An intercompany royalty is the rate one company in a group pays another for the right to use a patent, a trademark, software, know-how, or another intangible, and in a dispute it is measured against what an unrelated licensor and licensee would have agreed to for the same rights on the same terms. The rate is benchmarked in three ways: directly, from licenses of comparable intangibles between unrelated parties; indirectly, by testing whether the licensee keeps the routine return its own functions would earn after paying the royalty; and, where both sides contribute unique intangibles, by splitting the combined profit according to each side's contribution. This guide explains how the licensed rights and their owner are identified, how each benchmark is built and where it fails, the commensurate-with-income rule and the realistic alternatives the parties had, how royalty rates arise in tax, commercial, shareholder, and divorce disputes, and where opposing analyses go wrong.
An intercompany royalty is the payment one affiliate makes to another for the right to use an intangible, and the first task is to establish exactly what that intangible is. The Treasury regulations treat as intangibles patents, inventions, formulae, processes, designs, and know-how; copyrights; trademarks, trade names, and brand names; franchises, licenses, and contracts; methods, programs, systems, customer lists, and technical data; and similar items that derive their value from intellectual content rather than physical attributes and have substantial value independent of any individual's services. A license grants defined rights in one or more of them, and its terms set the scope a comparable has to match: the field of use, the territory, whether the grant is exclusive, its duration and termination rights, the right to sublicense, the right to improvements, and any technical support bundled with it. The royalty base matters as much as the rate, since a rate on gross sales and the same rate on net sales after returns, freight, and discounts are different prices.
Ownership is the second task. The regulations generally treat the legal owner under intellectual property law, or the holder of the rights under the license itself, as the owner, unless that ownership is inconsistent with the economic substance of the transactions. The OECD Guidelines ask, in addition, which companies perform and control the functions of developing, enhancing, maintaining, protecting, and exploiting the intangible, and allocate the return accordingly, so that legal title alone does not earn the profit the intangible produces. A royalty paid to an affiliate that holds the patents but employs no one who develops or defends them is where the two views are most likely to part, and the transfer pricing disputes guide explains the forums in which the question is decided.
The most direct benchmark is the price unrelated parties agreed for the same or similar rights, which the regulations call the comparable uncontrolled transaction method. Its strongest form is internal: the group licenses the same intangible to an unrelated company under substantially the same circumstances, and the rate in that license is generally the most direct and reliable measure of the arm's length rate, after adjustment for any minor differences in terms. Where no internal license exists, the economist looks for licenses of comparable intangibles under comparable circumstances, and the regulations set two threshold tests: the intangibles must be used with similar products or processes in the same general industry or market, and they must have similar profit potential, which is most reliably measured by the net present value of the benefits the licensee expects from the license. Beyond the threshold, the factors that move a rate are the exclusivity and territory of the grant, the stage of development, including any regulatory approval still needed, the right to updates, the uniqueness of the intangible and how long its protection lasts, the duration and termination rights, the risks the licensee assumes, and the services or other dealings that come with the license.
The candidate licenses come from agreements between unrelated parties disclosed in public filings and collected in commercial licensing databases, and from the group's own licenses with third parties. Each candidate is read in full rather than taken from a database summary of its rate, because the summary does not show the base, the exclusivity, the bundled know-how, the minimum payments, or the upfront fees that change what the stated rate means. Licenses granted to settle infringement litigation are a weaker comparable, since their terms reflect each side's view of validity and the cost of the case as much as the value of the rights. The result is a set of rates that survive the review, adjusted where a difference can be measured, and a range against which the controlled rate is tested.
Where reliable comparable licenses cannot be found, or as a check on those that were, the royalty is tested indirectly through the profit the licensee keeps. Under the comparable profits method, or its OECD counterpart, the transactional net margin method, the licensee is the tested party when it is the less complex participant: a distributor or manufacturer that performs routine functions and owns no valuable intangibles of its own. Its operating profit after paying the royalty is compared with the operating profit of independent companies that perform similar functions without licensed intangibles. If the licensee earns more than those comparables, the royalty may be too low; if it earns persistent losses its functions and markets cannot explain, the royalty may be too high. The IRS's annual report on its advance pricing agreement program describes this kind of post-royalty margin test being used to confirm a royalty set from comparable licenses.
The method works only where its premise holds. A licensee that built its own customer base, ran its own marketing, or adapted the technology to its market may hold local intangibles that the comparables lack, and treating it as routine hands the return on those intangibles to the licensor. The test also measures the whole profit of the licensee's business activity, so a downturn, a start-up phase, or a product failure moves the result for reasons unrelated to the royalty, which is one reason the comparables are measured over several years. The transfer pricing methodology page describes the tested party and profit level indicator choices that drive the result.
When licensor and licensee each contribute valuable, nonroutine intangibles, neither is a routine tested party and comparable licenses rarely capture the combination, so the profit split method takes their place. Under the residual profit split the regulations describe, the combined operating profit of the relevant business activity is allocated in two steps. First, each party receives a market return on its routine contributions, its manufacturing, distribution, and support functions, measured from comparable independent companies. Second, the residual that remains, the return to the unique intangibles, is divided according to the relative value of each party's nonroutine contributions, which the regulations allow to be measured by external market benchmarks or estimated from the capitalized cost of developing the intangibles, less amortization over their useful lives, such as accumulated research and development or marketing spending.
The method follows the economics of a joint contribution more closely than any one-sided test, but it asks more of the data. It needs reliable combined financial statements for the business activity, segmented from the parties' other operations and stated on consistent accounting and in a common currency, and its result is sensitive to the measure of relative contribution: the useful lives assigned to capitalized research and marketing spending, the years included, and the treatment of spending that created no lasting value. A royalty implied by a profit split is only as reliable as those choices, and the report shows the result under the alternatives the other side will press. The same useful-life and capitalization questions arise when the valuation approaches are applied to the intangibles themselves.
Section 482 adds a rule for intangibles that has no counterpart for goods: the income from a transfer or license of intangible property must be commensurate with the income attributable to the intangible. Under the regulations, a royalty in an arrangement covering more than one year can be adjusted in later years so that it stays commensurate with the income the intangible actually produces, and a determination that the rate was arm's length in an earlier year does not prevent an adjustment in a later one, subject to exceptions that include a rate set from an unrelated license of the same intangible and results that stayed within a band around the projections the rate was based on. A royalty that was reasonable on the projections at signing can therefore be revisited in a tax dispute once the product's actual profitability is known.
The statute also directs that a transfer of intangibles be valued on an aggregate basis, or on the basis of the realistic alternatives to the transfer, where that is the most reliable means of valuation, and the regulations build the same principle into their unspecified methods: independent parties enter a transaction only if no realistic alternative is preferable. In economic terms, the licensor would not accept a royalty lower than what it could earn by exploiting the intangible itself, and the licensee would not pay one that leaves it less than it could earn without the license, so the arm's length rate lies within the range those two alternatives bound. The OECD Guidelines add an approach for hard-to-value intangibles under which, in defined circumstances, the outcomes actually realized can be used as evidence of what the pricing should have been at the outset. Commercial litigation sets a different question: the license agreement and the claim determine whether the parties' rate is enforced, reformed, or measured against an arm's length benchmark, which is a matter of contract and fiduciary law, and later profitability matters only to the extent the agreement or the claim makes it matter.
In a federal tax dispute the question is whether the royalty a U.S. company paid to a foreign affiliate, or received from one, shifted income across the border, and the analysis is presented in an examination, in the Tax Court, or to the competent authorities. State tax disputes turn on the same question at the state level, for example over royalties paid to an affiliate that holds the group's trademarks or patents, where the state disallows the deduction or requires it to be added back unless the charge is shown to be at arm's length or another exception applies.
In commercial litigation an intercompany license can outlive the relationship it was written for. A division is sold and keeps using its former parent's technology under a license drafted for tax purposes, or a joint venture partner contributes intangibles in exchange for a royalty the other partner later disputes, and a contract dispute over the rate, the royalty base, or the audit rights follows. The arm's length rate is then evidence of a reasonable rate where the agreement is silent, ambiguous, or terminated, and the commercial damages analysis carries the difference into the claim. A reasonable royalty for patent infringement is a different measure: it asks what the parties would have agreed in a hypothetical negotiation when the infringement began, with validity and infringement assumed, and while it draws on the same kinds of comparable licenses, it answers a question the transfer pricing analysis does not, the one an intellectual property damages analysis answers.
In a shareholder dispute the claim may be that the controlling owner moved intangibles into an entity it holds separately and then charged the operating company to use them; the arm's length rate measures how much of the royalty exceeded what an independent licensor could have charged, and the excess restates the company's earnings for the valuation or the damages claim. In a divorce, royalties that the owner spouse's company pays to an affiliate the owner controls, sometimes offshore, reduce the income the company reports, and the economist restates them at an arm's length rate for the income available for support and for the marital financial analysis of the business.
The recurring errors are visible on the face of a report. A rate taken from a database summary without reading the agreement behind it. Comparables drawn from a different industry, or from intangibles with very different profit potential, adopted without the profit potential test the regulations call for. Exclusive and nonexclusive, worldwide and single-country, and early-stage and proven licenses mixed in one set without adjustment. Settlement licenses treated as ordinary market transactions. A royalty base that differs between the controlled license and the comparables. A post-royalty margin test applied to a licensee that holds local intangibles of its own, or a royalty that leaves the licensee with years of losses its routine functions cannot explain. A profit split that measures relative contribution with a choice of years or useful lives that happens to favor the retaining side. A trademark royalty charged on top of a distributor's purchase price that already pays for the brand, so that the same intangible is paid for twice.
On the other side, a report that rejects every comparable license as different without proposing a better benchmark has left the rate unanswered, and a report that relies on legal ownership alone, without engaging the functions the record shows, has answered a question the tribunal may not ask. The rebuttal guide lists the questions that expose an unsupported input, the report review service describes how an opposing royalty analysis is tested one comparable and one assumption at a time, and the transfer pricing expert witness page describes the affirmative analysis a royalty benchmark is built from.
Only with care. A litigated reasonable royalty assumes validity and infringement and is shaped by the facts of that case, and a settlement license reflects the cost and risk of the litigation. Either can inform the analysis as one data point, but a transfer pricing benchmark rests on licenses agreed between unrelated parties for comparable rights in the ordinary course of business.
For tax purposes it can. The commensurate-with-income rule lets the rate in a license covering several years be adjusted in later years to track the income the intangible actually produces, subject to exceptions the regulations list. In a commercial or shareholder dispute, the agreement and the claim decide whether later profitability matters at all.
Under the U.S. regulations the legal owner is generally treated as the owner unless that is inconsistent with the economic substance, and the developer is compensated for the functions it performs. The OECD Guidelines look to which companies perform and control the development, enhancement, maintenance, protection, and exploitation functions, so the two frameworks can allocate the return differently.
Request a consultation on Intercompany Royalty Rates in Litigation: How the Rate Is Benchmarked or call (201) 343-0700. Plaintiff and defense counsel.