Playing the Principal Role in Creating a Permanent Establishment

Bloomberg BNA’s Premier International Tax Library is a comprehensive global tax resource. Trust Bloomberg BNA's Premier International Tax Library for the guidance you need on...

By Gary D. Sprague, Esq.

Baker & McKenzie LLP, Palo Alto, CA

Now that the final report on BEPS Action 7 has been released, "Preventing the Artificial Avoidance of Permanent Establishment Status" (Final Report), taxpayers are able to consider the final language to be incorporated into Article 5(5) of the OECD Model Tax Convention.1 This language also will be incorporated into the many bilateral income tax treaties around the world by those countries which choose to do so, through their accession to the multilateral instrument or otherwise.2 This commentary notes how the agreed text has evolved from prior proposals, and offers some thoughts on whether the final text offers more or less clarity to taxpayers and tax administrations than what many thought would be the final language as published as recently as last May.3

While Action 7 addresses various other elements of Article 5 besides the dependent person deemed PE rule of Article 5(5), a main focus of Action 7 from the start has been on modifying Article 5(5) to address "commissionnaire arrangements and similar strategies."4 The first Action 7 Discussion Draft was published on October 31, 2014.5 It proposed four options to supplement Article 5(5), using as building blocks two expressions that described the interaction between the dependent person and the customer and two that described the consequences of the work of the dependent person.  For the first element, describing the activity of the dependent person, the alternatives were as follows: (1) "habitually engages with specific persons in a way that results in the conclusion of contracts;" or (2) "habitually … negotiates the material elements of contracts."

The alternatives proposed for the second element, describing the commercial result of the dependent person's activity, were as follows: (1) "contracts … for the transfer of the ownership of, or for the granting of the right to use, property owned by that enterprise or that the enterprise has the right to use, or … for the provision of services by that enterprise;" or (2) "contracts which, by virtue of the legal relationship between that person and the enterprise, are on the account and risk of the enterprise."

These proposals generated a significant number of business comments, and were the subject of a Public Consultation on January 21, 2015.6 Given the available choices, the large majority of public comments favored the combination of "habitually … negotiates the material elements of contracts … for the transfer of the ownership of, or for the granting of the right to use, property owned by that enterprise or that the enterprise has the right to use, or … for the provision of services by that enterprise" on the basis that this expression was the most precise and thus gave the best guidance to taxpayers and tax administrations. Indeed, that proposal was adopted as the single proposal in the Action 7 Revised Discussion Draft ("RDD") issued on May 15, 2015.7

It was something of a surprise, therefore, that the Final Report did not adopt the "negotiates the material elements" standard, but instead substituted entirely new language.  The Final Report states that a PE will exist under Article 5(5) where a person "habitually plays the principal role leading to the conclusion of contracts that are routinely concluded without material modification by the enterprise" if the contract is: (1) in the name of the nonresident enterprise; (2) for the transfer of the ownership of, or for the granting of the right to use, property owned by that enterprise or that the enterprise has the right to use; or (3) for the provision of services by that enterprise.

So is the new language an improvement or not? Putting aside the tax policy issues of whether this sort of activity should create a local tax nexus at all, the principal concern for taxpayers is whether the operative rule and its clarifying Commentary provide sufficient certainty for taxpayers to understand what activities will and will not create a PE. The main reason that business preferred the "negotiates the material elements" standard as proposed in the May 15 RDD is that the commercial activity of "negotiation" can be defined. It is not solicitation or demand generation. It is not after sales support. It means the give and take over contractual terms. Frankly, it seemed like a standard that business could understand and thus live with.

At first blush, the "plays the principal role" standard seems less precise than the "negotiates the material elements" standard.8 With an eye towards interpretative issues, it is useful to compare both expressions to the "contract conclusion" test, which is the core concept of current Article 5(5). The "contract conclusion" test has as a principal virtue its relative precision. The formation of a contract is a precise activity, requiring offer and acceptance.  Lawyers normally would be able to pinpoint the communications which constitute the offer and acceptance which cause a contract to be formed.9 Similarly, "negotiation" describes a much more discrete activity than "playing the principal role."

On the other hand, a standard which refers to "the principal role" perhaps removes the point of uncertainty raised by some commentators that the RDD formulation could result in multiple PEs for single contracts, if several persons in an organization located in different countries were involved in negotiating material terms.  It would seem that the final proposal requires the taxpayer and the tax administration to point to a single person who plays "the principal role."

The second conceptual element of the new standard — i.e., that the person plays the principal role "leading to the conclusion of contracts that are routinely concluded without material modification by the enterprise," — also at first blush seems to narrow the scope of the rule compared to the RDD, since this requirement appears in the revised Article 5(5) as a clear prerequisite for the rule to apply. Applying that test to different business contexts, however, can give rise to curious results.

One business context the new rule apparently is meant to describe is when the local "sales force" is given the autonomy to negotiate the terms and conditions of the contract.  It could be expected that an enterprise that authorizes a person to engage in negotiation over material terms normally would not renegotiate those terms after the negotiator had finished his or her work. These cases, therefore, would tend to result as a practical matter in "routine conclusion" by the nonresident enterprise. Indeed, the more efficient the nonresident becomes in setting guidelines for its negotiator, the more likely a PE will arise, as there will be less need for the nonresident to countermand agreements reached by its representative.

On the other hand, it is also clear that the new standard is meant to cover certain cases involving stardardized contracts, including on-line contracting. In the standard contract context, normally all terms and conditions have been set in advance by the nonresident, and the "sales force" has no negotiating authority at all. As demonstrated by an example discussed below, these cases also may result in the routine conclusion of contracts without material modification by the nonresident.

The RDD noted that one commentator pointed out that the "negotiates the material elements of contracts" test of the RDD would not cover the conclusion of standardized contracts, as there is no negotiation over such contracts. Given the focus in various elements of the BEPS process on business models involving remote selling through standard contracts, perhaps the delegates were responding to the desire to create PEs for such suppliers when they decided to deviate from the RDD proposal and add the "routinely concluded without material modification" standard.

As a result, this standard seems to apply both to some cases where a dependent person has full powers to negotiate, and to some cases where the person has none. Somewhere between those cases lies the safe zone where local personnel may engage in significant solicitation and even negotiation in country, but the nonresident intercedes in the process to engage in sufficiently substantive negotiations so that its acceptance cannot be seen as routine. That doesn't describe a particularly efficient business model. It may encourage groups to relocate some negotiating authority into regional centers to ensure that the nonresident can be seen as actively participating in negotiations.  It is remarkable that the explanatory material included in the Final Report gives no explanation for this significant change of language from the RDD.

The Final Report includes a few examples of the application of the new standard, which examples will be incorporated into the Article 5 Commentary. It seems clear that these examples will become the focus of intense interest by taxpayers and tax administrations, as they will constitute the principal interpretative guidance for the Final Report standard. Suffice it to say, these examples leave a lot of questions unanswered.

There are two principal examples, to be included in ¶32.5 and ¶32.6 of the Article 5 Commentary. These examples provide some additional texture to the legal principle that is meant to be described in the "plays the principal role" test.  The new Commentary states that this test describes cases where the conclusion of contracts is "clearly the direct result" of the in-country activities, suggesting that there must be some proximate cause between the activity and the contract. The Commentary also describes the in-country activity as conducted by persons who "act as the sales force of the enterprise," which suggests that sales solicitation activity is the most directly relevant activity. Further, the targeted activity is that which "convinced the third party to enter into a contract with the enterprise," suggesting that some inquiry into how the customer made its purchasing decision is a legally relevant analysis. Those sales-oriented activities are contrasted with the work of a person who "merely promotes and markets goods or services of an enterprise in a way that does not directly result in the conclusion of contracts."10

One example describes a company (RCO) which "distributes various products and services worldwide through its websites." The example describes several facts relating to activities performed by an affiliate located in the market jurisdiction (SCO), all of which presumably would be legally relevant to some degree, although there is no indication of the relative significance of the various facts.  The example describes the activity of SCO employees as follows:SCO's employees send emails, make telephone calls to, or visit large organisations in order to convince them to buy RCO's products and services and are therefore responsible for large accounts in State S; SCO's employees, whose remuneration is partially based on the revenues derived by RCO from the holders of these accounts, use their relationship building skills to try to anticipate the needs of these account holders and to convince them to acquire the products and services offered by RCO. When one of these account holders is persuaded by an employee of SCO to purchase a given quantity of goods or services, the employee indicates the price that will be payable for that quantity, indicates that a contract must be concluded online with RCO before the goods or services can be provided by RCO and explains the standard terms of RCO's contracts, including the fixed price structure used by RCO, which the employee is not authorised to modify. The account holder subsequently concludes that contract online for the quantity discussed with SCO's employee and in accordance with the price structure presented by that employee.11

Several elements of this are worth noting.  The legally relevant point seems to be that the actions of SCO employees "convince" the customer to purchase. The facts that the employees are compensated by commission, that they have "relationship building skills," and that they are responsible for "large accounts" would seem to be circumstances that the drafters of the example believe would be attributes of persons who perform the role of "convincing" customers to purchase.

Given that this is only one example, there are, of course, many cases that will not be clearly answered by this example.  For example, it is common for multinational enterprises to deal with each other through "global deals," i.e. sales agreements negotiated and agreed between global or regional headquarters of both parties, which provide for affiliates of the purchasing group to place purchase orders on the local affiliate of the supplier group. Sales personnel employed in local affiliates of the supplier group presumably would have all the attributes mentioned in the Commentary example of relationship building skills and the like, but in this case they may not be the person in the global or regional headquarters who ultimately convinced the customer to purchase. In cases where the sales revenue is booked in each of the supplier group affiliates, it certainly is to be hoped that the global or regional headquarters jurisdiction will not assert a "principal role" PE of all of the supplier group affiliates.

It is also worth noting that this example states that no discussion over terms took place until the customer already was "persuaded" to purchase. This would seem to be an unusual circumstance, perhaps existing principally where contracting is done on-line, but would not typically be the case in other business negotiations. This unusual chronology would seem to be a distraction from the main technical point being illustrated, i.e., however the contractual terms are proposed or worked out, the nonresident enterprise must routinely conclude the contract without material modification.  The point of this element of the example is probably to emphasize the message that on-line contracting results in contracts routinely concluded without material modification, so that the principal factual analysis in cases involving on-line contracting will be over who plays the principal role to convince customers to purchase.

The second example is meant to show the contrasting case of a person who "merely promotes and markets goods" in a way that does not directly result in the conclusion of contracts.  Again, the example is worth quoting in its entirety:Where, for example, representatives of a pharmaceutical enterprise actively promote drugs produced by that enterprise by contacting doctors that subsequently prescribe these drugs, that marketing activity does not directly result in the conclusion of contracts between the doctors and the enterprise so that the paragraph does not apply even though the sales of these drugs may significantly increase as a result of that marketing activity.12

The interpretative challenge here is to identify the legally relevant facts that distinguish the two cases. One suspects that the pharmaceutical manufacturer's representative described in this example would have many of the same attributes as colleagues in the on-line sector; they probably manage large accounts, enjoy commission-based compensation, and exercise "relationship building skills" to encourage the doctors practicing in their sales territory to loyally prescribe their employer's drugs. Indeed, there doesn't seem to be any other person in this business model who does more than the company's representative to "convince" the counterparty to purchase. The pharma example, however, doesn't describe whether the representative does or doesn't have those attributes, and doesn't point to any other person who "convinces" the purchaser to buy, creating ambiguity as to which of the facts in the online sales case are the most significant. The facts emphasized in the online example that the sales representative having no authority to negotiate over terms of sale would also seem to be present in the pharma case, as there the representative also has no authority to negotiate the terms of sale.

Unless further work is done on this issue during 2016, these two examples will be the only comprehensive examples included in the revised Article 5 Commentary. Taxpayers will face the challenge of interpreting these examples to determine which of the various facts and relationships described in the examples are the most relevant or determinative. The following themes, however, seem to be the ones the examples are meant to emphasize:

  •  Nature of activity: It is the activity of convincing a customer to purchase that indicates when a person plays the principal role leading to the conclusion of contracts, as opposed to promoting goods or services, even when the promotion results in an increase in sales.
  •  Identity of counterparty: The commercial capacity of the person with whom the local representatives are interacting is relevant, in that activity not directed towards those persons prepared to make a purchase may not "directly result" in the conclusion of a contract.  Promotional activity directed at someone other than the ultimate purchaser can reduce the PE risk, as a purchase decision will not be made by the person being promoted. The pharma example is unusual, however, because promotional activities generally are directed towards those who will buy the product or service, and as a practical matter the patient exercises far less discretion over what drugs to purchase than does the doctor (or the insurance company or government health service).
  •  Temporal: The promotion in the pharma case does not result in a contract immediately; rather, it requires that subsequent steps be taken by the doctor and patient. Online sales, on the other hand, can occur immediately, without such additional steps.

There is plenty of room for disputes to arise here. Groups that are the most directly affected by these changes presumably will either convert to low-risk resellers (as foreshadowed by the Final Report and the Commentary itself) or withdraw the relevant sales personnel from the jurisdiction.13 Others more on the edge will need to carefully consider these examples as the treaty revisions wend their way through the adoption process of the multilateral instrument in order to decide whether to restructure their sales organization.

This commentary also appears in the January 2016 issue of the  Tax Management International Journal. For more information, in the Tax Management Portfolios, see Katz, Plambeck, and Ring, 908 T.M., U.S. Income Taxation of Foreign Corporations, Nauheim and Scott, 938 T.M., U.S. Income Tax Treaties – Income Not Attributable to a Permanent Establishment,and in Tax Practice Series, see ¶7130, Foreign Persons -- Effectively-Connected Income, ¶7160, U.S. Income Tax Treaties.


  1 Http://

  2 For example, Germany and Australia signed a new treaty on November 12, 2015, which incorporates this new deemed PE test.

  3 The Explanatory Statement published with the final reports on October 5 hints that some further clarification may be forthcoming, although the scope of the further guidance may be limited to particular business models. The Explanatory Statement provides at paragraph 17: "Follow-up work will also be needed in 2016 to incorporate the changes resulting from the report on Action 7 into the Model Tax Convention through an update of the Model. This follow-up work will allow the Committee, where necessary, to provide additional clarification on the new treaty wording introduced by the report and to address any unintended consequences of the changes resulting from that report, notably by examining an issue related to the global trading of financial products" (

  4 Final Report, p. 15, para. 8.

  5 Http://

  6 Http://

  7  Public comments also were received on the RDD (

  8 Both are more precise than the "habitually engages with specific persons in a way that results in the conclusion of contracts" proposal from the October 31, 2014 Discussion Draft, so it is fortunate that the delegates did not revert to that expression.

  9 For example, see the interesting discussion of offer and acceptance in the early days of e-commerce included in several country reports in Main Subject I: Taxation of Income Derived from Electronic Commerce, Cahiers de Droit Fiscal International, Volume 86, IFA 2001 San Francisco Congress.

  10 New Art. 5 Commentary, ¶32.5.

  11 New Art. 5 Commentary, ¶32.6.

  12 New Art. 5 Commentary, ¶32.5.

  13 See new Art. 5 Commentary, ¶32.12, noting that a "low-risk distributor" is not intended to be covered by the revised standard for an Art. 5(5) deemed PE.