By Luis Schoueri (University of Sao Paulo; Lacaz Martins, Pereira Neto, Gurevich & Schoueri Advogados)
Introduction
There is no divine truth about what the Arm’s Length Standard (ALS) actually means. Its content can only be determined by a decision, which can be reached by a court or by means of political consensus. There is no international tax court with jurisdiction to promote harmonization among countries on the content of the ALS and all efforts in this direction are made by means of negotiation. Such decisions affect not only the extent to which double (non-)taxation will be avoided, but also concern the country to which income is allocated, which may render the issue controversial where countries present distinct patterns of capital in- and outflow[1].
The OECD has been responsible for consolidating some degree of consensus around the content of the ALS in the last decades. More than merely clarifying the meaning of an expression, it has adopted an evolutionary approach towards the ALS, often leading academics to perceive the outcomes as going way beyond the original intent of the standard[2]. It has promoted flexibility where comparables are not available and offered the theorization necessary to fill the gaps where the application of the standard was not immediate.
Parallelly, Brazil has gone its own way, designing a system which is much simpler than the OECD counterpart – for the good and for the bad. This simplicity seeks to strike a balance between preventing double taxation and ensuring ease of administration, considering the Brazilian institutional capacity – e.g. the available personnel vis-à-vis the size of our economy. The abrupt abandonment of the current system, suggested by a Joint Report of the Brazilian tax authorities (RFB) and the OECD[3] (hereinafter, the “Joint Report”) is expected to significantly increase uncertainty, punishing taxpayers in case a controversy arises, by means of practices that are not as progressist as one would expect from the intended reform.
Shifting gears in the Brazilian transfer pricing policy
In the 1995 Guidelines, following US legislative reforms[4], the OECD opened the path towards profit methods, which would afterwards become prevalent in transfer pricing practice. Indirect methods were incorporated into the Guidelines as measures of last resort, after years of discussions on whether they could be considered ALS-based in the first place. The Guidelines also provided important theorization on the remuneration of assets, risks and functions, which would become central to the evolution of the standard in subsequent years.
In 1996, Brazil enacted its first TP legislation, whose essential features have remained the same ever since[5], in spite of relevant improvements which have clearly reduced tax litigation initially observed in the country. The Brazilian transfer pricing methods are all inspired on the traditional transactional methods, but they adopt sectorial fixed margins instead of demanding the identification of comparables, and functional analysis is present only to a very limited extent[6]. Commodities are subject to a control based on publicly traded prices. No profit method is possible and the issue of intangibles is dealt with by means of significant restrictions to royalty deductions[7] – which have put tax authorities in the very comfortable position of not having to provide answers to the most complicated TP issues, but also have rendered the system particularly obsolete in the long run.
Despite the significant differences with the OECD Guidelines, it is a huge misconception to regard the Brazilian approach as a methodology of formulary apportionment. The Brazilian legislation does not take global profits into account: all existing methods are one-sided. The sectorial margins are intended to imitate the behavior of independent parties and, at the same time, grant certainty and practicability both to taxpayers and tax authorities. It is even inspired by the 1995 Guidelines, in the sense that it adopts simplified versions of the traditional methods. As far as Brazilian tax authorities and Brazilian treaty partners are concerned, the Brazilian system has been deemed as an ALS-based one during the last twenty years for the purpose of applying Article 9 of the signed treaties – which do not include an equivalent to Art. 9(2) of the OECD-MC.
In 2010, with widespread application of the TNMM, the Guidelines followed practice and acknowledged that profit methods were no longer methods of last resort[8]. Indirect methods progressively left the periphery and finally reached the center of the TP system, with further detail on its application. A year later, the UN Committee of Experts on International Cooperation in Tax Matters published the “Practical Manual on Transfer Pricing for Developing Countries”[9]. The Brazilian approach is described therein as an ALS-based approach. The document is much more aimed at presenting miscellaneous alternatives on the ALS than properly at reaching a more refined consensus on the topic. Instead of aiming at convergence, the Practical Manual gives space for countries to present divergent practices. The Practical Manual is not intended to pose itself as an antagonist to the OECD Guidelines. Being included therein has not attenuated the Brazilian isolation, as no country seem to have adopted something similar to the fixed margins.
In 2015, the BEPS Project brought the ultimate relativization of the need for comparables, enthroning the allocation of profits pursuant to “value creation” as the theoretical goal to be pursued by transfer pricing rules. Value chain analysis is now central and justifying individual transactions does not seem to cut it anymore. The allocation of residual profits under profit split methods is perhaps the most controversial outcome of the new ALS consensus. Even synergy rents, which were deemed to be the Achilles’ heel of the ALS, seem to have found their redemption in the value creation mantra. By means of abstractions on the behavior of independent parties, it is now possible to allocate synergy rents as independent parties would have, even though independent parties would never derive synergy rents in the first place. As part of the G20, Brazil had the first opportunity to present its dissonant perspective on transfer pricing at a negotiation table. Apparently, however, it only managed to remain isolated on transfer pricing issues, being the only country to which a footnote has been dedicated in the Actions 8-10 Final Report[10].
In relation to the digital economy, the OECD decided that the ALS can no longer be tweaked[11] and that the current allocation of taxing rights is not fair. Under Pillar One, new alternatives to the ALS are being discussed and it is acknowledged that, this time, some amendments to the Model Convention will be necessary, in order to make the application of the new policy possible. Within the Inclusive Framework, Brazil again could play a role, considering how central the current negotiations are to the future of the international tax regime. Brazil has not showed any particular initiative on the topic, even though the issue is one aimed at ensuring more source taxation along with simplicity concerns – which should be central for the Brazilian interests. India has taken the lead among emerging economies, but a strong position from Brazil has not followed.
In practice, Brazil has remained a complete stranger to the evolutionary process of the OECD Guidelines and is currently responsible for the most significant deviation from the OECD transfer pricing methodology, with no significant influence around the globe.
As the Brazilian accession to the OECD gains momentum, these separate ways are required to converge. Brazil is expected to adhere to the OECD consensus on the ALS. Since Brazil is the one requesting accession and not the other way around, convergence has taken the form of elimination of the Brazilian peculiarities. The Joint Report is an expression of such approach: it identifies the gaps between Brazil and the Guidelines and elaborates on how Brazil should adapt to the Guidelines. Both the traditional isolation and the recent tentative approximation strategies do not seem to have provided Brazil with any leverage to support its positions. Despite earlier criticism of Brazilian specialists[12], no partial alignment with the Guidelines is envisaged by the Joint Report and full alignment is presented as a deal-breaker for the Brazilian accession.
Can we converge to the full first world package?
Simply put, Brazil is expected to incorporate twenty years of evolutionary transfer pricing practice at once. Whilst the complete lack of precedents on the application of profit methodologies, and even on more basic aspects of functional analysis for the application of the traditional methods, full-alignment is demanded[13], also on a “gradual” version, based on the size of the companies or on the type of transaction[14]. If taken forward, this sort of alignment will surely be followed by a lot of uncertainty, considering the theoretical and institutional adaptations that will be necessary. When dealing with such uncertainty, the Joint Report has already set the tone and made clear that, despite the alleged modernization, some practices will remain medieval.
An example thereof is on the issue of penalties. The Joint Report states the difficulty of evaluating in abstract while a monetary penalty is excessive, making a general statement on the need for proportionate measures[15]. It further specifies, however, that “the imposition of sizable ‘no-fault’ penalty based on the mere existence of an understatement of a certain amount would be unduly harsh when it is attributable to good faith error rather than negligence or an actual intent to avoid tax”[16]. It also affirms that “it would be unfair to impose sizable penalties on taxpayers that made a reasonable effort in good faith to set the terms of their transactions with associated enterprises in a manner consistent with arm’s length principle”[17].
The Report then identifies the relevant Brazilian framework on the topic, according to which “as a general rule, the penalty for underpayment of federal taxes is 75% [of the amount due]”[18]. This is precisely the “no-fault” sort of penalty criticized in the preceding paragraphs of the Joint Report: a mere divergence of interpretation leads to the 75%-penalty. In case of fraud or sham, the penalty is increased to 150%.
In relation to the 75%-penalty, the Joint Report dedicates two conflicting statements. The first affirms that “the Brazilian framework does not necessarily deviate from the OECD Guidelines, since it is recognised therein that it is difficult to assess whether a particular penalty is fair or not”[19]. The second, considers that “the 75% penalty that is automatically applicable to a tax underpayment, irrespective of the reason, may be considered unduly harsh in some situations (e.g., good faith)”[20]. The conflict is further blurred by the following excerpt[21]:
This potential harshness may however be mitigated because the penalties resulting from an assessment by the tax authorities may be decreased by half if the taxpayer voluntarily pays the tax due, which also means that he gives up any administrative remedies.
The reasoning is therefore that a no-fault 75%-penalty is not “unduly harsh”, when it is reduced to 37,5%, if the taxpayer just agrees to pay the tax allegedly due and give up the relevant administrative remedies. This logic is not exactly sound, but leads one to believe that the no-fault 75%-penalty is acceptable under OECD standards. In practical terms, however, the unduly harsh penalty may reduce to ashes the efforts on mitigating double taxation. If maintained after the intended reform, there is no doubt that the 75%-penalty will remain being applied by tax authorities to any and every interpretative divergence – as it currently is. From a good faith taxpayer’s perspective there is no benefit in completely avoiding double taxation in a given transaction, but paying a 75%-fine (or 37,5%) to one of the states.
At the end of the day, the Brazilian divergence with the OECD is not one based on the fair allocation of taxing rights, which is a discussion to which Brazil has remained completely alienated and never achieved the necessary refinement to offer any opposition. The main problem is that the OECD Guidelines methodology is much more complicated than the current Brazilian system, which has also been drafted to be an ALS-based one, even though in a (much) rougher version. If there is no bargaining power to meet the OECD halfway, Brazil will end up with a complicated legislation with flavors of the current disproportionate administrative measures and practices – of which the 75%-penalty is only an example. The full first world package cannot be simply enacted as legislation and demands years of institutional development, which an abrupt reform will not be able to skip.
[1] See L. E. Schoueri & R. A. Galendi Júnior, ‘Justification and Implementation of the International Allocation of Taxing Rights: can we take one thing at a time?’, In: A. Christians; S. A. Rocha (ed.). Tax Sovereignty in the BEPS Era, Alphen aan den Rijn: Wolters Kluwer, 2017, pp. 47-72.
[2] See, for a discussion on the evolution of the standard, L. E. Schoueri, Arm’s Length: Beyond the Guidelines of the OECD, BIT, 69(12), 2015, pp. 690-726.
[3] OECD/Receita Federal do Brasil (2019), Transfer Pricing in Brazil: Towards Convergence with the OECD Standard, OECD, Paris, www.oecd.org/tax/transfer-pricing/transfer-pricing-in-brazil-towards-convergence-with-the-oecd-standard.htm.
[4] See, on the evolution of the methods in the U.S. tax system, R. Avi-Yonah, The Rise and Fall of Arm’s Length: A Study in the Evolution of U.S. International Taxation, Public and Legal Theory Working Paper Series, Working Paper No. 92 (September 2007).
[5] For an extensive analysis of the Brazilian legislation, see L. E. Schoueri, Preços de Transferência no, Direito Tributário Brasileiro (3rd ed., São Paulo, Dialética, 2013), 479p.
[6] See, on the Brazilian TP system, L. E. Schoueri; R. A. Galendi Júnior. Brazil. Cahiers de Droit Fiscal International – The future of transfer pricing, v. 102B, 2017, pp. 191-215.
[7] See L. E. Schoueri & R. A. Galendi Júnior, Challenges to Brazilian Transfer Pricing Rules upon Accession to the OECD, ITPJ, 26, 2019, pp. 433-441.
[8] See OECD, OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, July 22, 2010 (OECD Publishing, 2010), para. 2.2.
[9] United Nations, Practical Manual on Transfer Pricing for Developing Countries, (United Nations, New York, 2011). A second edition was published in 2017.
[10] The authors have commented the footnote on a previous article to Kluwer International Tax Blog. See L.E. Schoueri and R. A. Galendi Júnior, ‘The Brazilian Mysterious position on Actions 8-10: a blank check for cherry picking?’, Kluwer International Tax Blog, October 25 2016, http://kluwertaxblog.com/2016/10/25/the-brazilian-mysterious-position-on-actions-8-10-a-blank-check-for-cherry-picking/.
[11] For criticism on how the ALS has been “tweaked” in the last decades, see Yariv Brauner, BEPS: an interim evaluation, WTJ, 2014, p. 28.
[12] The first author has signed a public position on the topic, along with other Brazilian professors, which has been published by the Kluwer International Tax Blog. See L. E. Schoueri, Brazilian TP: Missed Opportunities Ahead, Kluwer International Tax Blog, July 30, 2019, http://kluwertaxblog.com/2019/07/30/brazilian-tp-missed-opportunities-ahead/.
[13] The authors have presented an alternative to the full-alignment. See L. E. Schoueri & R. A. Galendi Júnior, Challenges to Brazilian Transfer Pricing Rules upon Accession to the OECD, ITPJ, 26, 2019, pp. 433-441.
[14] OECD/RFB, Transfer Pricing in Brazil, para. 1054-1064.
[15] OECD/Receita Federal do Brasil (2019), Transfer Pricing in Brazil: Towards Convergence with the OECD Standard, OECD, Pariswww.oecd.org/tax/transfer-pricing/transfer-pricing-in-brazil-towards-convergence-with-the-oecd-standard.htm, para. 316.
[16] OECD/RFB, Transfer Pricing in Brazil, para. 318.
[17] OECD/RFB, Transfer Pricing in Brazil, para. 318.
[18] OECD/RFB, Transfer Pricing in Brazil, para. 363.
[19] OECD/RFB, Transfer Pricing in Brazil, para. 364.
[20] OECD/RFB, Transfer Pricing in Brazil, para. 365.
[21] OECD/RFB, Transfer Pricing in Brazil, para. 365.
The article was first published here.
