Spain vs BIOMERIEUX ESPAÑA SA, February 4 2021 English translation

NATIONAL COURT

Chamber for Contentious-Administrative Proceedings SECOND SECTION

Appeal No.: 0000658 /2017

Type of Appeal: ORDINARY PROCEDURE General Registry No.: 04256/2017 Applicant: BIOMERIEUX ESPAÑA SA

Attorney: ISABEL COVADONGA JULIA CORUJO

Defendant: TRIBUNAL ECONOMICO ADMINISTRATIVO CENTRAL (CENTRAL ECONOMIC ADMINISTRATIVE COURT)

Abogado Del Estado

Rapporteur: IImo. Mr MANUEL FERNÁNDEZ-LOMANA GARCIA

S E N T E N C E N C E Nº :

IImo. Mr. President:

D. MANUEL FERNÁNDEZ-LOMANA GARCÍA

Mr. Judges:

D. FRANCISCO GERARDO MARTINEZ TRISTAN

JAVIER EUGENIO LÓPEZ CANDELA

D. RAFAEL MOLINA YESTE

Madrid, on the fourth day of February, two thousand and twenty-one.

Before the Chamber for Contentious-Administrative Proceedings of the National High Court, appeal no. 658/2017, brought by BIOMERIEUX ESPAÑA SA, appearing represented by Attorney Isabel Covadonga Julia Corujo and assisted by Counsel Celso Cañizares Pacheco, against the Resolution of the Central Economic-Administrative Tribunal of 9 May 2017 (RG 4746/13); the Administration being represented and defended by the State Attorney. The amount has been ﬁxed at €128,643.28.

FACTUAL BACKGROUND

FIRST.- On 21 July 2017, a writ was filed filing a contentious-administrative appeal.

SECOND.- After several formalities, the claim was formalised on 5 March 2018. The State Attorney's Office filed a response on 24 May 2018.

THIRD - The trial was received for evidence. Briefs of conclusions were presented on 5 and 25 July 2018, and the vote and ruling was scheduled for 21 January 2021.

The Judge Judge who is the Judge-Rapporteur is Mr. MANUEL FERNANDEZ-LOMANA GARCIA, who expresses the opinion of the Chamber.

GROUNDS OF LAW

FIRST - The contested decision.

A contentious-administrative appeal is brought against the Resolution of the TEAC of 9 May 2017 (RG 4746/13), which dismisses the appeal brought in relation to the IS for the financial year 2008.

Before analysing the grounds of appeal, it should be borne in mind that the appellant itself filed a written statement requesting "the withdrawal of the claim filed in respect of the tax debt corresponding to the transactions between BM SPAIN and BIOMERIEUX SA (BM France), having initiated amicable proceedings" -pp. 9 and 10 of the TEAC Resolution-. The partial withdrawal was admitted by the TEAC - p. 11-. The debate is limited to the regularisation of the part of the tax debt corresponding to the transactions between BM ESPALA and BM INC (USA).

The grounds of challenge are twofold:

1.-Correct determination of the transfer prices of the transactions carried out between BM SPAIN and BM USA -pp. 8 to 35-.

Application to the transactions between BM SPAIN and BM USA of those agreed in the amicable procedure between the Spanish and French tax authorities in respect of the transactions between BM SPAIN and BM FRANCE.

SECOND.- Correct determination of the transfer prices of the transactions carried out between BM SPAIN and BM USA.

The most orderly way of exposition consists of setting out the regularisation carried out by the Inspectorate, focusing on the issues discussed by the appellant and responding to all of them.

1.- For the sake of clarity, we will refer to the non-conformity report (ID), which is assumed by the settlement agreement (AL). Without prejudice to adding, when the presentation so requires, the appellant's allegations and the response given to these questions by the AL and the TEAC. We will also set out the Chamber's position in relation to the issues debated. Logically, we will limit our presentation to what is necessary to resolve the regularisation in relation to BM USA.

BM SPAIN is part of the BM group, with all of its shares owned by the parent company of the bioMerux SA group, resident in France, which prepares consolidated financial statements.

Its business is essentially concerned with all aspects of clinical and biological analysis in the broadest sense, at all stages of production, distribution, training and technical assistance. Likewise, the provision of computer services and, in particular, the computer management of laboratories, including, inter alia, the sale, installation, programming, management and maintenance of computer programs and equipment. Without prejudice to further speciﬁcation throughout the report - p. 3 of the ID-. 3 of the ID.

This entity carries out commercial operations with BM FRANCE and BM USA. The latter entity BioMeriux INC is wholly owned by bioMerux SA through the US resident entity ABG Stella of which bioMerux SA holds 100%.

We are dealing with related entities -p. 4 of the ID-

BM SPAIN, during the 2008 financial year, carried out the distribution of finished and purchased products to BM FRANCE and BM USA. It also acquired all the shares of Biomedics SL, adding to the distribution activity two manufacturing activities: the manufacture and distribution of Biomedics branded products and the contract manufacturing of bioMerioux products, which it transferred in their entirety to BM FRANCE - p. 4 ID-. 4 ID-.

2.- The Inspectorate cites the Agreement with the USA -art 9.1-, art 16 TRLIS and art 16 RIS. It is important to note that the Inspectorate, like the appellant, also takes into account the OECD Guidelines, in their 2010 version and recalls that Law 36/2006, of 29 November, indicates that they will have interpretative value. The appellant does not dispute the application of those Guidelines. In fact, as tax operators usually do, it constantly refers to them, seeking interpretative criteria and criteria for the application of the rules - p 8 ID-. Moreover, the above is logical, since the legislator, as repeatedly indicated in the explanatory memoranda, has as one of its basic objectives to adapt Spanish legislation to the international context, and the OECD Guidelines are of great value in this respect, which cannot and should not be minimised.

The report indicates, and the plaintiff does not dispute this, that in the tax period under review the transfer pricing documentation criteria established by RD 1793/2008 were not in force. However, according to Guidelines 1.12, the Inspectorate may, in relation to the transfer price used, "require companies to demonstrate its consistency with the arm's length price". The Guidelines add that the taxpayer "should keep the information relating to its transfer pricing and be prepared to submit it" and "should be prepared to provide the documentation referred to" - p. 9 of the ID.

On the basis of these regulations, the Inspectorate asked the taxpayer to justify the transfer price applied and that it was in line with market values.

4.- The Inspectorate proceeded to describe in detail to justify the way in which it had determined the transfer price and which is described in detail on p. 10 of the ID, as well as the rest of the documentation presented. On pp. 11 et seq. he summarises the content of the report entitled "Informe precios 2008 bioMerieux España".

Without prejudice to the reading of these pages, it is stated therein that the amounts of the purchases of finished products in 2008 from MB USA were 5,697,672.24 €.

It is interesting to note, given the terms of the debate, that in relation to MB EUU "there is no agreement regulating the terms and conditions between bioMeriux Ince and bioMerioux Spain, but they are similar to those contained in the agreement between bioMerioux Spain and biomerioux France" -p.12 ID-.

5.- The method applied by the taxpayer is the so-called "net operating margin method" (OECD) or "net margin method of the set of operations" (art 16.4.2.b) of the TRLIS).

The profit indicator chosen by the taxpayer is the ROS (return on sales) which corresponds to the so-called "operating profit".

As regards the search for comparables, the taxpayer carried out a study analysing data from three financial years: 2008, 2009 and 2010 -ID p. 12-. P. 13 et seq. of the ID describes the selection process carried out by the taxpayer in search of independent entities carrying out similar activities.

P. 14 describes the 18 companies selected and explains that the companies considered have a "range from 0.17% to 49.83% and the interquartile range for the three years is 4.0%-11.9% with a median of 6.7%".

It is then reasoned that, based on the profit and loss account, an "operating result on sales volume of 4.65% is obtained, which is in the interquantile range of the median value of the European comparables in the period 2008-2010 (4.0%-11.9%) - p. 15 ID-.

It concludes with a transcription of the 2008 figure for revenues, expenses and distribution operating result -p. 15 ID-.

6.- The Inspectorate begins by indicating that the objective of the regulation is to make a "reasonable approximation" to market value and that this is not an "exact science, but requires value judgements by both the tax authorities and taxpayers". It then indicates that it will start from the report provided and will only make "a series of adjustments in those cases in which it considers that its content does not correspond to a correct application of the principle of free competition" -p. 16 and 17 ID-.

This way of reasoning seems to us to be correct. It is not argued that the application determined by the taxpayer, in this case, is arbitrary or manifestly irrational. What is said is that in relation to certain parameters there are others that are more reasonable and better correspond to the purpose sought. It will be necessary to analyse the reasons given by the Administration and by the appellant and to make a judgement of reasonableness from the ﬁnalidad sought.

Indeed, if what is sought through the transfer pricing ﬁgure is that the prices correspond to the market value that would have been agreed between independent parties -principle of free competition-, recognising that there is always a margin of uncertainty, the parameters that allow approaching with greater certainty to that ﬁnality will have to be taken into account.

Starting, therefore, from the method of the net margin of the transaction, which is used by the company and which consists of determining the market value of the transactions based on the operating profit, that is, on the EBIT or income before taxes and financial costs that would be obtained in transactions with independent third parties, we must analyse whether the application of the method carried out by the inspectorate is more reasonable than that carried out by the taxpayer and is closer to the desired objective.

7.- Well, without prejudice to referring to the reading of pp. 13 et seq. of the ID, the first discrepancy that arises between the appellant and the Administration is described on pp. 23 et seq. for the Inspectorate, following the OECD Guidelines: "Income and expenses not related to the related transaction under review should be excluded when they signiﬁcantly affect comparability with unrelated transactions. A certain degree of segmentation of the taxpayer's financial data is necessary to determine or test the net benefit to the taxpayer from a controlled transaction. It would therefore be inappropriate to apply the net operating margin method at the level of the whole enterprise if the enterprise is engaged in a number of different related party transactions that cannot be compared on an aggregate basis with those of a stand-alone enterprise".

This idea is developed, as indicated above on pp. 23 et seq. of the Remand Report and implies that the net margin achieved in 2008 was 4,47% and not 4,65%. Therefore, the ROS was 4.47% as rightly indicated by the Inspectorate.

Unless we are mistaken, nothing is said in the complaint on this issue.

8.- The second point of discrepancy is the time frame chosen. It is described on pp. 29 et seq. of the ID.

As stated in point 3.67 of the Guidelines, "comparability raises questions about the time frame in relation to the date of origin, collection and compilation of the information relating to the comparability factors and the comparable unrelated transactions used in the comparability analysis".

In paragraph 3.68 on the so-called "date of origin", the Guidelines state that: "In principle, one would expect that the most accurate information that can be used in a comparability analysis is information about the terms of comparable unrelated transactions entered into contemporaneously with the related transaction ("contemporaneous unrelated transactions"), since they reflect the behaviour of independent parties in an economic environment identical to that in which the taxpayer's related party transaction took place. However, practice shows a certain limitation in the availability of information on contemporaneous non-related transactions, depending on the date on which it is collected".

Therefore, what the Guidelines recommend is that comparability be made, to the extent possible, with contemporaneous transactions.

It is true that Guideline 3.69 adds that: "In some cases, taxpayers prepare transfer pricing documentation to demonstrate that they have made reasonable efforts to comply with the arm's length principle at the time the intra-group transactions took place, i.e. on an aprioristic time basis (hereinafter referred to as the 'arm's length pricing basis'), based on the information reasonably available to them at that time. This information includes not only information about comparable transactions in previous years, but also information about economic and market changes that may have occurred between those previous years and the year in which the controlled transaction took place. Similarly, independent parties in similar circumstances would not base their pricing solely on historical data".

Guideline 3.70 that "In other cases, taxpayers may need to verify the actual outcome of their controlled transactions to demonstrate that the terms of these transactions are consistent with the arm's length principle, i.e. with an ex post timing test (hereinafter referred to as the 'arm's length outcome verification test'). This veriﬁcation is normally carried out as part of the process of preparing the tax return at the end of the year".

Finally, Guideline 3.75, regulating what it calls "multi-year data", adds that: "In practice, the examination of multi-year data is often useful for comparability analysis, although it is not systematically required. Such multi-year data should be used where they add value to the transfer pricing analysis. It would not be appropriate to set standards for the number of years to be covered by an analysis of more than one year.

This implies that, in principle, in order to determine the transfer price, comparability should, as far as possible, be made with contemporaneous transactions, i.e. transactions carried out in the financial year. Apart from the fact that, in fact, it should not be ignored that it is sometimes difficult to obtain the necessary data for this purpose and, therefore, the possibility of taking into account, where necessary, transactions from previous and subsequent years, which confirm the reasonableness of the comparison made. A "multi-year" analysis should be used where this is useful for comparability.

As explained on pp. 32 et seq., the company has resorted to data referring to 2008, 2009 and 2010, remembering that we are judging the 2008 financial year. Well, on p. 33 the Inspectorate reasons that "the STR data for those years contained in the AMADEUS database for the sample of 18 companies selected by the taxpayer and shown in Appendix VIII of said Report are not correct for the following reasons: 1.- In general and for all companies except for those listed in the following points, the STR data for the years 2010, 2009 and 2008 actually correspond to the STR data for the years 2009, 2008 and 2007, respectively, in the AMADEUS database. In the case of the companies VAMVAS J. CHR SA and CO-MED GMBH (both rejected by the Inspectorate for insufficient data), financial data are only available from 2008 onwards. 3.- In the case of the companies LADAKIS, ENDOTECH SPA and INTERMED (the latter rejected by the Inspectorate for not complying with the independence requirement) the ROS data in the table for the years 2010, 2009 and 2008 coincide with the ROS data for the years 2010, 2009 and 2008 in the AMADEUS database".

But the essential point is that: "Even if the time criterion applied by the taxpayer to compare the transfer prices of its transactions with bioMérieux SA and bioMérieux Inc was the a posteriori time criterion, which the OECD Guidelines refer to as the "arm's length verification criterion" in paragraph 3.70, in no case can data from years later than those in which the transaction took place, 2008, be used, nor could the taxpayer reasonably have had them at the time of the correct application of that approach. Therefore, the data for 2009 cannot be used to determine the arm's length range of the ROS/EBIT for the distribution activities carried out in 2008, which will be used to determine whether the linked purchase and sale prices enabled bioMérieux España to obtain a market return on those activities.

That is why, contrary to the reasoning on pp. 9 et seq. of the application, the Board understands, with the Inspectorate, that, for the purposes of obtaining adequate comparability, it is more correct to resort to "the data known by bioMérieux España SA in 2008, referring to comparables for 2006 and 2007, which could have been used to value the related-party transactions for that year in an a priori approach and the data for 2007 and 2008 in an a posteriori temporal approach", as they are more "in line with the OECD Transfer Pricing Guidelines" - p. 31 ID-. 31 ID-.

9.- The departure from this approach has consequences for the sample selection of comparable companies, as described in detail on pp. 25 et seq. of the ID.

Indeed, the Inspectorate on pp. 25 et seq. explains, inter alia, that all the companies selected are foreign and adds that it is reasonable that comparable Spanish companies should also have been sought. On p. 29 it describes the sample of 17 companies which the Commission considers to be comparable.

The appellant disputes on p. 12 et seq. the selection made by the Inspectorate. The appellant argues that the Commission excludes data from three companies because there is no data available for the years which the Commission considers should be taken into account for comparability. The appellant does not dispute the lack of data, but argues that they do exist for the years 2008, 2009 and 2010. But we have already explained that, in our opinion, the approach followed by the Inspectorate is more in line with the Guidelines and the ﬁnality sought by the transfer pricing rules, at least in the case in question - this argument is answered on p. 53 of the AL.

The Inspectorate also excludes certain companies on the basis that the indicator 'D' appears in the database, which indicates that it is a dependent company. It is logical to exclude dependent undertakings when trying to ﬁx prices in free competition, as dependence can distort prices. The appellant states that it has consulted the data in April 2013 and that the 'D' does not appear, but says nothing about the years compared.

As regards ENFAREL, the Inspectorate excludes it on the grounds that it is not comparable with the products distributed by bioMérieux - 'unrelated products: cosmetics, pharmaceuticals. It does not have its own website to allow cross-checking of information". On p. 54 of the AL it is explained that the information on this company has been obtained from the AMADEUS database, and it is listed as "wholesale trade of fish, seafood and other food products", as well as "wholesale trade of perfumery and cosmetics". The Inspectorate also searched for information on the internet and found that the entity also trades in baby food, cosmetics, creams and veterinary equipment, which is not a comparable activity.

On the contrary, the Inspectorate included the company AERONASIS SA, "as it fulfils all the requirements of comparability". The appellant maintains that this is not the case and, to this effect, provides as document no. 1 a brochure describing the activity of this company, but the fact is that the only date that appears in the same is 2017. If the appellant excluded it, it should be in a position to prove with concrete evidence the reason why it excluded the entity when it drew up the report it took into account to ﬁx its transfer prices, which it does not do. On p. 54 of the AL it is explained that this entity "from the beginning of its activities focused on diagnostics, specifically on equipment and reagents for in vitro diagnostics, distributing high quality systems with reliability and precision, and having developed strong relationships with hospitals and laboratories, which makes it a comparable entity to the taxpayer".

In this regard, it should be noted that the State Attorney's Office attached a report from the AEAT with the claim, which analyses documents 1, 2 and 3 attached with the claim and in which it is indicated and the Chamber shares the AEAT's criteria that these documents "extracts do not include data on the page accessed or the date on which they were obtained, and it may be that there is a time lag of up to ten years in relation to the financial year to which the assessment appealed refers".

The Inspectorate added four Spanish entities: FRANCISO SORIA MELGUIZO SA, CAJAL SA, HEXACAT ESPAÑA SL and SUMEDES XA. The appellant accepts that the first two are comparable (p. 16 of the application). However, it adds that, with regard to the net margin recorded by the Inspectorate in respect of FRANCISO SORIA MELGUIZO SA, it has consulted the AMADEUS base update in April 2013 and the data do not correspond to those indicated by the Inspectorate. Well, page 29 of the ID contains the data relating to the aforementioned company and on page 28 it is explained by the Inspectorate that it obtained the data from the AMADEUS database and in response to this assertion the appellant merely states that it consulted the database and the inspectorate made a mistake without articulating any evidence to that effect.

As for HEXACAT ESPAÑA SL and SUMEDES XA, the appellant maintains that they are not comparable and attached with its claim some extracts which, as we have indicated, were analysed by the report drawn up by the Inspectorate at the request of the State Attorney's Office and for the reasons that we have indicated are not sufficient evidence.

The appellant, with the written statement of conclusions, tries to attach new evidence to support its version of the facts, which is opposed by the State Attorney's Office. In our opinion, the Abogacía del Estado is right. Indeed, the provision of evidence after the evidence phase is regulated in art 270 in relation to 56.4 LJCA and 271 of the LEC. It is clear that the documents provided do not fall under art. 271 of the LEC; but it is also clear that they do not fall under art. 270 of the LEC.

It is worth starting by pointing out that as art. 270 of the LEC regulates an exception, the party who provides the document has the burden, at least, of explaining the concurrence of one of the reasons expressed in the rule. It does not do so, it simply limits itself to the provision. The foregoing would be sufficient to refuse admission, but, furthermore, it is not a document that could not have been obtained prior to the claim and neither are we in a situation in which, in the defence to the claim, for the first time, disagreement with the facts was expressed.

Finally, however, the evidence adduced does not disprove the selection made by the Inspectorate. Thus, on p. 27 of the ID it is stated that "the codes used to identify Spanish companies engaged in the distribution of medical equipment and their consumable products have been the same as those held by bioMérieux España, in order to achieve the best degree of comparability: Companies included in the NACE Rev. 2 codes (primary codes only): 4646 - Wholesale trade of pharmaceutical products, 4690 - Non-specialised wholesale trade of the National Classification of Economic Activities of the European Community (CNAE, NACE Rev. 1.1). It can be seen from bioMérieux España's AMADEUS report on file (element 501) that these are the industry codes relating to the activities of bioMérieux España distribución'. Replying on p. 55 of the AL to the appellant's allegations.

Pages 34 et seq. of the IA describe how the Inspectorate has obtained the market value of the net margin of all the operations of the distribution activity for the 2008 tax period.

The first thing the Inspectorate does on p. 34 is to describe the EBIT of each of the companies and the average, as well as the maximum, the upper quartile, the median, the lower quartile and the minimum.

Well, as explained on p. 35: "The EBIT margin data, in percentage terms, for the companies comparable to bioMérieux España SA indicate that the interquartile range for 2008 is 6.71% to 15.64% with a median of 9.01%. Averaging the data over two years to mitigate any short-term ﬂuctuations in the price of related-party transactions (which, moreover, has also been done by the taxpayer) gives an arm's length interquartile range for the years 2007-2008, the time range that the Inspectorate has indicated in the previous section as applicable, of between 5.71% and 18.26% with a median of 8.75%. The operating margin of 4.47% obtained by bioMérieux España SA in the 2008 tax period relating to distribution activities is outside the arm's length range, being below the lower quartile figure for both 2008 and 2007-2008, and the Inspectorate therefore considers that the prices agreed on the acquisition of instruments and consumables between bioMérieux España and its related entities, bioMérieux SA and bioMérieux Inc, have not provided bioMérieux España with a market return on this activity".

Paragraph 3.61 of the Guideline states that: "If the relevant terms of the controlled transaction (e.g. price or margin) are outside the arm's length range determined by the tax administration, the taxpayer should be given the opportunity to argue how the terms of the controlled transaction satisfy the arm's length principle, and whether the result falls within the arm's length range (i.e. the arm's length range is different from that determined by the tax administration). If the taxpayer is not able to demonstrate these facts, the tax administration must determine the point within the arm's length range to which to adjust the condition of the controlled transaction".

Adding 3.62 that: "In determining this point, where the range comprises very ﬁable and relatively equal results, it may be argued that any one of them satisfies the arm's length principle. Where some defects in comparability persist, it may be appropriate to use measures of central tendency that allow this point to be determined (e.g. median, mean or weighted mean, depending on the speciﬁc characteristics of the data) in order to minimise the risk of error caused by defects in comparability that persist but are not known or cannot be quantiﬁed".

Based on the provisions of point 3.62, the Inspectorate, which does apply "measures of central tendency", although not in the way intended by the appellant, states that "it considers that the market value of the net margin of all the operations of the distribution activity carried out by bioMérieux España in 2008 in relation to the products (instruments and consumables) that it acquires from its related entities bioMérieux SA and bioMérieux Inc is represented by the median of the distribution that includes the average values of the EBIT margin for the years 2007-2008. This value is 8.75%.

As we already reasoned in our SAN (2nd) of 6 March 2019 (Rec. 353/2015 ), it is legitimate to resort to what the Guideline calls "measures of central tendency", but whoever resorts to them has the burden of reasoning and setting out the reasons that lead to their application.

In our opinion, the Inspectorate, in this case, does reason and state the reasons. Thus:

a.- " the 2008 results are in the middle of an increasing phase from the first data noted which is that of 2005. Thus, the results of 2005 are lower than those of 2006 and these are lower than those of 2007. The data for 2008 (the year in which the controlled transactions were carried out) are slightly lower than those for 2007 in terms of the median and upper quartile and higher in the lower quartile. On the other hand, 2009 continues the upward trend in the median and upper quartile and the result for the lower quartile decreases. Therefore, we can summarise that the data for this sample have been increasing since 2005 for the lower quartile with a small decrease in 2009 and for the median and upper quartile values they are also increasing, experiencing a small decrease in 2008 (10%) to obtain higher margins in 2009 than in 2008 and 2007. This evolution of market margins does not correspond to that experienced by bioMérieux, which has gone from an EBIT of 8.03% in 2007 to 4.47% in 2008, and serves as a contrast to the validity of the market value of the net margin of all operations valued by the Inspectorate at 8.75% insofar as this does not respond to an anomalous market situation but is fully in line with the evolution of its margins over time".

b.- "Similarly, the same calculations have been made as in the previous table for the sample of comparable Spanish companies. This information, which is shown in the table below, contains the EBIT margins for the years 2005 to 2009 as well as the average EBIT in 2006-2007 and 2007-2008 together with their respective interquartile ranges [see p. 39 of the ID] The EBIT margin data, in percentage terms, for the companies comparable with bioMérieux España SA indicate that the interquartile range for 2008 is 10.41% to 16.13% with a median of 13.35%. Taking averages of two years' data to mitigate any short-term ﬂuctuations in the price of linked transactions gives an interquartile range for the years 2006-2007 between 8.61% and 15.39% with a median of 12.12% and an interquartile range for the years 2007-2008 between 9.45% and 16.36% with a median of 13.13%. If the comparison had been made exclusively with Spanish companies, the operating margin of 4.47% obtained by bioMérieux España SA in the 2008 tax period for the distribution activities in question would also have been outside the market range in both 2008 and 2007-2008, the medians being much higher than those of all the comparables used by the Inspectorate to determine the market value giving rise to the proposed adjustment".

c.- On pp. 37 and 38 of the ID, the Inspectorate refers to a table with data obtained from the AMADEUS database drawn up in relation to "the distribution companies of the bioMérieux group that have the same activity codes as bioMérieux España (4646 and 4690) and reflects the EBIT margin in percentage terms obtained by these companies from 2005 to 2008". The report also states that "there is no evidence of a generalised fall in EBIT between 2007 and 2008 in the distribution activities as a result of any event common to the group, such as the activities of the subsidiaries having been restructured to perform fewer functions and with lower risks in 2008 (which is certainly not the case in Spain), which would imply lower remuneration". It is noted that "there are companies that increase margins between 2007 and 2008, but it is insisted that what is relevant in the information is that it does not reflect a situation of generalised and relevant decrease in the margins of the distributors of the group, regardless of the individual margins in each year of each of these companies which, it is reiterated, are not relevant".

d.- Lastly, the ID, on pp. 39 et seq., highlights "the contradiction between the drastic decrease in the results of bioMérieux España's distribution activities and the bioMérieux group's growing profit rates, which can be seen when analysing the information supplied to third parties by the latter through the bioMérieux group's 2008 Annual Report (pdf file named 2008 RAPP ANNU 08-GB, element no. 369 of the electronic file)". From the data extracted from the documentation provided by the company itself, it is stated that "the Group's sales growth rates are consistent with that obtained by bioMérieux España according to the accounting information provided by the taxpayer, which is 8.16% in relation to distribution activities from products acquired from related entities. The fundamental reason for the growth in the Group's results lies in the rapid growth in the sale of reagents, which, as we have seen, is one of the main products distributed in Spain, and yet, while the group's results grew by 12%, those of Spain fell by 41.40%".

In other words, "2008 was a year of outstanding economic results for the bioMérieux Group, as well as for bioMérieux Spain in terms of sales growth, according to the report. However, this situation of increased results for the Group is not reflected in the income statement of bioMérieux Spain's distribution business, whose profitability fell from 8% in 2007 to 4.47% in 2008. This is not consistent either with the Group's results or with the market remuneration for performing the same functions in 2007 and 2008, a market which has not been shown to have seen its margins of free competition reduced.

e.- Finally, the Inspectorate also proceeded to "verify whether the decrease in operating/operating costs had also occurred in bioMérieux España, and analysed the accounting information supplied by the taxpayer to determine the items or concepts that have undergone the greatest increase between the tax period 2007 and 2008" - pp. 40 et seq. 40 et seq., reaching the conclusion that "the decrease in results in the distribution activities of bioMérieux España SA in 2008 is due to the increase in sales costs, i.e. the increase in transfer prices at which the taxpayer purchases products from its related entities bioMérieux SA and bioMérieux Inc". In the same sense, the decision of the TEAC at pp. 35 et seq, which the Chamber agrees with.

11.- The appellant provided a table with alternative calculations - pp. 56 et seq. of the Settlement Agreement - but as explained on p. 57 of the AL: "The comparison made with this table, and therefore the conclusions drawn, are erroneous for the following reasons:

1.- The samples contain STR data from different years. Thus, the sample of the Inspectorate has data for the years 2007 and 2008 for all entities, while the sample of the taxpayer has data for the years 2007, 2008, 2009 and 2010 and, in addition, with data for different years for the companies: some have information for the years 2007, 2008 and 2009; others have information for the years 2008, 2009 and 2010 and others only have it for two years, either 2008 and 2009 or 2009 and 2010. The heterogeneity of the information contained in this sample means that it lacks representativeness and ﬁability and makes it unviable for comparison with any other sample that does not contain the same errors.

Some of the companies selected in the taxpayer's sample are not comparable with the latter, as they do not meet the comparability criteria, as the taxpayer himself has acknowledged in the allegations presented, so that a sample of selected companies that do not meet these criteria cannot be compared with one that does meet them, that of the Inspectorate.

Therefore, the interquartile ranges of these two distributions do not allow the conclusions that the taxpayer claims in his allegations to be drawn. In addition to the above, the data which it says correspond to the tax inspectorate's sample are not those used by the inspectorate in the tax assessment, but those calculated by the taxpayer with weighted data".

Pages 29 et seq. of the TEAC's ruling insist on the reasoning of the tax inspectorate and the Chamber shares the criterion set out.

In any case, and in relation to the application of the arithmetic or simple average, the TEAC reasons on p. 31 and its arguments are not refuted in the claim that "in the present case, in the 2008 tax year, the interquantile range is from 6.71% to 15.64%, resulting in a median of 9.01%. The Inspectorate, "in order to mitigate possible short-term ﬂuctuations in the price of related transactions" does not take this median but the median obtained on the basis of the arithmetic average of the data for 2007 and 2008. The application of the arithmetic average of the data for 2007 and 2008 determines an average median for 2007 and 2008 of 8.75%, i.e. lower than that for 2008. As a result, the taxpayer has benefited from the percentage applied by the Inspectorate - 8.75% - compared to what would have been the result if the Inspectorate had simply taken into consideration the median for the year 208 - 9.01% - without averaging the data obtained in 2007 and 2008".

As indicated on p. 32 of the TEAC's decision, the appellant argues that applying a weighted average is more appropriate than using an arithmetic average of the data for the two years compared. But as the TEAC reasons "the interested party itself in its initial report acted in a similar way to the inspection, calculating the arithmetic average, now claiming the use of the weighted average, and this without justifying the reason for this change of criterion, without justifying the inappropriateness of the arithmetic average, without justifying how the referred weighting is carried out and, in short, without accrediting why the weighting based on sales that it now deﬁnes compared to its initial criterion determines a percentage closer to the market value than that defended by the inspection".

12.- The taxpayer provided a new sample. In other words, when requested by the Tax Inspectorate to explain how it obtained the transfer price, it provided documentation on how it made the calculation. As the Inspectorate refuted the calculation made in the terms indicated, the appellant provided another document, providing a new sample which, in its opinion, is more reliable.

This argument is discussed at pp.58 et seq. of the AL and is maintained again in the complaint.

In essence, what the complainant company argues in order to justify its change of approach is that the entity was not obliged to document its transfer pricing transactions. It did indeed commission a transfer pricing valuation report which it submitted to the inspectorate. But as the inspectorate has contested this, it now provides another report making what it describes as "a greater effort in the search for suitable comparables".

In our opinion this argument is not admissible.

In fact, apart from the transparency of the inspection, which was limited to requiring the justification of the transparency price set, accepting it and refuting it only in the parts that it did not consider suitable. It is true that, as stated in point 1.13 of the Guidelines, the objective sought by the rule is "to arrive at a reasonable approximation of what would be an arm's length result based on reliable information. At this point, it should also be remembered that transfer pricing is not an exact science, but requires value judgements on the part of both the tax administration and taxpayers".

Precisely for this reason, the correct thing to do is to proceed as the inspectorate did, i.e. to ask the appellant to justify the price set and to analyse the reasonableness of the price obtained. In this sense, it is reasonable to require the appellant to keep the information regarding the criteria they have used to set the transfer price and the documentation that has justified them or, at least, to be able to precisely identify the sources from which they have obtained the information. This will allow for veriﬁcation. In this sense, paragraph 3.3 of the OECD Guidelines "considers it good practice for a taxpayer that uses comparables to justify its transfer prices ( ) to provide the other interested party with the supporting information that allows it to assess the reliability of the comparables used".

Well, as stated on p. 59 of the AL, what the AL does is to provide the other party with the information necessary to assess the comparability of the comparables used. 59 of the AL, what the appellant does "is not to make a greater or better documentary effort to prove that the transfer prices of its related transactions in 2008 complied with the arm's length principle, which, on the other hand, it could have done in the course of the verification proceedings if it considered that the documentation initially provided was insufficient or had not been adequately produced, but rather, in view of the proposal for regularisation of the IS 2008 made by the Inspectorate, it has submitted another sample of comparable companies, different from the one submitted during the verification proceedings, aimed at maintaining that the prices applied in 2008 were market prices and drawn up by the same company that carried out the first study, which it now seems to consider that it did not make a sufficient effort to prove what it was seeking".

To which it should be added - p. 60 of the LA - that "the initial study presented in the verification proceedings is more extensive than the one presented now (as can be seen from the summary of its contents in section V.1 of the non-conformity report), which is limited to a database search. Nor has the result of this search been a notable improvement on the selection already available, since of the companies selected (four) only two are comparable with the taxpayer and these had already been selected by the Inspectorate, one of them, Francisco Soria Melguizo, being fully known to the taxpayer prior to 2008". The AL criticises the report on pp. 61 et seq. which the Chamber agrees with and which are not refuted in the complaint.

In the same sense, the reasoning contained on pp. 32 et seq. of the TEAC's resolution.

All these reasons, assessed as a whole, lead us to conclude that the detailed analysis carried out by the Inspectorate allows us to conclude that the calculations made by the Inspectorate are closer to the purpose of the rule, that is to say, to the search for the price set at arm's length, than those provided by the appellant.

The plea is rejected.

THIRD - Application to the transactions between BM SPAIN and BM USA of those agreed in the amicable procedure between the Spanish and French tax authorities in respect of the transactions between BM SPAIN and BM FRANCE.

The applicant submits that the Spanish authorities have reached an amicable agreement with the French authorities and have ﬁxed the agreed mark-up as market rate at 6,20 %. What is sought is to apply the same margin in relation to the US company, in respect of which there is no amicable procedure.

The Abogacía del Estado opposes this argument, reasoning that the transfer price agreed with France in an amicable procedure is the result of a negotiation between sovereign entities involving considerations of international public law, and therefore its results cannot be extrapolated.

Essentially, there are three ways of resolving disputes between States: diplomatic channels, international arbitration or the decisions of international tribunals.

The amicable procedure in tax matters, regardless of its special features (e.g. the specialisation of the negotiating body, the speciﬁc regulation of the procedure, etc.), constitutes a channel of negotiation between two States which, by definition, must obey criteria of ﬂexibility and fairness. If States are not granted a reasonable margin for negotiation, the effectiveness of a dispute settlement instrument may be substantially diminished.

The amicable procedure obliges states - as it is required by international agreements - to initiate a negotiation process governed by the principle of good faith, but not to reach an agreement. The state therefore loses its absolute freedom of behaviour as a sovereign entity and is obliged to initiate a negotiation process under the terms established in the norm. However, although States have certain limits when negotiating, it is no less true that, during the negotiation process, different factors of opportunity come into play, which, in our opinion, means that the result obtained in the negotiation with one country cannot be extrapolated. The agreement obtained is an agreement that binds the negotiating States, but cannot extend its effects to relations with another State. The fact that the Kingdom of Spain, for reasons unknown to us, has reached an agreement with the Republic of France does not mean that the transfer price ﬁxed by the Spanish administration is not correct, but simply that the States have given in on their respective claims and reached an agreement, the effects of which cannot be extrapolated.

The appellant could have resorted to the amicable procedure with the United States and did not do so, and what it cannot do is claim that an agreement negotiated with the French authorities, the result of a negotiating process in which various interests had to be taken into account, should be applicable to it.

The plea is dismissed. FOURTH - Costs.

The applicant must be ordered to pay the costs - Article 139 of the LJCA.

Having regard to the aforementioned legal precepts and other relevant and generally applicable provisions, the Court gives the following judgment

WE DECIDE

Dismiss the contentious-administrative appeal brought by the Attorney Isabel Covadonga Julia Corujo, in the name and on behalf of BIOMERIEUX ESPAÑA SA, against the Resolution of the Central Economic-Administrative Tribunal of 9 May 2017 (RG 4746/13); which we conﬁrm as being in accordance with the Law. The applicant is ordered to pay the costs.

Enter the judgment in the book of its class and once ﬁrmed, send a copy of the same together with the administrative file to the place of origin of the same.

The present judgment is subject to appeal in cassation which must be prepared before this Chamber within 30 days from the day following that of its notification; in the document preparing the appeal, compliance with the requirements established in article 89.2 of the Law on Jurisdiction must be accredited, justifying the objective appeal interest it presents.

PUBLICATION: The previous judgment has been read and published by the Illustrious Judge Mr. Manuel Fernandez-Lomana García, the Chamber being in public hearing on the same day of its date, of which, as Legal Secretary of the Administration of Justice, I certify.