Transfer pricing: it is not correct to adjust results to the median when there are no comparability defects

In the examined case, the audited company, engaged in wholesale distribution in Spain, had obtained in the fiscal year results below the interquartile range from the sample included in its transfer pricing documentation. In the following fiscal year (also reviewed in the same audit) the results fell within that range, however, but were below the median. The auditors had adjusted the results for both years to the median in the range. The Central Economic-Administrative Tribunal (TEAC) accepted the adjustment for the first year, but held that the adjustment to the median for the second year was not correct because the result for that year was within the range. The taxpayer submitted to the National Appellate Court (i) that its results in a multiyear period were within the interquartile range and (ii) an adjustment does not need to be made to the median but to the lower quartile in the range in the year when the result was below the range.

Basing itself on the OECD Transfer Pricing Guidelines, the tribunal took the view that a multiyear analysis may be valid to make the comparability study, but comparison with the taxpayer’s results must be done with regard to each year reviewed. However, in relation to the value that must be used in the value adjustment to the result, it accepted the company’s argument and concluded (also in line with the Guidelines) that to use the median there needs to be “comparability defects” in the taxable person's sample. Because TEAC concluded there were no comparability defects in the company's sample, the adjustment for the first year should have been made to the lowest point in the range of arm’s length values (lower quartile).