The General Court of the European Union has concluded in two recent judgments that the burden of proof for the existence of state aid in the tax rulings issued by the tax authorities to companies lies with the European Commission (EC).
As has become common knowledge, the European Commission (EC) concluded in its decisions on the Fiat Finance and Starbucks cases that certain tax rulings issued by the tax authorities in Luxembourg and the Netherlands, respectively, to certain multinationals granted them more favorable tax treatment than that conferred in the general legislation applicable in those states:
- In Fiat's case, the EC considered that the margin on its intra-group financial activity which was regarded as taxable in Luxembourg was below a rational level in view of the funds and capital available at the company.
- In the Starbucks case, the EC concluded that the remuneration of the Netherlands company (which produced and distributed coffee grains) was not rational, from the standpoint of the characteristics of this activity; this remuneration was affected, to a large extent, by the payment of an excessive and unjustified royalty to another group company resident in the UK.
The General Court accepted in its judgments that the EC has the power to check correct application of the arm’s length principle, requiring transactions between related parties to be priced according to market conditions. However, the court concluded as follows:
- Transfer pricing is a complex field and allows a margin of appreciation.
- Not every discrepancy over the method used to price transactions between related parties may give rise to state aid.
- To conclude correctly as to whether state aid exists the EC (which has the burden of proof) needs to provide sufficient arguments that the pricing method used by the company is not appropriate and that it grants an advantage. In the Starbucks case, because these elements had not been demonstrated, the EC’s decision was set aside (which did not happen in the Fiat case).
- Tax rulings are individual measures which allow a presumption of selectivity.