31 January 2019
Commission's Nike investigation - why it can be important for other Dutch tax rulings
![Frank Pötgens]( https://dwbxnuhxoazve.cloudfront.net/Portraits-colleagues/_576x576_crop_center-center_60_line/frank.potgens_2024-07-15-141738_jrbx.jpg 576w)
![Wiebe Dijkstra]( https://dwbxnuhxoazve.cloudfront.net/Portraits-colleagues/_576x576_crop_center-center_60_line/wiebe.dijkstra_2021-07-16-142454_aigc.jpg 576w)
![Arjan Kleinhout]( https://dwbxnuhxoazve.cloudfront.net/Portraits-colleagues/_576x576_crop_center-center_60_line/arjan.kleinhout.jpg 576w)
The European Commission recently announced that it had opened an in-depth investigation into possible illegal state aid granted by the Netherlands to Nike's Dutch subsidiaries. This adds another Dutch case to the list of pending state aid cases connected with tax rulings (such as the Starbucks case and the Commission's IKEA investigation). The Nike investigation focuses on five tax rulings issued by the Dutch tax authorities between 2006 and 2015. It is too early to tell where the case will go, but we expect it may impact certain existing and past Dutch rulings and ruling practices. Companies that have benefitted from these practices should pay attention to the outcome of the Nike case. Click on read more to read our summary of the key points to bear in mind.
What is the Nike case about?
Like the Starbucks and IKEA cases, the Nike investigation was preceded by considerable media attention. Where Starbucks caught the attention of the public through hearings organised by the Public Accounts Committee in the UK parliament, and IKEA was the subject of a report by the Green Party in the European Parliament, Nike's Dutch tax planning was mentioned in the "Paradise Papers" in November 2017. At that point, the "CV/BV" structure that Nike appeared to have used, had already caught the media's attention and been criticised by academics.
The Commission has only provided limited information in its press release. The public version of the opening decision is likely to be published in the coming months (in recent cases, it took between three to six months). The opening decision will give a lot more detail on the structure and the direction of the Commission's current thinking.
According to the press release, the investigation focuses on the Dutch tax treatment, as confirmed in rulings, of royalty payments made by two Dutch group companies for the use of intellectual property rights owned by two other group entities. Those other entities are not taxed in the Netherlands. The Commission is concerned that these royalty payments may not reflect economic reality and exceed arm's length compensation for the use of the IP rights. Circumstances highlighted by the Commission include the fact that the two Dutch companies combined have more than a thousand employees, are involved in the development, management and exploitation of the IP rights, and bear the associated costs for marketing and sales activities; in contrast the licensors have no employees and are not involved in any economic activity.
Certainty on whether the tax treatment agreed in the relevant rulings constitutes illegal state aid is likely to take considerable time. In other similar cases, the Commission issued its final decision some two years after the opening decision. In addition, if the Dutch tax authorities or Nike consider the outcome to be unacceptable, the Commission's decision may be followed by lengthy litigation before the European courts, like in the Apple, Starbucks and FIAT cases.
Why is this case relevant?
The Nike case reconfirms the Commission's persistence in challenging the past tax treatment of multinationals, based on tax deals with EU member states perceived as overly generous and selective. This is relevant for a number of reasons. Since the start of the Starbucks case (which, in essence, is also about a CV/BV structure) there have been multiple developments affecting relevant tax planning practices. These developments include some of the recent steps taken or announced by the Netherlands against facilitating some of these practices (see our recent article here); the anti-hybrid provision to be implemented by EU member states pursuant to the second EU Anti-Tax Avoidance Directive; the 2018 US Tax Reform; and, arguably, the "deterrent" effect created by the pending cases. While the Commission has explicitly mentioned that it welcomes the recent steps taken by the Netherlands, it clearly sees no reason to stop investigating tax rulings as applied as far back as ten years ago.
This emphasises that taxpayers which have benefitted from these rulings, should continue to be prepared for a Commission state aid review. Proper preparation may include lining up a defence team; preparing defence files; establishing and validating relevant legal facts, including important IP law considerations (as the legal facts form the starting point for Dutch tax purposes); preparing communication strategies and protocols; forming views on disclosure questions; considering potential implications for corporate transactions; and, where necessary, considering restructuring action.
What is a CV/BV structure?
In a CV/BV structure, the taxable revenues of a US group's Dutch subsidiary – for example, revenues from European sales activities – are limited to an operating margin through tax-deductible IP royalty payments to a Dutch limited partnership (CV). However, the corresponding royalty income of the CV is not taxed by the Netherlands nor immediately taxed by the US (at least, prior to the 2018 US tax reform). This is the result of the hybrid classification of the CV as transparent from a Dutch perspective and opaque from a US perspective (based on a check-the-box election). This structure has been widely used (and is sometimes still used) by US multinationals for non-US operations.
Do CV/BV structures have specific vulnerabilities compared to other IP structures?
Some CV/BV structures may have specific vulnerabilities, including the potential issues flagged in the academic publication referred to above. However, the Commission has focused only on transfer pricing matters that, so far, are not unique to CV/BV structures including in the Starbucks case, which can also be regarded as involving a CV/BV structure.
Foreign partners of the CV are, in principle, subject to (i) Dutch corporate income tax as a non-resident taxpayer if the royalties received by the CV qualify as a ''share in the profits of a Dutch enterprise'' and (ii) Dutch dividend withholding tax on dividends paid by the BV to the CV if the CV holds the shares in the BV. Generally, no exemption under tax treaties between the CV partners' country of residence and the Netherlands is available if that country treats the CV as opaque for tax purposes (hybrid classification). However, the Netherlands has unilaterally adopted a policy to extend treaty protection to US taxpayers. It has been suggested that because this unilateral policy only applies to US taxpayers, it could qualify as state aid.
The above is only relevant in structures where the BV has paid dividends to the CV or where royalties paid by the BV qualify as a "share in the profits of a Dutch enterprise". As to the latter, several lines of defence may be available depending on the facts. For instance, variable royalties which are based on revenue rather than on profit, generally do not qualify as a share in the profits of a Dutch enterprise. Also, if the royalties are actually linked to profits from IP-related activities (for example, sales, marketing, advertising) but the relevant BV also has significant other assets or operations (for example, shares in foreign subsidiaries or distribution activities), it may be more difficult to argue that the royalties still qualify as a share in the profits.
Besides the above issues, we generally expect key aspects of the Nike case to be centred around transfer pricing matters which are not unique to CV/BV structures. See below.
What else can be expected from the Nike investigation?
Although much is still unclear, discussions about fundamental transfer pricing topics can be expected, many of which will be similar to those in other cases. These topics will involve the relative contribution of the "bare" IP rights vs. the functions of the Dutch companies in the total profits and value creation from the IP-related activities. The Commission is likely to seek to closely scrutinise the transfer pricing documentation used to support the tax rulings in this respect.
Fundamental topics include questions on:
- The status of the various versions of the OECD transfer pricing guidelines, under both Dutch domestic law and EU law:
- Can they be considered binding for taxpayers and be applied as part of the reference system (the yardstick for determining the presence of a selective advantage)?
- Can they be overridden by EU law? Is there an EU arm's length principle that deviates from OECD standards?
- Should new transfer pricing concepts or insights, which were developed in later versions, be applied or not (dynamic vs. static interpretation)?
- What is the relevance of the "DEMPE functions" – development, enhancement, maintenance, protection and enforcement – which the OECD in the more recent versions of its guidelines considers so important for the allocation of profits from intangibles?
- How to apply this concept, including scoping and weighing the relative importance of those functions? Should functions performed by group companies that are not party to the relevant transactions be taken into account?
- The transfer pricing methods applied:
- Should the choice of transfer pricing method (for example, transactional net margin method, or "TNMM") be accepted?
- If "one-side" transfer pricing methods have been applied (such as a TNMM linked to costs or revenue), have the Dutch companies been correctly identified as the "least complex", taking into account functions performed, assets deployed and risks assumed?
- Should functions performed in the past be taken into account?
- Are the benchmarks that have been used reliable?
- The burden of proof.