It seems the OECD has finally decided to lead from the front, rather than waiting endlessly for consensus – a Public Consultation Document, Secretariat Proposal for a “Unified Approach” under Pillar One, released in Paris on 9 October proposes winding back the permanent establishment threshold, the arm’s length pricing principle for transfer pricing, and even the separate entity principle for taxing members of corporate groups. The approach in the Consultation Document would be a momentous change to positions the OECD has defended for decades, and to the international tax landscape more generally. It would consciously shift more of the corporate tax base from countries of production to the country of consumption.
When the BEPS project began in 2013, the problem of the digital economy was highlighted as its first item of business. By the time the Final Reports were issued in October 2015, the problem of the digital economy remained largely unsolved – the main recommendation on Action Item 1 was to keep working toward ‘a detailed mandate to be developed during 2016 [and] a report reflecting the outcome of the continued work in relation to the digital economy … by 2020.’ Work continued over the next 4 years with further documents released in March 2018, January 2019, February 2019 and May 2019.
In the meantime, several countries as well as the European Commission, perhaps out of frustration at the slow progress of the OECD’s work given domestic political pressure to act, or perhaps as part of a concerted campaign to get recalcitrant countries over the line, began developing proposals for unilateral domestic actions. Although in 2015 the OECD had not recommended any of specific options canvassed to deal with the digital economy except in relation to VAT/GST, the political agreement in 2015 pointedly provided ‘Countries could, however, introduce any of these options in their domestic laws as additional safeguards against BEPS, provided they respect existing treaty obligations, or in their bilateral tax treaties.’ Some countries decided to act, denying deductions for payments to foreign companies or imposing flat rate taxes on gross cash flows paid to foreign tech companies. A 3% digital services tax was enacted by France earlier this year, provoking immediate threats of retaliation from the US. The MAAL, introduced in 2016, is Australia's version of 'going-it-alone.' These indirect methods were used precisely because they avoided the obstacle provided by the existing bilateral tax treaty network.
The Consultation Document was produced against this background of discontent and is undoubtedly bolder because of it. The Document candidly admits,
... the stakes are very high. In the balance are: the allocation of taxing rights between jurisdictions; fundamental features of the international tax system, such as the traditional notions of permanent establishment and the applicability of the arm’s length principle; the future of multilateral tax co-operation; the prevention of aggressive unilateral measures; and the intense political pressure to tax highly digitalised MNEs.
Not only was there dissatisfaction with the slow progress on the digital economy, the impracticality of arm's length pricing has been an ongoing target for criticism. For as long as the major rule-making body comprised the 36 members of the OECD, they could apparently convince themselves of the soundness of their approach. But the decision in 2016 to expand the opportunities for other countries to influence rule-making in international tax by forming the Inclusive Framework (now 134 jurisdictions) has given the developing world the venue to argue for change. And even some long-time supporters of the arm’s length principle like the UK were restive. They are no doubt responsible for passages such as this:
... there are increasing doubts that the arm’s length principle can be relied on to give an appropriate result in all cases (such as, for example, cases involving non-routine profits from intangibles). Moreover, there seems to be agreement that the arm’s length principle is becoming an increasing source of complexity and that simplification would be desirable to contain the increasing administration and compliance costs of trying to apply it. Thus, an “administrable” solution is essential, especially for emerging and developing countries. And a simple system will lower the risks of disputes, which currently endanger the cohesion of the international tax system.
The proposal in the Document (labelled the 'Unified Approach') deals only with what is now known as Pillar One in the work on tax issues arising from digitalisation of the economy and covers the quintessential digital problem: that it is now possible to make large profits in a country without any presence there. The Unified Approach is still rather nebulous but it appears the general design would look like this: countries would impose tax --
- on the profits of multinational enterprises of sufficient size, once they are earning relatively modest amounts (varying by the country’s economic size) from sales to consumers in their jurisdiction
- whether the enterprise operates in the country by way of branch or subsidiary or neither
- the profits amenable to tax in the consumer's country would reference the profits of the entire group regardless of where those profits are currently seen as earned
- using a mixture of arm's length prices and formulary apportionment
- with this tax being protected by a new article to be inserted into bilateral income tax treaties (think, MLI 2.0).
The phrase 'operating in the digital economy' is noticeably absent from the formulation. If this proposal were limited to Big Tech -- Apple, Google, Amazon, Facebook, Microsoft, Netflix, Uber, Airbnb and so on -- its ramifications for most Australian businesses would be limited. But it is clear that the OECD expects the new regime to extend much more broadly: the Unified Approach would extend to '... large consumer-facing businesses, broadly defined ...' In other words, it seems any business of sufficient size, operating in retail markets, with foreign customers, could be in this regime. Users of digital platforms are regarded as consumers for this purpose (thus sweeping in Google and Facebook) but it may be that some tech giants, which are not in the public consciousness because they serve only business customers such as our own Atlassian, are not included.
The Document indicates extractive industries and agricultural commodities will likely be outside the scope of the new regime -- China will not be offered the opportunity to tax more of the profits of Australian resource companies -- and it ponders whether financial services should be excluded, 'taking into account the tax policy rationale as well as other practicalities.' Australian banks may end up not being affected even if they have foreign operations.
But even with those exclusions, the Unified Approach could be very pervasive because, where it is triggered, it will 'change the nexus and profit allocation rules not just for situations where there is no physical presence, but also for those where there is.' The taxation of all multinational enterprises operating in retail markets could now be taxed without regard to physical presence and on a basis which incorporates elements of formulary apportionment. It might, for example, affect a large retailer with foreign stores, a manufacturer who distributes offshore through a network of local sales agents, a publisher that sells online access to its products, or a University with an offshore campus. Banks might be spared, but maybe not fund managers.
And it seems being in the new regime might not be an all-or-nothing decision: the Document refers to, 'work on the possible use of business line ...' hinting that even enterprises which sell predominantly in wholesale markets might find themselves included for the consumer-facing part of their business. It is also clear that attempts to avoid the new regime by interposing unrelated local distributors facing the consumer will be dealt with. Regional segmentation is another possibility to recognise different profitability in different regions.
The profits taxable in the destination country would not be determined on a cost-plus basis or using TNMM. Instead, the tax claim of the customer's jurisdiction would comprise three elements:
- a first component representing a share of the group's 'residual profit ... potentially calculated on a business line basis' -- this is more commonly known as formulary apportionment. Apparently the size of the residual profit will be determined by reducing total profit by 'a deemed routine profit on activities [in] the countries where the activities are performed.' This factor is yet to be negotiated. The next critical question -- just what factor is used to apportion residual profit -- is: 'a formula based on sales',
- a second component which reflects the activities actually undertaken in the destination jurisdiction. This might be determined 'according to existing rules' (meaning arm’s length pricing) or, 'using fixed remunerations ... reflecting an assumed baseline activity' -- this is more commonly known as an arbitrary number. There is also a suggestion this number might be, 'negotiated ...', and
- to the extent that the second component is based on fixed remunerations, a third component where local activities have additional value-adding characteristics, with likely compulsory use of advance pricing arrangements and arbitration when the additional amount is disputed between or among countries.
So the new vision is not formulary apportionment instead of arm's length pricing; rather the new world would be formulary apportionment as well as some form of arm's length pricing, but with less emphasis on specific facts and circumstances.
Will it fly?
As we noted at the outset, the Unified Approach would consciously shift more of the corporate tax base from the country of production to the country of consumption. It is not clear just how enthusiastic the US and other countries will be to that development, although the Document does allude to, 'extensive consultations to develop a “Unified Approach” ...' And the Document indicates that a parallel proposal on Pillar Two will appear in November – Pillar Two being a global minimum corporate income tax to make sure all profits of large multinational companies are taxed somewhere. The current Document makes clear that significant additional corporate tax revenues are expected out of the two Pillars combined, so it is not a zero sum game for countries (with winners inevitably offset by losers).
But it is worth noting the Document is expressly presented as 'prepared by the Secretariat, and [does] not represent the consensus views of the Inclusive Framework [or] the Committee on Fiscal Affairs (CFA)', perhaps indicating the OECD Secretariat has decided someone has to cut the Gordian knot. This is the OECD Secretariat's best effort at crafting a compromise between competing positions.
The Document acknowledges,
... the re-allocation of taxing rights raises important political considerations. A crucial one is that these changes would need to be implemented simultaneously by all jurisdictions, to ensure a level playing field.
The Document is open for comments until 12 November, with a Consultation Meeting in Paris to follow. The obvious goal of the new proposals on the two Pillars is to have a high level consensus agreed at critical global meetings in January and February 2020 so the rest of 2020 can be spent on developing the details.