EBIT’s Members1 thank the OECD for the opportunity to provide comments on the OECD’s public consultation document: Pillar One – Amount A: Draft Model Rules for Nexus and Revenue Sourcing 4 February 2022 – 18 February 2022. Below are a number of issues and open questions that EBIT believes are important for the OECD to take into account. At the same time, we regret the very short timeframe allowed under the public consultation for sending in comments.
Given the urgency and short time frame, EBIT prefers raising its issues and concerns with the nexus and source rules in a summarized bullet point format, keeping it relatively short. At the same time, we do wish to emphasize that the listed issues and concerns in this document are not exhaustive.
I. General comments
- EBIT's Members understand the fragmentary release of the Amount A model rules because of timing constraints. At the same time, we regret that no opportunity will be given to analyse the comprehensive set of Amount A rules and their interdependence.
- The public consultation document indicates that the model rules for nexus have been designed to limit compliance costs. It also indicates the model rules for sourcing balance the need for accuracy with the need to limit compliance cost. EBIT’s Members believe that the rules are complex and will create significant difficulties in terms of compliance, administrability and application for both MNEs and tax authorities. The complexity of the rules and the relatively low thresholds for the nexus test will increase the compliance burden. A higher threshold is necessary to justify the steep rise in compliance costs to both taxpayers and tax administrations. EBIT Members also suggest using a run-in period to test the applicability of the nexus and source rules.
- We consider that the nexus and source rules require further clarification. The relationship between the nexus and source rules with the domestic law and tax treaties (and the related well- known principles and approaches) should be made clearer as well as further interaction with indirect taxes (VAT, sales taxes and alike) provisions are to be thoroughly verified. In general, a change to the new envisaged allocation rules may cause significant profit shifts for MNEs and thus possible exits in certain countries, especially when it comes to IP. This should not lead to double tax costs, for example when the entrance country does not recognize a step-up.
II. Comments on the nexus test
- The nexus test provides for a threshold of 1 million euro (or 250,000 euro for countries with a GDP not exceeding 40 billion euro) of revenue in a country derived from third parties. To determine whether the threshold is passed, reference is made to the consolidated financial statements. Detailed rules including adjustments will be needed to determine whether the threshold has been passed, including whether it is a single year test or, like Pillar Two, a multi- year test. Similar complexities as with the GloBE model rules may arise.
III. Comments on the Source rules
III.1 In General
- The source rules require information from (end) customers under certain circumstances. In particular in the case of sales through unrelated distributors such information will not be available to the MNE. Requesting such information from a third party can be highly sensitive and have an impact on the relationship with the independent distributor. Further, EBIT's Members would like to mention that the request of such information from customers or users could be viewed as a violation of antitrust legislation, and disclosure of confidential business information which will ultimately lead to impacting competition. Finally, it may prove very difficult to assess the accuracy of the information received from (end) customers / users under these circumstances for both MNEs and tax administrations. It cannot be the intention of these proposals that MNEs will be required to use non-verifiable information for the purposes of their Amount A returns.
III.2 Comments on Sourcing rules on a Transaction-by-transaction basis
- The main criterion for the source rules seems to be the transaction-by-transaction basis. The transaction-by-transaction granularity may not be readily available within the MNE based on consolidated financial statements. If such information would be available, it would need to be supplemented and cross-referenced by local data adding to complexity and the compliance burden. This would effectively mean that the main rule would become the exception.
III.3 Sourcing rules on the basis of categories of revenues earned from the transaction and overlapping predominance/supplementary transactions
- The rules contain a potential for inconsistent/contradictory application. For example, the ‘‘predominant” category of revenue and separate country rules (e.g., one invoice) should be clarified. Alternatively, a standard could be established creating a rule order when one or the other criteria shall apply first.
- The transaction-by-transaction approach raises questions on how the 'predominant test' applies to embedded transactions where, for instance, goods and services are provided on a package basis instead of an itemised basis.
III.4 Detailed sourcing rules
- The definition of ‘reliable indicator’ is not clear. Indicators that a MNE uses for other tax compliance obligations should be accepted. In general, the starting point of the sourcing exercise should be the information and methods for sourcing that are available within the MNE, in particular in light of the fairly limited number of MNEs that are in scope of the Amount A rules. The notion of a reliable indicator contains both a subjective (reliability test) and objective (results in line with the sourcing rules) element that may lead to disputes, controversy and uncertainty. A (fast-track) dispute prevention and resolution mechanism when countries disagree on the indicators should be put in place. One possible - and workable - solution could consist in developing safe harbour elections into the allocation keys. Such safe harbour rules would enhance simplicity and certainty. Safe harbours could be construed as a right balance between accuracy, complexity and detail on the one hand and administrability and certainty on the other.
- In many instances, the reliability indicators refer to ‘use’. It is not always clear what is meant by ‘use’; for example, is the use of a service the place of performance of the service or the place of benefit of the service?
- Under part 2, reliable method: § 7 describes an exceptional situation where the Covered Group must use either the Allocation Key or, in the absence of such an Allocation Key, the Global Allocation Key. These situations have been listed in subparagraphs (letters) a-c. However, it is conceivable that these types of situations will only occur in relation to certain jurisdictions, while for other jurisdictions subparagraphs a-c would not be applicable (for instance because there is a reliable indicator for these jurisdictions, etc). To make sure that these situations would not be covered under the exception of § 7, and the mandatory application of the (Global) Allocation Key, EBIT recommends adding some wording to the first sentence of § 7 (for example by adding ‘to the extent that’).
- Under part 2, reliable method: § 8 indicates that the source rules state that a Covered Group must demonstrate that its internal control framework ensures that a Reliable Method is used in accordance with this Part. We recommend adding specific language to ensure that (i) facts and circumstances can be taken into account when assessing this requirement; and (ii) less substantial issues or shortcomings will not have the (unintended) consequence/ result that the requirements of § 8 would be deemed not to be met.
- Under point B - revenues from finished goods sold to final customers through an independent distributor: § 3(b), seems to indicate that all (remaining) tail-end revenue will be allocated to low-income jurisdictions, unless the Covered Group demonstrates that revenues did not arise in any low-income jurisdiction. Arguably there could be a question how to deal with the situation where tail-end revenue did not only arise in low-income jurisdictions. A literal application of § 3(b) could lead to the conclusion that all tail-end revenue needs to be allocated to low-income jurisdictions. This would arguably represent an unbalanced outcome, potentially even an over-allocation to low-income jurisdictions in cases where it is clear that tail-end revenue did not only arise in those low-income jurisdictions. Sourcing of end customers or users may prove to be burdensome and difficult.
- Further under point B: §(4) provides that the Covered Group must take reasonable steps to reduce the size of the Tail-End Revenues [5% max]. Footnote 16 indicates that there will be provisions on non-compliance. It seems that 5% is rather low, in particular when a Covered Group is predominantly active in developing countries. In EBIT’s view it would be better to first gain further experience with the practical application of this particular requirement. We therefore suggest removing this requirement, or including a higher percentage, for instance 15%, until further experience is gained with the practical application of this particular requirement. Also, for the avoidance of doubt, it would be good if it could be clarified in the Commentary that this does not include any “Non-customer Revenues”, which can be deemed to arise in a jurisdiction in relation (proportion) to certain revenues, including Finished Goods (see for instance Page 25, Part 9).
- With regard to rule H - revenue from business to business: § 5 necessitates that in order to apply the Headcount Allocation Key, the Covered Group must take reasonable steps to obtain the jurisdictional breakdown of headcount of the Large Business Customer. This is usually not information that will be available or that the LBC will be ready to give. Such information will be available however to the tax authorities under Country-by-Country Reporting but cannot be shared with the relevant taxpayers because of confidentiality issues. It would seem therefore inappropriate to impose this allocation key when the MNE has no access to this information.
- Under Part 4, Components: § 1 indicates that revenues from components are sourced in the place of delivery of the final customer of the finished product into which the component is incorporated. For the MNE producing the component, the destination of the end product may not be known. Under other circumstances, the same component can have different applications and be integrated into different end products. In such a case, the component provider might not know in which product the component ends up, let alone its destination. Further, theinformation needed would come from third parties and is not under control of the component provider.
- Part 6 intangible property: under § 2 it would seem that the place of use of IP covered in these cases is the same place of use as the goods in which the IP is embedded. One element of simplification could consist in combining the approaches instead of designing a separate rule.
- Part 7, real property: the use of the term real property is confusing as the definition refers to immovable property. It would be appropriate to align the language, for example with the OECD MTC. As to determining the location of real property on the basis of the jurisdiction granting the right to exploit the real property, it is not entirely clear how this criterion relates to the market state (user or consumer) and what would be the difference to where the immovable property is situated.
- With regard to (online) services being provided to a customer who is operational in different jurisdictions and utilises these (online) services simultaneously in a range of jurisdictions, this raises questions as to how the revenues earned from these services should be sourced to individual markets under the envisaged sourcing rules.
- Application of the knock-out rule: it would seem that the knock-out rule as defined in the source rules should identify a jurisdiction or group of jurisdictions where it can be reasonably assumed that revenues did not arise. This rule will have as a result that the MNE has to demonstrate that sales did not happen in those jurisdictions and render a proof of not being present through sales. This negative approach seems awkward and will be burdensome.
- Definition of categorising transactions: Part 1 provides further clarity in respect of the categories “Main Transaction” and “Supplementary Transaction”. For the latter it is stated that the gross receipts should not exceed 5% of the total gross receipts from the Supplementary Transaction(s) and Main Transaction(s) combined. It seems to EBIT’s Members that 5% is rather low, in particular when more than one Supplementary Transaction(s) takes place.
- Part 5 - Definition of services: § 36 states that “If the total invoice amount for services provided to a Business Customer does not exceed EUR [1-3] million in the Period, the Covered Group may treat the Business Customer as not being a Large Business Customer.” Footnote 47 further explains that if the invoice amount for services provided to the Business Customer exceeds EUR [1-3] million in the Period, reasonable steps could require the Covered Group to ask the Business Customer if it meets the definition of a Large Business Customer. In this respect it is noted that the Covered Group could be expected to undertake online research to view the financial statements of that Customer and whether such statements record consolidated group revenue exceeding EUR 750 million. In addition, once Country-by-Country Reporting data becomes publicly available in EU jurisdictions, a Covered Group would be expected to check whether the business customer has made such information publicly available.
IV. Other points
- One approach for simplification could consist in allowing the dispute avoidance and resolution panel to determine the source rules (for example on a formulaic approach), based on the general or global data with regard to Amount A provided by the MNE to the panel, for example based on an internal control framework. Once the information is provided, the allocation based on source rules is determined by the panel. This could also be done on a formulaic approach (eg. GDP of the source states) for simplicity.
- The necessary adaptations and alterations to the MNE recording and reporting systems to comply with the nexus and source rules should not be underestimated. In order to enhance certainty and simplicity, a pre-approval of data collection techniques could be useful.
- The rules should be applied consistently within the relevant jurisdictions. Indeed, sourcing issues may arise when certain countries accept the indicator, while others do not. For such cases, fast track dispute avoidance mechanisms should be available.
V. Overall Conclusions
- The nexus and source rules are very complex and will lead to uncertainty, in particular when different countries apply different sourcing rules.
- The administrative compliance burden of these proposals for MNEs, their contractual third parties and tax authorities must not be underestimated.
- The absence of a comprehensive document including the nexus and source rules, as well as the absence of consensus, increases the difficulty of commenting on these rules.
EBIT’s Members hope these comments are taken into account by the OECD. We are always keen to engage in further discussion and public consultations that will be required if matters are to be implemented successfully.
European Business Initiative on Taxation – 18 February 2022