Category Archives: Action 7

Comments on Draft Additional Guidance on Attribution of Profits to a Permanent Establishment

14 September 2017

The BMG has made a submission on Attribution of Profits to a Permanent Establishment in response to the OECD Discussion Draft.


A major motivator in initiating the entire BEPS project was to end BEPS motivated planning by centrally managed groups. Such planning often attributes sales to zero or low-taxed entities and separates sales through fragmentation from related core functions such as marketing, order fulfilment, and customer support performed by other group entities. Under Action 7 of the BEPS project some modest changes were agreed, so that in defined circumstances a non-resident entity could now be found to have a taxable presence (permanent establishment – PE) in a country in which it makes sales. The current proposals aim to clarify how profits should be attributed to such a PE.

We agree that attribution of profits depends on an analysis of the functions performed by the PE, but in our view this must not be done in isolation. A holistic approach should be adopted, which considers all the activities carried out in the country by the relevant entities in conjunction. Where a multinational chooses to carry out itself activities such as marketing, sales, order fulfilment, and customer support, it does so in order to take advantage of the synergies so created, thereby giving the customer a seamless experience and itself (i.e., the group) a significant market advantage. Hence, it is the cumulative importance of all group activities that should be considered when evaluating the value which is created in the country.

Due to this cumulative importance, our view is still that article 7 should be applied prior to article 9, since this would result in both better focus by taxpayers and tax authorities, and a practical reduction in the resources needed by both tax authorities and taxpayers for compliance.

A holistic approach will also lead in some circumstances to a different transfer pricing method being the most appropriate method. In particular, where such related functions are performed by highly integrated associated entities and are viewed holistically, the profit-split method is likely to prove more appropriate than one-sided methods.

A holistic approach is also important since the DD is meant to apply to all versions of article 7 of the model convention, and whether or not a state has accepted the changes adopted by a majority of OECD states in 2010, described as the authorized OECD approach (AOA). While the AOA has some merits, it has been used to further exacerbate a fragmented approach to the attribution of profits, which (along with the independent entity principle in general) has been a principal enabler of BEPS. Adoption of the holistic approach which we suggest could, we believe, allow some of those helpful features of the AOA to be retained, while ensuring that BEPS structures are not allowed to continue due to a narrow interpretation applying the independent entity principle to an entity which is not even legally separate.

Our Specific Comments section includes a number of concrete suggestions to make the DD more internally consistent and effective in its application.


Attribution of Profits to Permanent Establishments

The BMG has made a Submission to the OECD on its consultation on revisions to the guidance on Attribution of Profits to Permanent Establishments.


This discussion draft (DD) deals with attribution of profits to a host country resulting from changes to the taxable presence requirement in the definition of a permanent establishment (PE) in BEPS project’s Action 7. Although generally clear and well reasoned, it is of limited usefulness in our view, for two main reasons. These comments explain these shortcomings and suggest how they could be corrected.

First, it applies only to the 2010 version of the OECD model convention, which introduced the ‘authorised OECD approach’ (the AOA) for attribution of profits to a PE. The AOA attempts to extend to PEs as far as possible the independent entity principle as applied to associated enterprises within a multinational enterprise (MNE). A number of OECD countries have not accepted the AOA, and it has also been generally rejected by developing countries. One reason for this is that the independent entity principle is especially inappropriate for a PE, which by definition is part of the same legal entity. Hence, few actual treaties are based on the AOA, and this is also true for most national tax law rules which would apply to entities resident in non-treaty countries. States, especially developing countries (whether or not they decide to join the Inclusive Framework for BEPS), should not be pressurised into adopting the AOA. Instead, the UN Committee of Tax Experts, in liaison with the OECD, should develop its own revisions to the commentary to the UN treaty model consequent on the changes to the PE definition introduced by Action 7. Further work is clearly necessary, by a wider range of countries, and adopting a broader approach, to produce guidance that would be of use to tax payers and tax authorities, especially in the bulk of cases where the AOA is not applicable.

Secondly, the examples provided in the DD adopt a very restricted approach, which assumes that all or most significant people functions take place in the non-resident entity, and hence attribute only limited profits to the PE. They include some illustrations of when aspects of inventory and credit risk management may take place in a PE, but significantly the examples include no discussion of other sales-related functions such as marketing and advertising, which are instead assumed to be controlled by the non-resident entity, with no relevant local input. Similarly, the examples are silent regarding core business functions conducted in host countries that are often found in modern MNE business models. These simple examples may be relevant to relatively small firms based almost entirely in their home countries, which employ a foreign sales agent.  But they are entirely unrealistic in relation to most large MNEs and their modern business models, which aim to be both global and local. No MNE can operate effectively by centralising virtually all its significant people functions and all its core business functions at a distance from its customers and suppliers, as is assumed in the examples provided here. Indeed, there are many well known examples of MNEs which employ significant staff in host countries engaged in both customer-facing and many core business functions. The failure of this DD to discuss such situations suggests a lack of consensus on how to deal with them, which may regrettably exacerbate the likelihood of conflicts even between OECD countries.

As the DD is now drafted with its focus on the AOA and its unrealistically simple examples, its effect is to strengthen the BEPS mechanisms used by many MNEs. This contradicts the mandate for the BEPS project, which is to align taxation and value creation.

Comments on BEPS Action 7: Revised Discussion Draft on
Preventing Artificial Avoidance of Permanent Establishment Status

Under international tax rules, whether a foreign company can be taxed on profits earned from activities in a country depends on whether it has a Permanent Establishment (PE). Devised a century ago, this concept has become easy to avoid, and is clearly inappropriate for the post-industrial age. The BEPS Action Plan proposed only a relatively modest reconsideration of some aspects of the rules, although it should also be considered under Action 1 relating to the Digital Economy.

The BMG has now published its comments on the OECD Revised Discussion Draft, the latest and final proposals under this action point.

Summary of BMG Comments

We are concerned that this revised discussion draft (RDD) seems to have given too much weight to the quantity of comments received on the initial draft, failing to take into account that since the vast majority came from the same community of MNEs and their paid tax advisers, it is hardly surprising that they tend to agree. In particular the RDD picks out the comments which defended ‘the fundamental concept of the independence of legal entities’, without mentioning our submission pointing out that it flies in the face of business reality to claim that associated entities within a multinational corporate group operate as independent parties, a view which is supported by economic theory and practice, and by most independent commentators.

We welcome that nevertheless a majority of the Working Party agreed to the proposed anti-fragmentation rule, although it has been given very limited scope. Indeed, the Working Party has interpreted its mandate very narrowly, focusing essentially on commissionaire arrangements and sales-related activity, so that its proposals will only deal with a few types of corporate fragmentation, those related to delivery of goods to customers. The proposals on commissionaire arrangements leave unclear whether they will apply even to activities such as marketing, while failing to deal with many other types of functional fragmentation which facilitate BEPS.

The RDD does not deal at all with the reality of the modern world in which real value is created through scientific research and the development and testing of products and services, in continuous processes of innovation and improvement. This is not just a ‘digital economy’ issue. Spending on innovation is key to the success of many businesses today. This must be reflected in Article 5 with sufficient nuance that MNEs will not continue to undervalue for source countries the value that is truly created within their borders.

The report also does not deal with the key issue of attribution of profits, which has been deferred to be dealt with after the G20 deadline for the BEPS project, perhaps in conjunction with the continued work on the Digital Economy. The weakness of these proposals is confirmed by the introduction by some OECD countries (e.g. Australia, UK) of unilateral measures to tax ‘diverted profits’. Developing countries should also take measures to protect their tax base, e.g. through withholding taxes on fees for services, but these are blunt instruments (applying to gross payments rather than profits). An internationally coordinated approach would clearly be far better both for business and revenue authorities.

The BEPS project is in our view seriously weakened by this failure to reconsider the permanent establishment concept to provide a definition of taxable presence more suited to the 21st century.

Reform of the Criteria for Taxable Presence (Permanent Establishment)

The BMG has published its comments on the OECD proposals on Preventing Artificial Avoidance of Permanent Establishment Status, under Action 7 of the BEPS Project, which deal with the criteria for taxable presence of a foreign company.

Summary of BMG Comments

We welcome the move away from the independent entity principle, which allows and encourages multinationals to create complex corporate structures, fragmenting functions among different affiliates to avoid tax. Regrettably, however, these proposals are limited to (i) situations where a multinational selling in a country has an affiliate or agent involved with conclusion of contracts; and (ii) a proposed anti-fragmentation rule which only covers pre-sales-related activities, and not e.g. delivery or customer support. Many countries already have stronger rules, or are introducing them, notably the UK’s new Diverted Profits Tax. To avoid increasing conflicts, multilateral agreement is needed on a broader rule.

We suggest revision of Article 5(7) of tax treaties to reverse the presumption that the fact that two companies are under common control shall not ‘of itself’ create a taxable presence for the multinational as a whole. To align income from sales with expenses, greater use should be made of the profit split method in transfer pricing, or the apportionment approach which is still allowed for a PE in most treaties, especially with developing countries.