For quite some time now, but particularly over recent years, the direct taxation of business profits has been undergoing significant transformations. The origin of these changes lies mainly in the need to establish a fair framework for the taxation of multinational enterprises (MNEs). Nonetheless, by extension (tax laws are usually targeted at the majority of cases) these transformations can have an impact on all enterprises.
Well known in this context are the initiatives of the OECD Action Plan on BEPS, aimed at limiting the opportunities for aggressive tax planning offered by the global economy. The essential objective of the plan is to guarantee the fairness of the tax system, ensuring that taxation takes place where the economic substance of the transactions is actually located.
Whereas the OECD must trust in the effectiveness of the above “soft law”, supported in recent years by a significant political consensus, the EU has to rely on its own channels to create positive law, fundamentally in the form of harmonisation directives. These directives can have a direct impact on “aggressive tax planning” but must also comply with the fundamental principles of the single market.
Attention is drawn, for example, to the recent reform of the EU Directive on the exchange of information, which is now to include a system of automatic exchange of administrative rulings and advance pricing agreements (APAs) which could be of interest to authorities that are not party to such agreements.
It is also to be noted that in order to supplement the above measure (or perhaps in order to prepare for its effects and pass a message to MNEs), the European Commission (EC) has, in recent years, redirected its efforts to combat harmful tax competition towards the application of the provisions of competition law (state aid rules). The “Luxleaks” phenomenon and the consequent instigation of formal state aid investigation proceedings against certain MNEs allegedly benefiting from preferential tax treatment (Apple in Ireland, Starbucks in the Netherlands, Fiat and Amazon in Luxembourg) are a good example of this trend.
These competition law measures – “hard law” indeed – are undoubtedly more effective as deterrents, but they also involve certain technical factors which could make their suitability for this purpose questionable. The existence of rules which create tax advantages in certain countries is not sufficient under state aid rules; the advantages created must be selective within those countries.
This is the context of the two decisions issued on October 21, 2015, in which the EC concluded that, by issuing certain transfer pricing rulings, Luxembourg and the Netherlands had granted selective tax advantages to Fiat Finance and Trade and Starbucks, respectively. According to the EC, those rulings embody complex methods that lead to the establishment of transfer prices with no economic justification and for which the only explanation is the wish to transfer profits to more favourable tax jurisdictions.
Furthermore, according to the EC, these rulings confer on the companies that benefit from them a competitive advantage over other enterprises (eg, SMEs that do not belong to a corporate group), which are taxed on their actual profits because they pay market prices. They are therefore contrary to the EU rules on state aid.
As a consequence, the EC has ordered Luxembourg and the Netherlands to recover the taxes not paid by Fiat and Starbucks due to the application of the rulings, which amount to €20-30m (£14-21m) per company. It has also prohibited the future application of the content of the rulings in question.
The reasoning of the EC contains a few obscure points. If the system of reference (scope of comparison) used to identify the possible advantage consists of all enterprises rather than all multinationals, the comparison will always show a clear advantage: by definition, companies that do not belong to multinational groups are not affected/favoured by transfer pricing agreements. The EC claims that it is not questioning the rulings mechanism, but its analysis appears to indicate the contrary.
Based on the above – and on the assumption that only some rulings, and not others, will be problematic – it would appear that only rulings that apply the arm’s length principle in what the EC regards as the correct manner will be acceptable. Nonetheless, selectivity consists not of doing or failing to do what the EC regards as technically correct, but of treating certain enterprises better than others that, in de iure and de facto terms, are in the same situation.
In other words, in order for a selective advantage to exist (why do some rulings constitute state aid and others not?), the tax measure in question needs to be compared with the standard approach that the tax authorities in the country concerned apply to multinational groups in general, and not with the approach that they should apply according to an ideal model unilaterally designed by the EC.
The two decisions can be appealed at the General Court of the EU and are expected to be appealed by the two member states and by the companies concerned. ?
Rafael Calvo Salinero (Taxand Spain), Jose Luis Buendia (Taxand Belgium) and Alfonso Lamadrid (Taxand Belgium)