European Commission decides that tax advantages given to Fiat in Luxembourg and Starbucks in the Netherlands breach EU State aid law

Publication November 2015


Introduction

On October 21, 2015, the European Commission (Commission) announced that it had adopted two decisions concluding investigations into whether tax rulings by national tax authorities in Luxembourg and the Netherlands conferred illegal State aid on Starbucks Manufacturing EMEA BV (Starbucks Manufacturing) and Fiat Finance and Trade (FFT).

The Commission concluded that Dutch and Luxembourg authorities did violate EU State aid law, because their rulings allowed those companies to reduce their tax base through the use of transfer pricing arrangements that, according to the Commission, “do not reflect economic reality.” According to the Commission, transfer pricing arrangements risk infringing State aid law if they sanction a methodology for allocating profits between different parts of a group that is not consistent with the “arm’s length” principle. This practice, in the Commission’s view, accords companies an “an unfair competitive advantage over other companies (typically SMEs) that are taxed on their actual profits because they pay market prices for the goods and services they use.” The Commission ordered Luxembourg and the Netherlands to recover the alleged illegal aid, which it said would be in the range of €20-30 million each.

The Commission’s application of State aid law principles to tax rulings is highly controversial. Many have questioned whether it is correct, and the full impact of the Commission’s decisions will likely become clear only after years of litigation. Meanwhile, it appears that the Fiat decision may have broader implications for multinational groups than the Starbucks decision. Starbucks’ Dutch structure is highly fact specific, and Starbucks has already announced that the Commission got its facts wrong in a key respect. The Fiat decision could have broader implications for the calculation of profits by multinational groups’ finance subsidiaries and potentially investment fund structures.

The full texts of the Commission’s decisions are not yet public, but further background on the Commission’s approach can be gleaned from the decisions it published when it opened its in-depth investigations into Starbucks Manufacturing (Starbucks Opening Decision) and FFT (Fiat Opening Decision). The opening of these and other investigations was the subject of a Norton Rose Fulbright briefing earlier this year. Separately, future decisions are expected in Commission investigations into transfer pricing tax rulings involving Apple in Ireland and Amazon in Luxembourg, as well as the Belgian “excess profits” tax regime.

The Starbucks Decision

In the Starbucks decision, the Commission found that transfer pricing arrangements agreed with Dutch tax authorities allowed Starbucks Manufacturing to reduce its tax liability in the Netherlands through two intra-group agreements, an agreement for the purchase of green coffee beans with a Swiss affiliate and a license agreement with a UK affiliate.

In the Starbucks Opening Decision, the Commission noted that the Dutch tax ruling was based on the so-called transactional net margin method (TNMM) of calculating the net profit that a taxpayer realises from intra-group transactions. That method, which is approved in the OECD Transfer Pricing Guidelines, involves establishing the net margin on internal “controlled” transactions with a internal comparator, such as an internal “uncontrolled” transaction, or an external comparator, such as evidence of margins achieved by independent enterprises in comparable transactions. In the case of comparisons with independent enterprises, the relevant margin can be adjusted to take account of limited functional differences between the companies being compared.

In the Starbucks Opening Decision, the Commission questioned whether the Dutch tax authorities were correct to:

  1. Classify Starbucks Manufacturing a “toll manufacturer” (a company which processes raw materials or semi-finished goods for another group company) that may be subject to lower margins than other group companies;
  2. Accept certain margin adjustments applied to take account of apparent functional differences between Starbucks Manufacturing and comparator companies; and
  3. Accept Starbucks Manufacturing’s calculation of royalties paid to other group companies, where those royalties were not linked to the value of the intellectual property in question.

The Commission’s press release announcing its decision did not discuss these specific points, but identified the following as illegal selective advantages:

  1. The payment of substantial royalties to a UK affiliate for coffee roasting intellectual property although (according to the Commission) other roasters did not pay comparable royalties; and
  2. The payment of inflated prices to Starbucks Coffee Trading SARL, a Swiss affiliate, for green coffee beans, leading to a trebling of the affiliate’s margins since 2011.

The Commission’s State aid decisions can be appealed to the Court of Justice of the EU, and Starbucks has already announced its intention to do so. According to press reports, Starbucks argues that the transfer pricing agreements it reached with the Dutch authorities are consistent with international best practice as set out in the OECD Transfer Pricing Guidelines and that the Commission ignored evidence that Alki did charge royalties to a wholly independent coffee roaster.

The Fiat Decision

FFT is a Luxembourg company providing financial services, including intra-group loans, to other Fiat group companies. The investigation centred on a 2012 ruling by Luxembourg authorities which, in the Commission’s view, provided a selective tax advantage to FFT by allowing it to significantly reduce its Luxembourg tax base.

As in the Starbucks case, the Luxembourg tax ruling approved FFT’s use of the TNMM to calculate its taxable profit. The Commission’s opening statement did not challenge the use of TNMM (although the Commission suggested that the comparable uncontrolled price method, which consists in observing a comparable transaction between two independent companies and applying the same price for a comparable transaction between group companies, might have been more appropriate), but took issue with how TNMM was applied in that case.

In its press release, the Commission identified the following aspects of the calculation of FFT’s profitability as inconsistent with the arm’s-length principle:

  1. The use of a number of economically unjustifiable assumptions and downward adjustments to reduce FFT’s capital base. In the Fiat Opening Decision, the Commission argued that the tax ruling allowed FFT to hold insufficient capital aside to cover its intra-group exposure, particularly when compared to finance companies adhering to minimum capital requirements under the Basel II accords. As this “at risk” capital is remunerated at a relatively high interest rate (to reflect the risks involved), its reduction underestimated FFT’s overall profitability.
  2. The estimated remuneration applied to the already artificially reduced tax base fell below what would be expected under normal market conditions. In particular, the Commission noted in the Fiat Opening Decision that tax advice provided in relation to the Luxembourg ruling compared FFT’s operations against a “list of [66] comparable companies engaged in financial services.” It placed FFT in the lower 25 percentile of those entities, reflecting the fact (in their view) that FFT “bears limited risks.” The Commission challenged (i) the list of comparators (for example, two of them were central banks), and (ii) the fact that FFT was placed at the lower end of that risk scale. Moreover, the return on the capital backing FFT’s operations was considered to be exceptionally generous (being 0.87%, versus 2.85% which was the “risk-free rate” in 2011 (based on German government bonds)).

As a consequence of the assumptions in the Luxembourg tax ruling, the Commission estimates that FFT’s tax base was only one-twentieth of what it would have been had the arm’s-length principle been applied.

In the Fiat Opening Decision, the Commission also criticized the fact that the Luxembourg authorities fixed FFT’s tax base (within a 10% margin) over the five-year term of the ruling. In other words, FFT’s tax liability would not change appreciably even if there was a significant increase in its activities during that term. This point was not addressed in the Commission’s press release, and it remains to be seen if it was included in the final text of the decision.

Although Luxembourg and FFT have not yet announced their intentions, it seems likely that the Fiat Decision will also be appealed to the Court of Justice.

Recent publications

Subscribe and stay up to date with the latest legal news, information and events...