Transfer pricing methodology is the set of methods used to test whether a price charged between companies under common control matches what unrelated parties would have agreed to. Some methods compare the price itself with prices in comparable uncontrolled transactions; some compare the gross margin a reseller or a manufacturer earned with the margins of comparable independent firms; some compare the operating profit of the simpler party with the profits of comparable companies; and the profit split divides the combined profit by each party's contribution. The best method rule selects the method that gives the most reliable measure of an arm's length result on the facts, and where several comparables are equally reliable the result is a range, narrowed to the interquartile range when the comparables are inexact.
The methodology applies wherever an intercompany price is in question: an IRS examination or a transfer pricing dispute in the Tax Court, a state audit of related-party royalties and fees, a mutual agreement proceeding between two tax authorities, and commercial litigation in which a price between affiliates moved profit away from a minority owner, a joint venture partner, a buyer, or a spouse, as in a shareholder dispute or a divorce involving a business with foreign affiliates. The regulations specify the methods by the kind of transaction: for tangible goods, the comparable uncontrolled price, resale price, cost plus, comparable profits, and profit split methods; for intangibles, the comparable uncontrolled transaction, comparable profits, and profit split methods, applied to royalties as the intercompany royalty guide describes; and for services, a parallel set that includes a services cost method under which certain routine support services may be charged at cost. Once the arm's length price is established, the restated profit feeds a business valuation or a commercial damages calculation.
Every method rests on comparables that differ from the controlled transaction in some respect, and the reliability of the result falls as the differences, and the adjustments needed to bridge them, grow. The comparable profits method is the one the IRS's own advance pricing agreement reports show applied most often, because data on independent public companies are available, but it tests the profit of a whole tested party rather than the price of a specific transaction, so a tested party whose results were driven by a downturn, a start-up phase, or a product failure can look mispriced when it is not, and a mispriced one can look reasonable in a good year. The comparable set is only as good as its screens, and a set assembled to place the tested party within the range proves nothing. Ranges can be wide, and the point within the range that a dispute turns on is a question of method and law as much as of data. The profit split depends on a measure of each party's relative contribution that is rarely observable directly, and capitalized spending used as a proxy for the value of intangibles carries its own assumptions about useful lives. Finally, the regulations and the OECD Guidelines are tax rules: outside a tax forum an arm's length result is evidence of what unrelated parties would have done, not the measure a contract or a fiduciary duty prescribes, as the transfer pricing disputes guide explains.
The methods are prescribed by the Treasury regulations and described in the OECD Guidelines, so a challenge seldom attacks a method as such; it attacks the choice of method under the best method rule, the tested party, the comparables, the adjustments, or the point in the range. In the Tax Court, transfer pricing cases are tried to a judge, and each expert's written report is received as that expert's direct testimony, so the report itself is the evidence the court weighs, and the court may adopt either side's analysis, adjust it, or reach its own result from the evidence. In a district court, a commercial dispute that relies on a transfer pricing analysis meets the ordinary reliability standard for expert testimony, and the questions are the familiar ones: whether the method fits the transaction, whether the comparables are comparable, and whether the method was reliably applied to the facts of the case. A report that states the methods rejected and why, shows every screen, and presents the result under the opposing expert's tested party or comparable set is positioned to be examined on its inputs rather than excluded, in a New York commercial contract dispute as in any venue. The documentation versus expert report comparison explains why the tax documentation and the litigation report are judged differently, and the rebuttal guide lists the questions an opposing expert asks.
The requirement in the Treasury regulations that the arm's length result be determined under whichever method gives the most reliable measure on the facts. There is no fixed order of methods; the two primary factors are how comparable the uncontrolled transactions are and the quality of the data and assumptions. A method may be applied without first showing that the others are inapplicable, but a method later shown to be more reliable must be used. Describing the alternatives considered and why they were rejected is a requirement of the penalty documentation, and a litigation report that does the same shows the tribunal why the method chosen is the more reliable one.
When the comparables are not close enough for every material difference to be identified and adjusted, the regulations call for a statistical method to make the range more reliable, and the interquartile range, the middle half of the comparables' results, ordinarily serves. For a federal tax adjustment, a result outside it is ordinarily moved to the median of the comparables.
It is the participant whose operating profit from the controlled transactions can be verified with the most reliable data and the fewest adjustments, and for which reliable comparables can be found. In most cases that is the least complex company, one that owns no valuable intangibles or unique assets; choosing a party that does makes the comparables unreliable.
Request a consultation on Transfer Pricing Methodology or call (201) 343-0700. Plaintiff and defense counsel.