On July 27, 2015, the U.S. Tax Court issued a stunning rebuke to the IRS by invalidating the part of the Internal Revenue Services’ (IRS) cost-sharing regulations under code section 482 that says taxpayers have to take into account, among other costs, the costs of stock-based compensation. The Altera decision should also support the many taxpayers who have questioned the separate section 482 regulatory requirement that cost-based transfer pricing (e.g., cost-plus pricing for services) must include the cost of stock-based compensation.

Taxpayers with cost-sharing agreements or other transfer pricing that previously took into account such stock-based compensation costs should consider amending their returns and filing claims for refunds for open years. Taxpayers who followed the approach sustained in Altera may want to review their tax provisions.

The opinion, reviewed by the full Tax Court, is important for transfer pricing beyond cost sharing and stock-based compensation costs. The Tax Court held that, as an interpretation of the “arm’s length standard,” the regulation is invalid because there is no empirical basis for the proposition that third parties share such costs. The opinion concludes that the regulation “lacks a basis in fact…” 

In other words, according to the full Tax Court, the IRS cannot issue a valid transfer pricing regulation under the arm’s length standard that attempts to interpret how related parties “should behave” if there is no empirical evidence that unrelated parties actually behave in that fashion. (In Altera there was ample evidence that unrelated parties did not charge each other for the costs of stock-based compensation.) Even though the words “arm’s length standard” do not appear in section 482, U.S. courts and the U.S. Treasury’s comments on treaties (noted in Altera) have long said that the statute incorporates this standard, thereby limiting the IRS’ ability to reallocate among affiliates.

The IRS argued that Altera cannot object that the regulations are inconsistent with the arm’s length standard because they are part of an “elective” regulatory regime for cost sharing. The Tax Court, however, dismissed this IRS attempt to recharacterize the relevant provisions as “elective,” noting that in the regulatory process the IRS rejected taxpayer suggestions to make the requirement a true safe harbor. There are other parts of the pricing regulations that taxpayers can truly elect and that Altera should not disturb. 

The Tax Court decision may comfort those taxpayers already challenging in court the validity of other provisions of the section 482 regulations that “overreach” (e.g., 3M’s pending challenge to the “foreign legal requirements” provision in the regulations).

The holding has relevance to ongoing debates about whether certain Base Erosion and Profit Shifting (BEPS) proposals, such as those asserting that “control” over certain activities is a prerequisite for a related party to claim intangibles profits (or losses), could be adopted administratively in the United States. Altera lends support to those who believe the IRS and Treasury could not adopt these proposals as valid regulatory interpretations of the arm’s length standard of section 482; instead adoption could only occur through a statutory amendment by Congress.

If the government does not acquiesce in Altera, and does not remove the stock-based compensation provisions from the section 482 regulations, then taxpayers will have choices. Those who benefit from including the costs of stock-based compensation in the contexts of cost sharing and transfer pricing, such as a foreign company with significant stock-based compensation costs that provides services or goods to a U.S. affiliate, can continue to rely on the regulations. Those taxpayers who, like Altera, do not benefit from including such costs now have a significant case upon which to rely.

Unfortunately, it would not be surprising if the government appeals Altera even though sound tax policy and administrative considerations would argue that it should accept the decision and move on to other issues. However, given the record on this issue, the government will likely have an uphill battle should it appeal.

In conclusion, one might speculate whether Altera could mark the beginning of the end of the last 20 years of frequent theory-based (not empirical-based) approaches in regulations as to how related parties should behave with respect to their transfer pricing. Before the 1994 regulations, and after, the IRS has lost cases in court that were based on theories IRS developed in the litigation (essentially all the cases it has litigated). It became apparent to the IRS that facts, not theories, convince judges (at least under the current statute). The IRS has stated that it is focusing on improving its track record in court in part by doing a better job of mastering the facts in litigating a case. Will Altera cause IRS and Treasury to reexamine and revise the theory-based parts of the pricing regulations from a similar perspective? Only time will tell.