By Luís Eduardo Schoueri (Full Professor of Tax Law at University of São Paulo & Senior partner at Lacaz Martins, Pereira Neto, Gurevich & Schoueri Advogados) & Mateus Calicchio Barbosa (PhD Candidate and M.Sc. at University of São Paulo & Tax partner at Lacaz Martins, Pereira Neto, Gurevich & Schoueri Advogados)
It is said that in every crisis lies an opportunity. If the quote means that possibilities may emerge, in the tax realm taxpayers also have a new momentum to the danger component of the notion. In Brazil, outdated – not to say dangerous – tax alternatives have been put on the table to meet the recent budgetary needs. Certain wealth and capital taxes on both companies and individuals, despite previous and frustrated propositions since mid-90s, have been discussed while the government seeks a way out of an unprecedented public debt in the years to come.
A pressed legislature might also feel encouraged to step into digital taxation. An attempt was concretely made in recent Legislative Proposal No. 2,358 of 2020, which proposes to include a digital levy within the scope of “CIDE”, a contribution intended to operate as economic correction or to finance specific budgets for federal programs on strategical economic sectors. The proposal has not been drafted by government and it is not clear whether it shall endorse it.
The collection of proposed CIDE would be assigned to the National Fund for Scientific and Technological Development (FNDCT). The levy would be charged on total gross revenues earned by both resident and non-resident companies, provided that the relevant group (i) global turnover would exceed BRL 3 Bil and (ii) local turnover would exceed BRL 100 Mio. The tax would be calculated at progressive rates ranging from 1% (up to BRL 150 Mio revenue) to 5% (above BRL 300 Mio revenue). The services that the CIDE expect to cover include digital advertising to Brazilian users, transaction of data collected from Brazilian users and maintenance of business-oriented digital platforms.
The proposal as such kicked Brazil into the ongoing hassle about taxation of the digital economy. Despite a long (and uncertain) way ahead in the Congress (most legislative proposals never succeed, especially if the Executive is not engaged), it was enough for the United States Trade Representative to include it among other “digital services taxes” being investigated in view of their alleged unfair and discriminatory effects over certain US based multinationals. The notice issued by the USTR office states that the digital tax policy so adopted may diverge from the “international tax system in several respects”, including “extraterritoriality; taxing revenue not income; and a purpose of penalizing particular technology companies”.
It is surprising that a taxation of services reaching users within a country’s territory may be claimed as extraterritorial. It is true, however, that the Brazilian proposal joins the interim measures that other countries have adopted while the international fora strive for any consensus on how to distribute taxing rights on income from new business models. As a unilateral action, the CIDE levy is one more challenge put to the perspective and coordination required for a single universal solution. The proposal is well aware of its wider picture: its reasoning refers to BEPS Action 1, mentions the French and Italian levies, and acknowledges that a “better solution will undoubtedly be an international solution” – this being the moment when one should “defend the adhesion of Brazil and terminate the CIDE”.
Like its international counterparts, the proposed CIDE ring-fences the digital economy, either when it selects its taxpayers (tech giants) or defines the taxable events (internet advertising and users data collection). The CIDE is charged on revenue, thus raising the same concerns of over-taxation when combined with withholding tax levies and reasonable doubt on its compatibility with tax treaties. If any, the major difference to other digital taxes around rests on the progressive rates adopted under a vague claim of tax justice.
All in all, is it time to introduce the digital taxation in Brazil? Besides the usual claims above, detractors of “quick fixes” and the like unilateral levies might find an additional argument to sustain the inconvenience of the proposal. That is: to a considerable extent, Brazil has already introduced digital taxation.
Indeed, examining the existence of a local PE – digital or not – has never been on the Brazilian tax agenda. The reason attention has never been particularly directed is that, where payments from Brazil are remitted abroad, a rough source-of-payment criterion already makes it sure the country always has its (25%) share in services paid out of the pocket of its residents. The underlying notion tells that if the Brazilian payer is allowed a deduction, then the payment is to be taxed, be there a PE of the foreign supplier or not. Concretely, however, the operation of the source-of-payment rule does not depend on the effective deductibility by the payer.
This current source-of-payment criterion is very pragmatic. It curtails base erosion with no deference to the difficulties related to the PE, therefore relieving tax authorities from tricky auditing. The Brazilian solution is even more convenient for tax authorities if one combines the source-of-payment criterion with rigid foreign exchange control exercised by the local Central Bank. Accordingly, there is an additional tax consistently imposed at moderate rates on most of the currency exchange transactions involving BRL, which operates as a sort of equalization levy where the seller has no presence in the country.
The side effect of the current system is obviously well-known – and unwelcomed – by business: non-resident service suppliers see their gross remuneration taxed regardless of any meaningful presence in the country. Needless to say, usually the burden shall be transferred to the payer, discouraging the search for better/efficient service suppliers abroad. In any case, this policy of taxation has been authorized by courts and enforced for decades in Brazil. The system is settled for good.
Of course the source-of-payment criterion as above does not capture the income from certain transactions. Mainly in B2B, no monies might flow out of the country despite its user base creates value to the business being carried out abroad. The wide withholding tax system in force, however, demonstrates that the need to capture digital profits in Brazil is not as pressing as elsewhere. If the OECD is correct in that the digital levies should apply only where the relevant “income would otherwise be untaxed or subject only to a very low rate of tax”, then the Brazilian system may be evidence enough for the inconvenience of the recent proposal.
Since a digital tax is arguably less optimal than a withholding tax alternative, the proposed CIDE should not to be of any topical interest in the Brazilian context. Time will tell if sound policy considerations will outweigh budgetary concerns in the heat of a crisis.