Attribution of Profits to Permanent Establishments. Globalization has led to enterprises doing cross-border business operating through foreign permanent establishments. However, the allocation of profits to such permanent establishments has always been problematic. This article analyses the authorized OECD approach as compared with the previous method of calculating the profit attributions to permanent establishments and thereby addresses the question of whether the authorized OECD approach should be maintained as an OECD standard.
The calculation of a company’s business profits involving cross-border activities has always been a problematic area in international tax, in particular in situations in which international trade is conducted through a permanent establishment (PE) of a foreign enterprise in a host state. Historically, the OECD has introduced two methods to achieve inter-nation equity, with the new “authorized OECD approach” (AOA) replacing its previously enshrined approach, . This article endeavours to analyse both the advantages and disadvantages of the AOA against the background of the pre-AOA in order to subsequently assess whether the AOA should be maintained as the OECD standard. Although there are contrary perspectives, that the AOA has apparently not become the object of global consensus and therefore needs to be improved .
On the basis of Mitchell B. Caroll’s report from 1930, the OECD included article 7, with respect to business profits, in its draft model tax convention (1963) in order to provide clarity on and symmetry regarding how a company should calculate its business profits arising from a foreign PE. Though article 7 of the OECD Model did not change significantly from 1963 to 2010, its Commentary was amended several times, with the most important amendments occurring in 1994 and 2008. In 2010, however, the OECD, in order to achieve consistency, introduced a new article 7 based on the 2010 OECD report on the Attribution of Profits to Permanent Establishments (the 2010 PE Report) and, at the same time, kept the significantly different pre-2010 version of article 7 of the OECD Model as well. It is worth noting that the pre-2010 version of article 7 of the OECD Model, as it reads up to 2008, provided as follows:
The post-2010 version of article 7 of the OECD Model provides as follows:
Effectively, the AOA moves the taxation of the PEs nearer to that of subsidiaries, as at a first step considering the PE as a “separate and independent enterprise engaged in the same or similar activities under the same or similar conditions” and, as a second step, applying the OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (OECD Guidelines) by analogy – “taking into account the functions performed, assets used and risks assumed by the enterprise” – in this respect. For example, in the case that the ownership of an intangible asset such as intellectual property is attributable to the PE and, at the same time, other parts of the foreign enterprise use such intellectual property, an internal royalty paid to the PE would be recognized. Table 1 provides a summary of the most critical discrepancies between the post-2010 and pre-2010 versions of article 7 of the OECD Model.
Table 1: Discrepancies between the post-2010 and pre-2010 versions of article 7 of the OECD Model
|Post-2010 article 7 OECD Model||Pre-2010 article 7 OECD Model|
|Royalties||Royalties at arm’s length price||Shared cost|
|Interest||Arm’s length interest, subject to free capital||Shared cost, subject to capital (but notional interest deductions are allowed for banks)|
|Services||Arm’s length charges||Shared cost, with markup in certain circumstances|
|Good management||Arm’s length charges||Non-deductible|
|Transfer of assets for sale||Arm’s length price||Arm’s length price|
|Temporary transfer of assets||Rental fee at arm’s length price||Shared cost|
The pre-2010 version of article 7 of the OECD Model provided limited guidance with respect to calculating the profits attributable to a PE, and therefore countries used different methods in attributing profits to PEs, among them the “functionally separate entity approach” and the “relevant business activity approach”. The result of this lack of uniformity was that, with states using different rules of calculating the profits attributable to a PE, double taxation – or sometimes no taxation – could occur. Consequently, in order to achieve certainty and consistency, the AOA introduced new rules in article 7(2) and (3) of the OECD Model (2010) (see section 2.) that clearly rejected the relevant business activity approach and emphasized the functionally separate entity approach. Furthermore, the new rules in effect provided a uniform method and straightforward structure as compared to the more complicated mixed structure of the pre-2010 versions of article 7 of the OECD Model. Nevertheless, countries raised various interpretative concerns as to whether any new amendments might have a direct impact in respect of their internal law or might face rejection under their constitutions on the grounds that the AOA arguably violates the ability-to-pay principle. Further issues arose in civil law jurisdictions, because the transfer of economic ownership of assets seems infeasible. Nevertheless, the OECD did not clarify this relationship, as the AOA does not raise charge to tax but somewhat limits the profits attributable to the PE.,
A similar concern was also raised in respect of the pre-2010 version of article 7(4) and (6) of the OECD Model, as those paragraphs allowed countries to follow an alternative method, namely the “indirect” or “empirical” method, which triggered problems of double or no taxation. This issue was resolved with the introduction of the AOA, as the pre-2010 versions of article 7(4) and (6) of the OECD Model were abolished, and article 7(2) of the OECD Model (2010) accepts only the direct method.
Another significant advantage of the AOA is that it manages to provide symmetry in the calculation of profits between the head office state and the host state. In the pre-2010 version of article 7 of the OECD Model, at least two problems were identified. First and foremost, there was no mechanism to provide for corresponding adjustments similar to article 9(2)of the OECD Model, as one state could adjust the profits while the other state had no obligation to act accordingly. Second, there was an asymmetry in the approach taken by different states, as noted above in this section, and conflicts of qualification arose from differing interpretations of tax treaties and domestic law. In essence, the pre-2010 version of article 7 of the OECD Model provides that, on the one hand, the head office state needs to calculate the profits in order to provide relief and, on the other hand, the host state needs to tax the profits attributable to the PE. As a consequence, the symmetrical application of article 7 of the OECD Model is instrumental in avoiding double or no taxation. Effectively, these issues are likely to be solved, as the wording of article 7(3) of the OECD Model (2010) maintains explicitly that the host state needs to provide for relief under the mechanism of article 23. Even if there is a difference between the interpretations of the contracting states, the corresponding adjustment mechanism, similar to the mechanism of article 9(2) of the OECD Model, is also applicable for solving this issue. Hence, certainty is provided for the head office state in respect of eliminating international double taxation.
An additional issue that was further tackled by the AOA is tax planning techniques involving transferring assets. For instance, one could consider that an enterprise could transfer its assets to its PE, and the PE might then decide to sell these assets to third parties. Consider further that the head office state provides relief via the exemption method for the PE’s profits. Under the pre-2010 version of article 7 of the OECD Model, no profits are released in relation to the transfer of assets from the head office to the PE, but profits would only arise when the PE sells the assets to third parties. Such profits would remain untaxed by the head office state, because it provides relief via the exemption method. Under article 7 of the OECD Model (2010), these tax planning techniques would be minimized, as internal dealings between the head office and the (see Table 1) are now identified and measured.
Finally, the AOA has also clarified both the issue of the force-of-attraction rule, followed mainly by developing countries, and the issue of whether the head office can allocate profits to a PE when the enterprise as a whole has losses and vice versa. Article 7(1) of the OECD Model (2010) clarifies this issue, and the head office can now attribute profits to a PE even if the enterprise as a whole has incurred worldwide losses. As a consequence, double or no taxation seems likely to be eliminated.
Primarily, the AOA aims at attaining consistency among OECD and non-OECD member countries. Nevertheless, it has enjoyed only partial success, as many OECD and non-OECD member countries have raised reservations about using article 7 of the OECD Model (2010). Moreover, this issue is further compounded by the fact that the UN Model has rejected article 7 of the OECD Model (2010). By contrast, only a few countries, like the Netherlands, have fully adopted article 7 of the OECD Model (2010), either by including it in new double tax treaties or by changing their internal law. In general, one is left wondering whether consensus has been achieved by the OECD’s introducing of the AOA, as a large number of states have clearly rejected article 7 of the OECD Model (2010). In effect, many scholars believe that the AOA will disappear in a number of years, because a large number of important jurisdictions have rejected it as the Common Consolidated Corporate Tax Base (CCCTB) in the European Union and the destination-based cash flow tax in a global context.
Nevertheless, even if consensus were to be achieved among states, there are numerous difficulties that arise in respect of successfully implementing the AOA worldwide. For the AOA to be effective, article 7 of the OECD Model (2010) needs to be included in the approximately 3,000 bilateral tax treaties now existing. Though states have been able to adopt article 7 of the OECD Model (2010) when concluding new bilateral tax treaties, has not proven to be the case. between the years 2002-2013, for example, demonstrated that there was only limited adoption of article 7 of the OECD Model (2010), whether by amending their existing bilateral tax treaties or when concluding new bilateral tax treaties. Similarly, in any new bilateral tax treaties signed between 2010 and 2014. A case in point is Germany. As a result, it might take decades for the AOA to be widely incorporated in bilateral tax treaties. This ineffectiveness in respect of implementing the AOA leaves open the interesting questions of why the OECD did not include it in the BEPS agenda in an effort to produce a consensus among states or include article 7 of the OECD Model (2010) in the multilateral instrument (MLI) in order to speed up the process of amending existing bilateral tax treaties.
Another major difficulty arises in respect of interpreting the revised Commentary on Article 7 of the OECD Model (2008). Specifically, although the OECD recommends a dynamic rather a static approach in interpreting follow-up modifications to the Commentaries, these changes should not be “different in substance”. They should instead provide further clarification on the pre-2010 version of article 7 of the OECD Model. Nonetheless, there are solid arguments for opposing that the amendments , in essence, do not merely clarify the pre-2008 Commentary but, in effect, go beyond what was included in it., Thus, for example, one of the significant differences is the rule of the attribution of an arm’s length “free” capital in calculating the amount of debt attributed to PEs with regard to their functions , assets used and risk assumed. In practice, however, later Commentaries could be followed. A case in point is Irish Bank Resolution (2017), in which, even though the bilateral tax treaty between Ireland and the United Kingdom was signed in 1976, the 2008 PE Report was considered relevant. Again, the issue is further compounded by the fact that the UN Model has mostly rejected the AOA amendments in the Commentary on Article 7 of the OECD Model (2008).
Another significant concern arises from the step-two requirement to apply the OECD Guidelines by analogy . Put simply, the AOA moves article 7 of the OECD Model (2010) closer to article 9, causing the OECD Guidelines to play a significant role in calculating the profits attributable to PEs. Yet several issues may arise. To begin with, the weaknesses of the arm’s length principle, which are many and have been consistently criticized, are incorporated into the AOA. In this context, PEs have neither capital nor from the head office, whereas article 9 of the OECD Model is about the relationship between two separate entities. Furthermore, transactions are difficult to identify, as there are no actual contracts between the head office and the PE. Thus, calculating the free capital by analogy to the already problematic way provided in the OECD Guidelines becomes even more challenging, thereby leading to inconsistent results, such as in insurance services. Such findings would first give rise to uncertainty in relation to the outcome and second permit tax planning opportunities to shift profits into tax-advantageous jurisdictions.,
In addition, the AOA appears to be difficult to apply in practice, particularly in developing countries, just as , the OECD Guidelines. Accordingly, the UN Model has rejected this approach, because the AOA requires states to follow both the 2010 PE Report, which is 240 pages long, and the OECD Guidelines, which are 612 pages long, by analogy. In essence, the AOA seems like the most problematic impasse within the whole OECD Model. Undoubtedly, the AOA is hard to deploy, placing huge administrative burdens on both tax authorities and taxpayers. Consequently, developing countries would, in general, have neither the resources nor the capacity required to follow such rules and would inevitably permit multinational enterprises (MNEs) to take advantage of this lack of knowledge and shift profits away from their jurisdictions.
Finally, an unresolved area is the dependent agent permanent establishment (DAPE). For instance, in the case of Rolls Royce Plc (2011), Rolls Royce UK was found to have a DAPE in India based on the marketing activities of its local subsidiary there. The Indian tax authorities claimed that, apart from the DAPE commission, there should have been additional profits attributed to the Indian PE arising from the fact that the DAPE contributed substantially to the profits of Rolls Royce UK in India. As a consequence, 35% of the global profits of Rolls Royce were taxed in India.
Following the OECD BEPS Project, in its efforts to eliminate tax planning techniques using PEs such as commissionaire arrangements, the OECD proposed a lower PE threshold in accordance with Action 7. Specifically, the OECD lowered the PE threshold in relation to a DAPE by amending article 5(5) and (6) of the OECD Model. In addition, it modified article 5(4) of the OECD Model in order to include a new anti-fragmentation rule. In this context, in order to provide further guidance on profit attribution to Permanent Establishments, in March 2018 the OECD published a final report (the 2018 PE Attribution Report), two public discussion drafts (in July 2016 and June 2017) creating adverse results, however. In particular, the OECD did not clarify the overlap between articles 7 and 9 of the OECD Model. That is, under step one of the AOA, someone needs to identify the significant people functions in order to allocate assets and risks, while the OECD Guidelines (2017) provide a different risk control framework. This issue becomes even more problematic under step two of the AOA, given that taxpayers need to apply the OECD Guidelines (2017) by analogy, creating manifold inherent difficulties (see also section 4.). In addition, some authors argue that this lower PE threshold will make the already problematic zero-sum game even more complicated, because there could be no additional profits to attribute to the host state. Presumably, one could argue that some countries, like the United Kingdom, are trying to solve this issue unilaterally by introducing diverted profits taxes. It is worth noting that, although the 2018 Profit Establishment Attribution Report indicates that the guidance provided does not extend to the AOA, the fundamental principle contained in the examples the AOA method, if it were a separate and independent enterprise.
The OECD’s original intent for the AOA was to provide a consistent interpretation and application of article 7 of the OECD Model, bringing together both OECD and non-OECD member countries to establish a consensus as regards a preferred approach for the sake of simplicity, administrability and sound tax policy. Hence, to assess whether the AOA should be maintained as the OECD standard, it is necessary to closely look into the consensus achieved, as well as into the fundamental principles that fortify a sound tax policy, as follows: (i) ease of administration, which includes both simplicity and transparency; (ii) economic efficiency, together with tax neutrality and to abuse; (iii) overall systematic stability; and (iv) fairness, including inter-nation equity.,,
First, the AOA seems to have mostly failed to provide a simple way to calculate profits attributable to PEs, causing further difficulties in administering the method. The AOA is presumably becoming the most problematic and challenging exercise as taxpayers need to apply both the 2010 PE Report and the OECD Guidelines by analogy. Thus, although transparency has been partly achieved, the lack of simplicity is a significant shortcoming.
Second, efficiency and neutrality have been partially achieved as well. While the AOA introduced a uniform approach and many areas have been further clarified, resulting in certainty for both taxpayers and tax administrators, the AOA appears to be a challenge. On the one hand, economic efficiency deteriorates, as double taxation could arise in the absence of a single method of calculating the risk-free capital. On the other hand, the robustness of abuse is further weakened, as, where there are strictly speaking no contracts between PEs and the enterprise as a whole, tax planning opportunities, such as risk-stripping structures using contractual arrangements in a captive insurance case, are still possible. These techniques, used mainly by MNEs to benefit from the lack of clarity in the OECD Guidelines, can further flourish given that the PE is part of the same enterprise, which makes it easier to be abused.
Third, overall systemic stability is a significant weakness in the AOA. Even though some OECD member countries have adopted the AOA, it has failed both to find consensus among either OECD member countries or non-OECD member countries. Moreover, the AOA created additional interpretative issues by having two different articles 7 of the OECD Model, including their corresponding Commentaries. Consequently, as the majority of countries follow the pre-2010 version of article 7 of the OECD Model, two issues may result. On the one hand, these countries would be in conflict with countries, such as Germany, that have adopted article 7 of the OECD Model (2010) in their domestic legislation, fostering uncertainty among both taxpayers and tax administrators in terms of double taxation or no taxation. On the other hand, similar concerns would arise based on the fact that the OECD has expressed its intention not to update the Commentary on Article 7 of the OECD Model (2008), and therefore interpretative difficulties will remain in the future. As a result, the OECD will be unable to solve this impasse.
Fourth and finally, considering the policy considerations noted in this section, fairness and inter-nation equity have been met with mixed results, in both developed and developing countries, primarily due to the monitoring problems related to the AOA, which has proved insufficient to deter MNEs from using PE structures to erode their tax bases and decrease their tax revenues.
As a consequence, although the AOA has improved on the pre-AOA methodology, it cannot survive in its current form and, therefore, needs be redeveloped to achieve both consensus and stability within a global context. The OECD can perhaps do at least three things to improve the calculation of profits attributable to PEs. First, it can keep the advantages of the AOA and move towards to a more functional and more restricted independence approach, in particular with respect to the global trading sector, while maintaining the AOA in respect of the rather more straightforward financial sector. In this way, the OECD could achieve more consensus with countries opposing the AOA. Alternatively, the OECD could move to absolute independence, where the PE operates as a subsidiary to be treated as a resident of the host state., Nonetheless, this move might create further complications with regard to bilateral tax treaties, as it could completely change both the distributive rules as regards the income articles of the OECD Model and article 4. This needs further study and analysis. By contrast, if the AOA were to prove to be unsustainable, another possible solution might be to reconsider the attribution of profits to PEs, going back to its origin with the Mitchel B. Carol report (see section 2.), re-examining the three suggested solutions and bringing them up to date. As Baker notes, “it is far better to do that than to continue ‘flogging the dead horse’ of the AOA”.
In conclusion, the AOA has replaced the problematic but assumed-to-be-workable pre-AOA approach with mixed results. The main purpose of the OECD Model is to provide a common solution that is “to provide a means of settling on a uniform basis the most common problems that arise in the field of international juridical double taxation”. While the AOA has provided a single uniform approach resulting in clarity and consistency that minimizes double taxation or , it has failed to foster consensus among OECD and non-OECD member countries and therefore does not seem to be the preferred OECD method regarding the attribution of profits to a PE. The OECD seems to have failed to provide a sustainable method, but instead of having multiple approaches, this uniform approach could be a starting point for the OECD to find a way to achieve global consensus among states, mainly in the global trading sector, and further overcome the disadvantages of the AOA.
 P. Baker & R. Collier, The Attribution of Profits to Permanent Establishments p. 28 (IFA Cahiers vol. 91B, 2006), Books IBFD.
 P. Baker, Double Taxation Conventions, sec. A7B.03 (3rd edn., Sweet & Maxwell 2003).
 R. Vann & J. Sasseville, Article 7: Business Profits – Global Tax Treaty Commentaries secs. 1.2.2-1.2.3, Global Topics IBFD (accessed 15/09/2019.
 OECD Model Tax Convention on Income and on Capital: Commentary on Article 7 para. 5 (21 Nov. 2017), Treaties & Models IBFD.
 OECD, Report on the Attribution of Profits to Permanent Establishments (OECD 2010), Primary Sources IBFD [hereinafter 2010 PE Report].
 OECD Model Tax Convention on Income and on Capital (17 July 2008), Treaties & Models IBFD.
 OECD Model Tax Convention on Income and on Capital (21 Nov. 2017), Treaties & Models IBFD.
 S.B. Law, Profit Attribution to Permanent Establishments – A Tax Treaty Perspective on the “Single Taxpayer” Approach, 72 Bull. Intl. Taxn. 3 (2018), Journal Articles & Papers IBFD.
 Baker & Collier, supra n. 1, at p. 30.
 R. Collier & J. Andrus, Transfer Pricing and the Arm’s Length Principle After BEPS p. 87 (Oxford University Press 2017).
 Vann & Sasseville, supra n. 3, at sec. 18.104.22.168.
 Klaus Vogel on Double Tax Conventions, p. 504 (4th edn, E. Reimer & A. Rust eds., Wolters Kluwer 2015).
 Baker & Collier, supra n. 1, at p. 28.
 Taxation of Business Profits in the 21st Century: Selected Issues under Tax Treaties – Essays in Commemoration of IBFD’s 75th Anniversary Written by Tax Research Staff of IBFD p. 173 (C. Gutiérrez Puente & A. Perdelwitz eds., IBFD 2013), Books IBFD.
 Vann & Sasseville, supra n. 3, at sec. 5.1.4.
 Reimer & Rust eds., supra n. 12, at pp. 514 and 520.
 Baker, supra n. 2, at sec. A7B.26.
 Reimer & Rust eds., supra n. 12, at pp. 519, 526 and 527.
 Baker, supra n. 2, at sec. A7B.14.
 Baker & Collier, supra n. 1, at p. 34.
 Vann & Sasseville, supra n. 3, at sec. 4.1.
 Reimer & Rust eds., supra n. 12, at pp. 528-529.
 Baker, supra n. 2, at sec. A7B.20.
 Vann & Sasseville, supra n. 3, at sec. 22.214.171.124.2.
 Baker, supra n. 2, at sec. A7B.33.
 Para. 5 OECD Model: Commentary on Article 7.
 Vann & Sasseville, supra n. 3, at sec. 4.2.2.
 Paras. 95-100 OECD Model: Commentary on Article 7.
 R. Collier & J. Vella, Five Core Problems in the Attribution of Profits to Permanent Establishments, 11 World Tax J. 2, sec. 2.1.3. (2019), Journal Articles & Papers IBFD.
 Gutiérrez Puente & Perdelwitz eds., supra n. 14, at p. 173.
 A.J. Auerbach et al., Destination-Based Cash-Flow Taxation, Oxford University Centre for Business Taxation Working Paper 17/01 (2017).
 Brazil, Russia, India, China and South Africa.
 Gutiérrez Puente & Perdelwitz eds., supra n. 14, at pp. 315-318.
 Vann & Sasseville, supra n. 3, at sec. 126.96.36.199.
 A. Waltrich, Cross-Border Taxation of Permanent Establishments: An International Comparison p. 62 (Wolters Kluwer 2016).
 Para. 35 OECD Model: Commentary on Article 7.
 Gutiérrez Puente & Perdelwitz eds., supra n. 14, at p. 266.
 Reimer & Rust eds., supra n. 12, at p. 521.
 Gutiérrez Puente & Perdelwitz eds., supra n. 14, at p. 268.
 P. Baker & R.S. Collier, 2008 OECD Model: Changes to the Commentary on Article 7 and the Attribution of Profits to Permanent Establishments, 63 Bull. Intl. Taxn. 5, pp. 199 and 202 (2009), Journal Articles & Papers IBFD.
 UK: UKFT, 22 Sept. 2017, Irish Bank Resolution; Irish Nationwide Building Society v. Commissioners for Her Majesty’s Revenue and Customs,  UKFTT 0702 (TC), Case Law IBFD; see J.S. Schwarz, Permanent Establishment Profit Attribution: United Kingdom Courts Examine the “Authorised OECD Approach”, 73 Bull. Intl. Taxn. 6/7, p. 362 (2019), Journal Articles & Papers IBFD.
 Vann & Sasseville, supra n. 3, at sec. 188.8.131.52.
 Collier & Andrus, supra n. 10, at pp. 90, 124.
 Reimer & Rust eds., supra n. 12, at p. 506.
 Baker & Collier, supra n. 1, at p. 26.
 Collier & Andrus, supra n. 10, at p. 119.
 Vann & Sasseville, supra n. 3, at sec. 184.108.40.206.
 Vann & Sasseville, supra n. 3, at sec. 220.127.116.11.
 Reimer & Rust eds., supra n. 12, at p. 507.
 Vann & Sasseville, supra n. 3, at sec. 18.104.22.168.2.
 IN: HC Delhi, 30 Aug. 2011, ITA No. 493/ 2008, Rolls Royce Plc v. DIT, Case Law IBFD.
 OECD, Additional Guidance on the Attribution of Profits to a Permanent Establishment under BEPS Action 7 (OECD 2018), Primary Sources IBFD.
 B. Stack & J. Hawes, INSIGHT: OECD Guidance Doesn’t Resolve Fundamental Issue for Attribution of Profits to Permanent Establishments, Transfer Pricing Report (10 Aug. 2018).
 Collier & Vella, supra n. 29, at p. 178.
 Collier & Andrus, supra n. 10, at pp. 222 and 252.
 Law, supra n. 8.
 Baker & Collier, supra n. 1, at p. 54.
 Collier & Andrus, supra n. 10, at p. 3.
 Waltrich, supra n. 35, at p. 181.
 Tax by Design: The Mirrlees Review p. 22 (A. Mirrlees ed., Oxford University Press 2011).
 Collier & Andrus, supra n. 10, at pp. 120 and 280.
 Baker & Collier, supra n. 1, at pp. 62, 64 and 65.
 T. Kulcsár, How Fiction Becomes Reality: Equal Treatment for Permanent Establishments and Resident Enterprises, in Gutiérrez Puente & Perdelwitz eds., supra n. 14.
 P. Baker, The League of Nations’ Draft Convention for the Allocation of Business Income between States – A New Starting Point for the Attribution of Profits to Permanent Establishments pp. 514-520 (BTR 2018).
 Paras. 2 and 3 OECD Model: Commentary – Introduction (2017).