Menu
Tax Notes logo

Did Coca-Cola Make the Right Play in Its Motion for Reconsideration?

Posted on Aug. 16, 2021
Andrew Hughes
Andrew Hughes

Andrew Hughes is an economist specializing in transfer pricing, valuation, and risk management. He is based in Brussels.

In this article, the author outlines the strategy Coca-Cola chose in its motion for reconsideration before the U.S. Tax Court, summarizing the key issues and examining whether the company chose the best strategy, given the economic arguments before it.

By now, most of the tax world is well aware that the IRS has finally caught a big fish, barring a motion for reconsideration or challenge in another court. On November 18, 2020, the U.S. Tax Court found The Coca-Cola Co. (Coke) liable for more than $3.4 billion in tax deficiencies between 2007 and 2009.1

There are many factors at play in the dispute. First, Coke relied on a 1996 closing agreement to settle a dispute between it and the IRS regarding its tax liabilities between 1987 and 1995. Coke’s tax structure was such that the company was the legal owner of intellectual property that was necessary to manufacture and distribute its well-known beverage products. It had a global network of supply points for manufacturing beverage concentrate and selling it to bottlers. The bottlers, which typically were separate legal entities, were responsible for using the concentrate to manufacture the beverages, bottle them, and then send them to other distributors.

There was also a network of service companies in Coke’s supply chain providing various services, including marketing and advertising, in local jurisdictions to related parties. In the 1996 closing agreement, the IRS and Coke agreed on a 10-50-50 method to calculate the income that the various foreign supply points should retain: Supply points were to keep 10 percent of their gross sales and split the remaining income 50-50 with Coke. Coke continued to use the 10-50-50 method after the period specified in the closing agreement, although that is a key point it contests.

After Coke went through several audit cycles with no problems, the IRS challenged the 10-50-50 method as not arm’s length because the supply points did not own IP of their own. The intercompany agreements between Coke and the supply points confirmed the absence of IP ownership by the supply points and failed to mention the 10-50-50 method as remuneration for the supply points’ functions.

Per the economic analysis detailed in the Tax Court decision, the supply points essentially had guaranteed customers because they were required to sell to the bottlers Coke contracted with and had predefined sourcing because they were allowed to source ingredients only from suppliers identified by the sourcing team, which they did not employ.

Also, separate marketing services companies contracted directly with Coke and stated in their intercompany agreements that any market advice provided was within the strategic guidelines established by Coke and that all marketing concepts developed by the service companies or third-party vendors working on their behalf were the property of Coke. That would become an important point because the supply points were effectively cross-charged and bore the costs of the service companies performing marketing functions.

The IRS relied on an analysis by Scott Newlon, who determined that the comparable profits method was the best method for remuneration of the supply points, with 18 independent Coca-Cola bottlers used as comparable companies. Newlon determined that the return on assets (ROA) was the best profit-level indicator for the intercompany transaction at hand. The table summarizes some of his findings.

ROA of the Supply Points and Comparable Bottlers

Party

Weighted Average ROA (2007-2009)

Ireland supply point

214.4%

Brazil supply point

179.7%

Chile supply point

148.6%

Costa Rica supply point

143%

Swaziland supply point

128.5%

Mexico supply point

94.1%

Egypt supply point

-4.3%

Comparable bottlers (upper quartile)

31.8%

Comparable bottlers (median)

18%

Comparable bottlers (lower quartile)

7.4%

As demonstrated in the table, Newlon’s analysis implied that the supply points, aside from the one in Egypt, were overcompensated based on their functions performed, assets owned, and risks borne. Coke’s various experts proposed other methods, such as the comparable uncontrolled transaction and residual profit-split methods. Ultimately, the Tax Court sided with the IRS in determining that Coke was liable for more than $3.4 billion in tax deficiencies based on the overcompensation of the supply points as determined by the CPM.

Coca-Cola’s Response

On June 2 Coke filed a motion for reconsideration in the Tax Court, arguing two points.

The first point is essentially that the closing agreement provided IRS approval for Coke to use the 10-50-50 method indefinitely, barring any changes in the underlying conditions on which the closing agreement was based. Coke noted in its reconsideration that the IRS’s challenge of the method came as a complete surprise because the agency:

approved and encouraged Coca-Cola to calculate its taxes using the 10-50-50 method and for more than a decade repeatedly audited and approved the Company’s taxes to ensure the Company’s compliance with the 10-50-50 method to calculate the Company’s taxes before switching its position without notice.

Coke further argues that the closing agreement was extraordinary in that it referenced the arm’s-length nature of the transaction and allowed Coke to apply the 10-50-50 method indefinitely.

One part of the closing agreement is a particular point of contention:

For taxable years after 1995, to the extent the Taxpayer applies the [10-50-50] method to determine the amount of its reported Product Royalty income with respect to existing or any future Supply Points, the Taxpayer shall be considered to have met the reasonable cause and good faith exception of sections 6664(c) and 6662(e)(3)(D) * * * and shall not be subject to the accuracy-related penalty under section 6662 * * * with respect to the portion of any underpayment that is attributable to an adjustment of such Product Royalty.

According to Coke, that passage, along with others not fully cited in the available documents, provided it with approval for indefinite use of the 10-50-50 method to calculate the arm’s-length pricing for the supply points. However, a careful reading of the passage shows that it mentions only penalty protection when applying the 10-50-50 method. The Tax Court saw the passage as hurting Coke’s argument because it shows that the parties could have addressed the definitive nature of the 10-50-50 method in the future but did not. The court also said the passage recognizes that the IRS might make future transfer pricing adjustments, which implies that it did not agree to the application of the 10-50-50 method in perpetuity.

For the record, there is a somewhat ambiguous nature to closing agreements and how they are described by the IRS. For example, Internal Revenue Manual section 32.3.4 states that “the direct or indirect impact of the determination of a specific matter upon other years or related cases should be given careful consideration” and that “the agreement itself will be the primary basis for future action.” It is surprising that the IRS and Coke did not address future applicability in their closing agreement, given the passage on penalty protection and the IRM’s warning on taking into consideration applicability in future years. Without any direct citation in the closing agreement showing otherwise, it’s not likely that Coke has much of a chance of changing the Tax Court’s mind on this point.

The second point in Coke’s motion is that the IRS and Tax Court erred in determining that the supply points had no intangible assets and that the licenses given by Coke to the supply points to produce concentrate constituted valuable intangibles in and of themselves, implying that the CPM could not be the best method. Coke also argued that the supply points spent billions of dollars marketing Coke’s brands abroad, which would imply that the CPM analysis the court relied on would not adequately compensate the supply points for their intangible assets and unique contributions to the supply chain.

As discussed, the supply points were effectively allocated marketing expenses incurred by various service companies that were working under Coke’s direction rather than their own. It is clear why Coke chose to challenge this point. If it can show that the supply points owned valuable intangibles, it can move away from a CPM-based approach that would effectively limit the amount of income a tested party would receive, which is an extremely costly conclusion for Coke. However, as detailed in other articles in this publication, the idea that licenses to use intangibles constitute valuable intangibles in and of themselves — and Coke’s arguments on this point — might not be very compelling.2

For example, the Tax Court noted that from 1986 to 2009, Coke “closed (or shifted production away from) 18 supply points, but no supply point received any compensation for this loss of production and income,” including for any supposed intangible assets they built up over time. The court cited the closing of a Peruvian supply point from 2007 to 2009 that had been operating for many years. All the Peruvian operations were transferred to other supply points without compensating the Peruvian supply point for any supposed intangible assets.

Other Options for Coca-Cola

There are two other points that Coke could have made in its motion for reconsideration. First, Newlon’s selection of comparable companies was questionable, given their close relationships with Coke. To start, the bottlers were required to purchase product from Coke in their operations. That might not be entirely out of the ordinary for some unrelated-party relationships; however, Coke also brought bottlers within its control into what was known as the “bottler hospital” to increase their operational and financial performance.

When looking at those companies individually, many show a mutual dependence with Coke. For example, Coca-Cola FEMSA (one of the bottlers) uses Coke’s branding and logos in its website and annual report. Its annual report also states that Coke effectively owns about 33 percent of the voting rights in the company. Because the CPM requires selecting independent comparable companies, perhaps Coke should have explored this point further, including whether the bottlers could even be considered independent enough from Coke.

However, even if Coke had made that argument, it would have likely found itself subject to a different set of comparables, albeit under the CPM. That would still have greatly limited the amount of profit at the supply points and led to a large tax liability for Coke.

Another potential avenue for reconsideration is the CUT method. In discussing the reasonableness of the CPM analysis, the Tax Court left the door open in saying that “comparable external transactions involving high-value intangibles may occasionally exist.” One of Coke’s experts did propose using the CUT method. However, the Tax Court criticized that because the expert had only one comparable CUT license agreement and the amount of adjustments he needed to make to arrive at an end royalty rate led to that agreement being incomparable.

It would have made sense for Coke to explore that avenue further in its motion for reconsideration. Convincing the Tax Court of the comparability of one or multiple CUT agreements would have allowed the company to retain a similar structure whereby the supply points compensate it for use of intangible assets without needing to focus as much on whether the supply points owned valuable assets like that. Certainly, Coke would still need to convince the court that CUT is the best method, but as already mentioned by the Tax Court, there is precedent for arm’s-length pricing by CUT agreements for valuable IP.

Takeaways

The ruling offers many interesting takeaways for practitioners to keep in mind in their own intercompany structures. One of the most basic messages is that a closing agreement does not necessarily provide the tax certainty that an advance pricing agreement would. There is a gray zone in whether closing agreements might apply for future transactions, so they should clearly state that, instead of relying on implication by lack of specificity.

Further, just because the IRS did not challenge or sue you over an issue, that does not mean that it will not or cannot in the future. That may be obvious to many practitioners, but it is a point strongly argued by Coke. The company said that because the IRS had signed off on its approach in the past, the agency pulled the rug from underneath its feet. As the Tax Court pointed out, there has long been precedent that the mere fact that a taxpayer obtained a windfall in prior years does not entitle it to like treatment for other tax years.

Another takeaway is that taxpayers should take another look at their intercompany agreements and make sure they are in line with their tax strategies. Coke’s agreements made no mention of the 10-50-50 method. They also said Coke was the sole owner of all intangible assets and contracted directly with service companies to perform marketing-related activities under its direction and guidelines.

Although Coke argued that its contracts were inconsistent with the economic reality of the situation, the Tax Court ruled that only the IRS may set aside contract terms as being inconsistent with economic substance. It cited precedent that a taxpayer cannot apply the “substance over form” argument only after finding itself in an adverse situation.

The tax world now waits to see what the next step will be in this case because it could signal a changing tide for the IRS in transfer pricing matters.

FOOTNOTES

1 The Coca-Cola Co. v. Commissioner, 155 T.C. No. 10 (2020).

2 Ryan Finley, “Coca-Cola’s Reconsideration Motion: A Novel Take on Section 482,” Tax Notes Int’l, July 26, 2021, p. 425.

END FOOTNOTES

Copy RID