This article considers the rules implementing the OECD’s Pillar Two proposals, including consideration of the concepts of Effective Tax Rate, the Income Inclusion Rule, the Undertaxed Payments Rule and the Subject-to-Tax Rule, as well as highlighting some of the aspects that are likely to raise issues.
What is Pillar Two? How and why did we get here?
It has been something of a long and meandering journey, but the origins of Pillar Two and the global anti-base erosion (GloBE) rules can be traced back to the OECD’s base erosion and profit shifting (BEPS) project, which began in 2013 with the publication of Addressing Base Erosion and Profit Shifting. In this report, the OECD considered the use by multinational enterprises (MNEs) of legal constructs to shift profits cross-border, away from jurisdictions in which economic activities took place or value was created (typically, higher tax jurisdictions) and into lower tax jurisdictions.
The report led to a number of further proposals for action, the first of which looked to address the tax challenges arising from the digitalisation of the economy. There was a concern that large MNEs were contriving to ensure that profits arose in low-tax jurisdictions artificially, and that these issues were exacerbated by the new opportunities and freedoms conferred by a digital economy. Existing rules relying on concepts of residence and permanent establishment had been designed in the context of a more traditional and physical economy (most famously, the sale of champagne to thirsty customers in the UK), and it was felt that these were inadequate to deal with businesses operating remotely and with minimal physical presence.
In total, 15 BEPS Actions emerged. These prompted a great deal of work, and a raft of reforms were implemented globally as a result. However, the difficulties posed by digitalisation were considered to be particularly challenging, and work on these continued, ultimately leading to political agreement, in October 2021, on the so-called “two-pillar solution”. To date, 138 member jurisdictions of the OECD/G20 Inclusive Framework on BEPS (Inclusive Framework) have committed to the approach.
Broadly, Pillar One reallocates a portion of taxing rights over the residual profits of large MNEs in favour of market jurisdictions and Pillar Two (which is the focus of this article) looks to impose a 15% effective minimum corporation tax rate on the profits of in-scope MNEs.
Progress on this dynamic duo of reforms has, for a number of reasons, developed at different rates. So, while substantive agreement by the Inclusive Framework members on the scope and detail of Pillar One has yet to be reached, Pillar Two is now well on the road towards implementation.
This article is confined to the umbrella Pillar Two principles given by the OECD Model Rules, supplemented by the OECD Commentary published in March 2022 and Agreed Administrative Guidance published in February 2023. However, the reality is that the story does not end there.
Parallel EU developments exist in the EU Directive on Minimum Taxation Council Directive (EU) 2022/2523 of 14 December 2022 (EU Directive), which sets out similar (but not identical) reforms. Each of these will need to be implemented by individual jurisdictions within the context of their own tax frameworks. The UK and the Netherlands have each published draft legislation and other jurisdictions are following suit. There is so much more fun to be had.
What is Pillar Two trying to achieve?
The underlying intention of Pillar Two is reasonably straightforward – MNEs must pay a minimum effective tax rate (ETR) of at least 15% in every jurisdiction in which they operate. It is hoped that this will end the risk of a “race to the bottom”, with jurisdictions trying to compete to attract MNEs through lower tax rates and also lower the risk (as noted above) of MNEs structuring their affairs in a contrived manner to try to house profit-making activities in lower tax jurisdictions.
It is important to note at the outset that the OECD Model Rules are not mandatory – Inclusive Framework jurisdictions are not obliged to implement them. However, if they do so, they must adopt the common approach prescribed; where they do not, they must respect the application of the rules applied by other jurisdictions. The hope is that a common global standard will avoid a proliferation of conflicting domestic regimes. By contrast, the EU Directive is mandatory and member states must enact domestic law to implement its rules.
Broadly, the intention is that Pillar Two will be effected through a series of interlocking measures comprising an income inclusion rule (IIR) and an undertaxed payments rule (UTPR) (together, the GloBE rules), each described in further detail below. A subject-to-tax rule (STTR) will provide additional support (see Question 6 below).
Which MNEs are in scope?
In terms of scope, the Model Rules apply to those multinational groups and their constituent entities that have annual gross revenues equal to or exceeding EUR750 million in the consolidated financial statements of the ultimate parent entity (UPE). All entities, even Excluded Entities (see Question 7), are included for the purposes of calculating the gross revenue. Further, no deduction is allowed for amounts due to minority interest holders.
For determining whether a multinational group is in scope, it is the four fiscal years preceding the one being tested that are relevant; if the MNE’s gross revenues in at least two of the four preceding fiscal years meet the threshold, then it is within scope. In other words, the test is backwards-looking only. What happens in the current fiscal year – i.e. the year being tested - is not relevant.
Do I need to be an accountant to deal with this?
No. But you are going to have to read the accounts, since this is where you will find much of the information required.
Are you sitting comfortably..?
The initial figures for the calculation of the ETR will come from the consolidated financial statements. Where the MNE is not required to produce consolidated accounts (for example because local law does not require it), the Model Rules provide a deemed consolidated test which requires the preparation of consolidated accounts based on an authorized financial accounting standard (the MNE may choose the standard used).
Then we’ll begin.
The first stage is therefore to calculate, for each jurisdiction, the ETR. This is achieved by dividing “Adjusted Covered Taxes” by “GloBE Income or Loss”. Each of these concepts is determined, first, using information from the MNE’s consolidated financial statements, which is then subjected to certain prescriptive adjustments set out in the Model Rules. Where the ETR for a jurisdiction is less than 15%, a top-up tax is required.
The top-up tax is applied to the GloBE Income less an amount given for the Substance-Based Income Exclusion (SBIE). This gives the top-up tax which must be paid, although any Qualifying Domestic Minimum Top Up tax (QDMTT) can be credited against this (yet more acronyms and more on these (and the acronyms) later). That top-up tax must then be collected from the group, but the incidence of imposition can vary.
The Model Rules then set out a bespoke schema of adjustments to the figures given by the accounts. An article format that contemplates 20 questions does not permit a full analysis of the various permutations. However, in relation to the Covered Taxes, the adjustments that must be made include adjustments to the deferred tax provision and taxes reflected as OCI (Other Comprehensive Income) in the financial statements, uncertain tax positions and refundable tax credits.
In relation to the calculation of GloBE income, adjustments include deductions for certain excluded dividends (other than those in relation to short-term portfolio shareholdings), other modifications for stock-based compensation, fair value accounting, net asset gains and intra-group transactions, tax transparency elections for investment entities and industry-specific adjustments.
Many of the modifications can be adjusted by the MNE by written election. These elections are often irrevocable for a period of time (usually five years). These decisions therefore require careful consideration and often modelling and sophisticated software.
Who pays the top-up, and how?
The primary mechanism for collecting top-up tax is the IIR, and the starting position is that the top-up tax will be paid by the UPE. However, this is not always the case. For example, where the UPE is not in the jurisdiction applying the GloBE rules, it can be allocated among other group entities.
To the extent that the application of the IIR is not sufficient to impose the top-up tax payable, the UTPR provides a back-up. An insufficiency might arise, for example, if the UPE jurisdiction does not itself apply an ETR of at least 15%.
Any top-up tax due under the UTPR will be allocated among the countries in which the group has operations in proportion to the group’s payroll costs and tangible asset value in each. Those countries will then collect the tax by attributing a deemed additional cash tax expense.
There is considerable flexibility as to what form the UTPR may take, and in fact there need not be any payment, let alone a payment that is undertaxed. The UTPR could be effected either by denying deductions to, or imposing a separate charge on, group entities resident in the relevant territory. If necessary, the additional cash tax expense can be allocated to future fiscal years.
What is the Subject-to-tax rule?
The STTR is essentially a source-based rule that allows a source jurisdiction to impose a withholding tax, or deny treaty benefits, on certain cross-border payments, such as interest, royalties, or fees, if the payment is subject to an ETR below the minimum threshold in the recipient jurisdiction.
It is meant to address the situations where the IIR and the UTPR do not apply, such as when the payer and the recipient are not part of the same MNE group, or when the payment is made to a jurisdiction that has not enacted the Pillar Two framework.
In fact the STTR is not actually part of the GloBE rules (which strictly comprise the IIR and the UTPR only). However, it is part of the Pillar Two package.
It is anticipated that the STTR will be effected by means of bilateral tax agreement, but this is not necessarily the case, and implementing jurisdictions are given considerable flexibility in determining its structure.
Are any types of companies exempt?
Yes, there are a variety of exclusions that the scope of which, given the consequences of falling within the Pillar Two regime, may well be the subject of some close attention.
First, there is a list of Excluded Entities is set out in Article 1.5 of the Model Rules. These include:
- Government Entities (the Agreed Administrative Guidance confirms that these can include sovereign wealth funds).
- International Organisations.
- Non-profit Organisations.
- Pension Funds.
- Investment Funds and Real Estate Investment Vehicles that are also ultimate parents.
Other than in relation to pension funds, each of these Excluded Entities can also include their underlying holding vehicles (on the basis of either 95% or 85% ownership, depending on the nature of the entity’s income).
The significance of being an Excluded Entity is that the relevant entity is not subject to top-up tax under either the IIR or the UTPR. Neither is it subject to any administrative obligations, such as filing returns under the GloBE rules. However, the entity will have its revenue included in the group's revenue for the purpose of determining whether the MNE meets the EUR750 million threshold for the GloBE rules to apply.
It is possible for the filing entity to elect not to be treated as an Excluded Entity.
There is also a de minimis exclusion, applicable on a jurisdiction-by-jurisdiction basis, which provides that there will be no top up tax for a fiscal year if the average GloBE revenue in that jurisdiction is less than EUR10 million and average net GloBE income or loss is a loss or is less than EUR1 million in the current and two preceding fiscal years. Again, it is possible for the filing entity to elect that the exclusion should not apply.
An SBIE rule and a QDMTT operate to reduce top-up tax in slightly different ways and are addressed in Questions 8 and 9 (respectively).
Finally, there is also a sector-specific exclusion relating to International Shipping Income (as defined) and a transitional exclusion for MNEs in the initial phase of international activity.
What is the Substance-Based Income Exclusion Rule?
The intention behind the SBIE is to carve out from the rules an amount of income in respect of an entity’s substantive activities (i.e. where they have an active physical presence). The rationale is that these types of activities are less likely to be associated with profit-shifting. The rule operates by reducing the GloBE income to which the top-up tax rate is applied by an amount calculated by reference to substantive activities. In other words, its effect is not to alter the ETR, but to reduce the amount of profit subject to top-up tax.
The deduction comprises two amounts, the first in respect of payroll costs of employees in the relevant jurisdiction and the second linked to the carrying value of eligible tangible assets.
Again, a filing entity can elect not to apply the SBIE.
What is a Qualified Domestic Minimum Top-Up Tax?
The idea of introducing domestic minimum top up taxes appeared early on in the development of Pillar Two. From the perspective of a particular jurisdiction, if the effect of Pillar Two is that the profits arising in relation to the activities of an MNE in that jurisdiction are going to be subject to a top up tax somewhere else (under the IIR or UTPR), why not introduce a mechanism to increase the tax rate in that jurisdiction and keep the additional tax revenues themselves? However, it was only relatively recently, when the administrative guidance was published in February 2023, that additional detail has emerged as to how what form a QDMTT should take.
Specifically, the administrative guidance states that a QDMTT must be consistent with the design of the GloBE rules and provide for consistent outcomes. The Inclusive Framework will develop a multilateral review process to determine this.
Tax paid by an entity under a QDMTT will be fully creditable against that entity’s GloBE liability. In effect, a QDMTT will take priority over any top-up tax imposed under the GloBE rules. The result is that a jurisdiction with a QDMTT will be the first to receive any top-up tax from entities within that jurisdiction. Absent the QDMTT, that tax liability might otherwise be payable in another jurisdiction under the GloBE rules. By contrast, a non-qualified domestic minimum top-up tax can only be taken into account as a Covered Tax, i.e. at the numerator level, which will have reduced significance in the calculation of any top-up tax.
It is also anticipated that a form of QDMTT safe harbour will be implemented, allowing compliance simplifications for MNEs operating in a jurisdiction that has adopted a QDMTT.
Given the advantages of implementing a QDMTT, it might appear odd that any jurisdiction would not choose to introduce one. Certainly it is likely that the tax administrations of most developed jurisdictions will seek to introduce a QDMTT. However, other less developed tax administrations may not have the resources to do so. This may result in other jurisdictions receiving the benefit of increased tax revenues and raises very interesting policy questions about the structure of Pillar Two in the context of lower income and developing jurisdictions. We understand that the OECD is considering these issues specifically.
The EU Directive on the implementation of Pillar Two allows for the imposition of domestic minimum taxes by member states. Both the UK and the Netherlands have confirmed that their domestic legislation will include the rule.
What about safe harbours?
The question of safe harbours has been, and is still, even at February 2023, a lingering uncertainty.
Again, there were hints of safe harbours relatively early on in the process, but these were not developed further until the publication by the OECD, in December 2022, of its Guidance on Safe Harbours and Penalty Relief. This document advanced the issue, but the full position is still not clear.
What has been offered is a Transitional Country-by-Country Reporting (CbCR) Safe Harbour, based on qualifying CbCR and financial data, covering (broadly) the first three years of the GloBE rules. During this transitional period, the top-up tax in a jurisdiction for a fiscal year is deemed to be zero provided that the relevant entities meet one of the following tests:
- A “de minimis test” (MNE has total revenue less than EUR10 million, and profit (loss) before tax less than EUR1 million based on CbCR).
- A “Simplified ETR test” (MNE has a simplified ETR greater than or equal to a Transition Rate starting at 15% and rising to 17%).
- A “routine profits test” (MNE has a profit or loss before tax less than or equal to the “Substance-based Income Exclusion amount” for constituent entities resident in that jurisdiction under CbCR, as calculated under the GloBE rules).
Looking beyond the Transitional CbCR Safe Harbour, the OECD has said that it is exploring options for permanent safe harbours (but these are not yet confirmed).
The first proposal for a permanent safe harbour is a “Simplified Calculations Safe Harbour”. This would simplify compliance with Pillar Two on an ongoing basis, by reducing the number and complexity of calculations MNE groups would be required to make. The proposal would follow a similar approach to that set out in the Transitional CbCR Safe Harbour, save that it would rely on alternative, simplified GloBE calculations.
Secondly, consideration is also being given to the implementation of a QDMTT Safe Harbour, which would remove the need for the MNE to perform a GloBE calculation where a QDMTT calculation had been required under local law.
Although the direction of travel is encouraging, we know that a number of MNEs are frustrated that planning cannot adequately be undertaken, even at this late stage.
Is Pillar Two just another CFC rule?
Yes. And No.
Certainly, there are similarities between Pillar Two and the types of controlled foreign companies (CFC) rules implemented by many jurisdictions as anti-avoidance measures designed to protect their tax bases (and not just in the Byzantine complexity of the drafting). Typically, a CFC will be defined as a foreign entity owned or controlled by residents of a certain jurisdiction, and that earn income from low-tax or preferential regimes. The purpose of CFC rules is generally to prevent residents from shifting profits to CFCs and deferring or reducing their domestic tax liability. To this extent, the GloBE rules are similar in effect.
CFC rules vary across jurisdictions in scope and effect, but generally they involve two steps. The first is to identify whether a foreign entity is a CFC; the second is to attributae a portion of its income to its resident owners or controllers and tax them accordingly. The rules often distinguish between different types of income.
By contrast, the GloBE rules do not seek to identify control. Instead, scope is determined by reference the consolidated financial statements of the MNE group. Further, CFC rules generally apply on an entity-by-entity basis while the top-up tax imposed by the GloBE rules applies on a per jurisdiction basis.
How are incentives and reliefs treated? (The surprise low-tax jurisdiction)
The effect of government reliefs and incentives means that there may be some nasty surprises. Depending on the nature of the relief and the relative quantum of relieved income, it is still possible for an MNE to be subject to a top-up tax even where the headline corporation tax rate is significantly higher than 15%.
In the context of the UK, an obvious example of where this might arise is the patent box which (where it applies) gives a tax rate of 10% on profits derived from patents. Essentially this is a UK government policy decision – it wants to encourage investment in innovation and will reward companies that develop and exploit IP in the UK.
However, if a very large part of a company’s income derives from patent income, this might produce an ETR of under 15%, thereby triggering a top-up tax. Clearly this will reduce or even eliminate the appeal of the incentive, contrary to the government policy.
However, the Model Rules do contemplate a “Qualified Refundable Tax Credit”. Broadly, these are defined as incentives which are structured as tax credits paid as cash or cash equivalent to the recipient within 4 years of satisfying the conditions for receiving it. These look more like subsidies – and the advantage is that these are treated as income in the ETR calculation and not as a repayment of or reduction in taxes, producing a more favourable outcome. It may be that jurisdictions move to structuring tax incentives in this way to avoid degrading the appeal of the incentive.
We will have to wait and see what commercial and political response this might produce in practice. Of course, Pillar Two will only apply to the largest companies, so the reliefs will still be available and effective for others in their current form.
How will the rules work for funds and investment companies?
The treatment of funds and investment companies under the Model Rules has been an area of focus for many. Clearly, investment vehicles are not all created equal and everything will depend on the facts. However, there are some general statements that can be made.
First by way of recap, the GloBE rules only apply to MNEs with annual revenue of at least EUR750 million in the consolidated accounts. The question of consolidation is therefore key. Many collective investment vehicles are not required to file accounts that are consolidated with their portfolio companies, so may well fall out of scope on that basis.
Even where investment entities are potentially within scope, they may be Excluded Entities, but only where they qualify as an “Investment Fund” (as defined) and also constitute the UPE (see Question 7 above). The definition of Investment Fund includes requirements that the entity is designed to pool assets from a number of investors, it invests in line with a defined investment policy and is managed by investment professionals.
If an investment entity is within scope and not excluded, there are a number of specific adjustments to the accounting values which may be of particular relevance. In particular, distributions made in respect of investments held over 12 months are excluded from GloBE income (although distributions payable from short-term portfolio holdings are included). It is also possible in certain cases in relation to the disposal of assets to make elections out of the fair value accounting position that applies by default, and instead elect for a realisation principle to apply. It will be important for investment companies to consider these provisions and assess whether an election might be merited in the circumstances.
How will developing nations respond?
Developing jurisdictions face a number of difficulties in implementing the GloBE rules.
Perhaps the most obvious is that their tax systems and infrastructure tend to be relatively underdeveloped, which makes it difficult and possibly prohibitively expensive to superimpose the very complex GloBE rules over existing tax systems. The STTR regime is intended to provide a more straightforward mechanism to allow them to increase their share of MNEs’ global tax payments if it is not practical to impose the GloBE rules. However, in general terms, it will undoubtedly present political challenges if the effect of Pillar Two is to increase the tax revenues in the headquarter jurisdictions of the largest MNEs without also benefiting the jurisdictions in the Global South in which they operate.
A further challenging issue in the context of extractive industries is the existence of stabilisation agreements between a host nation and (for example) an MNE investing in that jurisdiction. These agreements often “freeze” taxes for a certain period, or offer beneficial terms, so that the company will commit to the investment. This raises a number of issues in the context of Pillar Two.
In particular, the way that Pillar Two is structured means that the taxing rights to any additional tax payable (due as a result of low taxation in the host nation) are given first to the jurisdiction of the UPE – usually in a high tax jurisdiction. This means that the jurisdiction misses out unless it renegotiates the agreement and/or enacts complex domestic legislation. A further issue whether the agreement prohibits the jurisdiction from altering its tax code in any case.
How will disputes be resolved?
The Pillar Two implementation framework contains certain mechanisms that are designed to prevent inconsistent outcomes under the GloBE rules. The OECD Commentary and administrative guidance is also intended to ensure coordination and greater certainty in the application of the rules. Further administrative guidance is expected to be published on a “rolling basis”.
Nonetheless, the Pillar Two rules must be implemented through domestic legislation in each participating jurisdiction. Even with the OECD’s Model Rules as a base, there will undoubtedly be differences as between jurisdictions – both in how the rules themselves are drafted, but also in how they are interpreted and applied. This has been the case in relation to other regimes derived from the BEPS Action Plans, for example the hybrids rules introduced in accordance with BEPS Action 2 (and ATAD 2 in the EU). This creates the possibility, if not the inevitability, that disputes related to unintended consequences, or inconsistent / uncoordinated application, of the GloBE rules will arise. The OECD is currently exploring options to address the risk of disputes arising under the GloBE rules, and how best to resolve them.
At present, there are frameworks for the resolution of disputes arising under double tax treaties, including the Mutual Agreement Procedure (MAP). However, MAP provisions (at least those based on the OECD Model Tax Convention) generally allow taxpayers to initiate MAP proceedings only where it is claimed that the actions of one or more jurisdictions results in “taxation not in accordance with” a relevant tax treaty. It is not obvious that disputes concerning the GloBE rules will necessarily involve alleged taxation otherwise than in accordance with a tax treaty. In such cases, MAP would, therefore, be of limited assistance.
In some double tax treaties, provision is also made for taxing authorities to consult on the elimination of double taxation in cases not provided for in the relevant treaty. However, it is not necessarily the case that disputes involving the GloBE rules must involve an allegation of double taxation; a dispute might instead relate to over-taxation in a given jurisdiction, for example. In any event, this mechanism (if it is included in a given tax treaty at all) is discretionary and would not directly involve the affected MNE.
Accordingly, relying on existing double tax treaty dispute procedures alone is unlikely to be a complete solution to addressing disputes arising under Pillar Two.
In light of this, thought is being given to what alternative dispute resolution mechanisms, outside existing double tax treaty frameworks, might be available. One option being actively considered by the Inclusive Framework members is the creation of a multilateral convention. This convention would likely concern GloBE-related administration issues generally, but could specifically provide for a dispute resolution procedure. Such a procedure might allow an MNE to request that relevant taxing authorities resolve any issues related to unintended or uncoordinated taxation under the GloBE rules by reference to a common global standard (such as the OECD Model Rules and Commentary).
Other avenues being explored include the Convention on Mutual Administrative Assistance in Tax Matters, which facilitates jurisdictions entering into bilateral tax information exchange agreements. A further possibility is a standardised dispute resolution mechanism at the domestic law level, which could apply on a reciprocal basis – that is, where the relevant jurisdictions concerned have each incorporated the same provision in their domestic law.
How should MNEs prepare?
Quickly.
There has been a great deal of concern about the arrival of Pillar Two, and much of it is as a result of the administrative and compliance burdens that the rules entail. The volume of data required to run the calculations and develop a sensible strategy (for example in relation to elections etc) is considerable and the exercise will be expensive in terms of time and cost - even where no tax ultimately payable.
In particular, it may be the case that the data required for Pillar Two purposes is not required for any other purpose so that even sophisticated existing systems will be inadequate and new (and expensive) systems must be developed. Further, the data may not be readily available. Historically, accounting standards have not always required MNEs to collect data on a jurisdiction-by-jurisdiction basis.
System preparedness and functionality is not just a problem going forward, either. Whether a group is in scope is determined by considering the previous four years and so systems must be in place already.
What are the implications for current transactions?
There is also the even more immediate question of the allocation of risk in existing transactions, whether M&A, joint ventures, raising capital through debt or equity and anything else under the sun. The market is currently grappling with what a reasonable allocation of Pillar Two risk might look like in all of these very different scenarios. The difficulties are exacerbated by the fact that it is not yet known how different jurisdictions will implement the rules. In particular, is there a risk of secondary liabilities within a group? This lack of certainty makes it very difficult to min
