France vs BSA Finances, December 2021, Court of Appeal Versailles, Case No 20VE03249

« | »

In 2009, 2010 and 2011 BSA Finances received a total of five loans granted by the Luxembourg company Nethuns, which belongs to the same group (the “Lactalis group”). Depending on the date on which the loans were granted, they carried interest rates of respectively 6.196%, 3.98% and 4.52%.

Following an audit covering the FY 2009 to 2011, the tax authorities considered that BSA Finances did not justify that the interest rates thus charged should exceed the average effective rates charged by credit institutions for variable-rate loans to companies with an initial term of more than two years. Hence, the portion of interest exceeding these rates was considered non-deductible pursuant to the provisions of Article 212(I) of the General Tax Code.

In 2017, the  Administrative Court ruled in favor of BSA Finances and discharged the additional corporate tax. But this decision was appealed by the authorities to the Administrative Court of Appeal which in  June 2019 overturned the decision of the lower court.

The Judgement from the Administrative court of Appels was then appealed by BSA Finances to the French Supreme Administrative Court.

The Supreme Administrative Court overturned the decision from the Court of Appeal and remanded the case to the Court of Appeal.

In considering that the company had not established that the margin rates applied were in line with market rates for loans made under the same conditions, whereas the Riskcalc application, which it was not disputed was fed from the company’s balance sheets and profit and loss accounts over several years, had classified its level of risk as “BBB/BBB-” on the basis of comparative ratios established by Moody’s, that the refinancing contracts produced, which made it possible to determine the actual margin rate of the loans taken out by the applicant company itself, were accompanied by details making it possible to compare the main specific conditions with the clauses of the loans in dispute and that, lastly, the combination of these elements was such as to justify, in the absence of any element to the contrary, that the credit margins applied by Nethuns were in line with market practices, the Court distorted the documents in the file submitted for its assessment.”

Judgement of the Court of Appeal

The court ruled in favor of BSA stating that

  • a scoring obtained by using automated tools such as RiskCalc is inherently less accurate than the actual rating a proper rating agency
  • the fact such a scoring is less accurate does not mean that it can be disregarded systematically.
  • in the absence of any valid criticism of the scoring by the tax authorities, it was an acceptable proof

“10. After the Conseil d’Etat, in its decision of 11 December 2020, annulled the judgment of the Court of Appeal of 25 June 2019 on the grounds that the combination of elements attesting to a rating of the company’s risk by means of a publicly accessible financial tool, namely the RiskCalc software developed by the rating agency Moody’s, which it was not disputed was fed from the company’s balance sheets and profit and loss accounts over several years, and the syndicated contracts entered into with the company, which were not the subject of a complaint, was not sufficient to justify the decision, and the syndicated contracts concluded with financial organisations in 2010 and 2011 were such as to justify, in the absence of any evidence to the contrary, that the credit margins applied by Nethuns were in line with market practices, the Minister for Economic Affairs, The Minister of the Economy, Finance and Recovery reiterated his criticism of the RiskCalc software, arguing that it covers only a small fraction of the methodology used by the rating agencies, so that it provides only a measure of the probability of default that is meaningful only in relation to the scale of default probabilities created by the software itself. It questions the relevance of the model for entities such as SNC BSA Finances insofar as it is based on data from companies with gross assets of less than EUR 10 million, and is therefore not relevant to global groups. It also stresses the inadequacy of the information provided by the methodologies published by Moody’s regarding the adjustments allowed by the software and their potential impact on the rating, even though these adjustments may be significant, particularly in terms of taking into account the support of the parent company or the special treatment of shareholder loans.

11. However, on the one hand, SA BSA argues, without being challenged, that the ‘BBB/BBB-‘ ratings used correspond to a ‘conservative’ analysis based on ratings that are less downgraded than those of SNC BSA Finances with regard to the rating of its main partner, known as a corroborative economic analysis, so that, having itself made the necessary adjustments, the argument based on the failure to take account of the group’s support in determining the rating is, in any event, lacking in fact. The Minister does not mention, in detail, any other form of data restatement, in particular as regards a possible ‘special treatment of shareholder loans’, which would have been necessary in this case. On the other hand, SA BSA argues, again without being challenged, that if the model is established with regard to a sample drawn up by Moody’s showing the balance between small and large companies, it does not lead to an under-representation of the latter given their economic weight. Finally, in a more general way, it is certainly constant that the ratings obtained from tools of the trade make it possible to attribute a rating to a specific loan that is more approximate than a credit rating that could be carried out by a rating agency for a given borrower. Nevertheless, while SA BSA argues without being challenged that the use of a rating agency is not intended to apply, given its cost, in an intra-group transaction, the rating provided in this case by RiskCalc can be considered sufficiently reliable to justify the profile of SNC BSA Finances and to assess it, in the absence of detailed criticism, an arm’s length interval, the Minister recognising moreover a reliability rate of 70% and refraining from indicating whether and to what extent this rating would be erroneous, which is moreover consistent with that attributed by the same tool to SA BSA and the internal rating attributed by Société Générale to the latter.

12. Under these conditions, and while the economic study (benchmark) is produced by SA BSA only to support the internal comparables put forward as the main argument, the relevance of which the Minister does not dispute and which he even maintains must be considered as the ‘determining factor’, the interested party must be regarded as justifying, in the absence of any evidence to the contrary, that the credit margins charged by Nethuns were in line with market practices and, consequently, that the rates charged by Nethuns were in line with the market, that the rates charged by Nethuns were, in their three components, lower than or equal to those which it could have obtained from independent financial institutions or bodies under similar conditions, such a burden of proof being incumbent on it, contrary to what it maintains, pursuant to the abovementioned provisions and without, however, disregarding the principles laid down by the Test Claimants in the Thin Cap Group Litigation decision of 13 March 2007 of the Court of Justice of the European Union (Case C-524/04). C-524/04).

13. It follows from all the foregoing that the Minister for Economic Affairs, Finance and Recovery is not entitled to maintain that the Montreuil Administrative Court, by the judgment under appeal, wrongly upheld the application of SA BSA…”


Click here for English translation

Click here for other translation


France vs BSA Dec 2021 CAA de VERSAILLES 20VE03249

Related Guidelines

Leave a Reply

Your email address will not be published. Required fields are marked *