In 2011, Willink SAS issued two intercompany convertible bonds with a maturity of 10 years and an annual interest rate of 8%. The tax authorities found that the 8% interest rate had not been determined in accordance with the arm’s length principle. Willink appealed, but both the Administrative Court and later the Administrative Court of Appeal sided with the tax authorities. The case was then appealed to the Conseil d’Etat which in December 2022 overturned the decision and ruled predominantly in favour of Willink SAS finding that RiskCalc could be used to determine a company’s credit rating for transfer pricing purposes in a sufficiently reliable manner, notwithstanding its shortcomings and the differences in the business sectors of the comparables. On that basis the case was remanded to the Administrative Court of Appeal. Judgment After re-examination of the case, the Administrative Court of Appeal annulled the tax assessment and ruled in favour of Willink SAS. “12. It follows from the foregoing that SAS Willink provides the evidence incumbent on it that the interest rate applicable to the transactions at issue could not have been lower than that which would have been applicable to borrowing transactions of the same nature entered into by independent undertakings. It is therefore entitled to request the cancellation of the corporation tax adjustments that led to the reduction of its losses carried forward for the 2011, 2012 and 2013 financial years, and the reinstatement of those losses.” Click here for English translation Click here for other translation ...

In 2011, Willink SAS issued two intercompany convertible bonds with a maturity of 10 years and an annual interest rate of 8%. The tax authorities found that the 8% interest rate had not been determined in accordance with the arm’s length principle. Willink appealed, but both the Administrative Court and later the Administrative Court of Appeal sided with the tax authorities. Judgment of the Supreme Court The Conseil d’Etat overturned the decision and ruled in favour of Willink SAS. The court found that RiskCalc could be used to determine a company’s credit rating for transfer pricing purposes in a sufficiently reliable manner, notwithstanding its shortcomings and the differences in the business sectors of the comparables. Click here for English translation Click here for other translation ...

In 2011, Willink SAS issued two intercompany convertible bonds with a maturity of 10 years and an annual interest rate of 8%. The tax authorities found that the 8% interest rate had not been determined in accordance with the arm’s length principle. Willink appealed, but in a decision issued in 2019 the Administrative Court sided with the tax authorities. An appeal was then filed with the Court of Appeal. Judgment of the Supreme Court The Court of Appeal dismissed the appeal of Willink and upheld the decision of the Administrative Court. Excerpt “7. It is common ground that the funds Apax France VIII-A, Apax France VIII-B and the companies MidInvest and Telecom Online are linked to the company Willink, of which they are all partners, and that the rate of 8% exceeds the rate provided for in the first paragraph of 3° of 1 of Article 39 of the General Tax Code. To justify that this rate was not higher than the one it could have obtained from independent financial institutions or organisations under similar conditions, SAS Willink produced before the Court a comparative rate study carried out in 2020 using the Riskcalc software developed by Moody’s Analytics, a subsidiary of the rating agency Moody’s. This study is based on a model calculating the probability of default in the short term (one year) and the long term (five years) and then associates an implicit scoring. In order to select the most reliable and consistent scoring possible, this was determined on the basis of the applicant’s financial statements for the years 2011, 2012 and 2013. A search for comparable transactions on the open market was then carried out using the SetP Capital IQ database. Transactions were selected for which the issuing companies had a score comparable to Willink, issued by public or private companies across a range of industries during the relevant period. The sample was then refined, including transactions with a maturity close to each of the bonds to be compared. An interquartile range of arm’s length interest rates was then constructed on the basis of the bonds identified as comparables and used to identify median rates. 8. Although it is possible to assess the arm’s length rates by taking into account the yield on bonds, it is only on condition that, even if the loan is a realistic alternative to an intra-group loan, the reference companies are in comparable economic conditions. In the present case, this condition cannot be considered to be met for the companies selected in the sample of the report mentioned above. The level of risk used as a basis for comparison is based on a statistical model derived from historical quantitative data for companies that are not representative of the market, since defaulting companies are over-represented, and was determined on the basis of some ten financial data provided by the company itself. There is nothing to establish that this risk rating adequately takes into account all the factors recognised as forward-looking, and in particular the characteristics specific to the sector of activity concerned, even though this sector of activity is provided for the implementation of the model. Nor is it established that the so-called comparable companies in the study sample, which belonged to heterogeneous sectors of activity, would have presented the same level of risk for a banker as that with which the interested party was confronted at the same time. It follows that SAS Willink, which cannot usefully argue in these circumstances that the service cannot require a rating from a rating agency for each of the intra-group financing operations, cannot be regarded as providing the proof, which is incumbent on it, that it could have obtained a rate of 8 % from independent financial institutions or organisations under similar conditions. 9. It follows from all the foregoing that SAS Willink is not entitled to maintain that it was wrongly dismissed by the contested judgment of the Paris Administrative Court. Consequently, its claims for the application of Article L 761-1 of the Code of Administrative Justice can only be rejected.” Click here for English translation Click here for other translation ...