Much has been written and debated on the joint Brazil-OECD transfer pricing (TP) project, especially after the results of the project (the report) were published on December 18 2019. Little has, however, been discussed about what was outside the scope of the OECD/Receita Federal do Brasil's (RFB) TP report – be it in relation to TP or in respect to other international tax matters relevant for Brazil's accession to the OECD – other than what is referenced in the technical note published by Brazil's National Confederation of Industry (Confederação Nacional da Indústria, CNI) in December 2019.
CNI is the Brazilian representative of the Business at OECD network. It has supported Brazil to best reflect international practices and adopt the local rules of the OECD TP Guidelines (the Guidelines). CNI also defends other reforms in relation to Brazil's withholding tax policy, the widening of Brazil's treaty, and the use of best practices of international tax rules adopted widely by OECD member countries, particularly in the G7.
Transfer pricing convergence: Beyond the joint report
The uncompromised adoption of the arm's-length principle (ALP) under Brazilian law, and the availability of the OECD Guidelines in local regulations, should mitigate double taxation (and double non-taxation) risks in a wider spectrum of circumstances. It should also, therefore, allow the opportunity for a greater insertion of Brazil in global value chains, which will consequently increase foreign direct investment (FDI) in the country.
Transfer pricing convergence is a critical item in the tax agenda to pave the way for Brazil's accession to the OECD. However, some details such as how such a convergence could take place, how Brazil's experience could be used to enhance the OECD Guidelines, and why the Brazilian approach creates distortions, were not fully explored in the published report. Instead, the distortions of the current approach were taken as evidence that Brazil's existing methodology could never be salvaged, and could not be useful to its case going forward.
In our view, a hypothesised and enhanced version of the Brazilian approach could still be useful. In particular, we see potential for the continued use of optional safe harbours, which are consistent with the ALP and OECD Guidelines, yet cover more than low-value-adding (or low risk) functional profiles. They should cover more than routine functions, assets and activities, and potentially capture normal returns or even above-normal residual profits.
This ambitious use of safe harbours is what would make them uniquely Brazilian, rather than a replication of what may already be common within the OECD world. This hypothesis was assumed away in the joint report, even though at the launch of the event there were statements referencing the potential 'cross-pollination' between the Brazilian rules and the OECD Guidelines. Moreover, a somewhat similar safe harbour debacle is very much present in the OECD-US debate over pillar one.
In our view, the Brazilian approach can be revisited, remodelled and enhanced to permit such results without inherently causing double-taxation (or non-taxation) risks, and without triggering the risks that are observed and criticised of the existing framework in the joint OECD/RFB report.
Broadly, there could be a situation where the ALP becomes the governing rule under domestic law, while an 'opt-in' system of safe harbours would be developed and controlled by the RFB, and results are made subject to the discipline of Article 9(1) and (2) as per the OECD Model Convention.
These three variables would mean that other states could never be harmed by Brazil's practice, and that the RFB could protect the Brazilian Treasury from base erosion by maintaining the safe harbours adequately (i.e. not allowing fixed margins that are too low, or distanced from economic reality). It can be argued that the development of such an ALP-compliant system should not be an obstacle for Brazil's accession to the OECD. This debate must evolve, grounded on technical argumentation rather than political preferences.
To be clear, CNI's technical proposal has always been for the full adoption of the ALP under domestic law, with the corresponding introduction of the OECD Guidelines into the Brazilian legal system. Preferably, the ALP would be legislated in a manner consistent with Article 9 of the OECD Model Convention, while the OECD Guidelines would include detailed chapters within the infra-legal income tax regulation (Regulamento do Imposto de Renda), under Decree 9,580/18 (soon to be reformed) and further detailed in the normative instructions, edited by the RFB itself.
CNI's proposal would, therefore, be considered as a convergence of the maintenance and enhancement of the Brazilian methodology of fixed margins. This would ensure compliance with the ALP, functioning as optional safe harbours permissible by the OECD Guidelines, and available to all taxpayers irrespective of size, and as a permanent and not transitory rule.
Brazil remains a large, closed economy. High trade barriers and regulatory restraints create an internal market where relative prices, and returns on equity, do not easily compare with what can be observed in other countries. Furthermore, most large businesses operate as privately held limited liability companies that are not required to publish financial statements. Therefore, there is a relative scarcity of microdata available to taxpayers in respect to Brazil, while comparability adjustments on foreign data would be particularly challenging given Brazil's quasi-autarchy economy.
Conversely, Brazil does have an extremely sophisticated system of digital recordkeeping (SPED) for submission of general ledgers to government authorities and of numerous granular, transaction-level and entity-level information to the tax authorities. Therefore, the Brazilian tax authorities have a wealth of microdata at their disposal. If in other countries transfer pricing is plagued by asymmetry of information with the tax authorities knowing less about comparable data than taxpayers, in Brazil it is conceivable that while taxpayers would know more about their functional profiles, tax authorities would have substantially more comparable data. Therefore, it would be in the public interest for taxpayer-level microdata that is kept by the tax authorities to become useful to all, in aggregate form so as to preserve taxpayer secrecy. Aggregating data by business sector and risk profile is an achievable task to the Brazilian tax administration, but perhaps not to others around the world who might not be as digital or have taxpayers that are as transparent.
The safe harbours would function as a system of presumptions seeking normative simplification and objectivity, benefiting from the in-depth comparability analysis that could be done by the RFB with microdata not generally available, and not inherently harbouring base erosion and profit shifting or double taxation, as the current Brazilian methodology does. With our support, CNI has always defended that, along with the introduction of the methodology of the OECD Guidelines, an enhanced version of the Brazilian rules would be salvaged and maintained, in a manner coherent with the ALP, and properly administered by the RFB. This enhancement of the Brazilian rules could even serve other countries, so that the Brazilian experience in transfer pricing would be useful to the OECD and to non-OECD countries, as it would be fully governed by the ALP, and hence not be an obstacle to Brazil's accession to the OECD.
In our hypothesised safe harbour approach, it is important to ensure that taxpayers can fully apply the methodology of the Guidelines (as a standard rule and not only through rulings or the rebuttal of presumptions), or to opt into the simplified system of safe harbours with fixed margins offered by the RFB regulations. This system of safe harbours could provide standard functional and risk profiles, and fixed margins varying per sector and profile. These would be published and updated periodically by the RFB and based on market research, which is achievable given the high level of automation and artificial intelligence already used by the RFB.
It will be critical to develop a robust advance pricing agreement (APA) programme to diminish conflicts in the application of the Guidelines, and even to reconcile a local simplified methodology bilaterally or multilaterally. Furthermore, this would be crucial for Brazil to finally maintain a stable and responsive practice supporting mutual agreement procedures (MAPs). It will be necessary to invest in Brazil's tax administration, and both the RFB and the Federal Tax Attorneys' Office (Procuradoria Geral da Fazenda Nacional, PGFN) need to build capacity to be able to apply the OECD Guidelines and enter into APAs and MAPs in complex TP matters.
Beyond transfer pricing
With the prospect of implementing OECD Guidelines and its methods, and of the administration of the ALP by the RFB and the PGFN, the inception of MAP arbitration would be highly recommended. We find the concept to be permissible under Brazilian constitutional law given that tax assessments are inherently uncertain, and while interpreting the facts and the law, it can be understood that it does not constitute potential tax claims to be a public asset (if so, such assets could not be disposed by officials as per constitutional law). A similar approach to arbitration and public claims has been practiced in other areas of Brazilian administrative law with federal, state and local governments being subject to arbitration over uncertain ownership rights.
To this end, it would be imperative to extend and revisit Brazil's treaty network. This is not only to ensure the full adoption of Article 9, as Brazil does not use 9(2) in its treaties, but to restructure the MAP provision which allows for arbitration. This would secure the legal certainty effect of tax treaties and their positive effect on FDI. Nonetheless, in order to extend its network, Brazil needs to reformulate its withholding tax policy. At present, the 15% domestic rate is not reduced by most treaties, and when reduced it remains very high (10%), while there is still substantial confusion over the scope of the royalty article causing an overreach to tax service fees beyond business profits.
These policies hurt Brazil and create an environment where treaty negotiations with many countries are not viable, particularly in the absence of a dividend withholding tax under domestic law. Most importantly, these policies may represent a real obstacle for Brazil's accession to the OECD, as many members may push back on Brazil's convergence not only to the OECD standards in respect to Article 9, but for the entire OECD Model Convention and in particular to the taxation of services.
Finally, Brazil does have an opportunity to use OECD benchmarks and best practices as it pursues accession. A reduction to the corporate income tax rate to a level comparable or below the OECD average of 21.4% would be advisable to foster FDI. Likewise, the reintroduction of a dividend withholding tax would enable the extension of Brazil's treaty network if it changes its service and royalty withholding tax policy, and embraces the permanent establishment rule. With this configuration, Brazil could adopt controlled foreign corporation rules and anti-BEPS provisions consistent with (and not stricter than) OECD benchmarks such as Germany and Japan. These changes would not be required by the OECD Secretariat or by any OECD member for Brazil, but would be enhancements serving Brazil's unilateral interests to make businesses and the economy more competitive.
Romero J S Tavares |
|
|---|---|
|
Partner PwC Brazil Tel: +55 11 3674 2880 Romero J S Tavares is the lead international tax services (ITS)/transfer pricing (TP) partner at PwC Brazil and is a part of the PwC global team on tax policy and administration. He is an international tax attorney and tax policy consultant with a career spanning 26 years (10 of which were in the US and Europe). He has served as a tax policy advisor to Brazil's Confederação Nacional da Indústria (CNI) since 2012, and has been heavily engaged in Brazil's OECD accession. While on his mid-career academic sabbatical in Austria, he received his PhD from the Vienna University of Economics and Business, worked as a professor, and also served as a tax policy advisor to the World Bank Group. In addition to his PhD, he holds a master's degree from the University of Detroit, and has attended postgraduate programs at INSEAD, Leiden, and Harvard Law School. |
Priscila Vergueiro |
|
|---|---|
|
Partner PwC Brazil Tel: +55 11 3674 2115 Priscila Vergueiro is a partner at PwC Brazil's ITS/TP team. With almost 20 years of experience in tax consulting, she has acted for clients in several industries including manufacturing, services, telecommunications, agribusiness, advertising and automotive. She specialises in global structuring for Brazilian multinational firms and also serves numerous inbound clients. She holds a business administration degree from Mackenzie Presbyterian University in São Paulo and a master's degree in taxation from Fordham University. She is a part of the board of advisors of GETAP, an association of large Brazilian corporations that aim to improve the Brazilian tax system and economy through collaboration with the authorities. |
