Complexity or chaos?

In October, the OECD released a suite of documents outlining proposals to revolutionise the international tax framework, including two Blueprints (one on each ‘Pillar’ of its ‘Digitalisation of the economy’ project). The release was described by member countries of its Inclusive Framework as ‘a solid foundation for developing a global, consensus-based solution to the tax challenges of the digitalisation of the economy’. 

Despite the name, the proposals are not targeted solely at digitalised activities. The OECD explains that: ‘Due to digitalisation, globalisation and new business models, many MNEs are able to make large profits in countries without necessarily booking these profits in these countries. This is [because] they may operate business without establishing any physical presence …  [and] rules to allocate profits are no longer fit in a globalised, highly digitalised economy where value is concentrated on intangibles.’ 

Among other things, Pillar One seeks to reallocate a portion of ‘residual’ profits of consumer and digital businesses to where consumers or users are located. Pillar Two seeks an effective global minimum tax to resolve continuing base erosion and profit shifting (BEPS) concerns. 

The proposals are extremely complex (nearly 500 pages, alongside another 300 pages in related documents) and by their nature will need further refinement, which is why the OECD is undertaking such extensive consultation. The proposals would (at least at first) only apply directly to large businesses. Businesses have been analysing the proposals in great detail and have been vocal about their concerns (more on this below).

The period for written consultations ran to 14 December, and a public consultation was held on 14 and 15 January. The OECD has indicated that a successful conclusion must be reached by ‘mid 2021’, and while this deadline has already moved, some countries are anxious to move forward and may not wait forever for an agreement before extending unilateral actions. Elements of the proposals lay groundwork for the future even if not agreed globally; they will either revolutionise or put pressure on an international framework that has remained comparatively unchanged for decades.

Background

Action 1 of the OECD BEPS Project concluded in 2015 that it was not possible to ringfence the ‘digital economy’; that BEPS issues exacerbated by digitalisation would be curbed by the other BEPS recommendations; and that digitalisation posed broader challenges to the international tax system that would continue to be explored. Countries could introduce unilateral measures to safeguard against BEPS if they were consistent with their tax treaty obligations.

By 2017, the G20 had mandated that the OECD prepared an interim report by 2018 and a final report by 2020. While this work has continued, scepticism remained from some countries that an agreement could be reached. With a lack of consensus in 2019 for an EU-wide digital services tax (DST) (a levy on turnover based on the location of users of digital platforms), France, Italy, Spain and the UK implemented their own DSTs in lieu of an OECD agreement. Other EU countries opted for similar taxes on digital advertising.

Outside the EU, several countries (including India) implemented similar measures, and several (such as India, Taiwan and Hong Kong) implemented broader proposals to tax a wider range of businesses on deemed profits attributable to intangibles or digital services consumed or contributed to from their countries. 

The US Trade Representative swiftly opened investigations, and has concluded those in relation to France, Italy, India and Turkey. Sanctions are being deferred while other investigations continue.

Meanwhile, at the OECD, four proposals were suggested for global reform. A global minimum tax was put forward by France and Germany (which has become Pillar Two), while the G24 proposed formulary apportionment of profits, the UK proposed allocation of profits based on digital platform users’ contributions, and the US proposed an allocation of profits to consumer countries based on marketing intangibles (which have combined into Pillar One).

Concurrently, the UN Tax Committee is debating changes to its Model Tax Convention (the UN Model). A new Article 12B will be included in the next UN Model which allows new withholding taxes and/or allocation of 30% of profits from automated digital service products to source countries. Further consideration will also be given to expanding the definition of royalties to include software payments.

It may be challenging to convince countries favouring the UN Model that the new OECD proposals would satisfy their interests. The G24 formulary apportionment proposal to the OECD is not the same as the UN Model changes under discussion, but it is even further from what is outlined below. The operation of Amount B under Pillar One (see below), the Subject to Tax Rule under Pillar Two (also below), and the ability for resource and capacity constrained tax administrations to administer the regime will be key in securing their endorsement.

The Pillar One Blueprint

The Pillar One Blueprint puts forward three changes to overlay the existing international tax framework through identifying new rules for two ‘amounts’.

Amount A

Amount A would allocate a portion of global residual profits from consumer facing businesses and automated digital services businesses away from wherever it is currently allocated, to where end consumers/users are respectively. This is not limited to businesses that interact directly with consumers and limited exclusions exist for some extractive and financial services activities.

• Automated digital services are defined by a positive list of activities, a negative list of activities, and a general definition. Broadly, where services provided by an electronic network require minimal human intervention to service individual users, they would be in scope.
• Consumer facing businesses are those that generate revenue from the sale of goods and services of a type commonly sold to consumers, including indirectly through intermediaries (e.g. franchising, licensing, third party distribution).

Global and ‘in-scope’ revenue thresholds would exist and may be phased, and there may be a threshold to exclude groups with minimal foreign income. Some groups will be able to segment their results based on hallmarks (e.g. those found in IAS 14). For in-scope revenues, groups will need to calculate their residual profits. (This is not agreed but, for example, could be, 20% of all profits over a 10% margin, or more for automated digital services if ‘digital differentiator’ factors are met.) 

The reallocation of this portion of residual profit is complex. There would be ‘nexus’ thresholds to identify which jurisdictions should receive an Amount A allocation, but they would differ according to whether activities are automated digital services (low revenue threshold only) or consumer facing businesses (higher and possibly staggered revenue thresholds, plus other factors); and a safe-harbour may restrict the allocation where taxes are already paid there. Amount A would be taken from entities identified based on connectedness to the recipient jurisdictions and their residual profit levels. 

A binding dispute prevention process could be applied for before tax adjustments are made by tax administrations. The process includes a review panel of relevant tax authorities. For in-scope businesses, mandatory binding resolution processes could be developed to assist with transfer pricing and permanent establishment disputes. 

Amount B

Amount B (which is not limited to automated digital services and consumer facing businesses) seeks to simplify the returns for routine marketing and distribution activities to a fixed Return on Sales basis, although this is complicated somewhat by the many differentiating factors to be taken into account, such as industry, region and functional intensity.

The Pillar Two Blueprint

The Pillar Two Blueprint contains four rules to ensure that a globally agreed minimum tax is paid on profits for each jurisdiction. The minimum rate has not yet been agreed, although it is expected to be 10% to 15%. The rules will be designed to ‘co-exist’ with the US global intangible low-income tax regime (GILTI) rules.

1. The Income Inclusion Rule: This requires parent companies to top up the tax of their constituent entities where the cash tax paid (‘covered taxes’) divided by the accounting profit in the year (based on parent GAAP and subject to some adjustments) falls below the minimum rate. Blending will be permitted where multiple entities are in the same country. Covered taxes include some taxes suffered on profits in other countries, such as controlled foreign corporation charges, withholding tax on interest or royalties, and those imposed in lieu of income tax. Other adjustments seek to account for temporary differences on tangible assets (and some others), and accounting losses can be carried forward (and back in some cases). A mechanism to carry forward excess taxes will be included, but no carry back. Instead, a mechanism may allow cross-border utilisation of excess where (effectively) it would have resulted in a refund if carry-back was allowed. All these mechanisms would be time limited.
2. The Undertaxed Payments Rule: This acts as a backstop, allowing a denial of some tax deductions for intra-group payments made to low-taxed entities in a group that is parented in a jurisdiction that does not have an appropriate Income Inclusion Rule regime (or, presumably, a GILTI).
3. The Switch-over Rule: This requires each foreign subsidiary to allocate an appropriate portion of its income (together with the taxes on that income) to a permanent establishment that may be maintained in another jurisdiction.
4. The Subject to Tax Rule: This will be triggered when a payment (for interest, royalties and some others) is subject to a low nominal (or base-narrowed) tax rate in the recipient. While the Income Inclusion Rule is said to be the primary rule, the Subject to Tax Rule will in fact apply first, and Subject to Tax Rule tax paid will be taken into account in calculating Income Inclusion Rule or Undertaxed Payment Rule liabilities.

The Pillars form a package, and for agreement to be reached at the OECD, they need to be agreed together. Of course, this does not preclude unilateral adoption (of the elements that can be implemented unilaterally) in the absence of agreement.

Major challenges to come

The obvious challenge facing both Pillars is their enormity and complexity, and the corresponding risks of compliance overload and double taxation. The administrative burden is clear even where detailed technical questions remain unresolved. How to deal with nexus threshold where both automated digital services and customer facing business activities are undertaken? How to recognise existing substance and tax  the market jurisdictions? What is ‘connectedness’ with a market? How to deal with losses? The list – as demonstrated by responses to the OECD – is extensive.

For Pillar One, the challenge requires a solution from multilateral mechanisms to give taxpayers advance certainty. Despite significant strides made by the OECD in recent years, many jurisdictions have neither the capacity nor inclination to enter into such mechanisms, but genuine and effective multilateral processes will be needed to allocate rather than multiply global taxing rights. Equally, effective double tax relief mechanisms will be needed to deal with the duplication of Amount A, such as from existing market taxing rights and withholding tax rights. 

For Pillar Two, similar double taxation challenges will arise – particularly for businesses with significant timing differences over long investment cycles; for example, if transitional rules do not account for pre-regime losses, or where higher taxes paid later in the investment cycle cannot be smoothed via a carry-back or deferred tax solution. Not all timing differences will be included, yet they all interact (e.g. fixed asset timing differences can manifest as brought forward losses), and double taxation on a sufficient scale could impact investment decisions.

Another key challenge for businesses under both Pillars is in getting the relevant information to comply, especially where businesses engage with consumers but do not sell directly to them. Accounting systems often will not provide the relevant data for either Pillar One or Pillar Two. Simplification mechanisms (e.g. based on existing country by country reporting data) might limit the burden but realistically that requires risk-based simplification approaches rather than technical ones. 

For Pillar Two, additional challenges arise with respect to GILTI co-existence at both a political and technical level. At a political level, non-US businesses and countries will question whether they are disadvantaged by GILTI’s use of global blending (and surrounding US check the box) features, and by the asymmetries of Pillar Two imposing domestic effective tax rate constraints (e.g. via the Undertaxed Payments Rule), while GILTI imposes no US effective tax rate constraints. Whereas viewed in aggregate from jurisdictional rather than individual business profile perspectives, the GILTI and Pillar Two tax impacts may be broadly equivalent, trying to dovetail the operation of two differently imperfect regimes may prove very challenging. Indeed, at a technical level, non-US businesses investing in or through the US could find themselves subject to both Pillar Two and GILTI – and US law change would be required to address that.

Interested SMEs have also noted that if they are operating from high tax jurisdictions by selling into lower taxed market jurisdictions, despite the complexity they may well prefer to opt into Pillar One, which the Blueprint does not currently foresee.

Where are we headed?

The scale and ambition of the project cannot be overstated. The political challenges in reaching agreement seem as insurmountable as the technical challenges in bringing that agreement to fruition. However, many commentators said the same about the BEPS Project in 2013, and the OECD delivered a package in 2015 that met its mandate. The challenge here is greater – a wider range of countries are involved in agreeing wider reaching reforms which reallocate income between major economies with very different objectives.

Following President-elect Biden’s victory, there is renewed optimism that a deal may be possible. However, even if the US were to agree to a minimum tax proposal that grandfathered GILTI, and to reform the broader allocation of residual profits for a suitable range of businesses, it is always a lengthy and challenging process to get even bilateral tax agreements through the US Senate. There is significant technical work that must be done before that could even start.

The proliferation of digital services taxes, the US’s response, and the European Commission’s 2020 Workplan confirming that it ‘stands ready to act if no global agreement is reached’ suggest that tax multilateralism will be under threat if the OECD negotiations falter. The pandemic raises additional political bandwidth and fiscal pressures.

Tax practitioners should consider the impact of these proposals – and of a failure of them – in terms of unilateral or regional measures. For example, if agreement could not be reached and the EU were to go ahead with France and Germany’s preferred Pillar Two proposal, then unless other countries aligned they could find themselves subject to the Undertaxed Payment Rules. Or a proliferation of digital services taxes could provoke additional non-tax trading barriers with broader impacts. However complex the proposals are, a lack of agreement could spark tax chaos. It is not too late to engage but the window is closing fast.

See the CIOT’s response to OECD Blueprints for Pillars One and Two on p49 of this issue.