Menu
Tax Notes logo

Economic Substance Doctrine: Still Giving Perrigo Heartburn?

Posted on July 26, 2021
Christopher Yan
Christopher Yan
Benjamin Alarie
Benjamin Alarie

Benjamin Alarie is the Osler Chair in Business Law at the University of Toronto and the CEO of Blue J Legal Inc. Christopher Yan is a senior legal research associate at Blue J Legal.

In this article, Alarie and Yan use machine-learning models to evaluate the strength of the government’s economic substance arguments in the pending Perrigo case.

Copyright 2021 Benjamin Alarie and Christopher Yan.
All rights reserved.

I. Introduction

Tax practitioners frequently struggle when providing advice on some cases in areas of tax law that have limited statutory guidance and a sizable body of case law. The reason is that the case law is often internally inconsistent; the courts can generate obfuscating noise. Sometimes the signal in the accumulated case law is relatively faint. For those studying hundreds of cases, the noisiness of a complex web of interacting considerations makes it difficult to assess the importance of any given factor in a case. Fortunately, advances in computing power and machine learning offer an opportunity to amplify the signal and diminish the noise to better understand the law.

In last month’s inaugural installment of Blue J Predicts, we examined whether an appeal of a Tax Court decision to the D.C. Circuit would be dismissed on the issue of whether a partnership exists.1 This month’s installment evaluates the strength of the commissioner’s economic substance arguments in the pending Perrigo2 case, which was heard before the U.S. District Court for the Western District of Michigan in a nine-day bench trial that concluded on June 7. Post-trial submissions have not yet been filed by the parties, and proposed findings of fact and conclusions of law were unavailable at the time of writing.

Although there are multiple issues at play in Perrigo, here we focus primarily on the economic substance arguments advanced by the government and the taxpayer. In addition to defending its transactions as having economic substance, the taxpayer takes the position that the application of the economic substance and sham doctrines is improper in the circumstances, and the court should focus solely on remedying any alleged deficiencies through section 482 transfer pricing mechanisms designed to deal with disputes of this nature. Although it remains to be seen whether the court will bypass what would otherwise be a significant issue in cases like this one, any discussion of the issue by the court will necessitate some commentary on the nature of the transactions and whether they meet some de minimis threshold to overcome the government’s sham doctrine and economic substance attacks.

Unlike an appeal decision, parties before a court of first instance do not have the benefit of presumed findings of fact or conclusions of law. Instead, parties must contend with the uncertainties regarding sufficiency of evidence and vastly different characterizations of material facts. This makes understanding the legal significance of each factor more important and allows tax practitioners to focus their arguments on characterizations likely to matter most to courts.

Based on only the information available leading up to trial and without regard to the sufficiency of evidence, Blue J’s algorithm predicts with 68 percent confidence that a court will find that the transactions in question will survive an economic substance challenge. This prediction does not, however, speak to the extent to which the taxpayer’s income may be subject to recharacterization under section 482 transfer pricing rules. The economic substance issue could go either way, especially since the merits of the case turn on disputed material findings of fact and conclusions of law that are unavailable.

II. Background and History

The dispute involves a $163.5 million tax refund suit filed by Perrigo Co. and Subsidiaries (Perrigo) against the government. The government had reallocated income from an Israeli affiliate limited liability company (Perrigo Israel LLC) to Perrigo’s U.S.-based business (Perrigo U.S.). Since Perrigo Israel LLC was intended to be a disregarded entity for U.S. tax purposes, all assets, liabilities, and activities of Perrigo Israel LLC were to be treated as assets as part of Perrigo Israel Trading Limited Partnership (Perrigo Israel LP). For ease of reference, the collective activities of the Perrigo Israel LLC/LP will be jointly referred to as Perrigo Israel. The underlying revenue stream that generated the tax liability came from U.S.-based sales of omeprazole, a store-brand generic drug used to treat heartburn from acid reflux.

In August 2005 Perrigo U.S. entered into an agreement with an unrelated Israel-based pharmaceutical company (Dexcel) for the sale and distribution of omeprazole in the United States. In November 2006 the Dexcel agreement was transferred from Perrigo U.S. to the newly formed Perrigo Israel, which took on the rights and obligations of the Dexcel agreement by way of assignment. Perrigo U.S. received some consideration for the assignment. Perrigo Israel then entered into a contract with Perrigo U.S. (intercompany contract), under which Perrigo Israel would supply the product to Perrigo U.S. for distribution in the United States in exchange for a return on its distribution activities and a percentage of the residual profit.

Figure 1. Before November 2006 Assignment + Intercompany Contract
Figure 2. After November 2006 Assignment + Intercompany Contract

In other words, Perrigo Israel was interposed between Dexcel and Perrigo U.S. Perrigo Israel became responsible for fulfilling the terms of the Dexcel agreement on behalf of Perrigo (that is, purchasing omeprazole from and paying expenses to Dexcel). Perrigo Israel also acted as a supplier by reselling the drug to Perrigo U.S. Through this process, Perrigo Israel captured some of the income from the sale of the drug.

Based on the terms of the assignment and the intercompany contract, a substantial amount of the income from the sale of omeprazole in the United States was recorded as being earned by Perrigo Israel, so the income escaped U.S. taxation. As a result, the government reallocated nearly all of Perrigo Israel’s income to Perrigo U.S. for tax years 2009-2012, arguing that Perrigo Israel was merely an “overseas shoebox” that lacked economic substance. Perrigo applied for a refund, arguing that the tax implications that arise out of its global structuring of affiliate companies, relevant transactions, and allocation of revenue from sales should all be respected.

III. Government’s Pretrial Position

The government supports its reallocation of income by claiming that: (1) Perrigo Israel is a sham entity that should be disregarded for tax purposes; (2) the transactions involving Perrigo Israel that arose out of the assignment and intercompany contract lacked economic substance; (3) Perrigo improperly assigned income among affiliated entities; and (4) alternatively, the reallocation of income conforms with the commissioner’s reallocation powers under section 482 to accurately reflect the income of related parties. The first three issues invoke common law substance-over-form doctrines. The last issue is statutory in nature and asks whether the commissioner’s reallocation was arbitrary and capricious. There are also penalties that will not be discussed in this article. Here, we will focus primarily on the economic substance analysis.

Although conceptually related, the sham and economic substance doctrines are distinct. The sham doctrine looks at the legitimacy of the entity, while the economic substance doctrine looks at whether the transactions meaningfully changed the taxpayer’s economic position and existed for a purpose other than tax avoidance.

A. Disregarding Entity as a Sham

The government claims that Perrigo’s case rests on the unsupportable contention that most of Perrigo’s omeprazole income was not earned in the United States because of an assignment to the undercapitalized, offshore Perrigo Israel affiliate for a mere lump sum price of $877,832 (despite earning hundreds of millions of dollars in revenue and profits from that assignment). The government characterizes Perrigo Israel as a sham entity without employees, facilities, or tangible assets. The government contends that the corporate form may be disregarded when it is a sham or unreal and that the paperwork and bank accounts created to further Perrigo’s tax scheme do not amount to substantive business activity.

The government further argues that most of the activity resided in, and never left, the United States, where Perrigo U.S. generated purchase orders, sourced supplies, monitored inventory levels, evaluated supply chain risks, created customized artwork, ran production lines, mass-customized tablets, marked and sold the tablets, created promotional materials, handled delivery logistics, and performed all accounting and finance functions to track the activity. Meanwhile, Perrigo Israel merely held “flash title” to the omeprazole drugs before immediately reselling the tablets to Perrigo U.S. at a substantial markup, which resulted in earnings to Perrigo exceeding $211.6 million during the years at issue.

B. Lacking in Economic Substance

To demonstrate that transactions have economic substance, they must meet both prongs of a conjunctive test. The first prong is objective: A transaction must have practicable economic effects aside from tax benefits, and the nontax benefits must exceed the costs and must be appreciable. The second prong is subjective: Taxpayers must show that they were motivated by profit to participate in the transaction. Although not overtly referenced, this formulation of the economic substance test is similar to the statutory economic substance provision introduced in 2010 under section 7701(o).

There is also a potential limitations issue raised by Perrigo. But the government argues that the statute of limitations does not bar the IRS from recharacterizing, adjusting, recomputing, or even disregarding transactions from closed years (2007-2008) for the purposes of assessing tax on income earned during the years in dispute (2009-2012).

The government also argues that Perrigo Israel performed no functions, had no operational assets, and was without equity, and therefore was unable to mitigate purported risks. Perrigo Israel was not capitalized until March 2008, and although it was eventually capitalized by another related U.K. affiliate, the U.K. affiliate’s liability was limited to a nominal sum of less than $100. Ultimately, the government claims that the “evidence will prove that [Perrigo Israel] existed as a paper entity in a corner of the company, an uninhabited waystation for funneling domestic income overseas.”

C. Income Was Improperly Assigned

The government also invokes the assignment of income doctrine and asserts that Perrigo U.S. improperly assigned income that it earned to Perrigo Israel. The assignment of income doctrine states that if an assignor retains sufficient power and control over the assigned property or over receipt of the income, it is reasonable to treat the assignor as recipient of the income for tax purposes. Here, Perrigo U.S. retained dominion and control over the Dexcel contract it assigned to Perrigo Israel.

IV. Perrigo’s Pretrial Position

Perrigo frames the issue as the government’s attempt at overreaching on two fronts. First, Perrigo claims that the government is acting in hindsight, after omeprazole’s unanticipated levels of success in the marketplace, when reassessing the tax years in dispute from 2009-2012. In Perrigo’s eyes, the government’s position is especially problematic because the IRS conducted 2007 and 2008 tax year audits that left the assignment undisturbed, and the government is now going back to challenge transactions in closed years to recharacterize transactions in open years. Second, Perrigo takes the position that an application of the common law sham and economic substance doctrines is improper in the circumstance and that the court should instead apply section 482, which exists primarily to deal with the related-party transactions at issue.

A. Improper Application of Doctrine

Perrigo asserts that the application of the economic substance and sham doctrines is reserved for the creation of artificial tax deductions and other unintended benefits and should not be applied to restructuring between related business parties that are engaging in otherwise legitimate transactions that form a part of Perrigo’s core business. Perrigo cites various authorities for the propositions that taxpayers are (1) permitted to choose low-taxed foreign locations for income-producing assets and business opportunities; (2) restructure income-producing transactions to relocate income offshore; (3) use or assign contracts to shift income; and (4) structure real business transactions even for the sole purpose of avoiding U.S. tax, albeit all subject to arm’s-length pricing under section 482.

Perrigo Israel conducted real business activities, such as purchasing and selling Dexcel’s omeprazole after launch, borrowing money from and lending money to the U.K. affiliate, making payments to other parties for services, and transacting with other third-party entities in its own right. These are all bona fide business activities, and the formation of Perrigo Israel, the assignment, and the intercompany contract were all part of an international business strategy that resulted in a change in economic positions within its business. Based on the cases cited by Perrigo, the minimal capital, the lack of operational employees, and the existence of parental guarantees are not sufficient indicators for the government to invoke the sham and economic substance doctrines.

In essence, Perrigo claims that the deficiencies raised by the government are not the type that warrant disregarding otherwise legitimate entities altogether. To apply the sham and economic substance doctrines, rather than remedy any deficiencies through section 482 (which is designed to address circumstances such as this one), would be akin to “cracking walnuts with a sledgehammer.” In making these arguments about the types of activities that Perrigo Israel was engaged in, Perrigo is also suggesting that both the arrangements and the transactions have enough economic substance to survive proceeding to a section 482 analysis.

B. Timing of Audits

Perrigo contends that for the tax years in dispute between 2009 and 2012, Perrigo Israel could not have been a sham because it engaged in real economic transactions. Perrigo claims that the government is trying to recharacterize Perrigo Israel as a sham in tax years that were previously subject to an audit and left unchallenged (2007-2008 tax years) and now import the effects of findings from closed tax years into the 2009-2012 tax years in dispute. In taking this position, Perrigo also continued to maintain that the pre-2009 transactions had economic substance.

V. Applying Machine Learning

Blue J uses machine learning to assess and model a data set of approximately 250 cases that invoke the economic substance doctrine. At a high level, most of these factors concern information about the origin of the transaction, the structure of the transaction, the economic effects of the transaction, the effects of the transaction other than profit, and the taxpayer’s risk level — collectively, those broad categories are represented by 25 factors.

Among the 25 factors, Blue J’s data reveal two variables that are particularly salient for courts and seem to be the most influential in determining the results in economic substance doctrine cases: (1) transaction costs and (2) pretax profits. Both factors are hotly contested by the parties in Perrigo based on varying characterizations of the facts. Before we examine how the parties’ positions differ on these factors, we will go through what these factors represent.

Transaction Costs: Courts are interested in whether a taxpayer could have fulfilled the transaction’s ostensible business purpose for a lower amount of transactional costs (for example, commissions and professional fees). If there were cheaper ways of achieving the alleged business purpose, it is less likely that a transaction has economic substance.

Pretax Profits: Courts have demonstrated an interest in whether a taxpayer expects a pretax profit from a transaction. The expectation of a pretax profit, particularly if higher than the expected tax benefit, makes it empirically more likely that a court will conclude that a transaction has economic substance.

VI. Assessing Points of Contention

First and foremost, Perrigo and the government appear to disagree on the unit of analysis under which the economic substance analysis ought to be performed. Under the first branch of the economic substance test, the transaction must produce practicable economic effects, aside from tax benefits.

The government argues that, when examined as a whole, conducting the transactions through the assignment and the intercompany contract had no practicable economic effects; on the contrary, it ended up as a drain on Perrigo overall because of the transaction costs and fees it had to pay (relative to if Perrigo U.S. had engaged in the transactions directly with Dexcel, as set out in the original Dexcel agreement). The government relies on several authorities in support of the proposition that the series of transactions must be viewed as a whole, rather than narrowly with a focus on how Perrigo Israel performed individual components of a scheme. In other words, the government is engaging in a unit of analysis that considers the total effect of the assignment, intercompany contract, and transactions in dispute.

Perrigo engages in a different unit of analysis: the individual transaction level. Perrigo argues that each transaction Perrigo Israel engages in has practicable economic effects, including the buying, selling, lending, and distribution of profits, as well as subsequent commercial agreements that Perrigo Israel enters into on its own behalf (with a separate unrelated third party regarding the drug product fexofenadine). In other words, the question is not only whether the creation and use of an Israeli entity served a business purpose but also whether the Israeli entity engaged in activities that served a business purpose. On this view, Perrigo Israel cannot be found to be a sham, because it is engaged in substantive transactions rather than the artificial tax shelters that these doctrines are meant to attack.

Applying this point of contention to one of the most salient factors — pretax profits — the government would argue that the transactions that flow from Perrigo’s assignment and intercompany contract lack economic substance, because conducting the transactions through this form does not generate any additional pretax profits (and is, in fact, more costly than if they had not taken this form). Conversely, Perrigo would argue that the purchase and sale of drugs, as well as its other activities, such as lending and entering into other contracts, were conducted with generating pretax profits in mind (and in fact, they did generate tremendous pretax profits). Thus, the framing of the unit of analysis can be determinative to the outcome.

Overall, the question of which unit of analysis is proper to assess whether transactions have economic substance is an important one. Is economic substance determined on the level of the aggregated transactions (in which the net effect of corporate restructuring must be designed to generate nontax economic benefits)? Or is it permissible to engage in costly corporate restructuring if the types of transactions that the entities engage in are profit-seeking economic activities?

Second, Perrigo rebuts the government’s contention that the contracts, as a whole, are without economic effects, instead arguing that they form a part of a series of restructurings that align with its larger global expansion strategy. This goes to the heart of the subjective element of the test, which is that Perrigo intended to engage in profitable economic endeavours, and the arrangements are a part of this overall global strategy.

Perrigo claims that its global plan involved having affiliates that would acquire contracts and assume risks associated with the products at issue, as part of a more efficient international operating structure. Moreover, the U.K. affiliate that eventually capitalized Perrigo Israel functioned as an offshore finance and holding company. Perrigo would likely assert that global context should be considered when assessing the appropriateness of transaction costs. Overall, Perrigo would argue here that the transaction costs for arranging its affairs in this manner could not have been lower, considering Perrigo’s desire to further its ability to exploit and deploy its intellectual property globally to achieve the full synergies of its combined operations.

The government, on the other hand, argues that Perrigo Israel, the assignment, and the intercompany contract were all unnecessary parts of an abusive tax-saving scheme. No risk-shifting existed, and on the contrary, Dexcel required Perrigo U.S. to provide a “performance guaranty.” Further, Dexcel exacted a quarterly fee from Perrigo, incurring unnecessary transaction costs along the way to funnel its business through the LLC. In terms of other market risks, the government maintains that Perrigo Israel brought nothing to the table to mitigate any risks. Therefore, the government would argue here that any transaction costs that were incurred in service of these goals are more than Perrigo should have paid if these transactions were legitimate.

Applying those two factors alone to Blue J’s machine-learning algorithm reveals the impact each factor has in the circumstances of the case (holding the other 23 factors constant):

Table 1. Effect of Factors

 

Pretax Profits?

Transaction Costs

Result

Scenario 1

No

Could have been lower

85% no economic substance

Scenario 2

Yes

Could have been lower

76% no economic substance

Scenario 3

No

Could not have been lower

60% no economic substance

Scenario 4

Yes

Could not have been lower

68% economic substance

From a review of the results, it is clear that the algorithm does not assign static weights to either factor, but adapts those factors dynamically based on how other factors are interpreted by the courts.

An examination of these two factors is just one example of how practitioners can use a machine-learning algorithm to test the range of possibilities based on a scenario. It is important to note that the above range is not even close to the entire spectrum of possibilities based on the differing pretrial positions set out by the parties in Perrigo.

In fact, other less impactful disagreements between the parties are borne out in the briefs, creating a larger predictive range. One can imagine that leading up to trial (before testimony and evidence is presented), parties will disagree on considerably more than two major factors. Although the number and extent of these differences begin to narrow as the evidence is heard, the most impactful factors likely remain in dispute. This is when having access to even proposed findings of fact and conclusions of law can assist with distilling the nature of the dispute into the most determinative and high-impact factors.

Although the perception of variability can create some uncertainty at the extreme ends of the government’s most favorable position (85 percent confidence of no economic substance) and the taxpayer’s most favorable position (68 percent confidence of economic substance), this range is much narrower in reality, as practitioners exercise their judgment and expertise in determining the likelihood that a characterization may succeed. Moreover, using an algorithm to quantify the impact of disagreements between the parties illustrates the significance of marshaling the appropriate evidence to support a characterization, and it conserves resources and limited space to advance arguments and characterizations that matter most to courts.

VII. Conclusion

Although we began this installment of Blue J Predicts with a prediction of 68 percent confidence that Perrigo’s arrangement would narrowly survive an economic substance challenge, a more detailed understanding and nuanced discussion of the points at issue between the parties demonstrate that Perrigo (or any case before the courts) cannot be entirely reflected in the prediction’s confidence. Instead, the principal value in using an algorithm is derived from using a heartier, data-backed way to quantify the risk faced by parties, as well as the varying importance of each factor in the circumstances of a case.

FOOTNOTES

1 Benjamin Alarie, Bettina Xue Griffin, and Christopher Yan, “An Unprofitable Pretax Venture Can Still Be a Partnership,” Tax Notes Federal, June 21, 2021, p. 1951.

2 Perrigo Co. v. United States, No. 1:17-cv-00737 (W.D. Mich. 2021).

END FOOTNOTES

Copy RID