Tax Challenges Arising from Digitalisation – Report on Pillar One Blueprint

7. Elimination of double taxation

7.1. Overview

556. Amount A will apply as an overlay to the existing profit allocation rules. As the profit of an MNE group is already allocated under the existing profit allocation rules, a mechanism to reconcile the new taxing right (i.e. calculated at the level of a group or segment), and the existing profit allocation rules (i.e. calculated at an entity basis) is necessary to prevent double taxation.

557. This is the purpose of the mechanism to eliminate double taxation from Amount A described in this chapter. To reconcile the two profit allocation systems, it identifies which entity or entities within an MNE group bears the Amount A tax liability, which effectively determines which jurisdiction or jurisdictions need to relieve the double taxation arising from Amount A. This mechanism is based on two components: (i) the identification of the paying entity (or entities) within an MNE group (or segment, where relevant); and (ii) the methods to eliminate double taxation.1

Component 1: Identifying the paying entities

558. Amount A will be calculated at a group or segment-level using a simple formula, incorporating a profitability threshold and reallocation percentage. For this reason, it could be argued that a similar formulaic approach should be adopted to identify entities that will bear the Amount A tax liability (the paying entities2) for example, as those entities with a PBT to revenue ratio in excess of the agreed profitability threshold. The process to identify the paying entities will include a profitability test. However, there are two technical reasons why the profitability test applied at entity level cannot be the same test as that applied to the group:

• Using the same formula to calculate residual profit at a group and entity-level could be distortive: The most logical starting point for this profitability test would be to use the same formula used to calculate Amount A at a group or segment level (PBT / revenue, in excess of x%). However, there are two reasons why this approach is considered not appropriate. First, entity-level accounts include intragroup transactions that are eliminated on consolidation. As a result, the consolidated revenue of a group will be less (potentially substantially less) than the total (unconsolidated) revenue of entities within the group. This means that where a group has a profit margin in excess of x%, it is at least theoretically possible that there will not be any entities within that group with a profit margin in excess of x%. Second, the inclusion in entity-level accounts of intragroup transactions means that an entity’s revenue and hence profit margin can be easily distorted or manipulated by the way intragroup transactions are structured.

• Differences between local GAAPs of group entities: At a group or segment-level, the use of consolidated accounts means that for most MNE groups the financial information necessary to compute the Amount A tax base and apply the Amount A formula will be readily available and prepared under a comparable accounting standard. In contrast, at an entity-level financial information will be less readily available, or may be prepared under local accounting standards, which means, even within the same group, the financial statements for different entities may not be comparable. That said, as discussed below, this issue could be overcome by using the entity-level accounts used by a group to prepare its consolidated group accounts.

559. Conceptually, different approaches can be considered. For example, consistent with the formulaic nature of Amount A, one option would be to identify the paying entities using a quantitative approach similar to that used to calculate Amount A. This approach would prioritise simplicity, as it could limit the need to enter into detailed facts and circumstances functional and transactional analyses using transfer pricing concepts to identify the entities that should bear the Amount A tax liability. This would be consistent with the fact that Amount A itself is designed independently of these considerations. Another option would start from the basis that conceptually, the different features of Amount A suggest that paying entities should not only be profitable, but should be those that (individually or collectively) make material and sustained contributions to an MNE group’s residual profits,3 rather than those that perform only activities that generate routine profits. In addition, from a conceptual standpoint it could be argued that a paying entity should perform activities that relate to an MNE group’s engagement in a market jurisdiction for which it bears an Amount A tax liability, as opposed to non-market related activities or activities that relate to other markets. These options could also be combined, in a variety of different ways, as outlined in the four-step process discussed below.

560. The process to identify the paying entities could have up to four steps.4 These are to:

• Step 1: Identify the entities within an MNE group that perform activities that make a material and sustained contribution to the group’s ability to generate residual profits.

• Step 2: Apply a profitability test to ensure the entities identified have the capacity to bear the Amount A tax liability.

• Step 3: Allocate, in order of priority, the Amount A tax liability to the entities that have a connection with the market(s) where Amount A is allocated.

• Step 4: Allocate, on a pro-rata basis, where no sufficiently strong connection(s) is (or are) found or where those with a market connection lack the necessary amount of profit.

Step 1: Activities test

561. An activities test would require that taxpayers undertake a qualitative assessment to identify the entities within the MNE group that make material and sustained contributions to a group’s residual profits. This reflects the fact that conceptually, it is these entities within the MNE group collectively that should earn the residual profits corresponding to Amount A profit.

562. Practically, the activities test would include a general principle describing the type of relevant activities that a paying entity would be expected to perform, and this would be supplemented by a list of indicia that would support the application of this principle. These indicia would include the functions, asset and risk (FAR) profile of an entity, an entity’s characterisation for transfer pricing purposes and the transfer pricing method used to determine the remuneration of an entity. Where possible, the documentation that taxpayers are already required to collect and report and that tax administrations already review will be used for this purpose.5 These existing documents and any additional documents that may be required would be included in the single standardised documentation package that would be submitted as part of the Amount A tax certainty process. This should limit any additional compliance costs and administrative burden arising from this test.

563. As part of the activities test, where an MNE group computes the Amount A tax base on a segmented basis, it would be necessary to identify the potential paying entities that relate to each segment.

Step 2: Profitability test

564. The profitability test would ensure that the potential paying entities have the capacity to bear the Amount A tax liability. The profitability test would ensure that an entity making routine, low profits or losses is not identified as a paying entity, and is consistent with the decision of the Inclusive Framework to limit the application of Amount A to in-scope MNE groups that earn residual profits, rather than all in-scope MNE groups irrespective of their profitability. For simplicity, this profitability test could be aligned with the substance carve-out that is being developed for Pillar Two, where there will be a combined payroll and tangible assets profitability test.

565. In combination, the objective of the activities test and profitability test is to identify the entities that earn residual profits relevant to Amount A (as calculated at a group or segment level). The activities test, by incorporating concepts imported from transfer pricing, will identify entities within a group that derive residual profits from the performance of non-routine activities that relate to the engagement of the group in market jurisdictions. The profitability test, like the Amount A formula, will be applied to financial, rather than tax accounts. This will mean that the entities identified through these two steps will perform non-routine activities and report residual profits in their financial accounts.

Step 3: Market connection priority test

566. The application of the activities and profitability test (or the profitability test if applied on a standalone basis) will identify a pool of potential paying entities on an MNE group or segment basis (depending on the basis on which the Amount A tax base is calculated). However, there will be instances where these potential paying entities derive profits from only a limited number of market jurisdictions allocated taxing rights over Amount A. This is because an entity should only bear an Amount A tax liability that relates to a market jurisdiction(s) in which it is engaged. A market connection priority test could be introduced to require that, in the first instance, the Amount A tax liability for a market jurisdiction is allocated to a paying entity that is connected to a market jurisdiction through the performance of activities identified under the activities test. This test may establish a connection between a market jurisdiction and a single or multiple paying entities.

Step 4: Pro-rata allocation

567. Where the entity or (entities) identified under the market connection priority test do not have sufficient profits to bear the full Amount A tax liability for a given market jurisdiction(s), other paying entities within the group or segment will be required to bear the remaining portion of the Amount A tax liability. This portion of the Amount A tax liability would be apportioned between these entities on a formulaic pro-rata basis. This would provide a “back-stop” position where it is not possible to establish a sufficient connection between any paying entity and a market jurisdiction allocated Amount A. Consideration could also be given to alternative “back-stop” positions, as discussed in more detail below.

568. Under a two-step approach relying primarily on the profitability test, where more than one entity is found to meet the test, the Amount A tax liability would be apportioned among them on a pro-rata basis according to the amount of profits earned by each entity that exceeds the agreed profitability test.

Other issues

569. In addition, when identifying the paying entity, it may in some circumstances be necessary to take into account transfer pricing adjustments and entity-level losses as determined under domestic entity-level loss carry-forward regimes.

Component 2: Methods to eliminate double taxation

570. The second component of the mechanism to eliminate double taxation deals with the methods to eliminate double taxation. The application of these methods will ensure that a paying entity is not subject to tax twice on the same profits in different jurisdictions, once under the current CIT rules and once under the new Amount A system. Today, jurisdictions apply two main methods to eliminate international juridical double taxation, the exemption method and the credit method. Under the exemption method, a paying entity would simply exempt from taxation the portion of its profits that had been allocated to market jurisdictions under Amount A. Under the credit method, the residence jurisdiction of a paying entity would provide a credit against its own tax for the tax paid in another jurisdiction.

571. The Blueprint contemplates that tax on Amount A could be relieved by paying entities using either the exemption method or the credit method. Given the current tax rates in market jurisdictions, both methods may ultimately deliver the same outcome.

7.2. Component 1: Identifying the paying entities

572. Amount A will be calculated at a group or segment level using a simple formula, incorporating a profitability threshold and reallocation percentage. This section explains how this calculation of residual profit at a group or segment level (which is allocated to market jurisdictions under Amount A) is broken down to an entity level. This is necessary as the mechanism to relieve double taxation arising from Amount A operates at an entity rather than a group or segment level. The Blueprint uses an approach that combines tax and accounting concepts, as well as simplifying conventions, to identify the entities within the MNE group that earn the residual profit under existing tax rules corresponding to Amount A and that are therefore designated as paying entities for the purposes of Amount A.

573. It could be argued that the same or a similar formulaic approach used to determine Amount A should be adopted to identify entities that will bear the Amount A tax liability – the paying entities – for example as those entities with a PBT to revenue ratio in excess of the agreed profitability threshold.

574. The most logical starting point for this profitability test would be to use the same formula used to calculate Amount A at a group or segment level (PBT / revenue, in excess of x%). However, there are two reasons why this approach is not considered appropriate. First, entity-level accounts include intragroup transactions that are eliminated on consolidation. As a result, the consolidated revenue of a group will be less (potentially substantially less) than the total (unconsolidated) revenue of entities within the group. This means that where a group has a profit margin in excess of x%, it is at least theoretically possible that there will not be any entities within that group with a profit margin in excess of x%. Second, the inclusion in entity-level accounts of intragroup transactions means that an entity’s revenue and hence profit margin can be easily distorted or manipulated by the way intragroup transactions are structured.

575. In addition, at a group or segment level, the use of consolidated accounts means that for most MNE groups the financial information necessary to compute the Amount A tax base and apply the Amount A formula will be readily available and prepared under a comparable accounting standard. In contrast, at an entity-level, financial information will be less readily available or may be prepared under local accounting standards, which means that even within the same group, the financial statements for different entities may not be comparable. That said, as discussed below, this issue could be addressed by using the entity-level accounts used by a group to prepare its consolidated group accounts. These factors mean that the test of profitability used at entity level in designating paying entities must be structured somewhat differently from that applied at group level in the determination of Amount A.

576. Conceptually, different approaches can be considered. For example, consistent with the formulaic nature of Amount A, one option would be to identify the paying entities using a quantitative approach similar to that used to calculate Amount A. This approach would prioritise simplicity, as it could limit the need to enter into detailed facts and circumstances functional and transactional analyses using transfer pricing concepts to identify the entities that should bear the Amount A tax liability. This would be consistent with the fact that Amount A itself is designed independently of these considerations. Another option would start from the basis that conceptually, the different features of Amount A suggest that paying entities should not only be profitable, but should be those that (individually or collectively) make material and sustained contributions to an MNE group’s residual profits, rather than those that perform only activities that generate routine profits. In addition, from a conceptual standpoint it could be argued that a paying entity should perform activities that relate to an MNE group’s engagement in a market jurisdiction for which it bears an Amount A tax liability, as opposed to non-market related activities or activities that relate to other markets.

7.2.1. The process to identify the paying entities

577. There will be a process with potentially up to four steps to identify the paying entities. Step 1 will be an activities test that will identify entities within a group or segment that perform activities that make material and sustained contributions to an MNE group’s ability to generate residual profits. Under Step 2, an MNE group will apply a profitability test to ensure the entities identified under Step 1 have the capacity to bear the Amount A tax liability. Step 3 will require that these paying entities have a connection with the market jurisdictions allocated Amount A, and that as a priority rule, the Amount A tax liability for these jurisdictions is relieved against the profits of these entities. Finally, if the entities connected to a market jurisdiction do not have sufficient profits to relieve the full Amount A tax liability, then the remainder will be apportioned on a pro-rata basis between other paying entities within the group or segment that do not have a connection to the relevant market jurisdiction. This will in effect provide a “back stop” where it is not possible to establish a “sufficient connection” between a market jurisdiction, allocated Amount A, and a paying entity with sufficient profits to bear the Amount A tax liability.

578. A two-step process could also be envisaged, whereby: a profitability test would be used to identify entities in the group that earn residual profits; and where more than one entity is found to meet the test, the Amount A liability would be apportioned among each entity on a pro-rata basis according to the amount of profits earned by each that exceeds the agreed profitability threshold.

7.2.2. Step 1: Activities test

579. An activities test would require that taxpayers undertake a qualitative assessment to identify entities within the group that make material and sustained contributions to an MNE group’s ability to generate residual profits. This reflects the fact that conceptually these are the entities that should bear the Amount A tax liability It is important that the activities test is developed in a way that is clear and simple to apply in practice, and draws on the existing transfer pricing analysis and documentation prepared by MNE groups. By doing so it will limit the additional compliance costs of MNE groups. Nevertheless, any additional documents that may be required would be included in the standard Amount A documentation required for the tax certainty process. At the outset, it is important to emphasise that Amount A would only apply to large MNE groups, that meet or exceed a number of threshold tests and have in-scope revenue. This section outlines the concepts that could underpin the activities test and explains how this test could be applied in practice.

580. The transfer pricing master file and relevant local files prepared by MNE groups provide a first point of reference for the application of the activities test, taking into account that this documentation should describe an arm’s length allocation of assets and risks based on the functional analysis of the MNE group. For many groups, particularly those with centralised operating models that perform the activities that make material and sustained contributions to an MNE group’s ability to generate residual profits in only a handful of locations, the application of this test should be straightforward. For groups with a comparatively decentralised operating model, more entities may be identified under this test. However, where activities that make material and sustained contributions to an MNE group’s ability to generate residual profits are performed in market jurisdictions, the application of the marketing and distribution profits safe harbour may mean that these groups do not need to apply the mechanism to eliminate double taxation, or only need to do so for a limited number of jurisdictions. In any case, the application of the activities test, alongside all other parts of the mechanism to eliminate double taxation, will be covered by the Amount A tax certainty process. Within the Amount A tax certainty process, tax administrations would be able to request additional information from taxpayers and challenge their self-assessed application of the activities test. The Amount A tax certainty process will be important to ensure that any disputes arising from the application of this test are resolved in a timely manner and do not give rise to unrelieved double taxation.

Determining the activities

581. The conceptual development of the activities test requires the identification of the set of activities that are likely to make a material and sustained contribution to an MNE group’s ability to generate residual profits, and potentially to specifically to identify the activities that relate to their engagement in market jurisdictions. The activities that relate to an MNE group’s engagement in a market jurisdiction would extend beyond the ownership of (or entitlement to income derived from) relevant marketing intangibles and would, for example, include the ownership of (or entitlement to income derived from) product or technology intangibles that help to support sales as well as the oversight/control of economically significant risks and decisions relating to the business. As set out in more detail in section 7.2.4 the activities performed by an entity may relate to its engagement in a market jurisdiction, even where there is no direct transactional connection between a market jurisdiction and an entity.

582. The OECD Transfer Pricing Guidelines (TPG) identify a series of factors that may entitle an entity to participate in the residual profits generated by an MNE group for transfer pricing purposes that will also be relevant for identifying the paying entities. These include the following:6

• First, the paying entity should oversee some of the core strategic and operational activities that generate residual profits for the MNE group, and specifically exercise control and decision-making authority over key aspects of those activities. The activities identified should also relate to the MNE group’s engagement with the market.

• Second, the activities of the paying entity will likely consist of the performance of some or all of the important functions related to the development, enhancement, maintenance, protection and exploitation of intangible assets of the MNE group that are specific to the MNE group’s market engagement. This would include intangibles related to technology that facilitates market engagement, such as technology used in the ADS businesses to gather user data and content contributions.

• Third, a paying entity would perform the activities that would mean it should have the entitlement to some or all of the residual profits arising from valuable intangible assets that relate to the engagement of the MNE group with the market where they have been acquired and were not self-developed.7

• Fourth, the paying entity should both assume economically significant risks that relate to the MNE group’s engagement in the market, and exercise decision-making functions and control over the assumption of those economically significant risks.8 This would at least include (i) the capability to make decisions to take on, lay off, or decline a risk-bearing opportunity,9 together with the actual performance of that decision-making function, and (ii) the capability to make decisions on whether and how to respond to the risks associated with the opportunity, together with the actual performance of that decision-making function. It may also include (iii) the capability to mitigate risk, meaning the capability to take measures that affect risk outcomes, together with the actual performance of such risk mitigation.

583. It may be considered that the provision of intragroup financing is not an activity that makes material and sustained contributions an ADS or CFB business’ ability to generate residual profits, specifically as it relates to their engagement in market jurisdictions. Therefore, the provision of intragroup financing alone would not be sufficient to identify an entity as a paying entity. The provision of an intercompany loan or guarantee, while representing an activity that would result in the assumption of an economically significant risk, relates solely to the assumption of financial risks and results in residual profits associated with financial risks being generated. This return is typically fixed and does not vary after the instrument/guarantee is entered into. Further, the return does not sufficiently relate to the performance of functions and the materialisation of risks related to non-routine activities in the market itself.10 Alternatively, it could be argued that the provision of intragroup financing alone should be sufficient to identify an entity as a paying entity and that adopting this approach would require the design of the activities test to be revisited.

Practical application of the activities test

584. To facilitate ease of compliance and reduce complexity, the activities test is structured as follows. A general principle derived from the activities identified above will govern the identification of paying entities and will assist with clarifying the expected FAR profile of a paying entity. The application of this general principle, which will be based on concepts already included in the TPG, will be supported by a more specific list of indicia that will either positively identify features associated with a paying entity, or negatively identify features not associated with a paying entity. The identification itself will primarily leverage from the transfer pricing master file and local files of an in-scope MNE group.

585. The proposed general principle is to identify paying entities as “the member or members of an MNE group (or segment) that perform functions, use or own assets and/or assume risks that are economically significant, for which they are allocated residual profits relevant to Amount A”. This principle seeks to reflect the types of activities that would be regarded as giving an entity the right to participate in a group’s residual profit. As noted above, this general principle draws on a series of concepts that already form part of transfer pricing today.

586. The application of this general principle will be supported by a list of indicia that would make it easier for taxpayers and tax administrations to apply this general principle in practice. To limit compliance costs associated with applying these indicia, they will be based on information that taxpayers are already required to collect and report and that tax administrations already review. In particular, a group’s master file and local file already include information on the functions performed, assets used and risk assumed by most entities in an MNE group.11 These documents typically characterise the primary activities of an entity, information that is also collated for CbCR purposes. Finally, transfer pricing documentation identifies the transfer pricing method (or methods) that is used to establish whether the profits allocated to an entity are arm’s length given its FAR profile. Hence, there are at least three indicia that could readily be used as part of the activities test. These are:

• The functions performed, the assets used and risks assumed by an entity;

• The characterisation of an entity, derived from existing transfer pricing documentation; and

• The transfer pricing method used to determine an entity’s arm’s length profits.

587. How would these indicia be applied in practice? An entity that is entitled to the entrepreneurial profit from exploiting key intangibles, from assuming economically significant risks; which is characterised as an entrepreneurial principal entity; and which should receive the residual profit according to the transfer pricing method (typically, but not always, in the form of a variable return), from a group’s value chain would, under a positive approach, be identified as a potential paying entity. In this context, the characterisation of an entity and the transfer pricing method used to determine its arm’s length profits should reflect and will be indicators of its FAR profile.

588. For example, an entity that makes unique and valuable contributions to activities that generate residual profits for the MNE group, and/or is entitled under the ALP to exploit and derive non-routine profit from intangible assets, and/or assumes or shares the assumption of economically significant risks, would be caught under the first indicator. In the transfer pricing documentation, it would likely be characterised as an entrepreneurial principal entity, thus meeting the second indicator. The third indicator may show that the remuneration for such an entity is determined through a transactional profit split method (TPSM) or by taking residual profits after remunerating a routine-function, limited-risk entity through the TNMM or other methods. Such an entity would be identified as a potential paying entity and so would, through the positive approach, be included in the pool of potential paying entities.

589. In contrast, an entity that does not own key intangibles or manage economically significant risks; is characterised as a limited risk entity or contract service provider (such as, a sales agent); and receives a fixed, benchmarked return determined using, say, the TNMM or cost-plus method would not be identified as a potential paying entity and so would on a negative approach, be excluded from the pool of potential paying entities.

590. As part of the activities test, where an MNE group computes the Amount A tax base on a segmented basis it will be necessary to determine the segments within which a potential paying entity is part of. For example, if a pharmaceutical group applied Amount A separately for its pharmaceutical and consumer healthcare segments, then it will need to identify its paying entities on the same basis.

591. Further work will be undertaken by the Inclusive Framework to refine the activities test and, specifically, the general principle and list of indicia on which basis it would be applied. This will include determining the types of activities that a paying entity should perform or the intangibles it should own. This will include further work on specific issues, including:

• The guidance that should be included alongside the list of indicia, to facilitate the application of these indicia by taxpayers, including the interactions between the general principle and the positive and negative lists.

• The type of documentation that could be used to develop and ultimately apply the list of indicia, which will include the identification of any additional or improvements to documentation that may be requested as part of the standard Amount A documentation included within the tax certainty process and how this documentation can be shared through appropriate exchange of information (EOI) agreements.

• The feasibility and additional guidance that will be required to facilitate the segmentation on the basis of entity-level accounts, where required.

• The interactions between the application of the activities test and existing ALP-based profit allocation rules. For example, that there would be a requirement for taxpayers to take consistent positions for the purposes of the activities test and the application of ALP-based profit allocation rules, and whether tax administrations would retain the option to take inconsistent positions.

7.2.3. Step 2: Profitability test

592. The profitability test would ensure that the potential paying entities have the capacity to bear the Amount A tax liability. The profitability test would ensure that an entity making low profits (or losses) is not identified as a paying entity, and is consistent with the decision of the Inclusive Framework to limit the application of Amount A to in-scope MNE groups that earn residual profits (rather than all in-scope MNE groups).

593. In combination with the activities test, the objective of the profitability test is to identify the entities that earn residual profits relevant to Amount A (as calculated at a group or segment level). The activities test, by incorporating concepts imported from transfer pricing, will identify entities within a group that derive residual profits from the performance of non-routine activities that relate to the engagement of the group in market jurisdictions. The profitability test, like the Amount A formula, will be applied to financial, rather than tax accounts. This will mean that the entities identified through these two steps will perform non-routine activities and report residual profits in their financial accounts.

594. The Inclusive Framework is considering a profitability test aligned with the substance carve-out that is being developed for Pillar Two, which is based on expenditures for payroll and tangible assets (calculated by reference to depreciation expenses for certain assets including land).12 In a Pillar Two context, the policy objective behind the carve-out is to exclude a fixed percentage of income from substantive activities within a jurisdiction. The use of payroll and tangible assets as indicators of substantive activities is justified because these factors are generally expected to be less mobile and less susceptible to lead to tax induced distortions. Conceptually, excluding a fixed percentage of income from substantive activities focuses Pillar Two on “excess income”, such as intangible-related income, which is most susceptible to remaining BEPS risks. This objective may also be relevant when identifying the potential paying entities for Amount A, as Amount A has been designed to allow jurisdictions to retain sole taxing rights over the routine (or non-Amount A profits) of entities resident in their jurisdiction.

595. In an Amount A context, a profitability test would identify entities that earn income in excess of a fixed return for payroll and tangible assets. A combined test based on payroll and tangible assets would take into account the varying substance profiles of different types of entities, including labour-intensive and asset-intensive businesses. It would level the playing field by acknowledging the contributions of both employees and tangible assets to an entity’s routine profits. Such an approach is preferable to a profitability test based on a single factor, either payroll or tangible assets, given that a single factor approach would favour entities performing one type of activities over another, though the Inclusive Framework will continue to explore other potential profitability tests.

596. There are two sources of readily available financial information that could be used to apply the profitability test outlined above, namely tax accounts and financial accounts. It would not be appropriate, however, to apply this test using tax accounts. This is because there are significant jurisdictional variations in how tax bases are determined, which means that, as an example, the design of a jurisdiction’s R&D tax incentives would impact the reported profitability of an entity. Therefore, and to align with Pillar Two, the profitability test will apply based on entity-level financial accounts used by a group to prepare its consolidated accounts. These accounts will be prepared under IFRS or a comparable accounting standard used by the group to prepare its consolidated accounts. Such an approach will be more consistent than reliance on entity-level accounts prepared under domestic accounting standards (where there can be significant variation between jurisdictions).

597. Where a potential paying entity derives profits from more than one segment (as defined under the Amount A segmentation framework) it will be necessary to subdivide the entity-level accounts between these two or more segments in order to apply the profitability test. This is necessary to ensure that the different segments are not pooled together, which could result in an entity being inappropriately identified or not identified at alas a paying entity. Further, the definition of PBT developed to compute the Amount A tax base, will exclude dividend income and gains or losses from the disposal of shares, and may also exclude other types of income. To ensure symmetry between the computation of Amount A and the identification of the paying entities, the profitability test will need to be based on a comparable measure of PBT.

598. The Inclusive Framework will further review the relevance and application of the Pillar Two substance-based carve out, in the context of Pillar One, to determine whether any modifications to this test are required. It will also continue to consider other quantitative methods that could be used in combination with, or as an alternative to the profitability test outlined above. This includes a proposal to identify paying entities as those that earn in excess of an agreed percentage of an MNE group’s profit, either on an individual or aggregate basis.

7.2.4. Step 3: Market connection priority test

599. The application of the activities and profitability test would identify a pool of potential paying entities on an MNE group or segment basis (depending on the basis on which the Amount A tax base is calculated). There will be instances where these potential paying entities derive profits from only a limited number of market jurisdictions allocated taxing rights over Amount A. A simple example is a group that has two regional principals that are entitled to all the residual profits derived from their region. In this scenario, it could be argued that each regional principal should only bear the Amount A tax liability that arises in respect of the market jurisdictions from which it derives residual profits. Therefore, a market connection priority test could be introduced, whereby, in the first instance an Amount A tax liability for a given market jurisdiction would be relieved against the profits of a paying entity that performs, in relation to a given market, the activities identified as part of the activities test.

600. As Amount A will apply on a group or segment basis, without taking into account the profitability of different regions or sub-segments, this approach could result in situations arising where it is not possible to identify a paying entity that meets the profitability test. Returning to the example outlined above, there could be instances where one regional principal does not by itself earn the necessary quantum of Amount A. In this scenario, other paying entities within a group or segment will be required to bear the remaining portion of the Amount A tax liability (see Step 4: Pro-rata allocation below).

601. For each market jurisdiction allocated Amount A, a taxpayer will be required to determine which (if any) of the potential paying entities identified have a sufficient connection to be identified as a paying entity for that market jurisdiction. It will be possible to establish a direct or indirect connection between a market jurisdiction and a variety of different entities within a group. For example, an entity may sell goods or services to related or unrelated parties resident in that market; or, it may license intangible assets that are subsequently exploited in that jurisdiction. For the purpose of this test, it cannot be assumed that any connection between a paying entity and a jurisdiction will be sufficient to establish a connection for Step 3. Instead, this connection should be linked to the activities identified as part of the activities test.

602. For many groups it will be relatively straightforward to identify the paying entities connected to different markets. For example, where a single entity within the group has the global rights to exploit intangible assets and receives residual profits from every market jurisdiction, it will have a sufficient connection to every market jurisdiction. In contrast, where a group has a regionally centralised operating model, where different entities within the group have the right to exploit intangible assets and receive residual profits from the market jurisdictions within its region, then each of these entities will have a sufficient connection with the market jurisdictions in their region (but will not have a sufficient connection to market jurisdictions in other regions).

603. As a result, in many instances it should be relatively easy to establish a sufficient connection between a paying entity and a market jurisdiction. This is because most MNE groups will have legal entity structures and legal agreements that clearly document which entities in the group have the rights to exploit specific intangible assets in and receive residual profits from specific markets. For example, a legal agreement giving an entity the right to exploit an intangible in a region would provide evidence of a connection between that entity and market jurisdictions in that region. Similarly, the direct license of an intangible from an entity to a market jurisdiction would also provide evidence of a connection between a paying entity and a market. In many instances, this information will already be available to tax administrations through a group’s master file or other transfer pricing documentation.

604. In the development of this priority rule it is important to highlight a number of design principles that will be relevant when defining what constitutes a “sufficient connection”:

• The transactions that give rise to the connection should be linked to the activities identified under the activities test. For example, if an entity performs certain routine support services, in addition to non-routine activities that result in it being identified as a paying entity, there would not be a sufficient connection between that entity and a market jurisdiction that received only routine support services. Similarly, the provision of intra-group financing to an entity resident in a market jurisdiction would not be sufficient to establish a connection, as the performance of intra-group financing activities is not sufficient to identify an entity as a paying entity.

• A paying entity should derive profits that are ultimately connected to sales to third parties in a particular market jurisdiction. For example, if an entity receives a fee for the provision of manufacturing support services from a related party manufacturer, this would not be sufficient to establish a connection between the entity that receives the fee and the jurisdiction in which the manufacturer is resident (which may also be a market jurisdiction).

• It would not be necessary for a paying entity to have a direct transactional connection with a market jurisdiction. For example, where a royalty payment is made from a market jurisdiction to a conduit entity (that bundles together royalty payments from a variety of markets) and then pays them to an ultimate recipient, there could be a sufficient connection between the market jurisdiction and the ultimate recipient. The same would be true in the event of a chain of conduit entities.

605. The Inclusive Framework will undertake further work to develop this test and seek to analyse the types of arrangements and transactions that could be used to establish a sufficient connection between a paying entity and a market jurisdiction and the evidential basis through which these arrangements and transactions could be identified. It will also look at situations where a market jurisdiction may be sufficiently connected with multiple paying entities and explore whether there should be a hierarchy so that the Amount A tax liability would first be allocated to the paying entity with a direct or otherwise the “strongest” connection. To the extent that this paying entity did not have sufficient profits13 to bear the full Amount A tax liability, it would then be allocated to the entity with an indirect or otherwise less “strong” connection until the tax liability had been fully allocated. Conversely, the Inclusive Framework will also consider whether in this scenario, the fairest and simplest approach would be to apportion the relevant Amount A tax liability between the paying entities with a sufficient connection to a market jurisdiction on a pro-rata basis (as explored in more detail in section 7.2.5).

606. These two approaches would deliver different outcomes. For example, assume that an MNE group has an operating model where a regional trading principal in Country A is entitled to the majority of the residual profits from a market jurisdiction and is determined to have the strongest connection to that market, but the ultimate parent entity in Country B (which performs some DEMPE functions in relation to intangibles owned and exploited by the regional trading principal) retains an entitlement to a small proportion of the residual profits from the relevant market jurisdiction. Under a hierarchical approach, the relevant Amount A tax liability would first be allocated to regional principal and would only be allocated to the parent entity where the regional principal did not have sufficient profits to bear the full Amount A tax liability. In contrast, under a pro-rata approach the relevant Amount A tax liability would simply be apportioned between the regional principal and parent entity on a formulaic basis assuming both have a sufficient connection with the relevant market jurisdiction.

607. Some Inclusive Framework members believe that there is a tension between calculating Amount A on a global basis and then requiring that profit allocated to a market jurisdiction to be paid by entities with which there is a market connection. These members support removing the market connection priority test and apportioning the Amount A liability between entities identified in Steps 1 and 2 on a pro-rata basis, as outlined under Step 4. To the extent that there are concerns about the profits of a regional principal being allocated to markets with which the regional principal has no transactional connection, these members think that these concerns should be considered as part of the debate on regional segmentation i.e. whether Amount A should operate on a geographically segmented basis in certain situations.

7.2.5. Step 4: Pro-rata allocation

608. As paying entities will only be entities that earn above a routine return, as defined under the Step 2 profitability test, it can be argued that after applying the mechanism to eliminate double taxation a residence jurisdiction should, at a minimum, retain taxing rights over this portion of routine profit. This is consistent with the rationale underpinning the inclusion of a profitability threshold in the Amount A formula. Therefore, a paying entity will be deemed to have no profits to bear a further Amount A tax liability, once the taxing rights of the residence jurisdiction have been reduced to a routine return, as defined under the profitability test.

609. There is a risk that in some circumstances the paying entities that are sufficiently connected to a market jurisdiction (through the market connection priority test) will have insufficient profits to bear the full Amount A tax liability. This is in addition to situations (such as that outlined in the example above) where it may not be possible to identify a connection between any paying entity and a market jurisdiction. In order to ensure that the Amount A does not lead to double taxation, if the paying entities connected to a market do not have sufficient profits to bear the full Amount A tax liability, then as a “back-stop” any outstanding liability will be apportioned between all other potential paying entities within a segment.

610. This step could also be designed so that the residence jurisdiction retains taxing rights over a portion of a resident entity’s residual profits. This could be delivered through this step, for example by deeming a paying entity to have no profits to bear a further Amount A tax liability after x% of their residual profits (as defined through the application of Step 2) had been allocated to market jurisdictions under Amount A.

611. Under either approach, the apportionment will be on a formulaic pro-rata basis. This formula will based on a paying entity’s PBT in excess of routine profit (as defined under the profitability test) and/or the profitability of the entity (i.e. PBT / revenue). Both of these approaches would be designed so that the apportionment of the Amount A tax liability between paying entities will ensure that at a minimum the relevant residence jurisdictions retain taxing rights over the profits earned by any paying entity that are attributable to routine activities. Again, a paying entity will be deemed to have no profits to bear a further Amount A tax liability, once the taxing rights of the residence jurisdiction have been reduced to a routine return. If the market connection priority test were not adopted, due to perceived tensions with the global nature of Amount A, this pro-rata allocation could be used as the only method to apportion an Amount A tax liability between multiple paying entities.

612. This formulaic pro-rata approach (including the application of the market connection priority test) can be illustrated through two simple examples. First, assume for a given market jurisdiction two paying entities are identified under Step 3 with an equally strong connection to the relevant market and that after allocating each entity a routine return (based on the profitability test) the entities earn residual profits of 60 and 40 respectively. The first paying entity would bear 60% of the Amount A tax liability and the second paying entity would bear 40% of the Amount A tax liability for the jurisdiction in question. Second, assume that for a given market jurisdiction a single paying entity is identified under Step 3, but that after allocating this entity a routine return (based on the Step 2 profitability test) it has insufficient profits to bear the full Amount A tax liability of the market jurisdiction in question. In this scenario, the paying entity identified under Step 3 would bear the portion of Amount A tax liability that would reduce the taxing rights of the jurisdiction in which it was resident to the routine profits attributable to the entity. The remaining Amount A tax liability would then be apportioned between other paying entities within the group or segment that were not identified as having a connection to the market under Step 3 on a pro-rata basis.

613. There are other alternative or additional “back-stop” rules that could be considered and on which Inclusive Framework members have different views. One approach would be that where it is not possible to identify a paying entity connected to a market with sufficient profits to bear the full Amount A tax liability, the Amount A allocation to a market jurisdiction could be reduced or eliminated. However, others members consider that this would be conceptually inconsistent with the Amount A nexus and profit allocation rules, which would apply alongside and on an equal basis to the existing rules. Further, these members argue that this approach would place significant additional pressure on the application of Step 3 and would likely significantly increase the controversy surrounding the application of this step. An alternative could be that in this scenario the Amount A tax liability could also be relieved against the routine profits that would usually remain taxable in the residence jurisdiction, or that the Amount A tax liability would be carried forward and would only become payable in a market jurisdiction in future years when a paying entity connected to a market had sufficient profits to bear the outstanding liability. It is acknowledged however that this would exacerbate the tensions that some members perceive to exist between the market connection test and the global basis for calculating Amount A, as referred to in 607. A further alternative would be to designate an entity (or entities) within the group as the paying entity of last resort. This could for example be the ultimate parent entity or an entity identified under the Global Anti-base Erosion Proposal (GloBE) as being subject to below an agreed minimum level of taxation. This approach could also be used to identify a paying entity in the unlikely event that no potential paying entities were identified following the application of Step 1 and Step 2 (outlined above). Alternatively, consideration could be given to the design of the steps before the pro-rata allocation which ensures that identified entities always have sufficient profits to bear the Amount A tax liability. If the development of such process is achieved, back-stop rules would not be needed.

7.2.6. Other issues

614. Discussions in the Inclusive Framework are ongoing on various other issues notably transfer pricing adjustments and entity-level losses. These discussions reflect some of the challenges that arise from integrating the new Amount A tax right into the existing tax system.

Transfer pricing adjustments

615. Tax administrations can make transfer pricing adjustments to increase the profits allocated to an entity to reflect the profits that would have been earned at arm’s length. Typically, it can take tax administrations three to five years to make transfer pricing adjustments, due to the detailed audit work required to make an adjustment. When a tax administration makes a transfer pricing adjustment, a taxpayer can request, through the MAP, that a corresponding adjustment is made to reduce the taxable profits of the counterparty entity in another jurisdiction. The MAP process can again take a number of years to complete. This means there may be situations where a paying entity has its profits reduced through a corresponding adjustment. This may reduce its capacity to bear the Amount A tax liability (as determined under Step 3 or 4), or, in extreme cases, may mean it no longer satisfies the conditions required to identify it as a paying entity. 14 Transfer pricing audits may also lead to a reassessment of an entity’s FAR profile, its characterisation for transfer pricing purposes and the transfer pricing method used to determine its arm’s length profits, which may have an impact on whether the entity has been correctly identified (or not identified) as a paying entity.

616. In many instances, reductions to an entity’s profits as a result of a corresponding adjustment will be small in relative terms, and it will not be necessary to take account of these adjustments for the purposes of Amount A. In addition, as part of the Amount A tax certainty process, the Inclusive Framework is considering how existing compliance processes could be used to reduce the likelihood of transfer pricing adjustments being made that have a significant impact on the identification of the paying entity (see Chapter 9). However, where these adjustments are “material”, the mechanism to eliminate double taxation under Amount A will take this into account. It would be very challenging to make retrospective changes to the elimination of double taxation arising from Amount A, because this would require the reopening of the tax certainty process and the reassessment of the application of Amount A. However, it may be possible to make prospective adjustments, for example by adjusting the future Amount A tax liability of a paying entity based on a transfer pricing adjustment for a past period. Alternatively, in some instances it may be possible to consider whether the apportionment of the Amount A tax liability could be taken into account as part of the MAP, where for example a transfer pricing adjustment results in the transfer of profit between two paying entities. The Inclusive Framework will undertake further work to develop this mechanism, including a definition of what constitutes a “material” adjustment, (which may be framed as a relative threshold or an absolute amount – for example, a “material” adjustment may exceed x% of an entity’s profits or EUR Xm). This work will also consider how transfer pricing adjustments can be taken into account where changes are made to an entity’s tax accounts, but not their financial accounts.

Entity-level carried forward losses

617. The rules to compute the Amount A tax base will incorporate loss carry-forward rules that will be separate from existing domestic loss carry-forward rules. This group-level loss carry forward regime should ensure that most taxpayers continue to benefit from domestic loss carry-forward regimes at an entity-level. However, there may be situations where groups are required to allocate Amount A profits to market jurisdictions, even when some paying entities within the group have carried-forward losses under existing domestic rules. This may arise where a jurisdiction’s domestic loss regime provides for the accelerated expensing of depreciation or amortisation expenses, or simply because there are some loss-making principal entities within a group that are profitable on a consolidated level.

618. To address these situations, it could be argued that domestic carry-forward losses should be accounted for when determining the paying entities for the purposes of Amount A. This would require a modification of Steps 3 and 4, to deem an entity to have insufficient profits to bear an Amount A tax liability when it has domestic carried-forward losses. This would avoid the risk that an entity bears an Amount A tax liability, even when it does not pay tax in the jurisdiction where it is resident due to domestic carry-forward losses. This outcome could be perceived as inequitable as it would, in effect, give market jurisdictions taxing rights over an entity, before a resident jurisdiction. It would also reduce the value to an MNE group of carry-forward losses, potentially discouraging investment.

619. However, developing rules to account for entity-level losses specifically would equally create fairness concerns if it resulted in a paying entity in one jurisdiction bearing the full Amount A tax liability of a group, because a paying entity in another jurisdiction has large amounts of carried forward losses, especially where the difference is simply the result of different loss carry forward rules in the two jurisdictions concerned. It could also create opportunities for manipulation. The risk of manipulation arises because there are situations where MNE groups can artificially generate carry-forward losses at an entity level. Therefore, if entity-level losses were to be taken into account in the process of identifying the paying entity, it may be necessary to develop additional protections to prevent this opportunity for manipulation, creating significant additional complexity for taxpayers and tax administrations. The Inclusive Framework is continuing to examine this issue.

7.3. Component 2: Methods to eliminate double taxation

620. The second component of the mechanism to eliminate double taxation concerns the methods to eliminate double taxation. The application of this method (or methods) will ensure that a paying entity is not subject to tax twice on the same profits in different jurisdictions, once under the existing profit allocation rules and once under Amount A. The identification of paying entities under Component 1 is structured such that the entities which earn residual profits under existing rules are designated as paying entities. The result is that the double taxation that will arise is juridical in nature (the taxation by two jurisdictions of one person on the same income). These paying entities are potentially subject to tax on the profits reflected in Amount A by both their jurisdiction of residence under existing tax rules and by the market jurisdiction which is given a new taxing right with respect to Amount A. This suggests the use of either the exemption or credit method would be appropriate. If the double taxation was economic in nature (where two persons are taxed on the same economic income), the reallocation method used under transfer pricing rules (effected by corresponding adjustments) would likely be more appropriate.

621. Today, jurisdictions apply two main methods to eliminate international juridical double taxation: (i) the exemption method (a version of which is found in Article 23 A of the OECD and UN Models); and (ii) the credit method (Article 23 B of the OECD and UN Models).

622. The exemption and credit methods are not simply different ways to deliver the same mechanical outcome. Under the credit method, the residence jurisdiction retains secondary taxing rights over the profits of a paying entity where these profits are taxed in the market jurisdictions at a rate that is lower than the rate in the residence country. Whereas, under the exemption method, the residence jurisdiction would not retain secondary taxing rights over the profits of a paying entity because those profits are exempted or removed. However, as outlined below, the initial economic analysis of Pillar One estimates the current average rate of tax in market jurisdictions at 26% (noting, however, that rates may change over time) and, therefore, there will be meaningful secondary taxation rights only for jurisdictions with relatively high tax rates. The Blueprint contemplates that a jurisdiction will be able to opt for the exemption method or the credit method to relieve double tax arising on Amount A. Where both methods produce the same or a similar result, member jurisdictions would likely opt for the simplest method.

623. This chapter shows how either the exemption or credit methods could be used to eliminate double taxation arising from Amount A. In situations where a group has a local subsidiary in a market jurisdiction, an alternative “reallocation method” could be contemplated. Under this approach, the Amount A profit allocated to a market jurisdiction would be deemed to arise in the local subsidiary and an upward adjustment to its profits would be made for tax purposes, resulting in the local subsidiary being liable to pay Amount A. Double taxation would be relieved by providing a deduction or downward adjustment to the profits of the relieving entity. This deemed transfer of Amount A profits between these two entities would be similar to the economic double taxation relief mechanism under Article 9(2). For the relieving jurisdiction, this would deliver an equivalent outcome to the exemption method. For the local entity, however, it would be difficult to justify paying tax on the Amount A profits if the entity does not own those profits. Therefore, the MNE would likely want to make a secondary adjustment to actually transfer the Amount A profits to the local entity for legal and accounting purposes so that they align with the tax accounts. This accounting transfer, which might be characterised as a loan or a contribution of capital, and subsequent transactions to repatriate the profits may have secondary spill-over tax effects (for example, withholding tax) which would be considered undesirable. If the MNE group does not have a local subsidiary in the market jurisdiction, the relieving jurisdiction would still have to provide either a credit or exemption for Amount A tax in that jurisdiction. A reallocation method may also be difficult to adapt to (or undo many of the benefits associated with) the centralised and simplified administration system discussed in Chapter 10. For those reasons, the reallocation method does not appear to be appropriate to eliminate double taxation under the current design of Amount A and Component 1. But, as work progresses, new features may emerge which could demonstrate a benefit to using this method, and so it may be revisited in due course.

7.3.1. Exemption method

624. Under the exemption method, a paying entity would simply exempt from taxation the portion of its profits that had been allocated to market jurisdictions under Amount A.

625. This means that the exemption method would effectively eliminate double taxation arising from Amount A and would be relatively simple to apply. That is because a residence jurisdiction would not be required to determine the rate at which profits allocated under Amount A had been taxed, but simply identify the proportion of the relevant paying entity’s profits that had been reallocated under Amount A.

626. While the exemption method may have the advantage of simplicity in certain cases, there will be a broader set of considerations taken into account when deciding which method to apply, including the absence of secondary taxing rights under the exemption method15 and the possibility that taxpayers could have less tax to pay overall under the application of the Amount A system with the exemption method (considered below). Further technical work will be done to identify the advantages and disadvantages of these two methods.

7.3.2.  Credit method

627. Though more complex than the exemption method, certain jurisdictions may have a preference for using the credit method to relieve double taxation of Amount A.

628. Under the credit method, tax applied to Amount A in the market jurisdictions would be available as a tax credit to the paying entity or entities. The available credit would then be capped at the lower of the tax applied in the market jurisdiction and the tax that would have been paid on the Amount A allocation in the relieving jurisdiction. Considerations for a jurisdiction that would want to opt for the credit method are described below.

Per jurisdiction or blended limit

629. The Blueprint contemplates that the credit method could potentially be applied either jurisdiction by jurisdiction, or using a blended approach. Under a jurisdiction-by-jurisdiction approach, the limit on the available credit is determined by comparing the tax that would have been paid in the relieving jurisdiction to the tax applied to Amount A in each market jurisdiction separately. If a blended approach applies, the limit on the available credit would be determined by comparing the tax that would have been paid in the relieving jurisdiction to the tax applied on the total Amount A liability allocated to each paying entity. A blended approach would permit tax applied to Amount A in low tax markets to be blended with tax paid on Amount A in high tax markets (this approach could be referred to as a “blended cap”). Jurisdictions operating a blended cap will typically provide a higher level of relief for double taxation on Amount A than those operating a per jurisdiction cap. The net effect of applying a blended cap may not be dissimilar to the effect of applying the exemption method where the tax that would have been paid in the relieving jurisdiction is lower than the weighted average tax rate applied to Amount A in the market jurisdictions.

630. It could be argued that the operation of a blended cap more closely reflects the multilateral nature of Amount A, although it is noted that the operation of the market connection priority test should facilitate the operation of a per jurisdiction cap. In order for the credit method to apply on a per jurisdiction basis, it will be necessary to first identify specific market jurisdictions where a paying entity incurs an Amount A tax liability. This will be possible where a connection is established between a specific market jurisdiction and paying entity under the market connection priority test. In this scenario it would be possible to operate the cap on a per country basis.

631.  Where a market jurisdiction’s Amount A tax liability is simply allocated between paying entities on a pro-rata basis, the per jurisdiction cap would be applied by reference to the amount allocated to each paying entity on a per jurisdiction basis.

Secondary taxing right

632. One of the main benefits of applying the credit method for relieving jurisdictions is that secondary taxing rights remain available to the residence state. However, as mentioned above, the value of such a right is expected to be relatively limited as it is expected that the average rate applied to Amount A would be relatively high. With a view to testing this position, some initial analysis was undertaken to compute the average CIT rate across market jurisdictions. That initial analysis indicated that for ADS, the average CIT rate in market jurisdictions based on the data would be at present 26.0%, and for CFB it would be 26.3%. Under such scenario, and noting that rates may change overtime, a jurisdiction with a CIT rate below 26% would have no additional collected under the secondary taxing right in respect of Amount A if the credit method is applied using a blended cap.16 Where the credit method is applied using a per jurisdiction cap, some additional tax may be collected by relieving jurisdictions with lower tax rates.

Outcome for MNE

633. Another argument made in favour of the credit method over the exemption method is that the exemption method could put taxpayers in a better position than before if the profits allocated to market jurisdictions were taxed at a lower rate in the market jurisdiction. Under the exemption method, taxing rights on residual profits are moved from relieving jurisdictions to market jurisdictions. If the tax rates in relieving jurisdictions are higher than the tax rates in market jurisdictions, globally the MNE group may be paying less tax under Amount A with the exemption method than it would under existing rules. However, the initial economic analysis on average tax rates referenced above would tend to indicate that this should not be a significant risk. Further, since Amount A is intended to effect a transfer of taxing rights to market jurisdictions, and some jurisdictions do not view this transfer as conditional on the level of taxation in the market, if in a particular case it results in a reduction in tax, this is not necessarily an inappropriate result.

Differences in sourcing rules

634. Another consideration for applying the credit method is that domestic foreign tax credit regimes tend to have “sourcing” rules that can result in part of the income that was taxed in the source country being regarded as domestically sourced income in the country of residence. As a result, double taxation can arise and go unrelieved.

635. Amount A will be agreed multilaterally under a single system along with the Amount A allocations to market jurisdictions and the Amount A that should be relieved by each relieving jurisdiction. Given the multilateral nature of Amount A, there should be no opportunity for having different domestic legislation and sourcing rules that would lead to unrelieved double tax on Amount A.

Information requirements

636. Another consideration when using the credit method to relieve double taxation on Amount A is that more information is required by the jurisdiction providing the double tax relief in relation to the tax treatment of Amount A in the market jurisdiction (for example, the applicable rate in the market jurisdiction) than is required to apply the exemption method.

7.3.3. Facilitating use of both methods

637. Allowing jurisdictions to select which of the two methods they would apply to relieve double taxation would introduce some additional complexity.

638. Where jurisdictions are permitted to select the method, once the paying entities of an MNE group were identified (through the process outlined in Component 1), the method to eliminate double taxation adopted by the jurisdiction in which a paying entity was resident would apply. So if the group identifies two paying entities – the first resident in a jurisdiction that applies the exemption method and the second in a jurisdiction that applies the credit method – in the first jurisdiction, the profits of the paying entity would be exempt from tax. But the second jurisdiction would tax those profits and provide a credit against that tax for the tax paid in eligible market jurisdictions.

639. MNEs typically operate a multitude of credit and exemption methods in their tax returns and while the possibility of an MNE group having to operate two methods will involve an amount of additional complexity, it would be incremental only. No paying entity would be required to operate more than one method of relief in respect of Amount A as each jurisdiction would be required to select one method only.

640. It is noted that if an MNE considers that one method produces more favourable outcomes, there may be a risk that it could seek to manipulate the application of Component 1 so that paying entities in jurisdictions that operate that method are identified. At the same time, the operation of the rules outlined in Component 1 is not anticipated to provide scope for manipulation and will be subject to the tax certainty process which should go some way to alleviating any such concern. However, this issue will be further explored.

7.3.4. Conclusion on the choice of a method

641. The exemption and credit methods may ultimately deliver similar outcomes. But some jurisdictions have a strong preference for one method over the other and so it is expected that both the credit and the exemption method should be available to jurisdictions to eliminate double tax on Amount A.

7.3.5. Other technical issues

642. Technical issues that require further work include:

• Interaction of Pillar 2 and the application of the exemption method. Further analysis must be undertaken on how Pillar 2 might impact an MNE group where the exemption method is used to eliminate double taxation on Amount A.

• Interaction of the methods with existing domestic law. Further consideration must be given to how the methods will interact with existing domestic rules governing relief of double tax.

• Interaction of the method and the centralised and simplified administration system and the tax certainty process for Amount A. Further work will involve the examination of the application of each of the methods in the context of other work streams such as the centralised and simplified administration system and the tax certainty process for Amount A covered in Chapters 9 and 10.

• Interaction between Amount A and certain withholding taxes collected by market jurisdictions. As noted in Chapter 6 (see section 6.4), further work will consider the interaction between Amount A and its double tax relief system and withholding taxes by which market jurisdictions may already be taxing a share of residual profits and how any double counting that might arise can be resolved.

7.4. Application of the marketing and distribution profits safe harbour

643. The application of the marketing and distribution profits safe harbour would adjust (and in some cases reduce to zero) the quantum of Amount A allocated to market jurisdictions where an MNE group has an existing marketing and distribution presence. This may give rise to the question of whether an MNE group could choose to adjust their transfer pricing in order to access the safe harbour and use their transfer pricing system as an alternative mechanism to eliminate double taxation.

644. In addressing this question, it is important to emphasise that the marketing and distribution profits safe harbour is not an alternative way to allocate Amount A to a market jurisdiction, but rather a method to determine whether allocating Amount A to a market jurisdiction would give rise to double counting. Correspondingly, where an MNE group already allocates the safe harbour return to a market jurisdiction under the existing profit allocation rules, there would be no allocation of Amount A, and hence no need to apply the mechanism to eliminate double taxation.

645. The principle-based nature of ALP-based profit allocation rules means that the safe harbour contains some limited flexibility for an MNE group to increase the profits it allocates to a market jurisdiction and correspondingly reduce (or even eliminate) a potential Amount A tax liability. However, the safe harbour would not permit a group to use transfer pricing to allocate profits to a market in excess of an arm’s length amount. For market jurisdictions where the safe harbour return is not met, a taxpayer would be required to rely on the Amount A mechanism to eliminate double taxation in order to identify the paying entities and relieve double taxation. For example, where MNE groups have adopted limited risk distribution structures, it may not be feasible to increase the profits of such entities so that the quantum of Amount A would be allocated through the transfer pricing system, as the entity characterisation and transfer pricing method would not allow for non-routine profits to be allocated to a limited risk distribution entity.

646. It should also be noted that where an MNE group meets the marketing and distribution profits safe harbour in a given market jurisdiction, this should not prevent an entity resident in that jurisdiction being identified as a paying entity for other jurisdictions. Though the non-allocation of Amount A in this scenario could be seen to disadvantage a jurisdiction, this is in fact not the case.

7.5. Next steps

647. For Component 1, the process to identify the paying entities, the Inclusive Framework will need to make a final decision on the design of the tests outlined above. This will include agreeing the elements of the tests and the ordering and design of their various components. Further work will also be undertaken on specific aspects of Component 1, including:

• the set of activities included in the positive and negative lists;

• the practical application of the profitability test;

• the practical aspects of implementing a market connection priority test; and

• Consideration of other back-stop positions for any remaining Amount A tax liability not allocated after application of the process

648. For Component 2, the method to eliminate double taxation on Amount A, further work will be required to provide guidance to jurisdictions in selecting and applying each method and, in particular to:

• analyse the interaction of the application of methods to relieve tax on Amount A and Pillar 2;

• consider the interaction between the methods to eliminate double taxation and existing domestic law;

• consider the interactions between Amount A and certain withholding taxes collected by market jurisdictions; and

• examine the application of each of the methods with other work streams including the centralised and simplified administration system and the tax certainty process.

Notes

← 1.  Under the marketing and distribution profits safe harbour, an MNE group would not be required to allocate Amount A to market jurisdictions, where doing so would duplicate a jurisdiction’s existing taxing rights over the residual profits of the group. In this scenario, there would be no allocation of Amount A, and hence the mechanism to eliminate double taxation arising from Amount A would not apply. Instead, an MNE group that meets the safe-harbour with respect to a particular market jurisdiction remains subject to the existing profit allocation and nexus rules.

← 2.  The “paying entities” are the entities that will bear the Amount A tax liability and hence the jurisdictions in which they are resident will be required to relieve the double taxation arising from Amount A. These entities will not necessarily be required to physically pay the Amount A tax liability in each market jurisdiction, as the simplified administration system may allow a single entity within a group to pay the Amount A tax liability (as agent) to each market jurisdiction (see Chapter 10).

← 3.  At the level of a group or segment, the term “residual profit” for Amount A purposes refers to profit in excess of an agreed profitability threshold (see above Chapter 6). This differs from the transfer pricing concept of “residual profits”, which are the profits (or losses) that remain after remunerating activities that can be reliably benchmarked using comparables. Unless specifically noted, the term residual profits used in this chapter refers to profit in excess of the profitability threshold pursuant to the Amount A formula.

← 4.  For ease of reference, the rest of this Blueprint will refer to paying entities (in the plural), though in some instances there may be only one paying entity for a group or segment.

← 5.  This will include a group’s master file, relevant local files, CbCR and other transfer pricing documentation that groups may already prepare. It is recognised that at present this documentation may not be available in all jurisdictions that would be impacted by Amount A, but this can be addressed within the context of the tax certainty process contemplated for Amount A (as discussed in Chapter 10).

← 6.  OECD (2017), Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, OECD Paris, paragraphs 1.60, 1.61, 6.51 and 6.56 constitute primary sources of reference, together with further examples derived from the concepts outlined.

← 7.  The ownership of assets and/or provision of funding alone is not sufficient in this sense, but the entitlement to residual profits from exploiting intangible assets that have been acquired is an indicator of the performance of the value-adding activities, including the DEMPE functions.

← 8.  Risk is not considered in isolation, but is directly associated with decision-making authority over the potentially residual profit-generating activities of the group.

← 9.  Inclusive of the deployment of capital in relation to the same.

← 10.  The return to financing activities is affected by the performance of activities in the market if the performance of the borrower in the market is such that a default or other credit event materialises; however, the financing activity does not sufficiently relate to the actual performance of the in-market borrower. This is to be distinguished from situations where an entity provides decision-making control over commercial and operational functions and the assumption of risks that drive the performance of the in-market entity.

← 11.  It is recognised that not all jurisdictions require taxpayers to file this information with tax administrations contemporaneously, however, in most instances it would still be expect that they would hold much of the information required on file internally.

← 12.  See chapter 3 (section 3.7) of GloBE rules.

← 13.  It is not intended that under Amount A, taxing rights over an entity’s entire profits would be allocated to a market jurisdiction. Instead, as discussed further in paragraph 608 at a minimum a resident jurisdiction should retain taxing rights over an entity’s routine profits, as determined under the profitability test.

← 14.  While the profitability test starts with the financial accounts, corresponding (or tax) adjustments will typically be made in the tax account. Further work is therefore required to align this issue.

← 15.  For example, if the rate of tax in the relieving jurisdiction is 28% and the average rate of tax in market jurisdictions is 26%, an exemption from tax will, in effect, be worth 28% of the exempted income while a foreign tax credit would normally be capped at 26%.

← 16.  The rates used in the analysis were the statutory CIT rates, combining national and subnational rates, in 2019, based primarily on OECD Corporate Tax Statistics. The average was weighted by the estimated amount of MNE sales into each jurisdiction, so that larger markets have more influence over the average than smaller markets. This is an average computed across all MNEs, and individual MNEs could face lower or higher average rates depending on the exact location of their sales.