“Gas-Trader KtF” – a subsidiary in the E.ON group – had entered into loan agreements with other group companies and the related parties had determined the interest rate by application of the CUP method using the Thomson Reuters LoanConnector database. Comparable transactions was extracted from the database by searching for credit rating, type of debtor party, date of loan, maturity, transactions with completed status, and spread/provision fee. An audit was conducted by the tax authorities for FY 2012-2013 and the interest rate determined by the group was found to be incompliant with the arm’s length principle. The tax authorities applied the same method as Gas-Trader but added further search criteria in the selection of comparable transactions – credit purpose and insurance coverage. This resulted in a different range and an assessment of additional taxable income was issued. An appeal was filed by Gas-Trader KtF with the National Tax and Customs Board of Appeal where a judgement in favor of the tax authorities was issued. Then an appeal was then filed with the courts where the decision was annulled and the Board of Appeal ordered to initiate new proceedings. During these proceedings, an expert opinion was obtained which was in favor of Gas-Trader. However following objections from the tax authorities, the Board of Appeal dismissed the expert opinion and decided predominantly in favor of the tax authorities. An appeal was then filed with Supreme Administrative Court. Judgement of the Court The Supreme Administrative Court set aside the decision and issued a judgement in favor of Gas-Trader. In its judgment, the Court states “[26] A large amount of data is needed to determine the transfer price. However, the available information may be incomplete, difficult to interpret, difficult to obtain for reasons of confidentiality, or may not exist at all, or the relevant independent enterprise itself may be missing. In the practical application of the arm’s length principle, the objective is always to determine an acceptable estimate of arm’s length profit based on reliable data. The ‘estimated’ nature of the transfer price means that it is never an exact tax act, but requires both the taxpayer and the tax authorities to subsequently take evidence that is clearly identifiable and realistic. The Curia has stated in its judgment Kfv.I.35.550/2018/12 that the question of transfer pricing can be a technical question or a purely legal question depending on the underlying facts. In the case at hand, the defendant transformed the decision into a question of law by basing its decision not on an examination of the transfer pricing method, but on a different classification of the underlying legal relationship from that of the plaintiff in that case. In the present case, the Curia adds that, in the event of a substantive examination of the transfer pricing method by the tax authorities, the applicant may also submit a request for evidence on a technical point in the course of the judicial review. [27] Among the methods of transfer pricing, both the Directives and the Tao [Corporate Tax and Dividend Tax] Law recognise the method of comparative independent prices. Pursuant to Section 18(2)(a) of the Tao Law, the arm’s length price is to be determined by one of the following methods: the arm’s length price method, whereby the arm’s length price is the price that independent parties would apply for the sale of a comparable asset or service in an economically comparable market. The problem in applying this method is the identification of the ‘economically comparable market’, which is ultimately achieved by applying the correction/constraint criteria within the scope of the method. Indeed, an independent transaction can only be compared with a controlled transaction using the method of comparable independent prices if one of the following two conditions is met: (a) none of the differences, if any, between the transactions to be compared or between the undertakings entering into those transactions can materially affect the free market price, or (b) relatively accurate adjustments can be made to eliminate the distortive effect of such differences. Therefore, where distorting differences exist between controlled and unrelated transactions, adjustments should be made to at least broadly eliminate price influencing factors and enhance comparability. Each of the narrowing methods should be assessed for their relative accuracy and only those adjustments should be made that are likely to improve comparability. [28] In the case at bar, it is a fact that the defendant did not make a finding with respect to the plaintiff’s records that the plaintiff had developed what it considered to be an appropriate transfer price, that the defendant agreed to the use of the comparative independent pricing method. The court’s remedy resulted from a difference in the criteria considered by the parties to eliminate the distorting effect of the differences. It may also be noted that in the decision of the Court of Appeal ordering a new trial in the main proceedings, the defendant excluded the application of the interquartile range correction criterion of narrowing the range around the midpoint. [29] Defendant used the data extracted by the plaintiff from the LoanConnector database to verify the transfer pricing. In its procedure, it considered the relevant aspects of the Directives to be relevant, according to the review request: in the expected audit practice, tax auditors should be flexible in their approach, take into account the business considerations of taxpayers and start their analysis from the perspective of the pricing method chosen by the taxpayer. If the taxpayer’s screening strategy is reproducible and the screening steps are suitable to produce a suitable sample for the transaction under consideration, the tax administration will use the taxpayer’s database screening as a basis. If, for any reason, the tax administration disputes the screening steps, it will attempt to make the necessary adjustments based on the taxpayer’s research to ensure that the results calculated from the improved sample are consistent with the market price principle. As a starting point, the tax administration does not therefore seek to determine the price or range of prices applicable to ...

Selection of comparables. (i) To evaluate the arm’s length result of a controlled transaction between USSub, the United States taxpayer under review, and FP, its foreign parent, the district director considers applying the resale price method. The district director identifies ten potential uncontrolled transactions. The distributors in all ten uncontrolled transactions purchase and resell similar products and perform similar functions to those of USSub. (ii) Data with respect to three of the uncontrolled transactions is very limited, and although some material differences can be identified and adjusted for, the level of comparability of these three uncontrolled comparables is significantly lower than that of the other seven. Further, of those seven, adjustments for the identified material differences can be reliably made for only four of the uncontrolled transactions. Therefore, pursuant to § 1.482-1(e)(2)(ii) only these four uncontrolled comparables may be used to establish an arm’s length range ...

For purposes of this section, the interquartile range is the range from the 25th to the 75th percentile of the results derived from the uncontrolled comparables. For this purpose, the 25th percentile is the lowest result derived from an uncontrolled comparable such that at least 25 percent of the results are at or below the value of that result. However, if exactly 25 percent of the results are at or below a result, then the 25th percentile is equal to the average of that result and the next higher result derived from the uncontrolled comparables. The 75th percentile is determined analogously ...

If there are no uncontrolled comparables described in paragraph (e)(2)(iii)(A) of this section, the arm’s length range is derived from the results of all the uncontrolled comparables, selected pursuant to paragraph (e)(2)(ii) of this section, that achieve a similar level of comparability and reliability. In such cases the reliability of the analysis must be increased, where it is possible to do so, by adjusting the range through application of a valid statistical method to the results of all of the uncontrolled comparables so selected. The reliability of the analysis is increased when statistical methods are used to establish a range of results in which the limits of the range will be determined such that there is a 75 percent probability of a result falling above the lower end of the range and a 75 percent probability of a result falling below the upper end of the range. The interquartile range ordinarily provides an acceptable measure of this range; however a different statistical method may be applied if it provides a more reliable measure ...

The arm’s length range will consist of the results of all of the uncontrolled comparables that meet the following conditions: the information on the controlled transaction and the uncontrolled comparables is sufficiently complete that it is likely that all material differences have been identified, each such difference has a definite and reasonably ascertainable effect on price or profit, and an adjustment is made to eliminate the effect of each such difference ...

...

Uncontrolled comparables must be selected based upon the comparability criteria relevant to the method applied and must be sufficiently similar to the controlled transaction that they provide a reliable measure of an arm’s length result. If material differences exist between the controlled and uncontrolled transactions, adjustments must be made to the results of the uncontrolled transaction if the effect of such differences on price or profits can be ascertained with sufficient accuracy to improve the reliability of the results. See § 1.482-1(d)(2) (Standard of comparability). The arm’s length range will be derived only from those uncontrolled comparables that have, or through adjustments can be brought to, a similar level of comparability and reliability, and uncontrolled comparables that have a significantly lower level of comparability and reliability will not be used in establishing the arm’s length range ...

A Korean subsidiary (“Semicon-sales”) of a foreign group was active in distribution and sales of semiconductors for the automotive and industrial industry. Following an audit, the tax authorities found that the subsidiary had purchased semiconductors from a foreign affiliated company at a higher price than the arm’s length price. An assessment was issued where the the sum of the difference between the arm’s length price and the reported price had been included in the taxable income for FY 2015-2018. Both “Semicon-sales” and the tax authorities had applied the TNMM to find the arm’s length price, but the tax authorities had rejected the comparables selected by “Semicon” and replaced them with others. Not satisfied with the assessment “Semicon-sales” filed an appeal. Judgement of the Court The court remanded the case with an order to exclude from the benchmark comparables where the sales volume is significantly different from that of the “Semicon-sales”. Since the proportion of the taxpayers transactions with large companies is significant, the transaction stage, sales volume, customer, business environment should also be taken into consideration. Click here for English translation Click here for other translation ...