On 24 September 219, the General Court (“GC”) delivered its long awaited judgments on the European Commission’s (“Commission”) decisions finding that tax rulings granted to Starbucks and Fiat constituted State aid. The GC annulled the Commission’s decision on Starbucks but upheld the Commission’s decision on Fiat. The judgements confirm that State aid rules enable the Commission to review whether tax rulings endorsing transfer pricing arrangements are in line with the arm’s length principle. However, in order to find that such tax rulings constitute State aid, the Commission must clearly show that they reduced their beneficiaries tax burden and cannot limit itself to pointing out inaccuracies or mistakes in the methodology used to calculate transfer pricings.
The Commission started investigating tax rulings granted by Member States to multinational groups in mid-2013. Its investigation focuses in particular on the transfer pricing arrangements used in those rulings to determine the taxable profits of group companies.
The Commission has expressed concern that some Member States may have reduced the tax burden of group companies by allowing transfer pricing levels that are not in line with the “arm’s length” principle. According to the Commission, the arm’s length principle would require that transfer prices charged within multinational groups reflect those which would be charged in conditions of free competition, that is prices negotiated by independent undertakings negotiating under comparable circumstances. And that this is a general principle of equal treatment in taxation and falls within the remit of State aid control.
The Commission’s decisions of 21 October 2015 in the Starbucks and Fiat cases were the first decisions to sanction as incompatible State aid tax rulings allegedly infringing this principle.
In the Fiat case, the Commission considered that Luxembourg lowered the tax burden that should have normally been borne by Fiat Finance and Trade Ltd (“FFT”), a subsidiary of the group Fiat/Chrysler which provided treasury and financial services to other companies of the group (such as market funding and liquidity investments, intra-group loans, financial coordination, etc.). Luxembourg had agreed in a ruling of 2012 to determine FFT’s taxable profit on the basis of the OECD’s transactional net margin method (“TNMM”), by applying a profit ratio on the capital needed by FFT to perform its functions and bear the financial risks linked to these functions. The Commission considered that Luxembourg made several methodological mistakes in determining the adequate profit ratio and level of capital needed by FFT. This would have resulted in attributing to FFT lower taxable profit than under market conditions and reduced its tax burden by EUR 20 to 30 million between 2012 and 2015.
In the Starbucks case, the Commission considered that the Netherlands lowered the tax burden that should have normally been borne by Starbucks Manufacturing EMEA BV (“SMBV”), a subsidiary of the Starbucks group responsible for the production and distribution of roasted coffee beans and related products to Starbucks shops in the EMEA. In 2008, the Netherlands issued a ruling endorsing a methodology to calculate SMBV’s taxable profit as a percentage of its operating costs and excluding as deductible costs:
- the payment of a royalty for the use of IP related to roasting to another group company, Alki LP, and
- a remuneration for the supply of green coffee beans to Starbucks Coffee Trading SARL.
According to the Commission, the royalty paid to Alki should have been zero and the price paid to Starbucks Coffee Trading for green coffee beans was excessive. The Commission concluded that these amounts did not reflect economic reality and unduly lowered SMBV’s taxable profit in the Netherlands. As a result, Starbucks would have benefited from a tax advantage totaling EUR 20 to 30 million between 2008 and 2015.
The GC’s Judgments
The Court confirmed the existence of an arm’s length principle in the field of State aid
Both decisions were appealed to the GC by the Member States and undertakings concerned. In both appeals, the parties contested the existence of an arm’s length principle specific to State aid law.
In that regard, the Court first recalled that Article 107 (1) TFEU prohibits State measures which distort or threaten to distort competition by granting an advantage to certain undertakings. According to settled case law, a tax measure constitutes an advantage if it mitigates the tax burden of its beneficiary by derogating to the “normal” rules of taxation.
Further, the Court explained that where national rules on corporate tax do not distinguish between integrated and standalone companies for the determination of their taxable profit, these rules intend to tax the profits of integrated companies as if those companies carried out their activities on market terms.
On this basis, the Court considered that Article 107 (1) TFEU allows the Commission to verify whether a tax ruling derogates from the normal rules on corporate tax – and therefore entails an advantage for its beneficiary – by endorsing levels of transfer pricing that do not correspond to market terms. The arm’s length principle is therefore a “tool” that can be used by the Commission for the purpose of State aid control.
Member States however have a margin of appreciation in the determination of transfer prices
However, the Court also explained that methods to determine transfer pricing are necessarily approximate in nature. Therefore, when verifying whether transfer pricings endorsed in a ruling constitute a reliable approximation of market prices, the Commission can identify an advantage within the meaning of Article 107(1) TFEU only if the variation between the two comparables goes beyond the inaccuracies inherent in the methodology used to obtain that approximation.
Further, the Commission is required to justify the choice of methodology used to calculate the transfer pricing that it considers appropriate. In that regard, while the Commission is not formally bound by the OECD guidelines on transfer pricing, the GC stated that these guidelines “have a certain practical significance in the interpretation of issues relating to transfer pricing”.
Application in the cases at hand
On this basis, the GC reviewed whether the errors identified by the Commission in the rulings granted by Luxembourg and the Netherlands to Fiat and Starbucks went beyond the inaccuracies inherent to the methodologies used to calculate transfer prices.
In the Fiat case, the GC agreed with the Commission’s analysis and confirmed that Luxembourg made several methodological mistakes when applying the TNMM to determine FFT’s remuneration. In particular, Luxembourg underestimated the capital required to perform FFT’s function and bear its risks by a factor of ten. FFT’s remuneration, calculated as a return on capital under the ruling, did therefore not constitute a reliable approximation of a market based remuneration. This resulted in a reduction of FFT’s tax burden in Luxembourg and therefore in an advantage in the meaning of Article 107 (1) TFEU.
However, in the Starbucks case, the GC considered that the Commission did not demonstrate that the transfer prices endorsed by the Dutch ruling did not constitute a reliable approximation of market prices.
First, the Court disagreed with the Commission that the transfer pricing method chosen in the ruling, the TNMM, necessarily led to an estimate too low in the case at hand and that another method, the comparable uncontrolled price (CUP), should have been used. According to the Court the various methods to set transfer prices identified in the OECD guidelines all endeavor to attain profit levels reflecting arm’s length prices and “it cannot be concluded, as a rule, that one method does not allow a reliable approximation of an arm’s length outcome to be reached”.
Second, the GC considered that the Commission did not establish that the royalty paid by SMBV to Alki for the roasting IP should have been zero. Almost all evidence relied on by the Commission in its analysis was unavailable at the time the ruling was granted and should therefore be disregarded. Further, the Court considered that the roasting IP did have economic value as SMBV needed it to exercise its activity and that the payment of a royalty was therefore justified.
Third, the GC considered that the Commission did not show that the ruling endorsed transfer pricing for the supply of green coffee beans that was not in line with the arm’s length principle. In particular, the Court pointed out that the price of coffee beans was not determined in the ruling but in annual tax assessments. It was therefore outside of the scope of the measure identified as State aid by the Commission: the ruling. Further, the Court considered that in order to show that the price of green coffee beans paid by SMBV was excessive, the Commission should have carried out a comparison with the price of green coffee beans paid by standalone companies, which it did not do.
Finally, the Court considered that remaining errors identified by the Commission in its subsidiary argumentation were not sufficiently motivated and did not allow to show that the transfer pricing was not in line with the arm’s length principle.
The GC’s judgments in Fiat and Starbucks confirm that the Commission is entitled to review tax rulings in order to determine whether transfer pricing arrangements are in line with the arm’s length principle. They also confirm the Commission’s definition of the arm’s length principle under EU law, i.e. that transfer prices charged within a group should constitute a reliable approximation of prices charged on the market between independent companies. These judgments therefore constitute a green light to the Commission to continue the investigation started in 2013. It may therefore be expected that the Commission will open new tax ruling cases in the future.
However, these judgments also constitute a warning to the Commission, as the Court has shown that it will carry out an in-depth review of the Commission’s assessment. In order to sanction a tax ruling as illegal State aid, the Commission must clearly demonstrate that this ruling confers an advantage. Identifying mistakes in the methodology used to set transfer prices is not enough. The Commission must show that these mistakes go beyond the inaccuracies inherent to that methodology and that the transfer prices agreed in the ruling are in fact not a reliable approximation of prices charged by independent companies in conditions of free competition. Further, when reviewing tax rulings the Commission must take due account of the OECD guidelines on transfer pricing and cannot base its assessment on elements that are subsequent to the ruling.