3. Nexus

186. The new international taxation framework set forth in this Blueprint recognises that in an increasingly digital age, taxing rights can no longer be exclusively determined by reference to physical presence. It therefore contains new nexus rules for in-scope revenue referred to in Chapter 2.

187. As explained in Chapter 2, the scope tests seek to capture those large MNEs that are able to participate in an active and sustained manner in the economic life of market jurisdictions through engagement extending beyond the mere conclusion of sales, in order generate profits, without necessarily having a commensurate level of taxable presence in that market (as based on existing nexus rules). In this regard, Chapters 2 and 3 need to be read together as they translate a common underlying rationale.

188. The nexus rules design intends to protect the interests of smaller jurisdictions, and in particular developing economies, and their desire to benefit from the new taxing right. It recognises the need for low and proportionate compliance costs.

189. The new nexus rules determine entitlement of a market jurisdiction to an allocation of Amount A only. They do not alter the nexus for other tax purposes, customs duties or for any other non-tax area. The new nexus rules will be designed as a standalone provision to limit any unintended spill-over effects on other existing tax or non-tax rules.

190. The new nexus rules could apply differently for ADS and CFB. For ADS, exceeding a market revenue threshold could be the only test to establish nexus. The very nature of the ADS allows them to be provided remotely and such businesses generally have a significant and sustained engagement with the market even if there is not a physical presence, which is one of the key challenges in taxing the digitalising economy.

191. For CFB, the ability to participate remotely in a market jurisdiction is less pronounced. This, together with the additional complexity and compliance costs associated with sourcing revenue derived by CFB (e.g. third party distribution) and the broad acknowledgement that profit margins are typically lower for CFB compared with ADS, could justify a higher nexus standard for CFB. Although many Inclusive Framework members, seeking simplicity, would prefer a nexus threshold based solely on revenue, other members consider that a higher nexus standard for CFB is essential. One approach for satisfying this higher nexus standard is through a higher threshold and the presence of additional indicators (“plus” factors) which would evidence an active and sustained engagement in that jurisdiction beyond mere sales.

192. The Blueprint therefore sets out an approach based on the following elements. A nexus is achieved:

  • for ADS:

    • with revenues of more than [EUR X million] a year;

  • for CFB:

    • with revenues of more than [EUR X million] a year and;

      • a “plus factor” to indicate a significant and sustained engagement with the market. One plus factor could be a subsidiary, or a “fixed place of business” (e.g. a permanent establishment based on the commonalities of the UN and OECD Model definitions), with the requirement that the entity or PE is carrying out activities that are connected to in-scope sales. Consideration will be given to the possibility of treating revenues of more than EUR XX million as a plus factor (where the MNE would be deemed to have an active engagement beyond mere sales). Other plus factors may also be considered (which may be unconstrained by physical presence).

193. Depending on where threshold amounts are set, consideration will be given to using a lower nexus standard for smaller developing economies, with GDP below a certain level while also considering compliance simplifications.

194. For in-scope MNEs the new nexus rules would be based on indicators of a significant and sustained engagement with market jurisdictions. In other words, absent this engagement with a jurisdiction, none of a group’s profits would be reallocated to that jurisdiction under Amount A. The new nexus rules would operate in a standalone manner to limit any unintended spill-over effects on other existing tax or non-tax rules. Thus they would not be used as a basis to create a nexus for any other taxes, customs duties or for any other non-tax purpose. For example, the new rules will not affect the operation of the existing permanent establishment rules, which will continue to operate as at present. The concrete design of this standalone provision, outside existing tax treaties, will be undertaken when the substantive elements of scope and nexus have been settled.1

195. The market revenue thresholds will apply to the in-scope revenue of a group (or segment of a group where relevant) generated in a market jurisdiction. The market revenue will be identified and measured in accordance with the revenue sourcing rules (see Chapter 4), and could apply separately to ADS and CFB.2 The nexus rules will follow any wider segmentation approach, i.e. the nexus will be assessed at the level of a segment where a group segments its operations in computing its Amount A tax base (see Chapter 5).

196. Further work will need to be undertaken to decide whether to apply a temporal requirement. Such a requirement would avoid covering isolated or one-off transactions that might not demonstrate a sustained engagement with a market. A duration test could be designed by requiring that the market revenue threshold be exceeded over a period spanning more than one year before establishing a nexus. However, assessing nexus year by year is simple and would be more aligned with other testing periods throughout Amount A.

197. The Outline indicated that the market revenue threshold would be commensurate with the size of each market, measured, perhaps by GDP. But this is likely to add substantial complexity, so the nexus rules could use a simple monetary amount of revenue in the market – one for ADS and one for CFB. However, consideration is being given to using higher thresholds for large markets and lower thresholds for small, developing economies.3

198. In determining the level at which to set thresholds, several considerations will need to be taken into account. The data analysis prepared as part of the impact assessment, while subject to significant uncertainty, suggests that a market revenue threshold has an important impact on the Amount A allocation to smaller jurisdictions, in particular developing economies (unlike the global scope revenue threshold)4. For the smallest jurisdictions (e.g. jurisdictions with GDP less than USD 5 billion), the analysis suggests that many MNE groups may not have a nexus in these jurisdictions if a single threshold is set at EUR 5 million or above. On that basis, and the considerations further discussed below, a possibility is to have two separate market revenue thresholds: one for ADS, and one for CFB, noting that the amounts of such thresholds are expected to be set below EUR 5 million. As stated in paragraph 197, consideration is being given to using higher thresholds for large markets and lower thresholds for small, developing economies.

199. Different threshold amounts for ADS and CFB could be justified on the following grounds. First, the ability of CFB businesses to participate remotely in market jurisdictions seems less pronounced compared to ADS. Second, additional complexity and compliance costs associated with sourcing revenue derived by CFB (e.g. third party distribution) compared with ADS also points to a higher threshold amount for CFB. Finally, profit margins are typically lower for CFB compared with ADS, and hence that the threshold amount should be greater for CFB to ensure the same balance between tax benefits for market jurisdictions and overall compliance and administrative costs. However, some jurisdictions prefer that the thresholds be aligned.

200. Further work will be undertaken on the administration and co-ordination of the thresholds used for Amount A to ensure they remain appropriate over time.

201. For ADS, the market revenue threshold would be the only test to establish nexus.

202. For CFB, a level of sales just over the market revenue threshold may not denote the active and sustained engagement with the market beyond the mere conclusion of sales that is envisaged as the justification of the new taxing right. To demonstrate this level of engagement, the presence of additional indicators (“plus factors”) may be necessary.

203. Many Inclusive Framework members, seeking simplicity, would be prepared to accept a nexus threshold based solely on revenue, especially since the scope tests already seek to capture those MNEs that participate in a sustained and significant manner in the economic life of market jurisdictions. Other members, however, consider plus factors as essential to a nexus rule for CFB. In order to bridge the gap between the two groups this Blueprint suggests that plus factors could be required for a CFB to demonstrate a nexus. One option being considered is that, where sales in a market have passed a certain level, the MNE can be presumed to have an active engagement. In that case, plus factors would be deemed to exist. As explained later, for many groups there will be a level of sales over which they will in any event want to establish a local presence. The application of plus factors for small, developing economies may need to be considered with the level of the thresholds.

204. Several possible plus factors have been examined. They include having an existing physical presence in the form of a permanent establishment (PE) or the residence of a group entity; a physical presence that falls short of a PE; and the undertaking of material, targeted and sustained advertising and promotion (A&P) activities that support in-scope sales into the market jurisdiction.

205. In the interest of simplicity, the likely indicator used would be just this one. A physical presence test in the form of a subsidiary or PE seems the simplest way to establish a nexus (beyond mere selling), particularly as the PE or resident entity will likely already have filing and reporting requirements in that jurisdiction. In other words, the nexus applicable for one entity (its place of residence or having a local PE) will apply to the whole group.

206. To establish an “active and sustained physical presence”, there could be a requirement in this test that the activities carried out are connected to in-scope revenues. This requirement could include not only distribution activities but also activities directly supporting sales into the market, e.g. facilitating billing and payment in the local currency, collecting indirect taxes and duties, transportation, maintenance of local stock, cross border delivery, after-sales support, repair and maintenance, advertising and promoting and adapting product or services to the particular market. Activities without a sufficient connection to sales, and which would therefore not constitute the sustained physical presence envisaged in the test, could include, for example, research and development functions. However, further technical work will be undertaken to define the nature of such a connection.

207. In more detail, the group-PE test will be met if any entity in the group has a PE in the market (as defined under a standalone definition explained below). It will also be met if a group entity is resident in the jurisdiction. In both cases, the local activities must be connected with the tested sales. An entity established with a view to aiding the group’s future expansion would not trigger the nexus test if, for example, it had no premises and no employees.

208. The question is then how to assess whether a group-PE exists. This is an issue because several treaties with different PE definitions may be in play (or there may be no treaty at all).

209. One possibility is to take the PE definition in any tax treaty that exists between the state of residence of a group entity and the market jurisdiction that has triggered the existence of a PE in practice, provided the PE is connected to in-scope revenues. But this option could lead to issues of fairness and risk of distortions:

  • Where there is no applicable tax treaty, and the fall-back is the domestic law of a market jurisdiction, an MNE could have a taxable presence from merely selling services, and perhaps even goods, into that market.

  • Relying on different PE standards found in existing tax treaties may lead to unfairness where one market jurisdiction (with a standard PE threshold in its tax treaties) may not receive any tax under Amount A from a given MNE, while another jurisdiction – with a lower PE threshold – would collect Amount A taxes from that same MNE (and for the same activities). The same fairness issue arises if the definition relies on domestic legislation.

  • If Amount A is conditional on an MNE having a PE under existing treaties or domestic legislation, tax administrations may feel additional pressure to find a PE to exist, increasing the chances of PE disputes between jurisdictions with a potential impact beyond the scope of amount A (increasing the risk of spill over effects).

210. To remove the concerns mentioned above, the preferred approach, is to use a single self-standing group-PE definition, instead of relying on a PE definition in a tax treaty or domestic law. Such a standalone provision will not affect the application of PE rules in existing tax treaties or domestic legislation, which will continue to operate as at present for tax purposes, other than the allocation of Amount A. Ready-made possibilities exist (e.g. in the OECD and UN models). But it seems preferable to use one that is designed for the purpose of Amount A. The starting point for doing this is the two models.5

211. The intention, therefore, is to develop a group-PE test taking the basic PE principle shared between the UN and the OECD Models: a fixed place of business through which an in-scope CFB of the MNE group is wholly or partly carried on and applying this to the MNE group. Further work will be undertaken on other aspects of the existing PE definition such as Article 5(4) on preparatory and auxiliary activities and Article 5(5) on “dependent agent”. This group-PE definition is only for the purpose of establishing a new nexus for Amount A.

212. As noted above, beyond a certain level of sales, it is increasingly unlikely that a group will be selling into a market with no supporting activities. For many groups, there will be a level of sales over which they will want to establish a local presence, either in the form of a subsidiary or a PE. Depending on the industry, that might be when annual sales reach a certain level. But before then, it is likely that the group’s engagement will increase its sales. One approach could be, therefore, to assume that once a group’s CFB sales in a market reach a certain threshold it will no longer be necessary to establish the existence of plus factors (i.e. the group could be treated as having a nexus). This deemed engagement provision would avoid the complexities and uncertainties that more factual and expansive plus factors would create, but further policy discussion is needed.

213. Depending on where revenue thresholds are set, consideration may also need to be given to using a lower nexus standard for small, developing economies. Given the size of their economies within the context of determining a significant and sustained engagement in the market, and the additional complexity in applying and verifying the plus factors for tax administrations with often very limited resources, both principles and practicalities may argue for further reflection. This may also need to involve considerations relating to compliance simplifications and administrative costs, including for other jurisdictions.

214. A few Inclusive Framework Members have favoured the addition of a test based on a sustained presence of personnel in a market jurisdiction (e.g. 183 days in a year), which could be an alternative way of establishing nexus if the physical presence was not met, e.g. for smaller jurisdictions. Many members, however, questioned the relevance and administrability of this test.

215. A test of advertising and promotion (A&P) expenditure is another test that has been explored. If it were adopted, it would be an alternative way of establishing nexus if the physical presence test were not met.6 The test could be met where, for instance, A&P expenditure for a jurisdiction exceeded a certain percentage of the market revenue threshold, perhaps with some simplifications: e.g. focusing only on market-specific expenditures (no regional or global expenditures) but including expenditure incurred outside the market where it is directed at the market. The A&P expenditure test could also be useful for franchising and licensing business models. However, such a test may not necessarily be an accurate measure of an active and sustained engagement with a market jurisdiction as expenditures could, for example, fluctuate over the lifetime of a brand. Such a test may also apply differently between established businesses and new businesses, given that penetrating a market with a new brand requires a higher level of A&P expenditures. Further work will therefore be carried out on the technical aspects of this test, noting however, the administrative difficulties and challenges.

216. As a next step, decisions will be needed on the amount of the thresholds, the use of plus factors, and if so, which ones, and whether and how, the rules might be adapted to the needs of small, developing economies.

217. In addition, further technical work will be undertaken on:

  • Defining a possible temporal requirement (the period over which an MNE will have to satisfy the above tests: whether reaching the nexus threshold for one year will be sufficient or whether it will be necessary to apply the test over a longer period);

  • The precise details of a standalone “physical presence” definition based on the commonalities of the UN and the OECD models;

  • The issue of the connection between the plus factors and the in-scope sales;

  • Work on possible plus factors, with specific considerations for franchise and licensing business models; and

  • Consistent technical definitions and rules applicable to all thresholds along Amount A (currency, timing issues, etc.).


← 1. See Chapter 10 on implementation.

← 2. This separate approach was developed to prevent the distortions that an aggregated approach would create in the treatment of mixed revenues. Using thresholds based on revenue from a market is less reliant on complex factual determinations compared with other factors, and therefore will limit compliance and administration costs and provide certainty.

← 3. See below paragraph 213.

← 4. Due to the lack of comprehensive entity level data on MNE sales in each jurisdiction with which to assess a nexus revenue threshold, a probabilistic modelling approach has been developed to approximate the effect of an illustrative range of potential nexus revenue thresholds. To consider the implications for a range of jurisdictions according GDP size, a range of thresholds (EUR 1m, 3m, 5m and 7m) were considered for ADS and CFB combined. As this approach is inevitably assumption-dependent, results should be considered as illustrative of the orders of magnitude rather than precise estimates. The impact assessment of various nexus thresholds is contained in CTPA/CFA/WP2/NOE2(2020)5.

← 5. Although there are differences between them, these may not be material for Amount A purposes. The existence of an insurance PE provision in the UN model would not be relevant, for example, if insurance services are removed from the scope of Amount A. Nor does the presence of “delivery” activities in the OECD Model’s list of exemptions make any difference after the BEPS Action 7 changes.

← 6. A possibility favoured by some jurisdictions would be to require both the physical presence test and the A&P test to apply as evidence of an active and sustained engagement with the market beyond the mere conclusion of sales.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.