Whoever thinks that a resident company’s income is protected from taxation in countries where it has no PE, is in for a rude awakening; it is not. The allowance for CFC rules in the OECD Model Convention (hereafter “MC”) commentaries is so all encompassing that it covers far more than just the taxation of a subsidiary’s income.

According to paragraph 23 to the commentary on article 1 of the OECD MC, “The use of base companies may also be addressed through controlled foreign companies provisions. A significant number of member and non-member countries have now adopted such legislation. … a common feature of these rules, … is that they result in a Contracting State taxing its residents on income attributable to their participation in certain foreign entities. It has sometimes been argued, based on a certain interpretation of provisions of the Convention such as paragraph 1 of Article 7 … that this common feature of controlled foreign companies legislation conflicted with these provisions. For the reasons explained in paragraphs 14 of the Commentary on Article 7 … that interpretation does not accord with the text of the provisions. It also does not hold when these provisions are read in their context. … It is recognised that controlled foreign companies legislation structured in this way is not contrary to the provisions of the Convention.

Article 7’1 reads “Profits of an enterprise of a Contracting State shall be taxable only in that State unless the enterprise carries on business in the other Contracting State through a permanent establishment situated therein. …”  This seems clear.

A plain reading interpretation is also confirmed by paragraph 1 of the commentary to article 7 which reads “… It incorporates the basic principle that unless an enterprise of a Contracting State has a permanent establishment situated in the other State, the business profits of that enterprise may not be taxed by that other State unless these profits fall into special categories of income for which other Articles of the Convention give taxing rights to that other State.”  I will refer to this as the “guardian function” of article 7 going forward.

Nonetheless, paragraph 14 of the commentary on article 7 reads “… The paragraph does not limit the right of a Contracting State to tax its own residents under controlled foreign companies provisions found in its domestic law even though such tax imposed on these residents may be computed by reference to the part of the profits of an enterprise that is resident of the other Contracting State that is attributable to these residents’ participation in that enterprise. Tax so levied by a State on its own residents does not reduce the profits of the enterprise of the other State and may not, therefore, be said to have been levied on such profits … .”

This seems in direct contradiction to the commentary under paragraph 1 here before, just as it seems in direct contradiction of the very essence of CFC legislation.  Surely the whole point of CFC legislation is to tax the income of the foreign subsidiary, even before it gets distributed. It is all about the income, not the dividend.  That is why countries with a dividend exemption system switches over to a credit system (also for underlying taxes at the foreign subsidiary’s level):  it is to TAX that income, not to exempt it. I do not know a single country that would apply a CFC regime to a foreign subsidiary’s income to prevent deferral, and then exempt that CFC income as a tax exempt dividend under a participation exemption.

The story goes further. Both paragraphs 23 to article 1 and 14 to 7, also make reference to paragraph 37 to article 10’5.  Article 10’5 prohibits so-called secondary withholding tax on dividends. So it should come as no surprise that the commentary in paragraph 37 states that that paragraph does not forbid CFC rules. After all, these taxes go in opposite directions: in secondary withholding tax rules, the subsidiary country tries to tax the parent’s parent on dividends (it is in effect an anti-treaty shopping measure protecting source taxation by combating intermediate head offices or holding companies); in CFC rules, the parent country tries to tax the subsidiary’s profits (i.e. it has nothing directly to do with either protecting source at the subsidiary country level, or with treaty protection). So article 10’5 is irrelevant.

It would have been far more appropriate for the commentary to address CFC rules under article 10’1, if one were to address CFC rules under article 10 at all. Article 10’1 reads, “Dividends paid by a company which is a resident of a Contracting State to a resident of the other Contracting State may be taxed in that other State”.

To show just how far the current commentaries on CFC’s are removed from the logic of the rest of the MC commentary, consider the following texts from the commentary on article 10:

  • Paragraph 2: “The profits of a business carried on by a partnership are the partners’ profits …” vs
  • Paragraph 3: “The position is different for the shareholder; … the company’s profits are not his; so they cannot be attributed to him.”

To be fair, paragraph 3 does also say “He is personally taxable only on those profits which are distributed by the company (apart from the provisions in certain countries’ laws relating to the taxation of undistributed profits in special cases).”  But why specifically mention CFC legislation in all other places, except here, where it makes most sense?  Besides, “certain countries” sounds like an exception, much more than it sounds like the general rule, BEPS action 3 wants CFC’s to be.

Me thinks the commentary to 10’1 would be much more useful, if paragraph 3 hereabove (or a new 3.a) mentions that the shareholder country does not have to wait until a dividend is actually paid, but can tax the shareholder as if a dividend has been paid as an anti-abuse measure either in case of deferral, or BEPS.   This would be far clearer, far more to the point, far more relevant to the treaty paragraph commented, and thus technically far more robust, than the sticky tape approach of the current paragraph 37. It also would facilitate a far more targeted exception to the guardian rule of article 7’1.  This is the way exceptions should be (narrow), rather than the current wide wording which even makes article 9 superfluous (see hereafter).

There is a bigger problem: the current line of argument in paragraph 14 to the commentary on article 7, completely undermines the guardian function of 7’1 and paragraph 1 of its commentary.   After all, if taxing ACo in A, based on the profits of BCo in B, is not prohibited by article 7’1, then nothing stops reading paragraph 14 of the commentary on article 7’1 such that BCo could be ACo’s parent (e.g. as the CFC version of a secondary withholding tax where the actual secondary distribution is preempted), or BCo could be an ACo sister (e.g. in a TP dispute and who then still needs article 9)?

So all the above leads to the question why the commentary does not simply say it as it is.  Yes, CFC rules infringe on article 7, paragraph 1, or at least on the spirit of article 7, paragraph 1, but that infringement is justified because CFC rules are anti avoidance rules and article 7, paragraph 1, is not meant to protect passive income from taxation indefinitely.   Maybe the political climate was such in 2010 that it was not possible to say that national anti-abuse rules cannot override a treaty, but in a BEPS world, that is no longer the case.

Fixing the commentaries in paragraph 23 to article 1, paragraph 14 to article 7’1, and paragraphs 37-39 to article 10, will restore the guardian function of article 7’1 and the internal harmony with paragraph 1 of the commentary to article 7’1.  It is time for a bit of cleanup, and the OECD paper on Action 3 of the BEPS action plan provides an excellent opportunity thereto.



  1. Not sure one is supposed to comment your own writing, but here we go. Obviously, a crucial point left out of the above discussion, is the role of the OECD commentaries in the interpretation of bilateral treaties. The answers vary by court, and by professor.

    However, it should be clear that the more the commentaries stick to the Vienna Convention on the law of treaties, the bigger its role can be. This does however mean that the commentaries should stick to:
    1) the “ordinary meaning” of the words used in the treaty itself, AND
    2) a “good faith” interpretation of those words.

    I would dare to argue that simply saying that the treaty may be set aside in cases of abuse, is a closer to these Vienna yardsticks and the truth. It outperforms the “clever” argument of paragraph 14 of the commentary to article 7, which blows a hole through the treaty and its basic principles. We do not need clever: we need clarity; we do not need squirming: we need straight shooting.

  2. Johann, you make an important point. The OECD Commentary is only an aid to interpretation of treaties if the text in the Commentary is interpretative. In the case of CFCs and treaties, we all know that the French Conseil d’Etat ruled in Schneider Electric that the French CFC regime infringed Art 7 of the France-Switzerland Treaty. On the other hand, in Bricom, the English Court of Appeal found the UK CFC regime unaffected by the Netherlands-UK Treaty on the basis of the particular way in which the UK regime is drafted. Each system and treaty needs careful analysis.

    Nothing in BEPS changes this although the various actions make it clear that the OECD now recognises that the only way to change the meaning of a treaty is to amend its terms and not to just say what they would like it to mean in the Commentary.

    1. Actually Jonathan (long time overdue, I know), I specifically ask the OECD during the BEPS consultations on Action 3, to also address this issue and just say that this is a treaty anti-abuse rules. However, the chair lady had no appetite for that. Alas.

  3. If I may add few points.

    Actually, the majority of court verdicts around the world are in favour of saying the CFC legislation does not run against double tax treaties, and more precisely that this type of legislation is outside the scope of double tax treaties (the UK, Finland, Denmark, Japan, Mexico, Italy).

    In two cases (France and Brazil) court verdicts eventually said that CFC rules were against DTTs, which however was supposedly remedied by the national legislators by changing the approach used by the CFC rules (e.g. by switching to dividende repute in France). Thirdly, one court (Sweden) said that there could be a conflict but allowed for an override.

    Lastly, the BEPS project does bring some minor amendment. The BEPS Report no. 6 advises to add saving clause – akin to the US saving clause – which it thought to safeguard the rights to tax own residents the way the home country sees necessary.

  4. Thank you for the update Filip. Still think the political atmosphere around BEPS was such that the OECD could have opened up and just call CFC rules what they are: anti-abuse rules which can override a treaty (I agree with Sweden on that one). The Commentaries are often complicated and vague enough not to be further burdened with illogical reasoning :-).

    1. Johann,

      It’s very interesting what you’re saying about these discussions you participated in at the OECD.

      I guess that the US delegation might not agree for classifying their Subpart F rules purely as anti-abuse (or anti-avoidance) tool, taking into account their previous reports on Subpart F and their formal insistence on CEN rule, but surely the solution of treating the CFC rules as permissible treaty override would probably please both sides of the never-ending conflict of whether this type of legislation is compliant with double tax treaties. To me this would require some standardization of CFC rules by OECD though.

Leave a Reply

Your email address will not be published. Required fields are marked *