The current transfer pricing rules for cost sharing arrangements (CSAs) call for certain "buy-ins" (now called PCTs) when one party makes intangible property available for certain research and development purposes. Recently, however, various transactions have occurred in the "real world" in which one party makes technology available in commercial transactions for little or no consideration. This certainly creates (at least in this author's mind) some thoughts worth considering, many of which could present significant opportunities in terms of tax planning, audit defense and compliance.

Background: U.S. style cost sharing

Of all transfer pricing topics, cost sharing is the most controversial. From a tax perspective, it largely is a U.S. invention with its roots in the 1986 Tax Reform Act. Congress noted in the Act's legislative history that cost-sharing is a legitimate device by which to divide profits from exploiting intangible property. Congress, however, failed to specify exactly what cost sharing is or how it works.

The IRS and Treasury discussed and defined the concept in 1988 in the so-called "White Paper," which was a study of intercompany transfer pricing. There, the IRS and Treasury described in general terms a transaction in which two or more participants could share the costs of developing intangible property. The participants thereby acquire proportionate ownership interests in the intangible property that results from their cost-shared research and development (R&D) activities. Basically, each participant can exploit their proportionate interests — free from royalties to the other participant — for the simple reason that a party need not pay royalties on intangible property that the party owns.

The first set of comprehensive cost-sharing rules appeared in 1995: Treasury Regulations Section 1.482-7. The key provision is the so-called "buy-in" requirement of Section 1.482-7(g). This section mandates that when one participant "makes available" its intangible property "for purposes of research," the other participant(s) must pay for the value of that contribution to the research effort.

The concept is as follows. Suppose that Smith owns Version One of a Wheel. Smith and Jones decide to collaborate on the R&D for Version Two. Not wishing to reinvent Wheel One, Smith makes Wheel One technology (T) available for purposes of joint research. Section 1.482-7 mandates that Jones must pay Smith for the value of T for "purposes of research," as measured, for example, by the costs that Smith and Jones will save by not having to reinvent the Wheel.

Although sound in concept, placing a value on T proves difficult and controversial in practice. There are many reasons for this, but at the 30,000-foot level, the basic problem centers on the difficulty of applying the arm's-length standard to so-called CSAs. That's because it is very, very difficult to find reliable information (contracts, financial results, terms and conditions, etc.) about real-world CSA transactions. In fact, many people — and perhaps the majority — acknowledge that CSAs, at least as defined in Section 1-482-7, do not exist in the real world.

Xilinx and the arm's-length standard

This issue came to a head in the now-famous, or rather infamous, case of Xilinx v. Commissioner. There, although the parties settled the buy-in issue before trial, they battled over the cost base, specifically, whether employee stock options are a cost and if so, whether the CSA participants must share that cost. Xilinx is controversial and notable for many reasons, but perhaps most significant because the parties agreed that there are no true "comparables" for CSA transactions. However, the company contended that evidence from non-comparable transactions was sufficient, because "something is better than nothing." Ultimately, the taxpayer won the case after appeals, but whether the taxpayer's contention (and evidence) prevailed is still subject to debate among academics and practitioners.

The point here is not to debate the merits or conclusions of one specific case (Xilinx). The point is simply that the common wisdom is that CSAs, as defined in the Treasury Regulations, simply do not exist in the real world. Thus, how can the participants comply with the arm's-length standard if, in fact, the transactions do not have real-world counterparts?

New CSA regulations

From 2005 to 2008, the IRS and Treasury rewrote the 1995 regulations to provide specific transfer pricing methods by which to price and value the transactions involved in CSAs — at least, CSAs as specifically defined in the regulations. Because those regulations continue to adhere to the arm's-length standard, taxpayers and the IRS must seek data of "comparable" real-world transactions and information by which to value "T" (see above), all the while recognizing that there aren't really any true comparables for CSA transactions.

So what, you may ask, does all of this have to do with Chinese trains?

Chinese trains

On Thursday, Nov. 18, 2010, The Wall Street Journal ran a front page feature on Chinese trains: "Train Makers Rail Against High-Speed Designs." The article describes how Chinese train makers have produced high-speed trains that in many cases are faster than those produced by non-Chinese manufacturers. The article also describes how the Chinese manufacturers do so: mainly by taking, modifying, re-engineering or building upon pre-existing high-speed train "T," which in many cases was made available to them free of cost or otherwise for set fees. According to Wang Xinhong, a senior engineer for China South Locomotive & Rolling Stock Industry (Group) Corp., "We attained our achievements in high-speed train technology by standing on the shoulders of past pioneers."

The article describes several different business arrangements in which various "T's" have been made available or even pirated or reverse engineered:

  • "As China increasingly favors domestic suppliers, it's able to up the ante further, demanding that companies who want to do business transfer over more advanced technologies."
  • "Any company bringing new technology, innovation or ideas to China has to deal with shanzhai, what one could readily refer to as bandit culture ... From cell phones to automobiles, Chinese companies have taken pride in using others' intellectual property and either innovating or counterfeiting goods."
  • "Post-war Japan pulled off its transformation in part by reverse engineering foreign technologies, eventually developing a stable of tech companies, steel producers, shipbuilders and auto makers, including Honda and Toyota. South Korea followed a similar path."

Why would a non-Chinese manufacturer make its train T (or other T) available in this connection? According to the article, "What's unique about China is its vast domestic market, which makes foreign companies willing to hand over their technology know-how for a piece of the action." Whatever the case — and putting aside bandit and piracy issues — one thing is clear. Independent parties do, in fact, willingly make their intangible property available for R&D purposes, for no or minimal cost. Of course, others apparently have rejected the T-for-free concept. As stated in the article: (1) some find themselves competing against their own technology; and (2) "Some executives question the wisdom of dealing with China. 'We didn't take part in the export project to China,' says Yoshiuki Kasai, chairman of Central Japan Railway Co. 'The conditions were not favorable — they wanted all the technology to be transferred for free. That was not good for us.'"

Putting two and two together

Those familiar with the CSA area know that the IRS typical asserts materially high values for buy-in transactions involving T. Yet, as noted above in connection with Xilinx, despite the fact that both the IRS and the taxpayer agreed that no true comparable existed, the taxpayer contended that "something is better than nothing." If it indeed is true that some evidence of real-world comparables is better than no evidence — and if the real world tells us that transactions exist in which T is made available for nothing — then absent any concrete evidence to refute "nothing," why isn't "nothing" the right answer?

Given the history of transfer pricing, the answer usually can be predicted: "It depends on the facts." All transfer pricing cases are inherently factual, however, so that answer, although predictable, is not entirely satisfying. Further, one can predict with reasonable (perhaps absolute) certainty that the IRS would respond that the real-world Chinese transactions simply are not comparable or reliable. But that predictable answer also is wholly unsatisfying, if in fact "something is better than nothing."

So, what is the answer? Can Chinese or other commercial transactions be applied to your situation either in formal CSAs or in other intangible property transfer/licensing situations? That is a complex question. The real answer lies in performing a robust analysis of the CSA transactions with due consideration to the market realities of real-world commercial transactions. And that is where your Grant Thornton transfer pricing team — some of whom participated in developing the current CSA rules and one of whom (David Bowen) personally litigated the Xilinx case — can and will assist.

Tax professional standards statement

This document supports Grant Thornton LLP's marketing of professional services, and is not written tax advice directed at the particular facts and circumstances of any person. If you are interested in the subject of this document we encourage you to contact us or an independent tax advisor to discuss the potential application to your particular situation. Nothing herein shall be construed as imposing a limitation on any person from disclosing the tax treatment or tax structure of any matter addressed herein. To the extent this document may be considered to contain written tax advice, any written advice contained in, forwarded with, or attached to this document is not intended by Grant Thornton to be used, and cannot be used, by any person for the purpose of avoiding penalties that may be imposed under the Internal Revenue Code.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.