Impact of Pillar One on Brazil’s Transfer Pricing Regime

Brazil has always adopted different rules for transfer pricing controls when compared to the United States and the more generally accepted Organization for Economic Cooperation and Development standards. It has usually favored statutory profit margins and a mathematical analysis, instead of a more economic focus on the nature of the transactions themselves.

This has often led to cases of double taxation, where the same profit is allocated to a foreign jurisdiction and to Brazil or, less commonly, to cases of non-taxation, where the margins are allocated elsewhere despite the substantial activities of Brazilian entities in each value chain.

However, given the ongoing discussions regarding Brazil’s accession to the OECD, there are currently talks taking place to align Brazil’s transfer pricing rules with the OECD’s transfer pricing guidelines. Brazil would also have to adhere to Pillar One, part of a global initiative to deal with cross-border trade deriving (primarily) from transactions carried out by companies that do not have any physical presence in the country where their customers are located.

In this article, we provide some details about the transfer pricing changes that are being discussed, as well as considering the main topics of debate.

The differences between Brazilian transfer pricing rules and the OECD guidelines may be summarized as follows:

• Fixed profit margins: Instead of a functional analysis to ascertain the profitability attributable to the Brazilian party, Brazilian legislation has adopted fixed profit margins for some methods. Theoretically, the margins can be changed, but, in practice, no such change has ever been approved since Brazilian transfer pricing legislation entered into force.
• Most favorable method: Brazil usually allows taxpayers to use the most favorable method (i.e., the method that yields the least taxable adjustment), instead of compelling them to use the most accurate method (i.e., the method that results in a price closer to market prices). Also, the method may be chosen per item, meaning that similar items in similar transactions could be traded with different margins if a different method is chosen.
• Safe harbors: Brazil’s safe harbors (especially for exports) release taxpayers from applying transfer pricing controls; some of them are completely unrelated to the profitability of the transactions.
• Lack of corresponding (and secondary) adjustments: Brazilian treaties, like Brazilian domestic rules, do not include provisions regarding corresponding adjustments. Additionally, there is no provision for the requalification of income, which is simply deemed either as a non-deductible expense or additional taxable revenue.
• Royalties: These are not included in the scope of Brazil’s transfer pricing rules.

Extrapolating from these characteristics, it is possible to conclude that Brazilian transfer pricing rules have had a twofold purpose: (i) to create objective rules to protect Brazil’s tax base, and (ii) to provide some independence to tax authorities in relation to other jurisdictions.

This may change with Brazil’s accession to the OECD as, among several other changes, Brazilian transfer pricing rules would need to be adjusted to comply with the OECD’s transfer pricing guidelines and, very likely, with Pillar One of the two-pillar solution as agreed by the OECD/G20 Inclusive Framework.

In a nutshell, these two changes could mean that Brazil’s transfer pricing legislation would undergo a substantial paradigm shift. Brazil would lose its objective rules and would have to adopt a more flexible approach regarding transfer pricing for most companies that are not subject to Pillar One. Brazilian tax authorities should coordinate with foreign tax authorities regarding the taxable adjustments due to transfer pricing rules, while allowing for a corresponding and possibly secondary transfer pricing adjustments.

Based on a joint presentation by federal tax authorities and the OECD, it seems that Brazilian tax authorities are working on a project, started in 2018, that aims to adjust the transfer pricing rules. However, this project will have to be discussed further with the private sector, modifications (if deemed relevant) will then have to be compiled, and only then will the bill be presented for congressional approval, where it might be changed, as per the usual legislative process. More recently, there has been some additional discussion regarding the possibility of the executive branch enacting a provisional measure. This would have to be converted into law by the Congress in 2022 (to apply in 2023).

The implications regarding Pillar One reach even further. Pillar One proposes two main changes: (i) elimination of digital services taxes, and (ii) a revision of taxing rights regarding multinational enterprises (MNEs) that sell to several countries, including, and perhaps primarily, sales made without a physical presence in the country where the customer is located.

The elimination of “digital services taxes” (a term which is yet to be defined) is a concept that should be viewed with care. While Brazil has not yet enacted a specific tax on digital services (although there is a bill in the Congress—2358/2020—proposing a tax on certain digital services (CIDE-Digital)), there are several other taxes that could be viewed as digital services taxes. For example:

• Income tax withholding in respect of royalties and payments for virtually all service imports, with rates that generally vary from 10% to 25%, depending on the nature of the service and the jurisdiction of the recipient;
• IOF-Exchange, a tax on currency exchange transactions relating to international payments made via credit cards. The tax, currently imposed at a rate of 6.38%, will be lowered, by 1% per year, to 0% by the beginning of 2028;
• CIDE-Remittances, a 10% contribution paid by Brazilians when making international remittances for various services and royalties; and
• ISS, a municipal service tax that may be levied on the import of services.

Considering that Pillar One proposes a trade-off between the elimination of digital services taxes and the allocation of taxing rights, this modification should be viewed with care from a Brazilian tax law perspective.

While none of the taxes mentioned above are specifically digital services taxes, they are all likely to be considered as such under Pillar One. For instance, while the triggering event of the IOF-Exchange tax is technically the exchange conversion based on the international credit card payment, this applies to virtually all international digital transactions carried out for most individuals in foreign currency. Today, it is very unlikely that such transactions are cleared via checks, cash, or other form of non-digital configuration.

Brazil has always had a creative approach to taxing rights. Any limitations on those rights were not flagrantly violated but rather circumvented via novel interpretations or the imposition of new taxes that had a very similar economic (although not legal) impact.

For example, Brazil considers that the technical services that are taxed under the royalties’ article are virtually all services that require specialized knowledge, as opposed to the more general interpretation that technical services would comprise only services that transfer technology. A second example of such creative taxation is CIDE-Remittances tax, which is a 10% contribution over virtually all services. However, instead of being withheld from the payment made to the foreign entity, it is levied directly over the import and charged to the Brazilian entity itself, therefore avoiding legal double taxation while maintaining the same general level of taxation.

Depending on the actual definition of digital services taxes, the consequences, and implementation, of the Brazilian accession to the OECD may entail more than simply changing the Brazilian transfer pricing rules. It would require, at least in principle, a substantial review of import taxation if the new rules are to achieve their original goal.

But even any transfer pricing changes may not be straightforward. The Brazilian tax system is rigid: the activities of tax authorities are considered fully bound by law, and subjective criteria often face substantial challenges from taxpayers. For example, Brazil’s “disguised distribution of profits” rules, which are aimed at verifying that the prices of domestic transactions are in accordance with market prices so that, for instance, a given cost is deductible or a given revenue is sufficient for tax purposes. Such tax assessments were often challenged, and the discussions dragged on for years—often with very different outcomes—and did not provide a clear definitional criteria which could be implemented consistently. Depending on how the Brazilian transfer pricing rules are amended, similar discussions may also arise.

From a constitutional perspective, it may also be difficult to prevent taxpayers appealing controversial matters to judicial courts. Such issues can often take years, if not decades, to be resolved. These delays could hinder the ability, both in Brazil and abroad, of providing reliable corresponding or secondary adjustments.

Apart from those more general considerations, there are specific aspects of Pillar One that should be analyzed and considered with care. The practical impact of the rules are: (i) that distinctions that are not based on constitutional criteria may be disallowed from a tax perspective; and (ii) that tax reductions could be subject to fiscal responsibility rules, which require such reductions to provide for a source of replacement or clarify a corresponding reduction on expenses (at least in theory).

While it is possible to consider a global approach for transfer pricing and Pillar One changes, in our view the burden imposed on MNEs in Brazil raises two questions:

• Will the tax burden on the MNEs subject to Pillar One be reduced in Brazil?
• Will such a burden be different (and, in particular, lower) than a smaller company operating in the country and, if so, is such difference allowed based on the criteria currently envisaged in Pillar One, considering the Brazilian constitution?

From a policy perspective, there are a few elements of Pillar One that should also be viewed with care.

Firstly, the taxing rights allocated to “market” jurisdictions is subordinated to that of foreign entities. Market jurisdictions only get a taxing right if the profitability exceeds a threshold (10%). And even in such case, it may only receive 20% to 30% of the taxing rights attributable to deals in the market jurisdiction.

Assuming that this division is acceptable, whether this new taxing right “costs” or “generates” tax revenues will depend strongly on what Brazil has to give up in order to receive it. If none of the “not-digital-services-taxes” described above are excluded, and transfer pricing adjustments remain roughly the same, then probably there will be more tax revenues for Brazil. Otherwise, a very in-depth analysis should be undertaken, preferably jointly with the previous adoption of Pillar One.

Apart from a purely technical analysis, we also suggest an additional question: Will these changes in tax legislation serve as an incentive for companies to remove their local presence in Brazil (for example, on account of Pillar One taxation being less burdensome than having a local presence), and, if so, is this impact being considered accordingly? Naturally, this depends not only on transfer pricing rules, but on the more general context of doing business in Brazil. For example, in the past, some companies have decided to operate locally simply because it made carrying out its transactions easier for individual customers, which were unlikely to go into lengthy, expensive, or complicated procedures to make international acquisitions or payments.

Conclusion

All in all, the combination of changes to Brazil’s transfer pricing rules and the adoption of Pillar One are likely to be at the center of tax controversies and debate.

Unlike countries that already adopt the OECD’s transfer pricing guidelines, these changes will probably be in the context of an overhaul of the Brazilian transfer pricing (and international transactions in general) taxation.

Furthermore, such changes should be viewed in light of the substantial debates concerning fiscal policy currently occurring in Brazil, and also with a view to the presidential and congressional elections due to take place this year, which may impact the originally envisaged schedule.

Our views in this article derive from past experience, where changes that were theoretically aimed at reducing taxes ended up creating additional complexities and increasing the total tax burden. This was a result, as mentioned above, of creative interpretation or even circumvention of the original arrangement via the creation of additional rules that were not direct infringements, but that nevertheless produced the same economic effect as previously mentioned.

This article does not necessarily reflect the opinion of The Bureau of National Affairs, Inc., the publisher of Bloomberg Law and Bloomberg Tax, or its owners.

Author Information

Henrique Munia e Erbolato is a Partner with Santos Neto Advogados and Daniel G. P. Orsini Marcondes is a Partner with Schmid & Orsini Advogados.

The authors may be contacted at: henrique.erbolato@santosneto.com.br and daniel.orsini@oslaw.com.br