The Legislative and Regulatory Framework of IRC Section 41

To understand the profound implications of Betz, one must first examine the legal foundation upon which the case was built. The R&D tax credit is designed to reward taxpayers for increasing their research activities, calculated based on the incremental amount of QREs—including wages, supply costs, and contract research expenses—incurred during the tax year. For an activity to constitute “qualified research,” it must satisfy a rigorous four-part test established under Section 41(d)(1).

The Four-Part Test for Qualified Research

For the 2014 tax year considered in Betz, the Section 174 test required expenditures for research or experimentation intended to resolve uncertainty about capability, methodology, or appropriate design. The process of experimentation test is a separate requirement. Under Treasury Regulation Section 1.41-4(a)(5) and (6), at least 80% of the relevant research activities, measured on a cost or other consistently applied reasonable basis, must constitute elements of a process that evaluates alternatives to resolve uncertainty for a qualified purpose. Modeling, simulation, and systematic trial and error are possible methods. This activity-based test is distinct from the separate 80% rule for an individual employee’s qualified services.

Factual Matrix: Catalytic Products International and the 19-Project Dispute

In Betz v. Commissioner, the taxpayers were shareholders of Catalytic Products International, Inc. (CPI), a subchapter S corporation that designed and supplied custom-built air pollution control systems. These systems, primarily catalytic and thermal oxidizers, are used in manufacturing environments to eliminate volatile organic compounds (VOCs) and other hazardous airborne pollutants. For the tax year 2014, CPI claimed $501,531 in net R&D credits based on wage and supply expenditures associated with 19 distinct projects.

CPI oversaw fabrication and assembly at subcontractor facilities before installing systems at customer locations. The dispute concerned whether the claimed supply costs and wages satisfied the research credit requirements, whether CPI retained substantial rights under customer contracts, and whether penalties applied. Custom engineering is not automatically qualified research, but work does not fall within the adaptation exclusion merely because it is undertaken for a specific customer.

Detailed Analysis of the 19 Projects and Technical Failure Points

The court examined the evidence for the 19 projects and generally found that CPI had sufficient information to establish the systems’ overall designs before the activities for which the credit was claimed. Detailed proposals, prior comparable systems, engineering spreadsheets, and purchase commitments undermined broad assertions that uncertainty persisted throughout each project. These findings concerned the evidence and timing in CPI’s projects; using spreadsheets or established engineering principles does not itself disqualify research.

The Section 174 Uncertainty Test and the Court’s Economic Reasoning

CPI argued that uncertainty about appropriate design persisted until the systems passed testing at the customer’s site. The court rejected the inference that postproduction testing alone established continuing uncertainty about the entire system. It did not establish a categorical prohibition on testing full-scale models or on research occurring during fabrication.

Timing and Documentation of Uncertainty

Judge Nega considered CPI’s substantial commitments to specially sized components and fabrication under detailed drawings when evaluating its assertion that the overall design remained uncertain. The economic reasoning supported the court’s assessment of the particular record. It is not an independent statutory test requiring manufacturers to avoid substantial expenditure before conducting research.

For a taxpayer claiming the costs of a whole unit as a pilot model, the relevant issue is whether the unit was produced to evaluate and resolve uncertainty. A unit can have that research purpose and later be sold or used commercially. CPI failed to establish the requisite purpose for the systems as a whole. Qualifying research on a smaller component must be examined separately, and ordinary production costs cannot be included merely because research also occurred on the project.

The Pilot Model Analysis

Treasury Regulation Section 1.174-2(a)(4) defines a pilot model by its role in evaluating and resolving uncertainty during product development or improvement. A model may be fully functional, full scale, and ultimately sold. Its eventual sale or use does not determine whether the production expenditures were research expenditures. The taxpayer must establish the research purpose and satisfy the additional Section 41 requirements before claiming a credit.

CPI did not substantiate that the systems as a whole were produced to resolve uncertainty. Detailed fabrication specifications and the limited evidence explaining investigative activities weakened the pilot model claim. Testing that merely confirms compliance or performs ordinary quality control does not establish qualified research. Design alternatives, test results, revisions, and related correspondence can support an experimental purpose; no particular form of hypothesis log is universally mandatory.

The Adaptation Exclusion and the Discovery Test Critique

Further complicating CPI’s case was the application of the Section 41(d)(4)(B) adaptation exclusion. This rule states that research related to adapting an existing business component to a particular customer’s requirement is not qualified research. In the M&W Ireland wastewater project, the court found that CPI delivered what was explicitly described in the proposal as an “exact copy” of a previous system, with only minor site-specific modifications to meet European standards. The court ruled that an exact copy with minor modifications falls squarely within the plain meaning of “adaptation”.

Commentators have questioned whether parts of Betz’s reasoning resemble an older, more restrictive discovery standard. That criticism should be distinguished from a judicial holding that the obsolete standard governs. Treasury Regulation Section 1.41-4(a)(3) does not require research to expand knowledge in the field, and existing technologies may be used. Work being new to the taxpayer is not sufficient by itself: the other requirements and exclusions still apply.

Substantiation of Wage QREs and the Rejection of the Cohan Rule

CPI relied on a consultant-prepared research credit study and employee testimony to support wage allocations. The court found the evidence insufficient to establish qualified services and the amounts attributable to them. Retrospective interviews and estimates are not automatically prohibited, but unsupported allocations and testimony inconsistent with project records do not establish entitlement to the credit.

The Limitations of the Cohan Rule

The Cohan rule, from Cohan v. Commissioner, 39 F.2d 540 (2d Cir. 1930), permits reasonable estimation in appropriate circumstances when entitlement to an expense has been established but its exact amount is uncertain. It does not substitute for proof that research satisfies Section 41. Betz, Little Sandy Coal, and Moore illustrate the importance of distinguishing qualifying services from broader product development or managerial work.

Before wages can be estimated, the taxpayer must establish that the employee performed qualified services and provide a reasonable evidentiary basis for allocation. Betz also illustrates that evidence of a limited design uncertainty does not necessarily prove a structured process of experimentation or support the broader wage claim. The doctrine cannot cure a failure to establish the underlying qualification requirements.

Contemporaneous time records can strengthen a claim, but neither Betz nor Treasury Regulation Section 1.41-4(d) imposes a universal daily time-tracking requirement. Records must be sufficiently usable and detailed to substantiate eligibility. Supported estimates, technical records, and credible testimony must establish both qualifying activities and a reasonable allocation of costs.

The Funded Research Exclusion: Intellectual Property and Substantial Rights

A significant portion of the Betz ruling analyzed whether CPI’s research was “funded” by its customers. Under Section 41(d)(4)(H), research is excluded if the taxpayer does not retain “substantial rights” in the results or if the payment is not contingent on the success of the research.

Analysis of CPI’s Customer Contracts

The court considered the contracts associated with the funded research dispute and concluded that CPI lacked substantial rights for five projects. Exclusive ownership provisions and restrictions on CPI’s ability to use the results supported that conclusion. The analysis depends on the actual contract terms, related agreements, and applicable law, rather than the label placed on the arrangement.

Under Treasury Regulation Section 1.41-4A(d), general experience or proficiency gained from performing research does not alone constitute substantial rights in its results. A taxpayer that retains no substantial rights cannot qualify the work merely by bearing financial risk. Conversely, exclusive ownership is not required: meaningful nonexclusive rights can be sufficient, subject to the separate economic-risk inquiry.

Comparative Jurisprudence: Related Research Credit Decisions

Little Sandy Coal Co. v. Commissioner, 62 F.4th 287 (7th Cir. 2023), and Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113, provide useful comparisons with Betz. These decisions address related qualification and proof issues. They do not constitute a formally recognized trilogy or replace the statutory and regulatory tests.

Little Sandy Coal and the PoE Numerator

The Seventh Circuit affirmed the denial of credits in Little Sandy Coal while disagreeing with the Tax Court’s categorical exclusion of direct support and supervision from the experimentation analysis. Such activities can count when they are elements of the process of experimentation; their wage classification alone does not settle the issue. The taxpayer still failed to substantiate the substantially-all requirement with a principled account of the relevant activities.

Phoenix Design and Routine Design

Phoenix Design Group concerned mechanical, electrical, plumbing, and fire protection (MEPF) engineering. The Tax Court found that the three trial projects did not establish qualified research, including at the shrinking-back level. Describing recurring design phases and eventual solutions did not adequately explain the investigative work performed. The decision does not hold that all engineering calculations using established principles are disqualified; the question is whether the taxpayer actually evaluated alternatives to resolve relevant uncertainty.

Accuracy-Related Penalties: The Cost of Negligence

The Tax Court upheld 20% accuracy-related penalties under Section 6662(a), addressing negligence and substantial understatements. Inadequate substantiation and the absence of a successful reasonable-cause defense were important to the result. Disallowance of a research credit does not automatically establish liability for a penalty.

The taxpayers did not establish reasonable cause and good faith under Section 6664(c)(1). Reliance on a consultant is assessed in light of the adviser’s competence, the information supplied, and whether the taxpayer actually and reasonably relied on the advice. A consultant’s financial interest may be relevant to that assessment, but a percentage-based fee alone does not establish negligence or automatically defeat the defense.

Implications for Future R&D Tax Credit Applications

Betz identifies practical weaknesses that taxpayers can address before filing: unclear research boundaries, unsupported wage allocations, insufficient pilot model evidence, and restrictive customer contracts. Technical and financial records should explain what work occurred, why it met the qualification requirements, and how the claimed expenses were determined.

Implementation of Granular Time-Tracking and Technical Nexus

Project-level time tracking can make a claim easier to substantiate, particularly when employees perform both research and nonresearch work. It is a recommended control, not a new statutory condition created by Betz. A suitable system distinguishes experimental design, testing, direct support, and immediate supervision from routine fabrication and general administration, with records supporting the classifications and allocations.

A defensible claim should connect:

• The specific technical uncertainty identified at the outset.
• The investigative activities (experimentation) undertaken to resolve it.
• The specific expenditures (wages and supplies) incurred for those activities.

The “Shrinking-Back” Strategy as a Defensive Tool

The court in Betz acknowledged the possibility that certain wages or supplies might qualify at the component or sub-component level through the application of the “shrinking