In this edition: U.S. tax court rules in favor of Amazon, U.S. tax reform and the effects on transfer pricing, OECD releases toolkit for identifying financial data in developing countries, UK releases guidance on cash pooling, Australia releases risk framework for centralized operating models and New Zealand releases consultation documents addressing potential transfer pricing changes.
U.S. Tax Court Rules in Favor of Amazon in Latest Transfer Pricing Case
On March 23, 2017, the United States Tax Court issued its long-awaited opinion in the Amazon transfer pricing case (Amazon.com, Inc. v. Commissioner, 148 T.C. No. 8 (2017)). This case focused on buy-in and cost-sharing transaction payments made by a Luxembourg affiliate associated with Amazon's cost sharing arrangement covering its European operations in 2005 and 2006. Given the years at issue, this case was decided under the "old" cost sharing regulations (1.482-7A) which were substantially revised in 2009. This summary focuses only on determinations associated with the buy-in payment. The IRS had determined that the total buy-in payment should have totaled $3.5 billion, a large multiple of the $245 million total buy-in payment determined by Amazon.
With respect to the buy-in payment, the core issues in this case are very similar to those in the Veritas case (see Veritas Software Corp. v. Commissioner, 133 T.C. 297 (2009)), and many viewed this case as the IRS' attempt to re-litigate the findings in the Veritas case. In Veritas, the taxpayer had determined a buy-in payment based on the application of the Comparable Uncontrolled Transaction ("CUT") method using internal transactions, in combination with a royalty decay mechanism meant to reflect the shift in intangible ownership that occurred as a result of the cost sharing payments that were expected to be made by Veritas' Irish subsidiary. The IRS, on the other hand, had determined a buy-in payment based on a variant of the income method that became embedded in the cost sharing regulations in 2009. This method did not rely on CUTs to determine income attributable to the pre-existing intangibles, but rather identified that value based on discounted projected profits over an infinite period.
The Amazon case involved three buckets of intangibles (technology, trademarks and associated intangibles, and customer-based intangibles). The taxpayer's experts at Amazon had used CUT approaches with declining royalties for each of these buckets (and had used internal agreements for the technology and customer-based intangibles). The IRS' primary expert had used a discounted cash flow ("DCF") method to determine the value of the aggregate bundle of intangibles. This DCF approach shared many similar features to that employed by the IRS in Veritas. The similarities included an embedded assumption that the pre-existing intangibles had a perpetual life.
Amazon presented a great deal of information demonstrating that Amazon's European business was highly uncertain at the time of the buy-in and required constant innovation. Amazon also argued that its European affiliates made substantial contributions to the European business, and that longer-term projections included profits attributable to future (not pre-existing) intangibles. As a result, Amazon argued the approach taken by the IRS did not identify the value of pre-existing intangibles, as required by the regulations in place at the time, but rather treated the buy-in as being "akin to the sale" of a business - a line of argument that was successful in Veritas.
In the Amazon opinion, Judge Lauber writes, "One does not need a Ph.D. in economics to appreciate the essential similarity (between the approach applied by the Commission in Veritas vs. that applied in the Amazon case)." The court found, like it did in Veritas, that the IRS' approach was fatally flawed because it treated the...