Michelle Phillips just won an A… or at least a nice note to her professor. Michelle’s entry on transfer pricing is one of two winning entries in my legal writing contest.
Michelle is currently an LL.M. Taxation candidate at Boston University. Her transfer pricing tax prof is Richard Ainsworth.
As you may know from yesterday, I received a number of great entries and I couldn’t pick just one so I opted for two. I was particularly impressed by Michelle’s entry because she made tricky subject matter understandable and interesting – most practicing attorneys can’t talk about transfer pricing in such a smart, well thought out manner. Here’s what Michelle had to say:
The current transfer pricing rules and regulations put huge power in the hands of the IRS in much the same way as those ubiquitous, one-sided boilerplate form contracts. However, unlike with boilerplate, the IRS has no real market pressure or reputational concerns keeping it in check, and thus the self-appointed powers are inappropriate. First, I’ll walk through what the IRS can do, then I’ll explain some of the theory on how form contracts, although lop-sided, are actually useful. At the end, I’ll show how, if we liken the Tax Code and Regs to a contract between taxpayer and government, this theory on form contracts demonstrates how inappropriate the IRS’s position is.
Transfer pricing, in case you’re not familiar, deals with situations in which two related parties make a deal — such as a parent company selling or licensing some intangibles to its subsidiary. If left to their own devices, these parties (“controlled parties”, in the vernacular of the regs) would probably strike a deal totally different from what independent parties (“uncontrolled parties”) would negotiate. A lot of the non-market price manipulations would be based around tax consequences and finding ways to have less income in high-tax jurisdictions. For example, if a US parent company (“USParent”) licensed its patent to a Bermuda subsidiary (“BermSub”) for well below market rates, BermSub would be able to make a product very cheaply, selling to the world at a huge profit That profit would be kept in low-tax Bermuda, basically siphoning it out of USParent via the cheap patent license. To prevent this sort of evasion, the IRS has the virtually unfettered power through Code §482 and the related §1.482 Treasury Regs to adjust the incomes of controlled parties to make them more in line with what their incomes “should” have been. So, the IRS would make BermSub pay USParent an amount “commensurate with the income attributable to the intangible”, and no longer could BermSub make such a low-tax killing.
The idea of preventing parties from skirting taxes in this manner is not particularly troubling — the OECD has its own Guidelines for doing the same — but the amount of control the IRS wields through §482 is entirely inappropriate. This is particularly true with an outright sale of an intangible to a related party. In general, the IRS uses a “commensurate with income” (“CWI”) standard. What is this? Per a 2007 Memorandum:
The Treasury Department and IRS’ longstanding authoritative interpretation set forth in the regulations and other published guidance is that the commensurate with income standard must be applied consistently with the arm’s length standard.
The arm’s length standard requires that a controlled transaction be priced so as to realize results consistent with those uncontrolled taxpayers would have realized if they engaged in the same transaction under the same circumstances. – “Taxpayer Use of §482 and the Commensurate With Income Standard,” AM-2007-007 (Mar. 15, 2006), reprinted at 2007 WTD 59-27 (Mar. 27, 2007), Issues 1, 3 (“2007 CWI Memo”).
Further, Reg. §1.482–1(b) clearly provides that “the standard to be applied in every case is that of a taxpayer dealing at arm’s length with an uncontrolled taxpayer” (emphasis added). If the parties don’t behave in this manner, the IRS swoops in with their own income numbers, and taxes the parties on those instead.
So, CWI = arm’s length standard = what uncontrolled parties would have done in the taxpayer’s shoes. That means that if USParent instead sold the patent to BermSub, the IRS would check to see what uncontrolled parties would have done in that case, what deal they would have struck. Right?
Wrong. The IRS instead gets to look at the income actually attributable to the intangible (here, the patent), and tax the parties based on that income, even for a sale paid for in full immediately. Let’s pretend the patent was for a drug for hypertension. USParent and BermSub sat down with some really smart economists and industry experts and accountants, and everyone agreed that the patent was going to bring in about $50million per year for 10 years, after which the patent would expire. The IRS actually lets this figure be off by 20% (up or down) each year, both looking at a year alone and totaling up the income up through that year. If projections were off by more than 20%, the IRS will adjust the income so that it’s commensurate with the income. If BermSub makes an up-front, lump-sum payment, the IRS views that payment as all the future income stream amounts (the $50m/year payments), discounted back to present value. Uncontrolled parties might negotiate a price in the same way.
What uncontrolled parties would not do is say, “Remember that deal we made three years ago? Where I sold you that patent for hypertension, which turns out to cure erectile dysfunction and you now call it ‘Viagra’? You need to pay me more for that patent.” However, this is what the IRS does. They completely restructure the business transaction. (You can be certain that they don’t do this every time — only when the US government would be the one missing out. If the controlled party with the windfall is in the US and thus there is an increase in US tax revenue, the IRS will just leave it alone.)
The OECD doesn’t do this, and is actually pretty snarky about it, too.
Restructuring of legitimate business transactions would be a wholly arbitrary exercise the inequity of which could be compounded by double taxation created where the other tax administration does not share the same views as to how the transaction should be structured. [emphasis added, chapter downloads as a pdf]
Ha! The OECD Guidelines go on to say that in rare, exceptional cases it’s ok to rewrite the terms of the parties’ contracts — when the parties were not behaving as would uncontrolled parties. In that case, the OECD Guidelines say to go back to the time the deal was made, see what rational, uncontrolled parties would have done, and adjust the incomes according to that.
This is not what the IRS does. Going back to the 2007 Memo,
The regulations allow the IRS, in its discretion, provisionally to treat the income actually resulting from the transferred intangible as evidence of what should have been projected at the time of the transfer and to make periodic adjustments to reflect the pricing had such results been projected at such time. The regulations then allow taxpayers the ability to rebut such presumption, e.g., by showing that such results were beyond the control of the taxpayer and could not reasonably have been anticipated at the time the transaction. (2007 Memo, Issue 3 (footnotes omitted).)
So, the income actually received is what the taxpayers, in their infinite psychic wisdom, should have predicted. Never mind that life is unpredictable, and that black swans occur when least expected, and that the first dozen publishers to read the Harry Potter script didn’t think it was valuable enough, and so on and so forth. Sure, the taxpayer can still go through an arduous settlement/court case process to rebut the IRS, but that’s pretty terrible.
In fact, it’s a lot like a form contract, written by the government with the US taxpayer as the customer. And it’s fair to say that this contract’s terms are distinctly unfair. As mentioned above, in certain contexts a contract with severely one-sided terms can serve a social function. In some theories on boilerplate contracts, it’s posited that these terms allow the stronger party — the one providing the boilerplate document, here being the IRS, in whose favor the terms are generally drafted — to use discretion that would otherwise be hard to exercise. That is, most of the harshest terms are not meant to be enforced against the vast majority of applicable cases, but only to circumstances where the breach is egregious. “But if the provision were explicitly limited to those circumstances, enforcement would be a difficult undertaking for a court… The one-sided provision obviates this concern…” (Lucian A. Bebchuk and Richard A. Posner, “One-Sided Contracts in Competitive Consumer Markets” in Omri Ben–Shahar (ed.), Boileterplate: The Foundation of Market Contracts, (2007) Cambridge University Press: New York, at 6–7.) This theory, however, is premised on the stronger party having strong reputational concerns and little opportunity to effectively communicate in court the bounds of reasonableness or “egregiousness” that it wants to monitor. The IRS does not fit such description, and should not be permitted to exercise such raw discretion through the creation of legal uncertainty.
To rule through fear is a sign of cowardice; uncertainty is an enemy of the Rule of Law. These rules should be amended to follow the OECD Guidelines, allowing honest parties to avoid penalties for innovative success or uncontrollable external events.
Nice job, Michelle!
If you’d like to read some of the other quality entries, check them out later today on Facebook.