The Tax Times: Liberty Global, Check-the-Box, and the Economic Substance Doctrine: What Taxpayers Need to Know After Project Soy

In Liberty Global Inc. v. United States, the Tenth Circuit didn’t just apply the economic substance doctrine to a sophisticated cross-border structure—it walked straight through a check-the-box election that the taxpayer tried to treat as sacrosanct. The case is a shot across the bow for taxpayers relying on classification elections to manufacture earnings and profits (E&P) and access the section 245A deduction without corresponding economic change or substance.

This post unpacks how the check-the-box regulations featured in “Project Soy,” what Liberty Global argued, how the courts responded, and what that means for planning going forward.

The role of check-the-box in Project Soy

Liberty Global’s “Project Soy” was a multi-step internal restructuring designed to create artificial E&P and a deemed dividend to support a section 245A deduction and an associated refund claim. At the heart of that design was an entity classification change under the check-the-box regulations.

One of the key steps in Project Soy was a check-the-box election that changed the tax classification of a foreign affiliate.
That election triggered deemed transactions that, as modeled, generated the E&P needed to support a large section 245A dividends received deduction.
In effect, the classification change was the mechanism used to turn internal group movements and latent attributes into the “dividend” Liberty Global wanted to treat as eligible for the 100% DRD.

Liberty Global tried to isolate this step and characterize it as a pure tax classification election that, by design, has no non-tax legal effect and therefore should be analyzed differently—or not at all—under the economic substance doctrine.

Liberty Global’s theory: check-the-box as a safe zone

Liberty Global and supporting amici advanced a theory that will sound familiar to anyone who has used check-the-box in planning:

Check-the-box is “all form and no substance.”
The regulations allow taxpayers to elect how eligible entities are classified for federal tax purposes, with the election often having no collateral legal or commercial consequence. From the taxpayer’s perspective, that means the election is pure tax “form,” yet it has always been respected when properly made.
Classification choices are “basic business transactions.”
The legislative history to section 7701(o) suggests that the economic substance doctrine is not meant to police “basic business transactions” where Congress clearly allows taxpayers to choose among alternatives (debt vs. equity, corporate vs. partnership, etc.) based on tax considerations. Liberty Global tried to place check-the-box in that protected category.
Relevance and the “unit of analysis.”
Liberty Global pushed a narrow view of what counts as the “transaction” for economic substance purposes. The idea was:

The relevant “transaction” is the check-the-box election itself.
Because that election is a permissible regulatory choice that Congress and Treasury intended to be tax-driven, section 7701(o) is not “relevant” to it.
If the court adopts that framing, the deemed steps and resulting E&P sit behind a sort of shield: the election is respected, and the government cannot invoke economic substance to disregard the tax consequences.

In other words, the taxpayer wanted the court to treat check-the-box as a quasi–safe harbor: if the regulations authorize the election and you follow the rules, the resulting attributes should not be second-guessed under section 7701(o).

How the courts actually treated the check-the-box step

Both the district court in Colorado and the Tenth Circuit refused to treat the check-the-box election as immune from economic substance review.

No threshold carve-out for classification elections

The district court rejected the idea of a special “relevance” gate that would pre-screen certain categories of transactions—like entity classification choices—out of the doctrine. Instead, it effectively treated the statutory relevance requirement as bound up with the two-prong test:

If a step is part of a transaction designed to generate a tax benefit, and that overall transaction fails the objective/subjective tests of section 7701(o), the doctrine is “relevant.”
There is no separate, front-end inquiry that says, “check-the-box elections (or similar steps) are off limits to economic substance.”

That framing matters, because it prevents taxpayers from using labels like “election” or “classification” to avoid scrutiny.

The “transaction” is the integrated project, not one step

The court also rejected Liberty Global’s attempt to define the “transaction” narrowly as the check-the-box step that produced the E&P:

It treated Project Soy as a single, integrated transaction, analyzing the steps in the aggregate rather than isolating the election.
The fact that the check-the-box step was the one that technically generated E&P did not make it the only relevant transaction; it was one link in a deliberately designed chain.

This approach is consistent with longstanding economic substance case law: when the taxpayer orchestrates a multi-step structure to achieve a specific tax result, courts look at the plan as a whole rather than elevating one formalistic step.

No “basic business transaction” shelter for Project Soy

On the legislative history point, the courts drew a line between:

Routine, economically grounded classification choices (e.g., choosing to operate as a corporation vs a partnership because of liability, investor expectations, or regulatory constraints, even if tax is a major factor); and
Highly engineered, year-end structures designed primarily to manufacture tax attributes (E&P, dividends, basis, etc.) without changing real-world operations or risk.

Project Soy landed firmly in the second category. The courts were comfortable treating it as a tax shelter rather than “basic tax planning,” even though it used familiar tools like check-the-box and section 245A.

What Liberty Global signals about check-the-box going forward

Liberty Global does not say that check-the-box elections are always subject to economic substance or that entity classification elections are inherently suspect. But it sends several clear messages about how courts may view these elections in complex planning:

1. Classification is respected; manufactured attributes are not necessarily

Courts will still respect a valid election as a matter of entity status—if you elect to treat a foreign subsidiary as disregarded, the classification itself holds. The question is what happens next:

The deemed transactions and tax attributes (E&P, basis, gain, loss) that follow from the election can be brought within the economic substance analysis if they are part of an overall scheme that lacks real economic change and non-tax purpose.
In Project Soy, Liberty Global’s own concessions about the lack of economic substance in key steps made that an easy conclusion for the court once it looked at the transaction as a whole.

2. Check-the-box is not a per se “basic business transaction”

The case undercuts the idea that classification elections are automatically protected by the “basic business transaction” language in section 7701(o)’s legislative history:

When a check-the-box election is used in a straightforward way to align tax treatment with business reality (for example, ignoring an entity to match a single, integrated operating business), it is more likely to fit comfortably within that carve-out.
When the election is used as part of a layered, advisor-driven project whose main purpose is to create tax-favorable attributes (like “manufactured” E&P) and extract value without corresponding economic income, it is more likely to be swept into an economic substance challenge.

Liberty Global sits at the aggressive end of that spectrum, and that’s exactly why the court was willing to disregard the tax benefit.

3. The “unit of analysis” question just got more important

For planners, the key practical question is: what is the transaction a court will analyze under section 7701(o)?

Liberty Global suggests that:

Courts will favor an integrated view of multi-step planning, particularly when the taxpayer has treated the steps as part of a named internal “project” with a modeled tax outcome.
Taxpayers will have an uphill climb if they try to isolate a single check-the-box election, or similar election, and argue it should be analyzed independent of the broader structure.

This makes it more important to think about how a series of steps will look as a cohesive transaction, not just whether each element technically meets regulatory requirements in isolation.

Practical takeaways for tax planning and controversies

For practitioners using check-the-box elections in cross-border planning, Liberty Global offers several practical lessons:

Document real non-tax purpose where it exists.
If a classification election is part of a genuine business restructuring—consolidating operations, streamlining regulatory footprints, integrating cash management—build that record contemporaneously. It will be critical if the structure is later framed as “Project X” in a controversy.
Be wary of structures that exist almost entirely on paper.
The more the structure’s effects are confined to the tax balance sheet (E&P, basis, DRDs, timing) without operational change, the more it will look like Project Soy to a court.
Assume sophisticated planning can be viewed as a single “transaction.”
If the design involves multiple affiliated entities, rapid or year-end changes in classification, circular flows, and a model focused on tax outcomes, expect a court to treat it as an integrated transaction for economic substance purposes.
Don’t rely on “it’s in the regs” as a shield.
That a result technically follows from the check-the-box regulations, section 245A, or similar rules is necessary but no longer sufficient comfort in high-stakes planning. Courts are increasingly willing to apply section 7701(o) to override formal compliance where they view the result as inconsistent with anti–base erosion policy.
In controversy, think carefully about how you define the “transaction.”
How you frame the transaction in protests, briefs, and expert reports matters. Liberty Global shows that trying to define the transaction too narrowly (e.g., “only the election”) risks losing credibility with courts inclined to see the bigger picture.