How Leon Max, Little Sandy Coal, and Phoenix Design Group Redefined Qualified Research Under IRC Section 41

An in-depth analysis of Leon Max v. Commissioner (T.C. Memo. 2021-37), Little Sandy Coal Co. v. Commissioner, and Phoenix Design Group, Inc. v. Commissioner, and how these decisions established a new evidentiary baseline for the federal Credit for Increasing Research Activities under IRC Section 41.

The federal Credit for Increasing Research Activities, codified under Internal Revenue Code (IRC) Section 41, has historically functioned as a cornerstone of American industrial policy, designed to foster domestic innovation through fiscal incentives. However, the operational reality of claiming this credit has undergone a profound transformation over the last several years, moving away from a regime characterized by flexible “estimated” claims toward a highly scrutinized evidentiary environment. This shift is most prominently exemplified by the United States Tax Court’s decision in Leon Max v. Commissioner (T.C. Memo. 2021-37), a case that has redefined the boundaries of “qualified research” for creative and design-centric industries. The implications of this case, alongside the findings in Little Sandy Coal Co. v. Commissioner and Phoenix Design Group, Inc. v. Commissioner, have established a new baseline for the “hard science” requirement and the necessity of contemporaneous, activity-level documentation to withstand Internal Revenue Service (IRS) audits.

The Statutory Architecture and the Evolution of the Four-Part Test

To evaluate the impact of recent jurisprudence, one must first dissect the statutory framework of Section 41. The credit is not awarded for general creativity or business risk but is contingent upon satisfying a rigorous four-part test applied to each “business component.” A business component is defined as any product, process, computer software, technique, formula, or invention held for sale, lease, or license, or used by the taxpayer in its trade or business. For a plain-language walkthrough of how these elements interact, see the FAQ on explaining the four-part test for R&D tax credits in simple terms.

The Four-Part Test for Qualified Research Under IRC Section 41

The interaction between these tests creates a hierarchical barrier to entry. The Section 174 test serves as the gateway, requiring that the taxpayer encounter an uncertainty regarding the capability, method, or appropriate design of the business component at the outset of the project. In recent years, the IRS and the courts have increasingly narrowed the definition of “uncertainty,” distinguishing it from common design challenges or routine troubleshooting that can be resolved through standard industry expertise. This narrowing of the “uncertainty” window is a primary driver of the disallowances seen in contemporary litigation. Practitioners evaluating the statutory overlap should also review the FAQ on the difference between the Section 41 credit and Section 174 expenses.

Deep Dive: Leon Max v. Commissioner and the Fashion Industry Challenge

The case of Leon Max v. Commissioner (T.C. Memo. 2021-37) represents a pivotal moment in the history of the R&D tax credit, particularly for industries that blend aesthetic design with technical execution. Leon Max, Inc. (LMI), a high-end apparel company, sought to claim approximately $750,000 in research credits for the 2011 and 2012 tax years based on its extensive garment design and pre-production processes. The company argued that the iterative process of turning a conceptual sketch into a wearable, mass-produced garment involved resolving significant technical uncertainties related to fabric behavior, draping, and structural integrity. Apparel claimants should compare these holdings with the FAQ on apparel and fashion industry R&D tax credits.

Facts of the Case and the Alliantgroup Study

LMI’s first success in the industry was born of creative constraint; Leon Max utilized narrow widths of undyed poplin originally meant for typewriter ribbons to create highly successful small-sized jumpsuits, eventually netting $1 million in the company’s first year. By the period in question (2011–2012), the company had evolved into a sophisticated operation, producing monthly collections and private labels like “Chelsea and Violet” for retailers such as Dillard’s.

In 2013, LMI engaged a tax consulting firm, Alliantgroup, to conduct a formal R&D credit study covering the tax years 2009 through 2012. Alliantgroup employed a sampling methodology, examining 35 specific garments and concluding that 32 of them involved activities qualifying for the credit. This study formed the basis for LMI’s amended returns and the subsequent litigation when the IRS disallowed the credits in their entirety. Taxpayers considering a similar path should review the FAQ on amending past tax returns to claim missed R&D credits.

The Pre-Production Process and “Fit Testing”

The taxpayer’s claim rested on a methodical pre-production cycle that LMI characterized as a process of experimentation. After conceptualizing a collection and creating “mood boards,” designers selected textiles and produced sketches. Patternmakers then created “blueprints,” and sample makers sewed prototypes. The core of the experimental claim was the “fit testing” phase, where a live model would wear the sample while the team assessed its appearance, function, and movement.

During these fit sessions, the team observed how the garment moved with the model, asking her to mimic activities such as driving a car, holding a baby, or reaching for an item on a high shelf. They also tested for transparency by having the model stand in sunlight and assessed fabric durability through washing and tugging. LMI argued that these activities were investigative in nature, aimed at resolving uncertainties about the “appropriate design” of the garment. For how courts expect that experimental record to be written, see the FAQ on how to document process of experimentation for tax purposes.

The Failure of the Section 174 Test

The Tax Court, led by Judge Buch, found that LMI’s challenges did not rise to the level of technical uncertainty contemplated by Section 174. The court determined that issues such as how to drape fabric, choosing proper thread sizes, or modifying patterns for plus-size garments were common solutions to common problems. According to the court, LMI’s designers and patternmakers already possessed the information necessary to resolve these issues through standard industry practice and experienced judgment. A model technical-uncertainty write-up appears in the FAQ example of a technical uncertainty description for the IRS.

Crucially, the court found that LMI had encountered similar uncertainties in prior years and its employees already knew how to align prints, alter standard patterns, and cut fabrics along the proper grain, rendering these activities routine rather than experimental. The testing of fabric shrinkage through washing was deemed a standard quality assurance step, which is specifically excluded from being qualified research under Section 41. Activities that fail for similar reasons are catalogued in the FAQ on what expenses are excluded from R&D tax credits.

The Technological Information and “Hard Science” Barrier

A further insight from the Leon Max decision is the court’s strict adherence to a “hard science” definition of the technological information test. LMI argued that its fit testing relied on engineering principles and its fabric analysis relied on materials science and chemistry. The court rejected these arguments as defying the common understanding of those terms.

By relying on dictionary definitions of engineering — the application of science and mathematics by which the properties of matter and the sources of energy in nature are made useful to people — the court concluded that fit testing was not engineering. The ruling emphasized that receiving feedback from a model and enlarging an armhole is not an application of the properties of matter in a scientific sense. This effectively creates a judicial boundary that excludes “soft” or creative technical knowledge from the R&D credit, even when that knowledge is applied to solve complex structural problems in a business context.

The Mandate for the Scientific Method

A significant implication for future R&D applications stems from the court’s interpretation of the process of experimentation test. The court held that to satisfy this test, a project must follow a formalized scientific method. This was defined as a methodical plan involving:

1. Identifying a hypothesis.
2. Conducting a series of trials to test the hypothesis.
3. Analyzing data and results.
4. Refining the hypothesis and retesting.

Legal commentators have noted that this “scientific method” requirement may sit in tension with the 2004 Treasury Regulations (T.D. 9140), which were intended to provide a more flexible definition of experimentation that takes into account a taxpayer’s specific facts and circumstances. The court’s insistence on a lab-like protocol creates a significant hurdle for industries that traditionally resolve technical issues through systematic trial and error that does not follow the rigid structure of a university laboratory. Claimants who need to translate that protocol into a filing narrative can start with the FAQ on how to draft a project narrative for an R&D tax credit claim.

The 80 Percent Rule: Little Sandy Coal and the Fractional Controversy

While Leon Max focused on the nature of the research, Little Sandy Coal Co. v. Commissioner (T.C. Memo. 2021-15, aff’d 62 F.4th 287 (7th Cir. 2023)) provided a critical examination of the “substantially all” requirement. This case involved a shipbuilding subsidiary, Corn Island Shipyard (CIS), which claimed credits for the design and construction of 11 first-in-class vessels, including a tanker barge and a floating dry dock. Manufacturers evaluating similar process-improvement claims should also see the FAQ on whether manufacturing process improvements qualify for R&D credits.

The Numerator and Denominator Debate

The “substantially all” test requires that 80% or more of the research activities for a business component constitute elements of a process of experimentation. The Tax Court initially took a highly restrictive view of the mathematical calculation for this 80% threshold. It reasoned that while “direct research” (the actual engineering and testing) could be included in the numerator of the fraction, activities involving “direct supervision” and “direct support” had to be excluded from the numerator while still being included in the denominator. The computational steps are unpacked in the FAQ on how to calculate the substantially all rule for R&D.

Under this formula, a project with significant wages paid to supervisors or support staff (like draftsmen or welders building a prototype) would almost inevitably fail the 80% test because the numerator would be too small relative to the denominator. This exclusionary approach made it nearly impossible for large-scale manufacturing projects — where support and supervision are costly — to qualify for the credit.

The Seventh Circuit’s Taxpayer-Friendly Reversal

The March 2023 appellate ruling from the Seventh Circuit offered a significant, albeit limited, victory for taxpayers. The appellate court rejected the Tax Court’s categorical exclusion of support and supervision from the numerator. It held that the numerator is broad enough to encompass research activities that are not per se experimentation, and that if support or supervision activities are themselves research activities deductible under Section 174 and part of a methodical plan, they should be included in both the numerator and denominator.

Little Sandy Coal: Tax Court versus Seventh Circuit Treatment of the 80% Fraction

Despite this favorable formulaic change, Little Sandy Coal ultimately lost the case because it failed the substantiation requirement. The company did not offer a principled way to determine what portion of employee activities constituted experimentation, relying instead on arbitrary estimates and the general “newness” of the vessels. This underscores the point that even with a more favorable legal standard, a lack of contemporaneous data remains a terminal flaw for R&D claims. Taxpayers who never kept hour-level logs should review the FAQs on whether they need timesheets for R&D tax credits and whether they can reconstruct R&D time estimates if they did not track hours.

Phoenix Design Group and the Engineering Design-Phase Gap

The 2024 decision in Phoenix Design Group, Inc. v. Commissioner (T.C. Memo. 2024-113) provides a modern wake-up call for the engineering and architectural sectors. Phoenix Design Group (PDG), a multidisciplinary engineering firm specializing in MEPF (mechanical, electrical, plumbing, and fire protection) systems, claimed credits related to over 200 projects. The Tax Court sided with the IRS in denying all credits for the three sampled projects tried before it and upheld a 20% accuracy-related penalty. Architecture and engineering firms evaluating eligibility should consult the FAQs on whether architects can claim R&D tax credits and on R&D tax credits for architecture and engineering firms.

The Disconnect Between Design Phases and the Process of Experimentation

PDG’s primary defense was its adherence to its own six-stage internal design process — running from initial project requirements through schematic design, design development, and construction documents, broadly tracking the phases used across the architecture and engineering industry. PDG argued that following this professional framework was itself a process of experimentation.

The court dismissed this as superficial. It observed that a linear design process is not necessarily an iterative experimental process. For many of the activities, the court found that PDG engineers were performing routine calculations on available data — such as determining duct size based on airflow — where the necessary information was already known. Complying with building codes or adapting prior designs to a new footprint was deemed routine work rather than a systematic evaluation of alternatives to resolve a technological uncertainty.

Lessons from the Denied MEPF Projects

The failure of the sampled projects in Phoenix Design Group offers specific insights into how the IRS defines “uncertainty” in the built environment.

1. Baptist Memorial Hospital (BMHNMO): The taxpayer argued that not knowing the building’s footprint or the specific medical equipment to be installed created design uncertainty for the MEPF systems. The court ruled that a lack of project specifications from a client does not by itself constitute technological uncertainty.
2. Vanderbilt University Building: The complexity of integrating a hybrid operating room into a larger MEPF system was cited as an uncertainty. The court held that uncertainty in one component does not lead to uncertainty for the entire business component or project.
3. Gerald Champion Military Unit: Much of the work was viewed as standard drafting and code compliance rather than the application of advanced engineering principles to resolve unknowns.

The PDG ruling reinforces that the R&D credit is reserved for work that seeks to eliminate technical uncertainty, not just work that is technically complex or follows professional standards. Failed projects are not automatically disqualified on that ground alone; see the FAQ on whether failed development work counts for R&D tax credits.

The Shrinking-Back Rule: A Strategic but Evidentiary Tool

A recurring theme in Leon Max, Little Sandy Coal, and Phoenix Design Group is the application of the “shrinking-back” rule. Under Treasury Regulation § 1.41-4(b)(2), if an overall business component fails to satisfy the four-part test, the test may be applied to a subset of elements of the product. This process continues until either a subset satisfies the requirements or the most basic element is reached and fails.

Application of the Shrinking-Back Rule Across Project Levels

The Little Sandy Coal appeal highlighted that by choosing an “all or nothing” strategy and defining the whole vessel as the business component, the taxpayer risked total disallowance. Conversely, the Phoenix Design Group ruling showed that the shrinking-back rule is not a fail-safe; if the taxpayer’s records do not link expenses to the specific subcomponent level, the rule cannot be applied. Future practitioners must document research activities for subcomponents concurrently with the project to ensure that the shrinking-back rule remains a viable fall-back option during an audit. A practical inventory of those records is in the FAQ on what documents are needed for an R&D tax credit study.

Contemporary Compliance Standards: IRS Enforcement and Reporting Shifts

The judicial trend toward rigor has been matched by a sea change in IRS administrative requirements. An October 2021 Chief Counsel Advice Memorandum (Memorandum 20214101F, announced via IR-2021-203) initially required taxpayers filing a research-credit refund claim to identify, at the time of filing, all business components, the research activities performed for each, the individuals who performed each activity, and the information each individual sought to discover. In June 2024, the IRS eased this standard, waiving the requirement to name individual researchers and to describe what each sought to discover at the time of filing. Taxpayers must still identify business components, describe the research activities performed, and study qualified wage, supply, and contract research expenses for the claim year, and the fuller detail may still be requested if a claim is selected for examination. See the FAQs on the impact of Chief Counsel Memorandum 20214101F on R&D claims and on the latest IRS guidance on R&D tax refund claims.

The Role of Form 6765 Updates