Predetermined Margins in the Brazilian Transfer Pricing Rules and their Compatibility (or not) with
the World Trade System

As Margens Predeterminadas nas Regras Brasileiras de Preços de Transferência e sua Compatibilidade (ou não) com as Regras do Sistema Mundial do Comércio

Eric Moraes de Castro e Silva

Advogado. Professor de Direito Financeiro e Tributário da Universidade Federal de
Pernambuco (UFPE). Doutor em Direito Financeiro e Tributário pela USP.
Mestre em Direito pela Universidade de Cambridge. E-mail: eric@castroesilva.adv.br.

Abstract

This article compares the Brazilian transfer pricing rules, which are based on fixed predetermined profit margins instead of the arm’s length principle, with the multilateral trading obligations, particularly the rules on subsidies under the Subsidies and Countervailing Measure Agreement (SCM Agreement) and the national treatment principle under the General Agreement on Tariffs and Trade (GATT) as evolved through the relevant jurisprudence of the World Trade Organization.

Keywords: transfer pricing, predetermined margins, national treatment, subsidies, WTO.

Resumo

O presente artigo compara as regras brasileiras de preços de transferência, que são baseadas em margens predeterminadas de lucro em detrimento do princípio arm’s length, com as obrigações do sistema multilateral do comércio, particularmente as regras de subsídios previstas no Acordo sobre Subsídios e Medidas Compensatórias (SCM Agreement) e o princípio do tratamento nacional previsto no Acordo Geral sobre Tarifas e Comércio (GATT), de acordo com a evolução jurisprudencial da Organização Mundial do Comércio.

Palavras-chave: preços de transferência, margens predeterminadas, tratamento nacional, subsídios, OMC.

1. Introduction

The objective of this study is to compare Brazil’s transfer pricing rules that do not follow the arm’s length principle, as advocated by the Organisation for Economic Co-operation and Development (OECD), with the rules of the multilateral trading system, now under the auspices of the World Trade Organization (WTO), which impact the taxation of income derived from international trade between related companies.

Brazil is one of the few countries that do not adopt the arm’s length principle in their transfer pricing rules, replacing the search for economic reality inherent to this principle with a fictitious profit arbitrarily fixed by law.

The Brazilian methods, criticized by some for not following the world standard but applauded by others for being more practical for calculating the income tax, are also relevant to international trade because a taxation that does not adhere to the economic reality may be detrimental to the fair competition pursued by the WTO system.

Therefore, this study intends to examine whether the Brazilian legislature’s income tax option is compatible with Brazil’s obligations before the treaties of the WTO, particularly the Agreement on Subsidies and Countervailing Measures (SCM Agreement) and the national treatment rule provided for in art. III of the General Agreement on Tariffs and Trade (GATT).

The SCM Agreement subsidies rules prohibit taxation from being a disincentive factor for the import and/or export stimulus, classifying any incentive or tax barrier to achieve these commercial purposes as a prohibited subsidy.

In this regard, it is important to note that footnote 59 of Annex I of the SCM Agreement provides for the compulsory adoption of the arm’s length principle for purposes of income allocation in international trade transactions between related companies.

This footnote and the precedents of the WTO in tax legislation cases, engaged between the United States and Europe, guide the analysis of compatibility between the Brazilian transfer pricing rules that do not follow the arm’s length principle and the rules of the multilateral trading system for tax subsidies.

The same focus on the WTO jurisprudence is used to compare the Brazilian transfer pricing rules with the national treatment principle, which prohibits local taxation applied on imported products from being higher than that applied on similar foreign products, to the clear detriment of the competitiveness of the foreign product in the domestic market.

In short, the aim of this study is to examine Brazil’s transfer pricing rules through the joint perspective of the international trading system and International Tax Law, fields of knowledge that are typically treated as separate, despite their practical and regulatory overlaps.

2. Brazilian transfer pricing rules

Transfer pricing is understood as the sale or transfer of goods, services, or intangible property between related companies located in different jurisdictions1.

Because the companies are related, transactions between them occur outside the market, and therefore, it is possible that the prices charged do not correspond to the economic reality, resulting in a possible erosion to the tax base of one of the involved jurisdictions because the variation in the tax burden of each country may encourage companies to allocate revenues and expenses in a more favorable jurisdictions under the fiscal view2.

Thus, traditionally, legislation on transfer pricing aims to inhibit the overpricing of imports (“losses import”) and the underpricing of exports (“earnings export”) to avoid the artificial transfer of profits abroad and the corresponding reduction in local taxation via the income tax3.

The artificial transfer of profits, known in international doctrine as income shifting4, in addition to eroding the tax base of the countries involved, can have a direct impact on the international trade of goods by affecting the price of the product involved in the transaction between related companies5.

By acknowledging the impact that transfer pricing can have on the international trade of goods, it is possible that some countries assume a position that is diametrically opposed to the traditional approach of inhibiting income shifting, using the transfer pricing legislation as a means to attract productive investments to its jurisdiction.

This position is what Avi-Yonah calls production tax havens, countries that impose low taxation on profit from manufacturing transactions performed by multinational companies located in their territory, and transfer pricing rules can be a sophisticated tool for this objective6.

Hence, there is the need for transfer pricing rules to seek competitive neutrality, eliminating the possibility of income shifting, which can erode the tax base of the countries either via the international tax planning of taxpayers, which the international doctrine calls tax arbitrage7, or via the unfair tax competition practiced by production tax havens.

The criterion that is universally accepted to achieve neutrality in transfer pricing is the arm’s length principle, which, by comparing the transactions between related companies with similar transaction conducted by independent companies, seeks the effective allocation of economic income in the respective jurisdiction8.

There are few countries that do not adopt the arm’s length standard in their legislation on transfer pricing, including Brazil, which replaces the comparison inherent to the arm’s length principle with a predetermined profit margins established by law9.

In practice, the Brazilian predetermined margins require the establishment of a ceiling for deductible import expenses and a minimum gross export revenue in transactions between related companies, which distances Brazilian law from the arm’s length principle and, according to some, brings it close to the formulary apportionment method10, a comparison not accepted by this study11.

Therefore, rather than comparing the prices charged between related companies with the prices charged by independent entities (arm’s length), Brazilian legislation compares the intergroup price with a price set by the methods specified in the law, the so-called parameter price12.

Note that Brazil has not completely distanced itself from international standards spearheaded by the OECD. What Brazil has done is implement methods that are formally equivalent to those espoused by this organization, in this case, the resale price and the cost plus; however, unlike what occurs in most countries, the object of the comparison does not reflect the economic reality (arm’s length) but instead a fixed price established by law (parameter price)13.

Therefore, there will always be a minimum profit margin for the Brazilian company, making its artificial transfer to related entities abroad difficult, which would result in the preservation of the domestic tax base, a goal whose satisfaction is questionable14.

For the preservation of the Brazilian tax base, in the hypothesis that the price charged by the related parties is (i) higher than the parameter price in imports or (ii) lower than the parameter price in exports, the difference (adjustment) should be added to the net profit for determining the real profit and the basis of calculation of the Corporate Income Tax (Imposto sobre a Renda das Pessoas Jurídicas – IRPJ) and the Social Contribution on Net Profit (Contribuição Social sobre o Lucro Líquido – CSLL)15.

The predetermined margins for transfer pricing show the Brazilian tax legislator’s preference for seeking a tax based on the taxpayer’s revenue/billing instead of pursuing the effective (and elusive) economic concept of income16.

In the view of this study, what the Brazilian legislator does, for transfer pricing purposes, is set a legal concept of income, divorced from economic reality, to exchanges between related companies, embodied in the parameter price, which is of questionable constitutionality17.

Despite being a strong critic of the adoption of a legal concept of income, to the detriment of the respective economic concept, Schoueri understands that the adoption of predetermined margins by Brazil is justified by practicality, as opposed to the difficult and often innocuous methods that seek to materialize the arm’s length principle18, and serves as a useful tool for inducing the taxpayers’ behavior19.

In the view of this study, economic induction is not a goal of the Brazilian legislation on transfer pricing20 but instead a residual effect of the local income tax for setting a legal concept of income (parameter price) at the expense of the economic concept of income (arm’s length).

Consequently, in most cases, the Brazilian tax burden will be higher or lower than that which would be due based on the economic reality determined by the arm’s length principle and hence, the inducing residual effect of the Brazilian income tax.

This occurs because when the taxation on the export income, based on the predetermined margins, is below what would be the due based on the arm’s length principle, there will be an incentive for export activity. By contrast, when the income tax in import exceeds what would be due based on the economic reality, the purchase of products abroad will be discouraged21.

In this sense, the United Nations (UN) Practical Manual, which praises the Brazilian predetermined margins given the relatively low cost of compliance, does not fail to note as one of the weaknesses the fact that companies located in Brazil are likely to be taxed in disagreement with the effective profitability, in addition to the possibility that the predetermined margins will lead to the occurrence of double international taxation22.

To not distort the economic reality and hence honor tax competitive neutrality, the UN Manual suggests that countries interested in adopting predetermined margins should seek the greatest possible number of margins, which should be established on the basis of market surveys that bring predetermined margins close to the actual profit of the respective economic sector. It also suggests that instead of a static margin, countries adopt variable tax ranges, which also facilitates the approach to the reality of the market, honoring neutrality23.

The Brazilian legislation itself indirectly recognizes the distorting effect of predetermined margins, given that it allows the margins established by the legislature to be modified by the Finance Minister or at the request of the taxpayer, in accordance with art. 20 of Law no. 9,430/1996.

Art. 20 requires “justified circumstances” on the part of the Finance Minister to change the margins established in the law. In Schoueri’s view, such circumstances could be twofold: the fact that “[...] a comparative analysis shows that the percentages are no longer useful for determining an arm’s length price or another (inducing) principle extracted from the legal price”24.

Thus, it is important to examine whether the inductive effect inherent to the fixed margins adopted by Brazil is in accordance with the principles and rules of the multilateral trading system, now under the auspices of the WTO.

3. Predetermined margins and the SCM Agreement

The predetermined margins for parameter pricing setting purposes apply both to import and export and are used in Brazilian transfer pricing methods based on the acquisition/production cost of products (cost plus method) or the resale price of goods (resale price method).

According to the current wording of Law no. 9,430/1996, the methods that adopt fixed profit margins based on cost are the Production Cost Plus Profit (CPL)25 in the case of imports, whose profit margin is 20%, and the Acquisition or Production Cost Plus Taxes and Profit (CAP)26 for export operations, whose fixed margin is 15%.

For methods based on the cost, the parameter price is set by adding a predetermined cost to the price of the operation (cost plus markup), which will be equivalent to a maximum value in the case of imports or a minimum value in the case of exports27.

The methods that adopt a fixed income margin based on the product’s resale price are the Resale Price less Profit (PRL), in the case of imports, whose general profit margin is 20%28, and, for export operations, the methods of Wholesale Price in the Destination Country Less Profit (PVA) method29, with a fixed margin of 15%, and the Retail Price in the Destination Country Less Profit (PVV) method30, with a margin of 30%.

For methods based on the resale price, the final price charged by the related company to the end consumer is reduced by a percentage that reflects a profit margin fixed in the law in the percentages indicated above.

For practical reasons, methods based on the resale price are best suited for import operations because there are few companies that are willing to provide overseas production costs, even for related companies. Thus, neither the related company domiciled in Brazil nor the local tax authorities have access to the information to effectively calculate the parameter price established in the law31.

For the same reasons, methods based on the production or acquisition cost are more feasible for use in export operations. In such hypotheses, as the UN Practical Manual states,

“[…] the Brazilian manufacturing exporter uses its own account book costs to calculate the correct transfer price, with no need to request any data from the non-Brazilian affiliate. Furthermore, in the case of exports, all necessary information can be accessed and verified by the Brazilian tax administration”32.

The question that arises for the purposes of this study is whether the lack of symmetry between the economic reality of import/export operations and the amount of income tax due based on the predetermined margins discussed above constitutes a subsidy prohibited through taxation, as provided in the SCM Agreement.

At first, the difference in the transfer price paid by Brazilian taxpayers and the price that would be due in the event of an enforcement of the arm’s length principle, which, in the view of this study, is the unintentional inducing effect of Brazilian legislation, could constitute a tax subsidy that is prohibited by the multilateral trading system.

The reason for this is that in an import operation, if the transfer price determined by the PRL method, for example, was higher than the transfer price determined using the arm’s length principle, then Brazil would be providing a subsidy linked to the substitution of imported products by national products, which is a prohibited subsidy under art. 3, § 1, “b” of the SCM Agreement.

In this sense, Schoueri exemplifies the induction to the substitution of imported products manufactured in Brazil arising from a fixed margin adopted by the PRL method to the detriment of arm’s length methods because given that the profit margin of the related importer is always fixed, “[...] it is interesting to the taxpayer that all industrialization occurs in the country, as it would bring, in addition, other functions such as guarantee, advertising, etc.”33

Conversely, if in the export of a related company domiciled in Brazil, the transfer price supported by it is lower than the price determined using the arm’s length principle, then this difference could be perceived as an export subsidy classified as prohibited by art. 3, § 1, “a” of the SCM Agreement.

In the specific case of export, it should be remembered that the United States, in the Domestic International Sales Corporations (DISC) and Foreign Sales Corporations (FSC) legislation, has introduced formulary apportionment methods of transfer pricing34 based on fixed percentages, which were promptly questioned by Europe as export subsidies, resulting in disputes that had a significant impact on the taxation of the income of the countries involved and on international commercial regulations, with such disputes (the taxes legislation cases) even resulting in footnote 59 of the SCM Agreement35.

Footnote 59 SCM Agreement expressly states the following:

“[…] the Members reaffirm the principle that prices for goods in transactions between exporting enterprises and foreign buyers under their or under the same control should, for tax purposes, be the prices that would be charged between independent enterprises acting at arm’s length”.

Despite defending the residual inducing effect of Brazilian predetermined margins noted above, justifying it based on the principles of the constitutional economic order, Schoueri notes that such induction may be challenged in the WTO by Brazilian business partners exactly based on the above transcribed excerpt of footnote 59 of the SCM Agreement36.

For this work and according to the WTO’s jurisprudence, it is initially important to ascertain whether the tax measure in question falls under the general definition of subsidy provided in art. 1, § 1 of the SCM Agreement, i.e., whether it is a financial contribution granted by the government consisting of the foregoing of government revenue that would normally be due (“if government revenue that is otherwise due is foregone or not collected”).

For this analysis, it is worth remembering Brauner’s caveat that the mission of examining a subsidy by means of transfer pricing is arduous because it is very difficult to identify and measure the income derived from export37.

However, this difficulty does not prevent the application in the present hypothesis of the tests developed by the WTO in the US – FSC case to assess the existence of subsidy, namely, the “but for” test and the comparability test. In the “but for” test, first, the general rule of taxation is analyzed, which, because of the tax measure (but for), fails to materialize, representing foregone of revenue “otherwise due”38.

In the comparability test, the tax measure in question is compared to other equivalent tax measures of the same order imposed on a similar materiality. If the comparison results in an incentive characterized by the foregoing of government revenue, then there will be subsidy through taxation39.

The assumption of both tests developed by WTO jurisprudence is the identification of parameter taxation, the normative benchmark, against which the contested tax measure is excepted (“but for” test) or compared (comparability test)40.

In the Brazilian case, the normative benchmark is the predetermined margins. However, there is no Brazilian tax measure that excepts it or with which it can be compared; thus, there is no foregoing of government revenue that would otherwise be due resulting from this syllogism.

For this reason, it is understood that it is not possible to state that the comparison of the taxation based on predetermined margins to that which would be due based on arm’s length is an exception or valid comparison to the Brazilian standard taxation for identifying a subsidy. The reason for this is because the arm’s length principle was not adopted by the Brazilian transfer pricing methods based on the cost or resale of goods.

This feature of Brazilian law, i.e., the non-adoption of the arm’s length principle together with methods based on predetermined margins, for this work, eliminates the configuration of a subsidy. In this regard, the Brazilian situation differs from that which occurred with the transfer pricing rules introduced by the US in the DISC and FSC legislation.

In these American cases, Europe made an accusation of the existence of subsidy precisely as a result of the (lower) difference in taxation provided for exporters, which could choose the formulary apportionment methods of the DISC and FSC legislation, in comparison to the general standard in force in the U.S., which was the arm’s length41.

Such a situation does not occur in the Brazilian case because there is no general arm’s length standard. The Brazilian legislation opted to tax the transfer prices based on an artificial concept of income (the parameter price), denying the economic reality gaugeable by the arm’s length principle. This option is valid in the view of the multilateral system because it is applicable without distinction to all taxpayers subjected to the Brazilian methods based on predetermined margins, even because, as the WTO’s jurisprudence also reiterates, governments can theoretically tax or not tax any income42.

Consequently, for this study, the residual inducing effect of the predetermined margins of the Brazilian law on transfer pricing is not characterized as a subsidy.

For the same reasons, this study considers that the predetermined margins do not violate footnote 59 of the SCM Agreement, which requires compliance with the arm’s length principle in the transfer of products between related companies.

The reason for this is because footnote 59 is an exception to the general rule of prohibited export subsidies. In other words, WTO members may grant tax subsidies to export, provided that such measures are meant to avoid double taxation43.

In this regard, one cannot forget that footnote 59 was introduced into the SCM Agreement to make the exemption method adopted by the territorial systems to avoid double taxation, which are based on the economic theory of capital import neutrality (CIN)44, compatible with the general rules of subsidies that prohibit the exoneration of direct taxes on exports.

Because the Brazilian predetermined margins do not even fit the general concept of subsidies because they are not a foregoing of government revenue otherwise due, there is no need to discuss violation of the special rule provided in footnote 59, which excepts the ban on subsidies.

In this sense, emphasis is placed on the position taken by the Panel of the US – FSC case, which rejected the argument that footnote 59 constitutes an interpretive criterion for the definition of subsidy itself. For the Panel, it was initially important to ascertain that there is a subsidy as provided in art. 1 of the SCM Agreement. In the terms of this decision, “Footnote 59, on the other hand, relates to the Illustrative List of Export Subsidies and is obviously thus of greater relevance to determining when a measure is an export subsidy than to determining whether it is a subsidy as such”45.

Finally, it is interesting to highlight that a feature of the Brazilian legislation on transfer pricing may fit the concept of prohibited subsidy provided in the SCM Agreement. This feature is art. 20 of Law no. 9,430/1996.

As observed in the previous section, this device precisely allows that the act of the Finance Minister, ex officio or following an application by the interested party, can change the percentages of predetermined margins in justified circumstances.

§ 1 of art. 45 of the normative ruling of the Brazilian Internal Revenue Service no. 1,312/2012, which regulates changes in the percentages of the predetermined margins of Law no. 9,430/1996, provides that changes can be made for a specific taxpayer or for a specific economic sector.

If there is a concretization of the possibility of change, then there will be a tax measure that deviates from the general pattern of taxation (normative benchmark) of transfer pricing in Brazil.

In such a situation, it is understood that it will be possible to apply the syllogistic of the comparability test and the “but for” test because there will be a general rule of taxation in the law – the predetermined margins contained in Law no. 9,430/1996 – and a special rule that deviates from it – the Ordinance issued by the Finance Minister – from which the subsidy contrary to SCM Agreement rules may result.

In such a circumstance, there is the paradox that the very legal device that the local doctrine considers the “valve” for aligning the Brazilian transfer pricing rules to international standards, and even for ensuring their constitutionality46, could be viewed as conflicting with the principles and rules of the multilateral trading system.

4. Predetermined margins and the national treatment principle

Art. III of the GATT established the principle of national treatment, one of the few provisions of the treaties that comprise the multilateral trading system that expressly limits the taxation powers of the members, forbidding them to collect taxes or other internal charges that could be used to benefit domestic production to the detriment of foreign production47.

Today, it is already common sense in WTO jurisprudence that the national treatment rule also applies to direct taxes, such as income tax48; hence, the transfer pricing rules should respect this principle of the multilateral system.

In the case of Brazil, Gustavo Vettori notes that a possible incompatibility of the transfer pricing rules with the national treatment principle can occur because only imported products are subject to the adjustments required by Law no. 9,430/1996. In contrast, Vettori notes the following:

“[...] such adjustments do not apply in the case of the acquisition of goods in the Brazilian market. Thus, it may be argued that there would be an advantage (i.e., the non-adjustment of costs by transfer pricing rules) granted to Brazilian companies that acquire products in the Brazilian market, as opposed to the purchase of imported products”49.

However, Vettori believes that the Brazilian predetermined margins do not violate the national treatment principle because transfer pricing rules aim to seek the effective market reality by applying the arm’s length principle, which ultimately honors the principle of equality.

In Vettori’s words: “[...] transfer pricing rules lend themselves to adjust, for purposes of determining the basis for the calculation of taxes on income, the prices charged between related parties so that they come to reflect market prices. This is the application of the arm’s length principle”50.

This study disagrees with this positioning in relation to Brazilian predetermined margins, understanding that the adjustments that result from the application of gross profit margins on imports practiced by related companies domiciled in Brazil violate the principle of national treatment.

It is accepted, however, that the application of the arm’s length principle in the transfer pricing rules is the assumption of its compatibility with the principle of national treatment. However, this is not what occurs when Brazil determines the use of predetermined margins, which are established out of tune with the reality of the market.

Brazilian predetermined margins are set outside the arm’s length standard because they do not seek economic reality but only a minimal taxation in Brazil, based on the parameter price, and therefore do not reach the competitive neutrality advocated by the multilateral system51.

Thus, in practice, an imported product subject to the PRL method, for example, is subjected to an adjustment in the calculation of the income tax to mirror a fictitious profit margin, which a similar product produced in Brazil is not necessarily subject to, and therefore, it enjoys a competitive advantage in the domestic market.

If there was an application of the arm’s length principle, the exact opposite would occur: the imported product could be subject to adjustments in the calculation of the income tax to mirror the market reality, the same reality to which the locally manufactured product is subject.

It is from this difference in taxation between the domestic product, which is subject to the income tax based on the economic reality, and the similar imported product, which is subject to adjustments that mirror the artificial profit margin, that the inducing effect results, which, as observed in the previous sections, can be defended on the basis of principles of constitutional economic order, according to Schoueri, but which, for the present study, goes against the principle of national treatment.

This occurs when the domestic and imported products are similar (i.e., like products) for the purposes of the first sentence of § 2 or § 4 of art. III of the GATT because this requirement is always indispensable to starting the analysis of the violation of the national treatment principle based on these devices52.

Here, once again, we take the liberty to disagree with Vettori, according to whom:

“[...] in relation to transfer pricing, products imported by Brazilian companies from related parties abroad or from tax havens are comparable to only the national products exported by Brazilian companies to related parties overseas or tax havens”53.

In Vettori’s view, the assumption for violation of national treatment by the transfer pricing rules would be the asymmetry of the predetermined profit margins in import and export operations. For example, there is a violation when there are more onerous adjustments in imports than in exports54, something that occurs in the Brazilian legislation, which provides for higher predetermined margins in imports, in addition to safe harbor rules55 only in exports56.

Here, we disagree with the comparative criterion between products indicated by Vettori to establish the analysis of the alleged violation of national treatment by the transfer pricing rules (i.e., products imported by Brazilian companies vis-à-vis exported domestic products).

The reason for this is that according to the view with which this study disagrees, there is no violation of competitive neutrality in a particular internal market because the products identified by Vettori as comparable (i.e., products imported by Brazilian companies vis-à-vis exported domestic products) do not compete with each other within the national jurisdiction.

In the view of the present art., the products that should be compared are similar products imported for sale in the Brazilian market and products manufactured domestically and marketed in Brazil.

As Daly teaches, the principle of national treatment works internally in each State, prohibiting the importing State from discriminating between locally produced goods and those that come from abroad57, which is done to seek competitive neutrality in the domestic market of the member countries of the trading system.

Consequently, this article believes that the different taxation between similar products must be examined in relation to its impact on the competitive conditions of the products in a particular internal market.

It is this asymmetry observed in a particular market – and not the tax asymmetry in different markets – that, in the view of this study, should guide the interpreter in the analysis of a potential violation of national treatment.

After establishing that the asymmetry of taxation by the income tax on similar products must occur within a certain domestic market, one should then check whether the distinctly taxed products are like products because this is the first requirement mandated by the first sentence of § 2 and § 4 of art. III of the GATT.

The WTO’s jurisprudence is taken for granted in the sense that the mere fact that the good has a foreign origin does not make it distinct from the national competitor, with the presumption that both are like products for the purposes of the GATT58.

Because the predetermined margins, in the hypothesis analyzed here, are always applied on the importation of products, there is the assumption that they are similar to domestic competitors for the purposes of the national treatment rule.

Being like products, it is now important to examine whether the predetermined margins of the national transfer pricing violate the other requirements of the paragraphs of art. III of the GATT.

According to the first sentence of § 2 of art. III of the GATT, in addition to the imported product being similar to the national product, it is necessary for the former to suffer excess taxation in relation to that imposed on the latter59.

For the WTO’s jurisprudence, not only a vanishingly higher taxation for the imported product but also the mere risk of tax discrimination in favor of the local product, is enough to violate the national treatment rule60.

The fact that the predetermined profit margins do not reflect the economic reality is sufficient for the present study to characterize the “mere risk” that exists for the imported product to suffer excess taxation, when compared to the national product, which violates the national treatment rule contained in the first sentence of § 2 of art. III of the GATT.

5. Conclusion

From the joint perspective of the international trading system and Tax Law, the analysis of predetermined profit margins in the Brazilian transfer pricing rules reveals that they are compatible with the subsidy rules but violate the principle of national treatment.

Despite the lack of neutrality of the fixed profit margins, given that they do not reflect the economic reality and hence may result in over-taxation or under-taxation via income tax on international transactions between related companies, such an inducing effect does not violate the subsidy rules of the WTO.

The reason for this is that the Brazilian transfer pricing rules apply without distinction to all national taxpayers, constituting, according to the WTO’s consolidated jurisprudence, a standard taxation (normative benchmark) that is not excepted to favor the exporter or to be prejudicial to the importer.

The paradoxical aspect noted by the present article is that if such predetermined margins are specifically altered for a taxpayer or group of taxpayers, which is advocated to achieve economic reality (arm’s length), then such an alteration can align national rules with the standards of the OECD but infringes the rules of the WTO.

The reason for this is that after materializing the exception foreseen in Brazilian law on transfer pricing, there will be a general rule of taxation (normative benchmark) – the predetermined margins that do not follow the economic reality – and a specific norm for the taxpayer in “justified circumstances” that, although honoring the arm’s length principle, will represent under-taxation or over-taxation compared to the general rule.

The mere risk of excess taxation on the imported product, which is subject to the calculation of income tax in the domestic market on bases that do not reflect the economic reality, make the predetermined profit margins violate the principle of national treatment because the similar national product enjoys a comparative advantage in relation to its international competitors.

Such questions, scarcely addressed by scholars of Tax Law and of the multilateral trading system, must be included in discussions concerning the transfer pricing rules for those who both defend and reject the arm’s length principle because of the relevant consequences noted here for International, Tax and Trade Law.

6. References

AULT, Hugh J.; BRADFORD, David F. Taxing intenational income: an analysis of the U. S. System and its economic premises, 1989. Available at <http://www.nber. org/papers/w3056.pdf>. Accessed 24 Aug. 2015.

AVI-YONAH, Reuven S., SLEMROD, Joel. (How) should trade agreements deal with income tax issues? Available at <http://papers. ssrn.om/abstract=285345>. Accessed 29 Oct. 2014.

BOURGEOIS, Jacques. Direct taxation and the WTO: in or out? In: BOURGEOIS, J. Trade law experienced: pottering about the GATT and the WTO. London: Cameron May Ltd., 2005.

BRAUNER, Yariv. International trade and tax agreements may be coordinated, but not reconciled. Virginia Tax Review. 2005.

DALY, Michael. The WTO and direct taxation, 2005. Available at <http://www.wto. org/english/res_e/booksp_e/discussion_papers9_e.pdf>. Accessed 27 Oct. 2014.

FAJERSZTAJN, Bruno; SANTOS, Ramon Tomazela. Preços de transferência. frete, seguro e tributos devidos na importação e o método PRL. Revista Direito Tributário Atual v. 29. São Paulo: IBDT/Dialética, 2013.

FERNANDES, Edison Carlos. Constitucionalidade in thesi e in concreto do controle fiscal dos preços de transferência. In: FERNANDES, Edison Carlos (coord.). Preços de transferência. São Paulo: Quartier Latin, 2007.

GREGÓRIO, Ricardo M. Restrições da comparabilidade, margens predeterminadas e liberdade de escolhas de métodos. In: SCHOUERI, L. E. (coord.). Tributos e preços de transferência. São Paulo: Dialética, 2013. v. IV.

HAMAEKERS, Hubert. In arm’s length – how long? International Transfer Pricing Journal. Mar./Apr. 2001.

JACKSON, J. H.; DAVEY, W. J.; SKYKES, A. J. Legal problems of international economic relations, cases, materials and text. 3. ed. Saint Paul: West Group, 1995.

MIRSHAWAKA, Valeria Zimpleck. Preços de transferência: diferentes visões. Dissertation (Master’s Degree in Law) – Law School of USP. São Paulo, 2012, advisor: Alcides Jorge Costa.

RATHEKE, Alex Augusto Timm. Transfer pricing e income shifiting: evidências de empresas abertas brasileiras. Dissertation (Master’s Degree) – University of São Paulo, 2014, advisor: Carlos Alberto Pereira.

ROSENBLOOM, David H. International tax arbitrage and the “international tax system”. Tax Law Review v. 53, n. 137, 2000.

SCHOUERI, Luís Eduardo. O mito do lucro real na passagem da disponibilidade jurídica para a disponibilidade econômica. In: MOSQUERA, Roberto Quiroga; LOPES, Alexsandro Broedel (coord.). Controvérsias jurídico-contábeis (aproximações e distanciamentos). São Paulo: Dialética, 2010.

_____. Preços de transferência no direito tributário brasileiro. 3. ed. São Paulo: Dialética, 2013.

SHADIKHODJAEV, Sherzod. National treatment on internal taxation: revisiting GATT article III:2. Seoul: Korean Institute for International Economic Policy, 2008.

UNITED NATIONS. UN Practical Manual for Developing Countries. Available at <http://www.un.org/esa/ffd/tax/documents/bgrd_tp.htm>. Accessed 13 May 2016.

RECEITA FEDERAL. Preços de transferência. Available at <http://www.receita. fazenda.gov.br/pessoajuridica/dipj/2005/pergresp2005/pr672a733.htm>. Accessed 21 Jul. 2015.

VETTORI, Gustavo Gonçalves. Contribuição ao estudo sobre as influências recíprocas entre a tributação da renda e o comércio internacional. Thesis (PhD In Law) – Law School of USP. São Paulo, 2011.

VICENTE, Marcelo Álvares. Do controle fiscal dos preços de transferência: consequência da aplicação dos ajustes e hipóteses de não aplicação. Revista de Direito Tributário Internacional year 3, n. 9. São Paulo, 2008.

WHALLEY, John. Taxes and trade. Available at <http://www1.worlbank.org>. Accessed 29 Oct. 2014.