways to solve the problem (LANG, 2014, pp. 661-662):
The rejection of the benefit would go through the denial of the classification of the investment made, so that it would be taken as “social capital”, forcing its recognition by the State of Source as “dividends” and, thus, attracting the application of article 10 of the OECD MC, with withholding tax at 5% or 15%; or
The denial of the classification of “interest”, i.e. the departure from the application of Article 11 of the OECD MC, but without the immediate transposition of the definition of income as dividends, moving to Article 21 of the OECD MC, which applies to income not addressed by the other provisions of the Model Convention. As a result, taxation would be the exclusive responsibility of the State of Residence; or
In a more extreme measure, the dismissal of benefit could be realised in the sense that none of the Treaty’s distribution rules would apply so that the tax convention itself will be overruled. In this case, the national laws of the States will apply as if there were no Agreement in force (CHAND, 2014, p. 488). This position may result in double taxation and the ultimate irrelevance of the Tax Treaty.
The practical application of the PPT rule raises other problems, as the BEPS Final Action 06 Final Report is silent regarding the distribution of competence in the implementation of the PPT rule among States, as well as the situation of the other tax benefits obtained by the tax planning, and that was not hampered by the GAAR. As solutions to these issues
have not been aired by Action 06, there is room for the assumption that both Contracting States will be able to suppress the attainment of the tax benefit, in bis in idem, which would inevitably enact double taxation (LANG, 2014, p. 662; CHAND, 2015, p. 488). For example, if the Source State applies the PPT rule and thereby changes the income classification from “interest” to “dividends”, there will be an increase in withholding taxation. Should the State of Residence allow an adequate deduction to settle double taxation? Is the State of Residence obliged to increase the credit granted under Article 23 of OECD MC? It seems that both questions have positive answers. However, by contrast, if the State of Residence does not recognise the application of the PPT rule in the State of Source, should it apply the measures set out above? How will conflicts of tax jurisdiction between the two countries be resolved? The present study follows the path that, regarding the conflicts of competence between states in the application of the PPT rule, the countries concerned will have to enter into mutual agreements to remedy the situation (LANG, 2014, p. 662).
Cases Involving More Than Two States
Another issue that arises from the PPT rule concerns the cases in which there are more than two states involved in the abusive tax structure. For a better understanding, it is necessary to draw an example scenario: company “X”, resident in country “A”, wants to finance, through equity, the company “Y”, which is resident in State “C”. To do so, the company “X” establishes a subsidiary company, named “W”, in State “B”, which will, in turn, use the “BC” Tax Treaty to reduce the tax burden of this transaction. In this scenario, the PPT rule will be applied to cope with the treaty shopping, denying the tax benefit collected in that tax planning. However, it should be noted that the company that obtained the benefits was not in State “B”, but in State “A”. It is also emphasised that the “AC” Tax Treaty limits the taxation of dividends to a maximum of 15%. At this point, Michael Lang (2014, pp. 662-623) asks: State “C” – source country of income – will be required to apply, or at least extend the effectiveness, of the 15% limitation provided for in the “AC” Convention, notwithstanding the fact that the PPT rule cancelled a tax benefit provided for in the “BC” Treaty?
The BEPS Action 06 Final Report does not provide answers to this question. However, the present study works on the hypothesis that the effects of a Tax Treaty are not effective erga omnes but inter partes. This logic is anchored in the international principle of the relative effectiveness of treaties, in the sense that there is no correlation between different Agreements, which means that one cannot be applied to interpret another (COURINHA, 2015a, pp. 91-98). Thus, it makes no sense for State “C” to be obliged to act in a certain way concerning Convention “AC” because of the collateral effect of Treaty “BC”. Reverse thinking entails departing from the provisions of one tax agreement for the benefit of another, which is not in line with best international practice, as the “AC” Tax Treaty is not hierarchically superior and is not specific vis-à-vis the “BC” Convention, and vice versa. The
denial of benefits, therefore, in one treaty, cannot reverberate in the interpretation of another, so that the 15% limitation in Convention “AC” applies regardless of the application of the provisions expressed in Convention “BC”.
Compatibility with Domestic Law
Several countries have now introduced or are considering adding GAARs into their legal systems. In this sense, the position of the Multilateral Instrument to include the PPT rule in international treaties only reflects this current trend. However, it should be noted that this global movement for the incorporation of GAARs is not done in an entirely harmonic way. In contrast, there is not much particular international consensus on the form and effects of a general anti-avoidance rule. Thus, the introduction of the PPT rule can cause compatibility problems concerning domestic law. Therefore, it is imperative to reflect on the effects of including a GAAR on the international scene, as the benefits of this new standard should outweigh any possible harm that may arise from the incompatibility between the numerous global legislations (LANG, 2014, 656).
Initially, BEPS Action 06 highlighted that the PPT rule might not be appropriate for all countries. Indeed, individual states may: (a) have already explicitly disciplined anti- avoidance rules in their internal order which effectively combat abusive structures at the same level or even more rigorously than the PPT rule; or (b) through its Courts, have developed tools to withstand the abuse of Tax Treaties, such as formulating the need for the economic substance or the principle of substance over form. Regardless of the national choice, these legal instruments might make the inclusion of GAAR under Article 7 (1) of the Multilateral Convention unnecessary.
However, it is essential to note that even if an anti-avoidance standard is present in the domestic scenario, it will still be necessary to implement the PPT rule if the internal standard is not adequate or does not act in the same line as this GAAR. In short, flexibility is allowed in adopting the rule expressed in Article 7 (1) of the Multilateral Convention, but only in cases where domestic State law is advanced in combating abusive structures. The minimum that is required of these countries that have already included full anti-avoidance rules in their legal systems is that they change the preamble of their Tax Agreements, including the provision that the rules provided for in the Treaty should not be used in favour of tax evasion or avoidance schemes (OECD, 2015, p. 19).
With these preliminary considerations, the means for preventing compatibility conflicts between the anti-avoidance clauses adopted by the national legal system and the PPT rule should be investigated. In response, the domestic antiabuse rule will not be incompatible with the PPT rule, which is governed by a Tax Agreement, if the Tax Treaty: (i) provides for the overlap of national GAAR/SAAR to the detriment of the PPT rule; (ii) refers to the interpretation of some element of the general international anti-avoidance clause to the
definition provided by the domestic rule; or (iii) allows the domestic rule to deny the benefits expressed in the Treaty if such procedure is in accordance with international principles and/or domestic jurisprudence (CHAND, 2015, p. 486).
Yet, there are scenarios in which the conflict between national GAAR/SAAR and the PPT rule cannot be completely avoided or resolved. In this sense, it is emphasised that there is a possibility that the Supreme Court of the Contracting States may adopt different concepts for the subjective and objective elements of the PPT rule, which would only weaken the application of this anti-avoidance clause (LANG, 2014, pp. 659-660); or even invalidate the application of the PPT rule because they may consider it too broad and, consequently, contradictory with the principle of legal certainty (CUNHA, 2016, pp. 187-189). For example, in 2013, the Constitutional Court of France declared a rule similar to the PPT rule unconstitutional. In that case, the GAAR worked with the concept of “main purpose”, which was considered violating the constitutional text for disrespecting, among others, the principle of legal certainty (FRANCE, 2013, points 112-119):
“Considérant que, compte tenu des conséquences ainsi attachées à la procédure de l’abus de droit fiscal, le législateur ne pouvait, sans méconnaître les exigences constitutionnelles précitées, retenir que seraient constitutifs d’un abus de droit les actes ayant «pour motif principal» d’éluder ou d’atténuer les charges fiscales que l’intéressé aurait dû normalement supporter. Considérant qu’il résulte de ce qui précède que, sans qu’il soit besoin d’examiner les autres griefs, l’article 100 doit être déclaré contraire à la Constitution”5.
Compatibility with SAARs established in the MLI
Another point that deserves attention is the compatibility between the PPT rule – a GAAR
and the Specific Anti-Avoidance Rules – SAARs –, foreseen in BEPS Action 06 and the Multilateral Instrument, as the mutual application of these two types of clauses makes it possible to flourish ambiguities and disparities. In this light, the present study will analyse the compatibility between the LOB and PPT rules, to study the joint application of these two anti-avoidance rule modalities.
The issue of incompatibility arises from the fact that, on the one hand, Article 7 (1) of the Multilateral Instrument begins with the expression: “Notwithstanding the provisions of a Covered Tax Convention”, which means that the PPT rule shall apply irrespective of the other provisions of a Tax Agreement. Consequently, the specific anti-avoidance rules, such as the LOB rule, cannot be raised to exclude the application of the PPT rule – which could
“Considering that, in view of the consequences thus attached to the procedure regarding the abuse of tax law, the legislator could not, without disregarding the constitutional requirements mentioned above, retain that it should be considered abuse of law the acts having “for main purpose” to avoid or mitigate the tax burden that should have normally been incurred. Considering that it follows from the above that, without there being any need to examine the other complaints, Article 100 must be declared unconstitutional” – translated from original by the author.
even imply restricting the scope of SAARs. On the other hand, the same article 7 (1) provides that the application of the PPT rule should be “[…] in accordance with the object and purpose of the relevant provisions of the covered Tax Convention”, so that it must be understood that the application of specific rules in their particular cases appears within the scope of the Tax Treaties, so that the PPT rule cannot, therefore, invade the scope of SAARs, in compliance with the principle of speciality (LANG, 2014, p. 658).
In this trail of thought, it is emphasised that the combination of the LOB and PPT rules should be done in a complementary way, in the sense that one norm supersedes the weakness of the other (ELLIFFE, 2019, p. 7). The LOB rule is based on an objective criterion, providing legal certainty, and applies to clear cases so that it will not act beyond its scope. Thus, treaty shopping situations that the LOB rule cannot repel, such as those using conduit societies, as an example, should be tackled with the use of the PPT rule, which, while finding safe-haven on a more flexible criteria, not having, by comparison, such legal predictability, it will generally apply to all cases (OECD, 2015, pp. 18-20; COURINHA, 2004, p. 110). In summary, cases not explicitly covered by SAARs will be covered by the PPT rule.
One issue that needs to be addressed is the subsidiarity of the PPT rule compared to the LOB clause. In other words, the question is: if the LOB clause applies in this case, but proves insufficient to prevent abusive planning, will the PPT rule apply? On the one hand, it should be argued that the PPT rule should be rejected in respect of systematic integrity (COURINHA, 2004, p. 107) and the lex specialis derogat legi generali principle. On the other hand, it can be inferred that the PPT rule is a rule of autonomous application to the other standards, as stated in the first line of article 7 (1) of the MLI so that it applies to the specific case regardless of the particular rule. What is certain is that Action 06 of the BEPS project has already offered the answer to the question, arguing that the benefits accepted by SAARs might be later denied by the general rule (OECD, 2015, p. 55). In summary, situations covered by SAARs should transpose two antiabuse rules: the specific one first and, in the alternative, the PPT rule (CHAND, 2015, p. 486).
CONCLUSION
The PPT rule, due to its formulation, raises several problems, both conceptual and practical, which were developed throughout the study. It is noteworthy that the PPT rule, in the current contours, despite being the new trend of International Tax Law, arising from recent global initiatives to combat the abuse of conventions, presents clear issues in its conceptualisation, especially regarding its subjective element, which is characterised by being harsh, generic, susceptible to arbitrariness and even irrelevant, being overshadowed by the objective aspect of the PPT rule. This paper recommends the return to “the essential aim” criterion, already employed by international jurisprudence, to remedy the issues present in the subjective test, as well as the need to compile clear evidence of the existence of abuse, based on Comment No. 80 to Article 1 of the 2017 OECD MC.
After all, an anti-avoidance rule must prevent the abuse of rights by taxpayers, but not allow, for this instance, the abuse by the State. Besides the conceptual issues, the PPT rule also presents quandaries when practically applied, such as disputing conflicts of tax allocation rights between States, which may only be resolved through mutual agreement. Finally, the PPT rule will not eliminate tax planning but instead will require them to increase their levels of complexity to fit new antiabuse trends. In a way, the current context adds more importance to future tax planning, which will need to be in line with the PPT rule.
BIBLIOGRAPHY