Coca-Cola’s Cost-Sharing Defense Runs Into a Wall of Regulatory Reality
Coca-Cola’s long-running transfer-pricing battle with the Internal Revenue Service has reached a pivotal stage, and the company’s latest argument hinges on a comparison that the regulations never meant to accommodate. By likening its decades-old profit-splitting arrangement to a modern cost-sharing agreement, the beverage giant hopes to overturn a Tax Court decision that reallocated billions of dollars in income to the United States. Yet the very analogy Coca-Cola deploys exposes flaws so fundamental that, far from rescuing the company, it underscores why the Tax Court sided with the government in the first place.
Background of the Dispute
The case arose after the IRS rejected Coca-Cola’s 10-50-50 formulary method, a mechanism the company adopted through a 1996 closing agreement with the agency. Under that formula, the seven foreign “supply points” that manufacture beverage concentrate first received a routine 10 percent return on sales. They then shared equally with the U.S. parent the residual, systemwide profit. For many years the arrangement passed without incident, but by the mid-2000s the IRS concluded that the formula no longer reflected arm’s-length standards. The Service applied the comparable profits method (CPM) instead, dramatically boosting the U.S. tax base and leaving Coca-Cola facing an adjustment exceeding $9 billion for 2007-2009 alone.
The company petitioned the Tax Court, arguing that CPM ignored the substantial marketing investments the supply points allegedly shouldered. In 2020 the court upheld the IRS, finding that the foreign affiliates did little more than mix ingredients and had no plausible claim to half the global residual profit. Coca-Cola appealed, and its latest briefs frame the supply points as risk-bearing investors whose marketing outlays entitled them to a return, just as participants in today’s cost-sharing agreements (CSAs) receive a share of intangible income.
The 10-50-50 Formula Under Fire
At the heart of the controversy is whether the formulaic split blessed three decades ago can still bind the IRS. Coca-Cola insists that abandoning the 10-50-50 mechanism denies the supply points a fair reward on billions spent to nurture brand demand in their territories. The IRS replies that the affiliates were contract manufacturers, not entrepreneurial marketers, and that they already earned high margins that dwarfed any routine service fee.
The Tax Court agreed with the government, highlighting several facts:
- The supply points had no authority to develop, alter, or exploit trademarks.
- They incurred marketing costs only because the U.S. parent allocated expenses to them after year-end.
- Coca-Cola routinely shifted production among supply points to chase tax or tariff advantages, undermining the idea that any one entity assumed long-term market risk.
Cost-Sharing Analogy Emerges
On appeal, Coca-Cola sharpened its narrative. While conceding that no formal CSA existed, the company portrays the affiliates’ marketing spend as a de facto investment in intangible development. Under the investor theory embedded in the current cost-sharing regulations, a party that funds high-risk intangible creation should earn an expected return commensurate with that risk. Coca-Cola argues that CPM eliminates that upside, awarding all residual profit to the U.S. parent despite the affiliates’ “billions” in consumer advertising expenditures.
It is a clever rhetorical pivot. By invoking language familiar to modern transfer-pricing practitioners—platform contribution transactions, discount rates, reasonably anticipated benefits—the brief suggests that the supply points resemble the limited-risk “cash boxes” that populate real CSAs. If those cash boxes deserve a share of future intangible income, why not the concentrate makers?
Why the CSA Comparison Fails
The difficulty is that regulatory compliance requires far more than a label. The investor model in Treasury Regulation §1.482-7(g)(2)(ii)(A) demands ex ante discipline: participants must establish fixed divisional interests, share costs in proportion to anticipated benefits, and compensate one another through platform contribution transactions (PCTs) when pre-existing intangibles are transferred. None of those building blocks existed in Coca-Cola’s historical arrangement.
Consider the essentials the company lacks:
- Fixed Divisional Interests. CSAs divide the world into exclusive territories or fields of use so that each participant’s investment can be matched to its prospective revenue. Coca-Cola, by contrast, moved concentrate production among affiliates “dozens of times” between 1980 and 2009, often at short notice, and without compensating the losers. That fluidity severs any link between a given affiliate’s past spending and its future profit stream.
- Predictable Allocation Method. In an authentic CSA, participants agree up front on how to allocate intangible development costs. Coca-Cola retroactively booked more than 60 percent of the supply points’ operating expenses via internal service-fee markups from a web of 60 foreign marketing service companies. The allocation method changed over time and was not tied to the supply points’ actual benefit expectations.
- Platform Contribution Transactions. When a CSA participant contributes existing intangibles, regulations require compensation so that all parties attain a risk-adjusted return. Coca-Cola never priced, let alone paid, a PCT for the global brand value it licensed to the supply points.
Investor Model Requirements
The investor model equates an arm’s-length return with the discount rate suitable for the project’s risk profile—often approximated by a multinational’s weighted average cost of capital (WACC). For Coca-Cola in the late 2000s, that hurdle rate hovered near 7 percent. To validate its investor theory, the company would need to show that the affiliates’ reasonably anticipated returns aligned with that 7 percent benchmark. Instead, numbers in the record paint a wildly different picture.
Numbers Tell a Different Story
Coca-Cola’s brief mentions an $8.5 billion marketing “investment” for 2007-2009 across six supply points. Using a 7 percent discount rate and mid-year convention, the present value of those outlays equals roughly $7.7 billion. Yet after adding back the allocated marketing charges (because they reduce reported profit), the supply points posted operating profits of $6.2 billion in 2007, $7 billion in 2008, and another $7 billion in 2009. That stream delivers an internal rate of return of about 107 percent—fifteen times the 7 percent baseline a third-party investor would expect for comparable risk.
Even when analysts stress-test the assumptions—treating intercompany royalties as additional marketing investment or shifting to year-end cash flows—the implied return still dwarfs the company’s WACC. The gulf is so extreme that it cannot be justified by ex ante uncertainty, especially when the activity financed was ongoing consumer advertising intended merely to preserve a household-name brand.
Imagem: Internet
Absence of Ex Ante Discipline
Regulations further allow the IRS to disregard any arrangement that allocates risk after outcomes become knowable. Treasury Regulation §1.482-7(k)(1)(iv)(A) expressly targets post-hoc risk assignments that lack economic substance. The Tax Court found precisely that: marketing expenses were allocated ex post, production assets migrated among jurisdictions for tax reasons, and no participant could forecast its long-term revenues or costs in a manner consistent with arm’s-length behavior.
Moreover, the Irish supply point—a 1.4 percent tax-rate entity that supplied more than 90 markets and captured 65 percent of total revenue—illustrates how profit followed tax incentives, not marketing risk. To defend the Irish entity’s outsized income as a return on prior investment, Coca-Cola would have to prove that it funded disproportionate marketing spend in scores of countries before the company decided to shift production there. The record contains no evidence of such forward-looking funding. On the contrary, when facilities were downsized or closed, the displaced supply points received no compensation for their supposed long-term “investments.”
What It Means for Multinationals
The Coca-Cola appeal showcases how alluring—but perilous—it can be to retrofit an old structure into the mold of modern cost-sharing. CSAs remain a popular vehicle for managing intangible income, but their viability rests on meticulous, forward-looking compliance. Without fixed divisions, coherent cost-allocation rules, and measured PCTs, the investor model collapses. Multinationals contemplating similar defenses should heed three lessons:
- Retroactive analogies rarely survive scrutiny when contemporaneous documents and conduct point elsewhere.
- Excess returns far above a group’s WACC invite probing questions about risk, substance, and the timing of allocations.
- Regulations empower the IRS to recast or ignore arrangements that allocate income after results are visible, making robust ex ante planning essential.
For Coca-Cola, the road ahead in the appellate courts may yet deliver procedural twists, but the substantive hurdles remain steep. The company’s cost-sharing comparison not only fails to cure the deficiencies the Tax Court catalogued; it magnifies them by spotlighting the very regulatory safeguards the arrangement never satisfied.
FAQs
What is a cost-sharing agreement (CSA)?
A CSA is a contractual arrangement in which related parties jointly fund and develop intangible assets, such as technology or trademarks, sharing both costs and anticipated benefits in proportion. U.S. regulations require fixed divisional interests, proportional cost sharing, and compensation for pre-existing intangibles through a platform contribution transaction.
What was Coca-Cola’s original profit-splitting formula?
The 10-50-50 method granted the foreign supply points a 10 percent return on concentrate sales and then split residual profits 50-50 between the affiliates and the U.S. parent. The IRS later rejected this approach as inconsistent with arm’s-length standards.
Why did Coca-Cola invoke a cost-sharing analogy?
By portraying the supply points’ marketing expenditures as intangible development investments, Coca-Cola hopes to justify a larger share of residual profit, suggesting that arm’s-length investors would demand returns on those outlays.
How did the Tax Court respond to the analogy?
The court found that no qualified CSA existed, that expenses were allocated retroactively, and that the affiliates performed routine manufacturing rather than entrepreneurial marketing. Consequently, the analogy could not rescue the 10-50-50 formula.
What is the investor model’s benchmark rate of return?
Regulations link an arm’s-length expected return to the discount rate appropriate for the project’s risk—often proxied by the group’s weighted average cost of capital. For Coca-Cola, that was about 7 percent during the relevant years.
