In summary

This Tax Insight highlights some of the key design features of the GloBE proposal and the newly important subject to tax rule (STTR) as set out in the Report on the Pillar Two Blueprint released on 12 October 2020 (Pillar Two Blueprint). We also consider how these proposals might look from an Australian perspective, in particular how they are expected to interact with Australia’s litany of international tax integrity rules.

In detail

The income inclusion rule (IIR) and its backstop, the undertaxed payments rule (UTPR), together known as the “GloBE proposal”, are the key proposed changes to domestic law under the OECD’s Pillar Two proposals, aimed at subjecting multinational enterprise (MNE) groups to a minimum effective tax rate (ETR), the rate for which is yet to be politically agreed.

The other two proposals in Pillar Two, the switch-over rule (SOR) and subject to tax rule (STTR) are treaty based rules and have generally received less attention. The Pillar Two Blueprint clarifies that the SOR is merely part of the GloBE rather than a free standing rule, while the STTR is a free standing rule that has been elevated to front of the queue in the priority order of the rules, even though it is much less developed and discussed last in the Pillar Two Blueprint.

The Pillar Two Blueprint is incredibly comprehensive and shows that there has been much progress made since the GloBE public consultation documents were released in November 2019. Highly technical and complex, the Pillar Two Blueprint sets out a flow diagram intended to provide a high-level overview of the process for applying the GloBE in the simplest case of 100% ownership for the whole group, showing that no less than 17 steps and decision points are involved! Some of the key steps include:

  • Identifying which entities are in-scope;
  • The ETR calculation, including carry-forward adjustments and the substance-based carve-out; and
  • Identifying who pays the top-up tax.

This Tax Insight highlights some of these key design features as well as providing details on the newly important STTR. In addition, we also provide some commentary on how these Pillar Two proposals might look from an Australian perspective.

In-scope entities

MNE Groups subject to country by country reporting (CbCR) obligations under BEPS Action 13 will generally be subject to the GloBE proposal i.e. groups which are consolidated for financial accounting purposes that have annual consolidated revenue of EUR 750 million or more in the immediately preceding fiscal year.

However importantly, certain MNE Groups headed by investment and pension funds, and government entities such as sovereign wealth funds, international organisations and not-for-profit bodies are excluded from the GloBE proposal. The Pillar Two Blueprint sees this exception as important to ensure that the GloBE proposal does not undermine a domestic jurisdiction’s policy objectives for providing tax exemptions and/or ensuring tax neutrality for these types of entities.

‘Constituent Entities’ of an in-scope MNE Group are then required to be identified. A ‘Constituent Entity’ includes not only business units included in the MNE Group’s consolidated financial statements but also:

  • any separate business unit that would be so included in the MNE Group’s financial statements if the ultimate parent of the MNE Group were traded on a public securities exchange, or if the financial accounting materiality exclusion did not apply (additions only recently included in the Australian CbCR domestic legislation); and
  • permanent establishments (PEs) of such business units, but only if such PEs prepare separate financial statements for financial reporting, regulatory, tax reporting, or internal management control purposes.

A Constituent Entity will generally be located in the jurisdiction of its tax residence (and therefore its income will be allocated to that jurisdiction) – again, consistent with the CbCR rules. However, special rules are necessary for choosing which jurisdiction to assign income relating to PEs, joint operations, entities resident in a jurisdiction with no corporate tax system, tax transparent entities, hybrid and reverse-hybrid entities and dual resident entities.

The Pillar Two Blueprint also contemplates that where certain income of entities is deemed to be stateless, this may be allocated to a stateless jurisdiction requiring a stateless jurisdiction ETR and top-up tax to be calculated.

Effective tax rate calculation

The ETR calculation under the GloBE proposal is done on a jurisdictional basis and serves two purposes: first, it identifies which jurisdictions are ‘low-tax’ and therefore acts as a trigger point for the application of GloBE, and then if the relevant jurisdiction is a low-tax jurisdiction it is used to compute the top-up tax payable.

The basic ETR calculation is determined by dividing the amount of covered taxes by the amount of income/tax base for each Constituent Entity on a jurisdictional blending basis.

Covered tax

A covered tax is defined to mean “any tax on an entity’s income or profits (including a tax on distributed profits), and includes any taxes imposed in lieu of a generally applicable income tax.” Covered taxes also include taxes on retained earnings and corporate equity.

This definition of a covered tax is intended to not be overly technical by focussing on the underlying character of the tax rather than specific design features of the tax. For example:

  • Taxes imposed on the net income from specific activities such as banking or exploration and production of oil and gas would qualify as covered taxes (but not resource levies linked to the quantity, volume or value of resources extracted unless they are imposed ‘in lieu’ of a generally applicable income tax).
  • The “in lieu of” test generally includes interest, rent and royalty withholding taxes as well as other taxes on gross payments such as insurance premiums but on the proviso that such taxes are imposed in substitution for a generally applicable income tax. It also includes taxes imposed on an alternative basis, such as taxes on deemed return for investments in foreign equity or tonnage taxes.
  • Where a country has an imputation regime, such as Australia, taxes paid are generally not reduced where taxable resident shareholders are entitled to credits for or a refund of corporate taxes in respect of distributions paid.

The assignment of tax paid to the relevant jurisdiction will be important given the ETR is determined on a jurisdictional basis. Covered taxes are generally assigned to the jurisdiction where the income arises but not always. For example, dividend withholding taxes are assigned to the country where the paying company is resident but other withholding taxes paid are assigned to the jurisdiction where the withheld income is received; and taxes paid by parent entities in accordance with controlled foreign company (CFC) rules are assigned to the jurisdiction in which the underlying income arises.

Taxes that are excluded from being covered taxes include consumption and sales taxes, excise taxes, digital services taxes, stamp and other transfer taxes, payroll taxes and social security contributions, and property taxes.

Tax base

The Pillar Two Blueprint notes that the starting point for the GloBE tax base is the profit (or loss) before tax in the financial accounts of each Constituent Entity in the MNE Group prepared in accordance with the financial accounting standards used by the parent entity to prepare its consolidated financial statements. However, the Pillar Two Blueprint then proceeds to acknowledge that in many cases a subsidiary is unlikely to produce an income statement on a stand-alone basis that would comply with the parent entity’s accounting standards – accordingly if it is reasonable to do so, MNE Groups can rely on entity-level financial information instead.

For PEs, their profit or loss is determined based on the income and expenses that are treated as arising for tax purposes in the jurisdiction where they are located (which may include dealings between the permanent establishment and head office for countries which adopt the OECD’s functionally separate entity approach). The kinds of issues raised for hybrid mismatch rules by Australia’s somewhat peculiar position of not directly recognising PE dealings, but effectively doing so by a ‘proxy’ approach especially for local banks, are likely to come up again in Pillar Two.

Other adjustments to the tax base being contemplated include:

  • Intercompany items. It is noted that these should be recorded in accordance with the arm’s length principle. Further these are excluded only to the extent the transaction is between group members in the same jurisdiction i.e. transactions with group members outside of the jurisdiction are treated the same as transactions with external parties – a kind of domestic consolidation regime for the jurisdiction concerned.
  • Adjustments for permanent differences (note that adjustments for temporary differences are generally dealt with as separate ‘carry-forward adjustments’ discussed below though there will likely be tax base adjustments for the most common and significant timing differences relating to tax expensing or accelerated depreciation for tangible assets). The Pillar Two Blueprint considers that the pertinent principles for selecting permanent differences are materiality and commonality across jurisdictions. Some examples include dividends and equity method profit (or loss), dispositions of company shares and fair value accounting gain (or loss), covered taxes, stock-based compensation expense, bribes and penalties, and adjustment for Pillar One outcomes which effectively gives priority to Pillar One over Pillar Two.

Carry-forward adjustments

Where it has been determined that a jurisdiction is a ‘low-tax jurisdiction’ i.e. the jurisdictional ETR is less than the agreed minimum ETR, two further adjustments are made to the jurisdictional ETR calculation to determine whether top-up tax is payable.

The first adjustment is for carry-forward losses and excess taxes. This adjustment is intended to ensure that temporary differences in the timing of the recognition of income and expenses under financial and tax accounting rules do not result in a permanent GloBE tax liability. Due to integrity concerns with using deferred tax accounting to manage this issue, the carry-forward approach has been put forward as the preferred method as it relies on actual tax liabilities existing.

Under the carry-forward approach, where an entity pays excess taxes in a jurisdiction for a year, this may give rise to an IIR credit against future IIR tax payable if IIR tax has been paid previously in that jurisdiction, or a local carry-forward of excess tax expense to the extent that previous IIR tax has not been paid. A loss carry-forward is also allowed to reduce the GloBE tax base in a subsequent year, including certain losses incurred by a Constituent Entity prior to the MNE Group becoming subject to the rules.

‘Substance-based’ carve-out.

A substance-based carve-out further reduces the income of Constituent Entities in a low-tax jurisdiction by a routine return for ‘substantive activities’ within the jurisdiction i.e. the use of employees and tangible assets. Conceptually, this is aimed at focussing GloBE on income susceptible to BEPS risks, such as intangible-related income.

The payroll component of the substance-based carve-out is worked out by applying a fixed return by reference to the Constituent Entity’s employment costs. Identifying eligible employees, determining where they are located and calculating the relevant payroll expenses will be key for this component.

The tangible asset component of the carve-out is equal to the sum of a fixed return on:

  • depreciation of property, plant and equipment;
  • deemed depreciation of land;
  • depletion of natural resources; and
  • depreciation of a lessee’s right-of-use of a tangible asset.

The starting point for calculating these inputs to the tangible asset fixed return is conformity to the financial accounts used to compute the GloBE tax base, however there are some important exceptions to this - including in relation to the revaluation model, intercompany transactions and self-constructed assets.

It is likely that the covered taxes allocated to the jurisdiction will be scaled back proportionally to the extent that the substance-based carve-out reduces the tax base.

Who pays the top-up tax?

If, after the carry-forward and carve-out adjustments are made, the ETR for the jurisdiction is below the minimum rate, a top-up tax percentage is calculated, being the percentage that is necessary to bring that jurisdiction’s ETR up to the minimum ETR.

This top-up tax percentage is then applied to the GloBE income of each Constituent Entity in the low-tax jurisdiction to work out the top-up tax payable.

Under the IIR, a ‘top-down approach’ applies such that the ultimate parent entity at the top of the ownership chain in the MNE Group would pay the top-up tax. If the relevant jurisdiction has not implemented the IIR then responsibility for applying the IIR is given to the next entity down in the chain of ownership, and so on. However in the case of split-ownership structures where 10% or more of the equity interests in a group member are held by persons outside of the MNE Group, the obligation to pay the top-up tax is pushed down to the partially-owned intermediate parent and only to their share of the income.

In order to prevent a tax treaty requiring the jurisdiction levying the IIR to exempt the profits of a PE in another country which is a Constituent Entity, the SOR will apply to remove the exemption. This integration of the SOR into the GloBE means that the tax paid on PE profits will only be the top-up tax and not at the full tax rate in the parent jurisdiction. Consideration of applying the same approach to PEs of other Constituent Entities in the group and not just of the ultimate parent is ongoing. Where the PE exemption arises under domestic law as in Australia, signing up to the GloBE will likely mean amendment of that exemption in a similar way (see further below).

The UTPR only applies where an entity is not already subject to the IIR and that entity is located in a jurisdiction that has implemented the UTPR. The top-up tax is allocated based on two allocation keys:

  • The first allocation key allocates top-up tax to entities which make deductible payments to low-tax group members, in proportion to the total of deductible payments made to low-tax group members by all group entities subject to the UTPR; and
  • The remaining top-up tax unallocated under the first allocation key, is allocated to entities subject to the UTPR in proportion to the total amount of net intra-group expenditure incurred. This second allocation key is intended to prevent MNE Groups from structuring around the first allocation key.

If a Constituent Entity is resident in the same jurisdiction as the ultimate parent entity and the ETR in that jurisdiction is below the minimum rate, the UTPR also applies to the income in that jurisdiction if it has signed up to the GloBE. In this case a cap on the UTPR applies based on the amount of deductible intra-group payments received from Constituent Entities outside that jurisdiction by low-tax Constituent Entities within that jurisdiction, with a possible integrity rule to prevent structuring designed to take advantage of the cap.

Subject to tax rule

The STTR is described as the complement of the GloBE rules, as a treaty based rule targeting risks to source countries from intra-group payments subject to low nominal rates of taxation in the other country. The rationale offered for this additional rule is that source taxing rights ceded by tax treaty should be recoverable up to the agreed minimum tax rate where intra-group payments are subject to no or low taxation in the recipient’s jurisdiction.

The STTR is designed to assist source countries with lower administrative capacity and was insisted on by developing countries as part of making Pillar Two acceptable to them. It will be implemented by a separate treaty provision codifying the STTR.

The rule will essentially apply payment-by-payment which means no jurisdictional or entity blending. The definition of connected person in tax treaties (inserted by BEPS 1.0) will be used to identify intra-group payments and the payments covered will extend to interest, royalties, or for the provision of capital, assets or risk (eg payments for captive insurance). It is a top-up tax like the GloBE and will exclude similar recipients as the GloBE. A materiality rule for its application is under consideration.

The design details are in the early stages as the STTR has not attracted the same amount of work as the GloBE. It will be complex but less so than the GloBE and as noted earlier will take priority to the GloBE. It remains to be seen how important it will be in the context of Pillar Two. The GloBE gives the first taxing right to the ultimate parent jurisdiction of an MNE Group and so inevitably favours developed countries. The STTR has the potential to reverse that outcome in favour of source (including developing) countries though how far will depend on the extent to which the treaty networks of affected countries are being used to reduce source taxation.

Impacts for Australia

It remains to be seen how much Pillar Two will add to Australia’s tax coffers, assuming that Australia signs on to whatever consensus is reached as seems likely. The IIR could collect tax from the few significant MNEs headquartered in Australia but generally not from inbound MNEs, unless they are headquartered in low-tax jurisdictions so that Australia may be able to apply the UTPR – several such MNEs are headquartered in the Asian region. It is also possible that the STTR could apply to inbound MNEs as Australia does have a significant source tax base and the STTR takes priority over the IIR.

It is expected that implementation of Pillar Two in Australia will be similar to the implementation of the hybrid mismatch rules in BEPS 1.0 Action 2 where the OECD developed detailed legislation-like recommendations which Australia closely followed, though with differences in relation to CFC and PE mismatches. The OECD proposes to develop detailed model legislation and explanatory material for the GloBE and treaty provisions for the SOR and STTR. While it considers that the support of a new treaty is not essential for the GloBE proposal the OECD may promote a multilateral treaty setting out the basic GloBE principles to ensure more uniform domestic implementation. If this occurs Australia will have little room to manoeuvre but if not it may adopt a variation as occurred with hybrid mismatches. Timing is not specified for the start of Pillar Two but realistically it is likely to be 2024 at the earliest and perhaps later.

Australia already has several significant tax regimes targeted at international tax planning so issues will arise about the interaction of those regimes with Pillar Two, some of which are answered in the Pillar Two Blueprint. Although the IIR has similarities to a CFC regime, Pillar Two does not see any conflict as they have different purposes. As already noted any tax under Australia’s CFC regime will count in determining the ETR for foreign jurisdictions. In other words the CFC regime has priority over the GloBE. The same is true for any Australian tax collected on income of a foreign PE of an Australian company but as noted above to be consistent with the policy underlying the GloBE the extra tax should only be sufficient to bring the total tax up to the agreed minimum tax rate rather than the general corporate tax rate.

The general Australian PE exemption and CFC rules also already operate with in effect an IIR-like ‘substance-based’ carve-out, by broadly only requiring top-up tax to the Australian corporate tax rate to be paid where unlisted country companies and branches fail the active income test. One area in which Australia could see some upside from the GloBE is where such income is derived from listed countries (Canada, France, Germany, Japan, New Zealand, the UK and USA) but subject to low-tax there and still not taxed in Australia. The OECD economic impact assessment for BEPS 2.0 suggests there is a considerable amount of profits in low-tax pockets in otherwise high tax countries.

The diverted profits tax and multinational anti-avoidance law, which were introduced in Australia’s initial responses to BEPS 1.0 already serve to prevent MNE Groups from structuring out of Australia to take advantage of low-tax structures. These have elements of the UTPR about them in that they effectively collect tax on a source basis in response to tax planning structures which deny Australia’s taxing rights such as avoidance of PEs in treaty situations or stripping the Australian tax base by deductible payments to low-tax jurisdictions. Australia also effectively has a UTPR in the form of the hybrid mismatch targeted integrity rule introduced as an add-on to Australia’s implementation of the BEPS 1.0 Action 2 hybrid mismatch report, which prevents companies from claiming interest deductions paid via an interposed entity where they are not taxed at a rate exceeding 10%.

These existing rules are less formulaic than the Pillar Two proposals as they require a finding of an objective purpose to avoid tax and if left untouched by Pillar Two essentially reverse its priority ordering of the IIR and UTPR. Or perhaps they could be viewed more like an STTR which would also suggest giving them priority over the IIR. As noted above there is some discussion in the Pillar Two Blueprint about a treaty based approach to support the GloBE and one of the concerns seems to be countries’ effectively doing an end run around Pillar Two. Already the US has been politely requested in the Pillar Two Blueprint to not apply its Base Erosion Anti-Abuse Tax (or BEAT) if the IIR of another country applies to the same income and similar requests may be made to other countries including Australia.

Next steps

The Pillar Two Blueprint offers a comprehensive starting point for the design of the GloBE and STTR rules; however, aside from the obvious question of what the minimum ETR will be (note that the OECD has modelled preliminary economic outcomes based on a minimum ETR of 7.5% up to 17.5%), there remain a number of items flagged for consideration including:

  • Interaction of the GloBE rules with the US GILTI and BEAT rules;
  • Development of integrity and anti-avoidance rules to prevent MNE Groups from restructuring out of the GloBE rules;
  • Whether simplification options should be pursued and their design; and
  • What will the impact of the STTR be on the GloBE, which will depend on design and in particular the implementation details for the STTR.

These issues, as well as a number of other design aspects of GloBE rules, have been included in the public consultation document released at the same time as the Pillar One and Pillar Two Blueprints. Responses to the public consultation document are due by 14 December 2020 with public consultation meetings due to be held in mid-January 2021.