After *Liberty Global*: Technical Code Compliance No Longer Shields a Transaction From the Economic Substance Doctrine

Key takeaways

• The leading appellate precedent on codified Section 7701(o) is now adverse to taxpayers. A divided Tenth Circuit (Murphy, J., for the panel; Eid, J., dissenting) affirmed the District of Colorado and disallowed Liberty Global's roughly $2.4 billion Section 245A deduction, holding the doctrine is "relevant to attempts by taxpayers to mechanically utilize the provisions of the Tax Code to obtain a benefit not intended by Congress."
• Mechanical compliance is not a shield. The government conceded Project Soy technically complied with Title 26, and the court disallowed the benefit anyway. Pointing to the literal terms of the Code does not, by itself, take a transaction outside Section 7701(o).
• The unit of analysis is the entire integrated series — Section 7701(o)(5)(D) — not the favorable final step. A taxpayer cannot inoculate a plan by bundling "basic business transactions" such as Section 351 organizations or check-the-box elections inside it.
• The penalty exposure the doctrine triggers is strict-liability. A Section 7701(o) disallowance carries a 20% penalty under Section 6662(b)(6), rising to 40% for undisclosed transactions under Section 6662(i), with no reasonable-cause defense by operation of Section 6664(c)(2). Opinion letters do not avert it. (No penalty was litigated in *Liberty Global* itself; this is the regime any taxpayer faces in an ordinary deficiency posture.)
• Build the economic-substance file at deal time. Contemporaneous pre-tax return modeling and a business-purpose memorandum covering the whole structure — not just the final step — are now the price of defending an integrated transaction.

What the Tenth Circuit decided

*Liberty Global* arose from "Project Soy," a tightly integrated, four-step series Liberty Global, Inc. ("LGI") executed over a four-day period in December 2018. The plan was engineered to exploit what the court called a "last day of year rule/mismatch" in the international provisions of the 2017 Tax Cuts and Jobs Act (Pub. L. No. 115-97) — specifically, the timing seam between the GILTI regime of Section 951A and subpart F on one side and the new Section 245A dividends-received deduction on the other.

The mechanics, stripped to essentials: across the first three steps, LGI manufactured roughly $4.8 billion of earnings and profits in a controlled foreign corporation (TGH) using profit certificates and "springing-to-life" intercompany debt, accompanied by a change to the subsidiary's corporate form and capital structure through an entity-classification election. At step four, an LGI affiliate sold the TGH interest. Because sale proceeds of a CFC can be treated as a dividend to the extent of the CFC's earnings and profits under Sections 964(e)(1) and 1248(a), the manufactured E&P let LGI recharacterize a $2.4 billion gain as a dividend and offset the entire amount with a Section 245A deduction. LGI initially complied with Treasury's temporary regulation (Treas. Reg. § 1.245A-5T), reporting most of the E&P as GILTI and claiming a roughly $360 million deduction; it then amended its return to assert the regulation was invalid, report no GILTI, and claim the full $2.4 billion deduction, generating a refund claim of close to $110 million.

In the district court, *Liberty Global, Inc. v. United States* (ESD Order), No. 1:20-cv-03501-RBJ, 2023 WL 8062792 (D. Colo. Oct. 31, 2023), Senior Judge R. Brooke Jackson held the regulation procedurally invalid but disallowed the deduction under Section 7701(o). Critically, LGI conceded in discovery that the first three steps "did not change, in a meaningful way, LGI's economic position" and "served no substantial non-tax purpose." On appeal, LGI's sole argument was that the doctrine was not "relevant" to Project Soy at all.

Mechanical compliance does not take a transaction outside the doctrine

The Tenth Circuit rejected that framing. The panel held that Section 7701(o) "is relevant to attempts by taxpayers to mechanically utilize the provisions of the Tax Code to obtain a benefit not intended by Congress. That is exactly what LGI did with Project Soy." The court grounded the result in long-settled circuit law that transactions which "comply with the literal terms of the tax code" can still be disregarded if they are mere tax-avoidance schemes (citing *Blum v. Comm'r*, 737 F.3d 1303 (10th Cir. 2013), and *Sala v. United States*, 613 F.3d 1249 (10th Cir. 2010)). The government's concession that Project Soy mechanically complied with Title 26 did not save it.

The integrated series is the unit of analysis

For deal practitioners, *what* the court tested matters as much as how it ruled. Section 7701(o)(5)(D) defines "transaction" to "include a series of transactions," and the court held the appropriate unit of measurement was "the entirety of Project Soy," not the favorable step four standing alone. LGI argued that because the plan was partly composed of "basic business transactions" — Section 351 corporate organizations, entity selections, capitalization choices identified in the doctrine's legislative history — the doctrine could not apply. The court refused to let a taxpayer "inoculate" a complex transaction by salting it with otherwise ordinary steps, warning that the contrary rule would invite taxpayers to "quickly find ways" around Section 7701(o). It expressly declined to decide whether any particular basic transaction, standing alone, is exempt — leaving that question for another case — but held it does not matter when the step is one gear in a tax-motivated machine.

The dissent: a threshold "relevance" question the majority did not resolve

Judge Eid dissented. In her view, the prefatory clause of Section 7701(o)(1) — "[i]n the case of any transaction to which the economic substance doctrine is relevant" — requires a *threshold relevancy inquiry* separate from the two-prong test, and reading "relevant" out of the statute renders multiple Code provisions superfluous and contradicts decades of precedent. The disagreement matters: it flags that the precise scope of the codified doctrine, and how courts should screen for "relevance" before applying the two prongs, remains genuinely contested. Taxpayers in other circuits — and any subsequent en banc or Supreme Court review — will return to exactly this seam.

What it means for M&A and multinational structuring

*Liberty Global* removes a comfortable assumption: that a structure assembled from recognized, individually compliant Code provisions is safe because each piece checks out. The defensible question is no longer "does each step comply?" but "viewed as a whole, does the plan change the taxpayer's economic position in a meaningful way apart from tax, and is there a substantial non-tax purpose for doing it?" Those are the conjunctive prongs of Section 7701(o)(1)(A) and (B) — both must be satisfied — and the analysis runs against the entire series.

The practical exercise at deal time is to walk a proposed structure through that two-pronged screen before signing, with the integrated transaction as the unit, and to identify where the pre-tax economics and the business rationale actually live. The figure below frames that decision path.

The stakes turn on Section 6662(b)(6), enacted alongside Section 7701(o) by Section 1409 of the Health Care and Education Reconciliation Act of 2010 (Pub. L. No. 111-152), effective for transactions entered into after March 30, 2010. A disallowance under the economic substance doctrine carries a mandatory 20% accuracy-related penalty, increased to 40% where the transaction's relevant facts are not adequately disclosed on the return (Section 6662(i)). Section 6664(c)(2) removes the reasonable-cause-and-good-faith defense for that penalty. The consequence is blunt: a well-reasoned tax opinion can support a non-economic-substance position, but it cannot defeat the penalty once the doctrine applies. *Liberty Global* itself was a refund suit in which no penalty was asserted — but it widens the gate to the disallowance that triggers the penalty in the ordinary deficiency case.

Pointing to the literal terms of the Code is no longer an answer; the answer is a contemporaneous record that the deal made commercial sense without the tax result.

That reframes the deal-team checklist. Pre-tax return modeling should be run and retained for the structure as executed, showing meaningful profit potential independent of tax attributes. Business-purpose memoranda should address the integrated plan, not just the step that produces the deduction or exclusion. And representations and warranties in transaction documents should be re-papered to allocate economic-substance and disclosure risk between the parties — including disclosure covenants pointed at the 40% rate — rather than leaving that exposure to fall by default.

Frequently asked questions

Does *Liberty Global* mean Section 351 contributions or check-the-box elections are now risky?

No — used for genuine business reasons, they remain ordinary and unremarkable. The holding is narrower and more pointed: those provisions are not *safe harbors* that exempt a larger plan from Section 7701(o) simply because they appear within it. The court tested Project Soy as a whole and declined to exempt the series merely because it incorporated steps that, in isolation, might be "basic business transactions." Treat such provisions as in-scope building blocks of the integrated analysis, not as a shield around it.

We have a tax opinion at the "more likely than not" or "should" level. Doesn't that protect us from penalties?

Not from the economic substance penalty. Section 6664(c)(2) denies the reasonable-cause-and-good-faith defense for the Section 6662(b)(6) penalty, so an opinion that would otherwise establish reasonable cause does not abate the 20% (or 40%) charge once a transaction is found to lack economic substance. An opinion still has value in framing the position and in non-economic-substance penalty contexts, but it does not neutralize this specific exposure.

Is this binding outside the Tenth Circuit?

It binds within the Tenth Circuit and is, as of April 2026, the leading appellate decision construing codified Section 7701(o). Other circuits are not bound, and Judge Eid's dissent identifies a live "threshold relevance" question that could yet divide the courts. But examiners and Appeals nationally will cite it, and prudent planning should assume its logic — test the whole, not the step — applies wherever a transaction could be litigated.

What is the single most important documentation step?

A contemporaneous, deal-time file built around the *entire* structure: pre-tax economic modeling showing meaningful profit potential apart from tax, paired with a business-purpose memorandum that explains the non-tax rationale for the integrated plan. Documentation created after an examination opens carries far less weight than a record assembled when the deal was negotiated.

Bottom line

If you are structuring or signing an integrated, multi-step transaction in 2026, step-by-step Code compliance is no longer a defense. Before closing, run the deal through the conjunctive Section 7701(o)(1) test as a whole — meaningful pre-tax change in economic position and substantial non-tax purpose — and build the contemporaneous file (pre-tax modeling plus a business-purpose memorandum) at that point, not later. Re-paper representations, warranties, and disclosure covenants to allocate the economic-substance and 40%-rate risk between the parties. Where a structure cannot clear both prongs on the integrated facts, the right move is to redesign or disclose, because the Section 6662(b)(6) penalty that *Liberty Global* makes easier to trigger cannot be cured by an opinion after the fact.