1 marzo 2021
Allocation of profits in the digital economy era: is Pillar 1 a solution, or is it more of a compromise?
Il punto di Nupur Jalan* e Gianmaria Favaloro**
Overtime, taxation of digitalised businesses (DBs) has raised several issues mostly due to a systematical disconnection between the place of business from the actual place of consumption. Most notably, the progressive departure from a pure ‘physical presence’ connecting criteria to a ‘digital presence’ one is pushing the system towards formulary apportionment (FA) and/or destination-based right to tax (DBRtT). Tough, given an international tax framework based on out-of-date connecting principles, heading to a balanced division of the taxing rights (among the countries involved in the digital/data value chain) has become an onerous task to fix. Indeed, the entire debate revolves around the fair allocation of the profit of the DB-MNEs to all the market jurisdictions or the jurisdictions in which value creation takes place.
It is hard to say how good the principle of value creation holds in today’s situation where – on the one hand – Pillar Two digresses from the value chain and – on the other hand – various stakeholders are suggesting unitary FA for corporate tax regime overhaul. Particularly, the DB-MNEs (which mainly capitalise on user data) might have a strong presence in a market, to which they supply their digital services, without having ‘physical presence’ in that market. Consequently, since some of these players have a strong control over their data ecosystem and data value chain, they can obtain a competitive advantage over their counterparts. To this end, it is undeniable that network effects (along with data) play an important role for DBs which are mainly operating on platform-based model.
Coming to the existing profit allocation norms, several authors have raised concerns with the existing Arm ’s Length Principle (ALP). As already said, since the current international tax framework does not consider the demand side of the market, moving towards FA and/or some simpler destination-based approaches shall be seen as the way forward to aim for a balanced profit allocation. Rightly said:‘The oranges upon the trees in California are not acquired wealth until they are picked, and not even at that stage until they are packed, and not even at that stage until they are transported to the place where demand exists and until they are put where the consumer can use them.’ (League of Nations, ‘Report on Double Taxation’, E.F.S.73.F.19 1923). Hence, both demand and supply side play and important role in generating the profit of the organization.
With reference to FA, long back, the European Commission proposed both a ‘Common Corporate Tax Base’ (CCTB) and a ‘Common Consolidated Corporate Tax Base’ (CCCTB) for the EU Member States, to be implemented through a two-step approach which is still under discussion. FA has already been adopted by some federal countries as the US and Canada. Also, India’s Central Board of Direct Taxes (CBDT) on April 1st, 2019 issued a draft report on the attribution of profits to permanent establishments (PEs) of non-resident enterprises in India based on fractional apportionment method. The subnational experiences suggests that, as long as economic and political integration proceeds FA may be a better and simple solution. However, FA has its own downsides mainly in terms of the factors to be adopted and the percentages that can be allocated to these factors.
Talking about the current proposals, Pillar One (the so-called ‘Unified Approach’, UA) mainly focuses on the allocation of taxing rights and seeks to undertake a coherent and concurrent review of the profit allocation and nexus rules. UA does not cover not only highly digitalised business models but goes wider broadly focusing on consumer-facing businesses and automated digital businesses (ADS) with certain carve-outs. However, nexus rules are largely only based on sales (it is pertinent to mention that sales alone as a nexus factor may not be sufficient). In addition, the UN Committee of Experts on International Cooperation in Tax Matters has proposed a new Article 12B – Income from Automated Digital Services to be added to the UN Model tax treaty. Particularly, the release of the draft proposal named ‘Tax Treatment of payments for Digital services’ concerned a provision that would allow the source taxation of income from the rendering of automated digital services. In other words, it requires foreign providers of automated digital services to pay income tax at source by means of either (i) withholding on gross income (whose rate is to be agreed upon by the parties to the treaty), or (ii) on net income pursuant to an apportionment formula. Both proposals try to cover ‘Sale or other alienation of user data’ among others, yet the entire limb of data value chain still remains uncovered.
Conversely, the spreading of different unilateral measures (e.g. withholding taxes on royalty payments, the equalization levy and Significant Economic Presence (SEP), Digital Services Taxes (DSTs) etc.) has led to multiple taxation (i.e. in most cases with no tax credits available) due to the presence of different tax bases under each levy. Some of these unilateral measures cover some limb of data too. For example, the Indian SEP also includes sales of data collected from a person who resides or from a person who uses IP, sales of goods or services using data collected from a person who resides or who uses IP. In other words, transactions of data are addressed by the forms of collection and monetisation, but the elements of the data value chain and actors are not addressed, and it is left ambiguous.
In this regard, the OECD – as of today – has been trying to reach a consensus solution (under Pillar One and Pillar Two) to avoid the enforcement of multiple and uncoordinated unilateral measures. Tough, considering the UA, its complexity, and a small reliance on FA through Amount A (i.e only for the allocation of residual profit rather than a fair share of revenues to different jurisdictions) any formula and/or any mechanism devised should consider the data limb and the role of actors in the data value chain (Confluxdata 2020). Also, it is hard to foresee how a single formula (that shall be used in Amount A determination) would work well for all types of industry and business models even for the allocation of residual profits.
Separately, in the Authors opinion, the development of a bespoke global tool to calculate the actual amount (and the residual profits) to be apportioned among different jurisdictions – in the light of strong political and technical consensus (e.g. through the Inclusive Framework) – is seen as an important way forward. The access of the tool used by the taxpayers should be available to all the participating jurisdictions and – for example, it might be secured using blockchain technologies with a private key for the single taxpayer, and a public key for each participating jurisdiction. Similarly, tax authorities can use the same software tools to calculate the taxable profit at their end and create an automatic match (and the related ‘corresponding adjustments’) among the taxable profits calculated by the other jurisdictions involved. Since the tool would be developed on a global basis, there should be unanimity in the taxable profits calculated at both ends (UA) and less distributional conflicts. Thus, as of today, it seems that Pillar One represents – in any case – a mid-way solution.
* International Tax Professional; Research Fellow | Working Party on Tax & Legal Matters
** CTA-CFE, Postgraduate | University of Oxford – Faculty of Law, Research Fellow | Working Party on Tax & Legal Matters
