The U.S. Tax Court and a federal appeals court have ruled on the appropriateness of intercompany royalties between Coca-Cola and its foreign affiliates, and between the medical device manufacturer and its Puerto Rican affiliate.
On Nov. 16, 2020, the IRS prevailed in Coca-Cola Co. v. Commissioner as the tax court accepted the IRS’s application of the comparable profits method (CPM) in its evaluation of the appropriate royalty rate that the foreign affiliates of Coca Cola for the use of highly valuable intangible assets. On Aug. 16, 2018, the U.S. Court of Appeals for the Eighth Circuit vacated the June 9, 2016 Tax Court ruling in Medtronic v. Commissioner (900 F.3d 610 (8/16/18), vacating T.C. Memo. 2016-112).
The IRS again used a CPM approach whereas the taxpayer argued on the basis of a Comparable Uncontrolled Transaction (CUT).
There are often large differences between weak CUTs and mechanical applications of CPM. Faced with these differences, debating which method is better misses the entire point of economic analysis based on the facts and circumstances of the litigation. My earlier reviews of the issues in these two litigations suggested that the usual application of CPM is a flawed application of profits-based approaches, as it ignores the essence of licensee risk-taking in the presence of highly valuable intangible assets. With that said, the IRS is certainly correct to raise the profit-potential issue when the expected profitability in a controlled transaction is substantially higher than the profitability in the third-party transactions that taxpayers offer as alleged CUTs. But it is precisely in these situations when ignoring the role of licensee risk-taking leads to absurdly high results for the intercompany royalty rate.
My discussions of these two litigations were based on the premise that the related party licensee performed all routine functions including production and distribution. (“Coca Cola’s Intercompany Royalty Rate: An Intermediate View”, Journal of International Taxation, February 2020; “Medtronic’s Intercompany Royalty Rate: Bad CUT or Misleading CPM?”, Journal of International Taxation, February 2019). The trial court decision in the Coca Cola litigation notes that affiliates in different jurisdictions perform the production versus distribution functions presenting facts that lead to certain qualifications of what I have previously presented.
Section I summarizes my earlier arguments with respect to the Coca Cola case comparing them to the key portions of the tax court decision. Section II revisits the issues in light of the court’s discussion of how the functions, assets, and risks of the overall licensee were divided between the production versus the distribution affiliates. Section III summarizes my earlier arguments with respect to the Medtronic litigation as well as explore the implications of a separate legal entity approach,
I. The Coca Cola Litigation and an Intermediate View
The U.S. Tax Court ruled on Nov. 18 that Coca-Cola’s U.S.-based income should be raised by over $9 billion in a dispute over the appropriate royalties owed by its foreign based licensees for the years from 2007 to 2009. This ruling addresses a classic issue under tax code Section 482 with respect to the appropriate evaluation of intercompany royalties for highly profitable multinationals.
The relevant foreign production affiliates are located in Brazil, Chile, Costa Rica, Egypt, Ireland, Mexico, and Swaziland, while distribution affiliates and bottlers are located in various nations in Europe, Africa, and Latin America. The U.S. parent licenses intangible property, including trademarks, brand names, logos, patents, secret formulas, and proprietary manufacturing processes. Consolidated profitability in Europe, Africa, and Latin America was considerable, but the intercompany royalty rates were modest.
The testimony in this trial involved a multitude of witnesses and experts relying on various pieces of information. The taxpayer made several arguments for its lower intercompany royalty rates, including providing evidence of a 1996 closing agreement between the taxpayer and the IRS. Coca-Cola produces concentrate and sells it to third-party bottlers who convert the concentrate into soft drinks such as Coca-Cola, Fanta, and Sprite.
The Tax Court noted:
“For 2007-2009 petitioner reported income from its foreign supply points using the “10-50-50 method,” as it had done for the previous 11 years. This was a formulary apportionment method to which petitioner and the IRS had agreed in a closing agreement executed in 1996, which resolved petitioner’s tax liabilities for 1987-1995.”
This method could be seen as a crude form of the residual profit split method where the licensee received a routine return for production and distribution equal to 10% of sales with residual profits split evenly between the foreign licensee and the U.S. parent.
Table 1: Four Approaches to Coca Cola’s Intercompany Royalty Rate
Table 1 provides a simplified version of Coca-Cola’s and the IRS’s positions, showing a licensee in a nation such as France, Germany, or the U.K., where annual concentrate sales equal $1 billion. Table 1 also assumes production costs are 17.5% of sales, distribution costs are 27.5% of sales, and payments to third-party marketing firms equal to 10% of sales. Thus, consolidated profits are 45% of sales.
Table 1 considers four intercompany royalty rates. The 10-50-50 approach would suggest a 17.5% royalty rate, which Table 1 labels the taxpayer approach. The IRS position was that the royalty rate should be 40%. The other two columns represent intermediate positions, which we discuss below.
The taxpayer asserted that this closing agreement was binding for 2007 to 2009. The court disagreed, noting the 1996 agreement did not discuss any transfer pricing methodology or in any way bind the IRS for future years. The taxpayer attempted to argue the closing agreement was predicated on factual understandings that the U.S. parent owned only product intangibles while the foreign affiliates created marketing intangibles. The court also rejected this position. The IRS position rested on two propositions. The first proposition was that the overall routine return was only 5% of sales, which would suggest residual profits would be 40% of sales. While the taxpayer representatives severely criticized the specific approach used by the IRS expert witness, the taxpayer’s expert witnesses agreed that the routine returns for distribution and production were very modest.
The other proposition is that the U.S. parent is entitled to all of the residual profits in the form of intercompany royalties. This position more than doubles the intercompany royalty rate leaving income equal to only 5% of sales for both the production and distribution functions. The IRS position was an application of the comparable profits method (CPM). The court decision noted that there were no third-party license agreements that could be deemed as comparable to the related party license arrangements for purposes of the comparable uncontrolled transaction (CUT) approach.
The Tax Court also noted:
“This case is particularly susceptible to a CPM analysis because petitioner owned virtually all the intangible assets needed to produce and sell the Company’s beverages. Petitioner was the registered owner of virtually all trademarks covering the Coca-Cola, Fanta, and Sprite brands and of the most valuable trademarks covering the Company’s other products. Petitioner was the registered owner of nearly all of the Company’s patents, including patents covering aesthetic designs, packaging materials, beverage ingredients, and production processes. Petitioner owned all rights to the Company’s secret formulas and proprietary manufacturing protocols. Petitioner owned all intangible property resulting from the Company’s R&D concerning new products, ingredients, and packaging. Petitioner was the counterparty to all bottler agreements, giving it ultimate control over the distribution system for the Company’s beverages. And most ServCo agreements executed after 2003 explicitly provided that “any marketing concepts developed by third party vendors are the property of Export,” thus cementing petitioner’s ownership of marketing intangibles subsequently developed outside the U.S. The supply points, by contrast, owned few (if any) valuable intangibles. Their agreements with petitioner explicitly acknowledged that [The Coca-Cola Co.] owned the Company’s trademarks, giving the supply points only a limited right to use petitioner’s IP in connection with manufacturing and distributing concentrate.”
The IRS reasoned that the intercompany royalties capture all of the residual profits since the U.S. parent owned all valuable intangible assets.
Our table, above, also assumes that the European licensee owns $500 million in tangible assets, while the intangible assets located in the U.S. and utilized by this licensee have a value equal to $4 billion. The relative value of intangible assets to tangible assets in our illustration is consistent with the expert testimony in this litigation. If the intercompany royalty rate is less than 20%, the return to the licensor’s intangible assets is less 5%, while the return to the licensee’s tangible assets is greater than 50%. This result is implausible under arm’s-length pricing if the licensor owns all of the intangible assets.
In two earlier papers on intercompany royalties, I lay out this argument in terms of the capital asset pricing model, which holds that the expected return to any asset (R) is given by (“Arm’s-Length Royalties in Light of BEPS and Uniloc,” Journal of International Taxation, November 2015”; “Profits-Based Approaches to Evaluating Reasonable Royalties in Light of Uniloc v. Microsoft (Parts 1 and 2),” Intellectual Property & Technology Law Journal, September and October 2012):
R = Rf+ β(Rm- Rf),
where Rf = risk-free rate, Rm = expected return on the market portfolio of assets, and β = beta coefficient for this particular asset. The beta coefficient measures the tendency of the asset’s return to move with unexpected changes in the return to the market portfolio. Beta coefficients are often estimated for the equity of publicly traded companies. Equity betas reflect both operational risk and leverage risk. Since the task herein is to estimate the expected return to assets rather than the expected return to equity, the beta estimates must be on a debt-free, unlevered, or asset basis. Remove the effect of leverage on these equity betas (βe):
β = βe/(1 + D/E),
where D = debt and E = market value of equity. To the degree that a company engages in leverage risk by financing its assets through debt instead of equity, the equity beta would tend to exceed its asset or unlevered beta (β).
In our example, total assets are $4.5 billion that consist of $0.5 billion in tangible assets owned by the licensee and $4 billion in intangible assets owned by the licensor. Let the risk-free rate (Rf) be 5%, the premium for bearing overall market risk (Rm- Rf) is 5%, and the overall asset beta (β) is 1. Under these assumptions, the overall expected return to assets is 10%, implying that consolidated expected profits = $450 million.
CPM asserts that the owner of the tangible assets deserves only a 10% return or $50 million, which is often referred to as routine return. Residual profits would, therefore, be $4 billion, representing a 10% return on intangible assets. This approach, however, ignores the commercial risk that the licensee bears.
BEPS Action 9 (Assure That Transfer Pricing Outcomes Are in Line with Value Creation Risks and Capital), paragraph 63, notes:
“Risks should be analyzed with specificity, and it is not the case that risks and opportunities associated with the exploitation of an asset, for example, derive from asset ownership alone. Ownership brings specific investment risk that the value of the asset can increase or may be impaired, and there exists risk that the asset could be damaged, destroyed or lost (and such consequences can be insured against). However, the risk associated with the commercial opportunities potentially generated through the asset is not exploited by mere ownership.”
While this discussion relates to the leasing of tangible property, the same economic principle holds for the licensing of intangible assets. Also assume that the licensor affiliate deserves an expected return to its intangible assets equal to the risk-free rate plus a modest premium for bearing the risk of ownership. In my 2012 paper, I wrote this notion as follows:
Rl = Rf + βl(Rm - Rf),
where βl represents the beta appropriate for a pure play licensing firm.
CPM represents one extreme where βl = β. In this case, the licensee receives none of the residual profits. The other extreme is based on the premise that the licensor deserves only the risk-free rate as if βl = 0. In this case, the licensee deserves a high expected return on the assets that it owns formally because licensing involves off-balance-sheet financing or leveraging. The appropriate beta coefficient for the licensee is:
β’ = (β - kβl)/(1 – k),
where k = the ratio of intangible value to total value. Since the value of intangible asset is eight times the value of tangible assets, k is almost 90%.
Let’s reasonably assume that βl ranges from 0.25 to 0.5, which would imply that the licensor deserves a return to the value of intangible assets from 6.25% to 7.5%. Table 1 shows the low end of this range to be a royalty rate = 25%. Under these assumptions, the expected return to the tangible assets of the license = 40%. Table 1 shows the high end of this range to be a royalty rate = 30%. Under these assumptions, the expected return to the tangible assets of the license = 30%.
This economic model notes that licensees deserve a share of residual profits even if they do not own valuable intangible assets. Even if we accept the IRS premise that the U.S. parent in Coca Cola holds all of the valuable intangible assets, a lower royalty rate is suggested by a model grounded in sound financial economics that properly considers licensee risk and the implications for expected returns.
II. Implications of the Foreign Affiliates Splitting the Production and Selling Functions
The testimony of the taxpayer’s expert witnesses was based on a licensee that performed both production and distribution functions, which is our assumption in the previous. The Tax Court noted several facts that call into question the reasonable of this assumption. The facts indicated that the manufacturing affiliates incurred none of the selling or marketing functions or oversaw the operations for the franchise bottlers. These responsibilities were conducted either by the U.S. parent or the distribution affiliates.
Ernst & Young prepared transfer pricing documentation for the various distribution affiliates of Coca Cola. The U.S. parent reimbursed the third party marketing costs incurred by the affiliates at cost plus no markup. The compensation for the selling costs incurred by these affiliate were compensated by a markup over these costs consistent with a CPM analysis.
The Tax Court reasoned based on these facts that it was the U.S. parent that was responsible for developing and maintaining the marketing intangibles. While our analysis also makes this assumption, it calculates the return to tangible assets for a licensee rather than the separate production and distribution affiliates.
Table 2: Coca Cola Transfer Pricing with Three Relevant Affiliates
Table 2 remedies this oversimplification by casting our table 1 presentation using a structure where a manufacturer affiliate charges the distribution affiliate such that the latter receives a 41% gross margin. The gross margin compensates the distribution for its selling and marketing affiliate and leaves the distribution affiliate with a 3.5% operating margin. While table 1 assumed that the licensee held tangible assets equal to 50% of sales, table 2 splits these assets into manufacturer assets = 15% of sales and distributor assets = 35% of sales.
Under the taxpayer’s transfer pricing policies where the manufacturer affiliates pays the U.S. parent royalties = 17.5% of sales, the manufacturer’s profits represent a 160% return to its tangible assets. Our illustration roughly captures the economic presentation by the lead economic expert witness for the IRS.
While we have argued that a royalty rate between 25% and 30% might be appropriate for the entrepreneurial risk taking by the licensee, it is implausible that all of the residual profits captured by our licensee should reside with a contract manufacturer with few tangible assets and no responsibility for marketing and relationships with the bottling.
The transfer pricing approach used by Ernst & Young calls into question whether the distribution affiliates should receive any of these residual profits as the Tax Court notes that the typical agreement between the U.S. parent and the foreign distribution affiliate (known as ServCo) stated:
“ServCo acknowledges that it does not take entrepreneurial risk in developing marketing concepts because the marketing advice provided by ServCo is within the strategic guidelines established by Export for the brands. ServCo also acknowledges that any marketing concepts developed by third party vendors are the property of Export.”
While the foreign tax authorities in nations such as France, Germany, and the U.K. might argue that their distribution affiliates deserved a portion of residual profits, the taxpayer in fact trapped itself with these inconsistent positions with respect to which foreign affiliate (if any) deserves any of the residual profit either from ownership of marketing intangibles or from taking on licensee risk.
Medtronic produced and sold implantable cardiac pulse generators and neurological stimulators. The structure was such that the U.S. parent provided the intangible assets, produced the components, and distributed the finished products, which were assembled by its Puerto Rican affiliate Medtronic Puerto Rico Operations. These operations were highly profitable with much of the profits retained by the Puerto Rican assembly affiliate. The issue was whether the intercompany royalty paid by this affiliate to the U.S. parent should be 20% of sales or 50% of sales. Many practitioners have viewed this intercompany royalty rate issue as a battle between an application of a CUT approach versus an application of CPM with the assembly affiliate as the tested party.
The Tax Court reasoned that this intercompany royalty rate should be 30% of sales based on a rather tortured application of the CUT approach. The Appeals Court expressed serious concerns with respect to the CUT application, leading some to believe that the court was endorsing the IRS approach. The Appeals Court is certainly calling for a more substantive analysis than either the taxpayer or the IRS offered in the Tax Court.
Table 3 presents the type of model I noted in my 2018 paper on the Medtronic litigation, which is based on the following assumptions:
- U.S. distributor with sales = $6 billion and selling costs = $1.6 billion;
- U.S. components manufacturer with production costs = $400 million;
- Puerto Rican assembly affiliate with labor costs = $400 million; and
- U.S. parent that collects intercompany royalties from the Puerto Rican affiliate for the use of product and marketing intangibles.
Table 3 aggregates these three affiliates into a single licensee that owns $6 billion in tangible assets. Let’s also assume that the value of intangible assets owned by the U.S. affiliate = $30 million. Let’s retain our assumptions that the risk-free rate = 5%, the premium for bearing overall market risk (Rm- Rf) = percent, and the overall asset beta is 1. Under these assumptions, the overall expected return to assets is 10%, implying that consolidated expected profits = $3.6 billion. CPM would imply that the owner of the tangible assets deserves only a 10% return or $600 million, which is often referred to as routine return. Residual profits would, therefore, be $3 billion, representing a 10% return on intangible assets.
Table 3: Medtronic Licensee Income Statement under Alternative Royalty Rates
The taxpayer’s position that the royalty rate should be only 20% of sales would leave the owner of the intangible assets with only a 4% return, which is less than the risk-free rate. The Tax Court’s ruling that the royalty rate should be 30% is more plausible as the return to intangible assets would be 6%.
The IRS position is that the royalty rate should be set at 50% of sales so that the U.S. parent captures 100% of residual profits. If the royalty rate were set at 40%, the licensee would capture 20% of residual profits. This position would be consistent with βl = 0.8, so that the expected return to the licensor’s intangible assets would 8% percent. The expected return on the tangible assets of the licensee would be 20%.
The Tax Court based an application on the cross-license between Medtronic and Siemens-Pacesetter agreed to as part of a settlement of several lawsuits including patent infringement and anti-trust issues. Pacesetter agreed to pay a 7% royalty rate. The Appeals Court was not convinced that this starting point and the various adjustments were sufficient evidence that the royalty rate should be only 30%. The main concern of the Appeals Court is that the controlled transaction had a much higher profit potential than the Pacesetter operations.
Robert Pindyck in his report on behalf of the taxpayer assumed a 23% potential profit margin for the Pacesetter operations. His report argued that one could use the ratio of royalty income to consolidated profits for the Pacesetter operations as a metric for the appropriate split of consolidated profits between the related-party licensee and related-party licensor in the Medtronic issue. Consider the following model for the Pacesetter operations:
- Expected operating profits relative to sales = 20%;
- Tangible asset to sales ratio = 100%; and
- Overall cost of capital = 10% so routine returns = 10% of sales.
Residual profits represent 10% of sales. A 7% royalty rate implies that the licensor receives 70% of residual profits, leaving the licensee with the other 30% of residual profits. In terms of our expected return and risk approach herein, this result is consistent with profits with the licensee being afforded 30% of residual profits. This position would be consistent with βl = 0.6, so that the expected return to the licensor’s intangible assets would be 7%. In other words, this evidence suggests that β’ would be 4 and the expected return on the tangible assets of the licensee would be 25%. The evidence, therefore, supports the premise that the arm’s-length royalty rate should be 35%. Residual profits relative to sales in the controlled transaction are five times as great as the residual profit-to-sales ratio that the Pacesetter illustration suggested. It stands to reason that the appropriate arm’s-length royalty rate for the controlled transaction would also be five times that for the Pacesetter agreement.
Table 4 segments the financials for the licensee among the U.S. distributor, the U.S. manufacturer of components, and the Puerto Rican assembly affiliate. We have assumed that the tangible assets for the Puerto Rican assembly affiliate is 250% of its operating costs and that the tangible assets for the U.S. component manufacturer is 250% of its operating costs. We have also assumed that the tangible assets for the distribution division is also 250% of its selling costs. As such, the routine return for each of these functions represents a 25% markup over operating costs.
The U.S. manufacturer of components sells its product to the Puerto Rican assembly at an intercompany price (I/C price 1) = $500 million, thereby retaining a routine profit of $100 million. The U.S. distributor purchases finished goods from the Puerto Rican affiliate paying it an intercompany price (I/C price 2) of $4 billion, retaining gross profits of $2 billion and operating profits = $400 million. The Puerto Rican assembly retains operating profits before royalties (OPB4Royalty) of $3.1 billion.
The profits for the Puerto Rican affiliate depend on the royalties paid to the U.S. owner of intangible assets. We have assumed that the routine profits for the Puerto Rican affiliate are $100 million, which implies residual profits = $3000 billion. The IRS approach would have all of these residual profits accrue to the U.S. licensor. Under this approach, all entities would receive a 10% return on the assets that they formally own.
Under the taxpayer’s 20% royalty rate, the Puerto Rican affiliate retains $1.8 billion in residual profits and $1.9 billion in overall profits. Under the actual intercompany policy, the Puerto Rican affiliate’s return to tangible assets would be 190%.
Table 4’s presentation is based on the 35% royalty rate, which we argued is consistent with an application of the CUT approach that properly considered the differences in the profit potential between the controlled transaction and the Pacesetter operations. Table 3 notes that the licensor’s return to intangible assets would be 7% and the return to tangible assets for the various operating entities collectively would be 25%. Table 4 notes, however, that the U.S. distributor and the U.S. manufacturer of components received only a 10% return to their tangible assets, while the Puerto Rican affiliate would receive a 100% return to its tangible assets.
Table 4: Segmented Income Statement for Medtronic’s Operating Entities
A disconnect exists between the licensee/licensor model and the structure of Medtronic’s operations as it relates to the arm’s-length division of residual profits. If Puerto Rico conducted all of the functions of a true licensee including production and assembly of components and distribution, the model in the first section would allocate appropriately all profits including routine returns and residual returns.
The U.S. parent, however, performs two of these commercial functions. The question then becomes which entity bore the commercial risk. The taxpayer’s position makes two leaps of faith. First, it implicitly assumes that the licensor does not bear ownership risk. It also assumes that the U.S. operations do not bear any of the leverage risk from the use of valuable intangibles as if the Puerto Rican assembly affiliate bears the entirety of this risk. The appeals court called for a more detailed analysis of the risk issue said specifically:
“[T]he tax court did not decide the amount of risk and product liability expense that should be allocated between Medtronic U.S. and Medtronic Puerto Rico….The tax court rejected the Commissioner’s 11% valuation, concluding that it was unreasonably low because it did not give enough weight to the risks that Medtronic Puerto Rico incurred in its effort to ensure quality product manufacturing. Accordingly, the tax court allocated almost 50% of the device profits to Medtronic Puerto Rico. In doing so, the tax court also rejected the Commissioner’s comparable profits methods because it found that the comparable companies used by the Commissioner under this method did not incur the same amount of risk incurred by Medtronic Puerto Rico. Yet the tax court reached these conclusions without making a specific finding as to what amount of risk and product liability expense was properly attributable to Medtronic Puerto Rico. In the absence of such a finding, we lack sufficient information to determine whether the tax court’s profit allocation was appropriate.”
If the commercial risk for a medical device multinational goes beyond risk incurred ensuring quality product management, the Appeals Court’s call appears to go beyond simply segmenting overall systematic risk into commercial risk vs. risk of ownership. Thus, the analysis herein suggesting that the arm’s-length royalty should be only 35% may not compensate the U.S. operations sufficiently for their functions, assets, and risks even as the IRS abuse of CPM overcompensates U.S. operations.
This column doesn’t necessarily reflect the opinion of The Bureau of National Affairs Inc. or its owners.
Harold McClure has been involved in transfer pricing as an economist for 25 years.