The federal research and development (R&D) tax credit under Internal Revenue Code (IRC) § 41 requires evidence of qualifying research and eligible expenditures. Engineering complexity, professional qualifications, and a consultant’s credit study do not by themselves establish entitlement. Penalty defenses, research qualification, and contract funding must be evaluated separately.
The depreciation and penalty facts attributed to Magee v. Commissioner in the original study concern Johnson v. Commissioner, T.C. Memo. 2023-116. Johnson was not an R&D credit case and did not establish a special “Magee standard” for engineering firms. The relevant research-credit decisions discussed below are separate authorities.
The Statutory Architecture of Innovation Incentives
For the historical years examined in these cases, § 41 incorporated the research-expenditure standard under § 174. That threshold concerns research in the experimental or laboratory sense undertaken to eliminate uncertainty about capability, method, or appropriate design. The credit also imposes additional requirements concerning technological information, business components, experimentation, eligible costs, and exclusions.
The deduction rules must be distinguished by year. For tax years beginning in 2022 through 2024, § 174 generally required capitalization and amortization over five years for domestic research and fifteen years for foreign research, subject to later transition provisions. For tax years beginning after December 31, 2024, § 174A generally permits immediate deductions for domestic research or experimental expenditures, with an election to capitalize and amortize them over at least 60 months. Foreign research remains subject to § 174’s fifteen-year amortization rule. Current § 41(d)(1)(A) refers to § 174A; historical litigation must be evaluated under the law applicable to the years involved.
In Snow v. Commissioner, 416 U.S. 500 (1974), the Supreme Court interpreted the former § 174 deduction to encompass research connected with a business that had not yet begun selling its product. Snow did not decide eligibility for the subsequently enacted research credit. A research deduction does not automatically produce a § 41 credit.
Johnson v. Commissioner: Reasonable Reliance and Penalty Defense
In Johnson, the taxpayers conceded adjustments producing deficiencies of nearly $1 million for 2015–2018. The errors included seven-year depreciation of commercial buildings that required a 39-year period, duplicate mortgage-interest deductions, inadequately substantiated charitable contributions, and incorrect Social Security income. The remaining dispute concerned accuracy-related penalties.
The court found the CPA competent and the information supplied adequate for the professional-reliance test. The taxpayers nevertheless failed to establish actual, good-faith reliance on advice concerning the disputed treatments. The court also found that an experienced real-estate taxpayer should have noticed significant errors when reviewing the returns. Merely having a CPA prepare a return did not establish reasonable reliance.
Professional Experience and the Limits of the Analogy
Treasury Regulation § 1.6664-4 requires a case-specific assessment of reasonable cause and good faith. Relevant considerations include the taxpayer’s efforts to determine the correct liability, education and experience, the adviser’s competence, the information provided, and whether advice was actually relied upon. Technical expertise does not automatically prevent an engineer from reasonably relying on competent tax advice.
Applied to R&D claims, the practical lesson is to review the underlying facts, explain technical activities accurately, and obtain advice on uncertain tax positions. This is an inference from general penalty principles, not a holding that every rejected engineering claim attracts a penalty. The original study’s assertion that Magee created an IRS technical-adviser approval requirement for penalty relief is unsupported and should not be treated as an established rule.
| Penalty Component (IRC § 6662) | Johnson v. Commissioner Rationale | Implication for R&D Claims |
|---|---|---|
| Accuracy-Related Penalty | The court sustained penalties after rejecting the reasonable-cause defense. | A 20% penalty may apply to a qualifying underpayment; credit disallowance alone does not establish every penalty requirement. |
| Reliance on Professional Advice | Actual good-faith reliance on relevant advice was not established. | Keep evidence of substantive advice, the facts provided, and the basis for relying on the adviser. |
| Review of Tax Returns | The court considered significant errors and the taxpayer’s experience. | Review credit calculations and ensure technical narratives accurately describe the work. |
| Good Faith Defense | Return preparation alone did not demonstrate reasonable reliance. | Assess all relevant circumstances; missing project logs do not automatically negate good faith. |
Engineering Research: Phoenix Design Group, Inc. v. Commissioner
Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113, examined a firm designing mechanical, electrical, plumbing, and fire-protection systems. The credits concerned research undertaken during 2013–2016 and credit years 2015–2019. The parties tried a nonbinding sample of three projects from a larger disputed population. The court found that none of those three projects involved qualified research.
Routine Engineering and Technical Uncertainty
PDG relied on its six-stage design process and the possibility of later design revisions. The court required evidence explaining the uncertainties and the investigative activities undertaken to resolve them. A project’s complexity, changing requirements, or sequence of design phases did not establish qualifying experimentation. The decision does not categorically exclude HVAC, plumbing, hospital, or laboratory engineering from § 41.
The regulatory standard does not require information new to the entire industry. Existing technology can be used in qualifying experimentation. The relevant question is whether the taxpayer’s activities satisfy the statutory and regulatory tests, including uncertainty at the outset and evaluation of alternatives.
The Four-Part Test
- Research-expenditure threshold: Identify uncertainty concerning capability, method, or appropriate design, under the applicable research-expenditure provision.
- Technological information: The research must fundamentally rely on physical or biological sciences, engineering, or computer science.
- Business component and permitted purpose: The intended application must develop or improve a business component’s function, performance, reliability, or quality.
- Process of experimentation: Substantially all relevant research activities must constitute elements of a process evaluating alternatives to resolve uncertainty, potentially through modeling, simulation, or systematic trial and error.
PDG’s decisive failures concerned the research-expenditure and experimentation requirements. It is misleading to describe the opinion as finding separate failures of every element of the four-part test. Its penalty result followed the parties’ stipulation tied to the outcome of the trial projects.
| R&D Test Requirement | PDG Court Finding | Practical Recommendation |
|---|---|---|
| Technical Uncertainty | The evidence did not establish qualifying research for the trial projects or examined subsets. | Explain what information was unavailable at the outset and why investigation was necessary. |
| Process of Experimentation | Design phases and unexplained drawings did not demonstrate the required investigative process. | Connect alternatives, analysis, testing, and results to the uncertainty. |
| Substantially All | The regulatory experimentation requirement applies; the opinion did not establish a universal 35% experimentation ratio. | Support the at-least-80% calculation using cost or another consistently applied reasonable basis. |
| Shrinking-Back Rule | The court considered systems and smaller subsets, but the evidence remained insufficient. | Identify actual subcomponents and explain the associated activities and costs. |
The Funded Research Doctrine: Risk, Rights, and Fixed-Price Contracts
IRC § 41(d)(4)(H) excludes research to the extent funded by another person. Treasury Regulation § 1.41-4A(d), incorporated by § 1.41-4(c)(9), requires examination of the relevant agreements, payment conditions, and rights in research results. Payment contingent on research success is not treated as funding. If the researcher retains no substantial rights, the research is treated as fully funded. If substantial rights are retained, partial funding and allocation rules may leave eligible expenditures after funding is accounted for.
Meyer, Borgman & Johnson
In Meyer, Borgman & Johnson, Inc. v. Commissioner, No. 23-1523 (8th Cir. May 6, 2024), the appellate court affirmed a funded-research ruling against a structural engineering firm. Its contracts did not expressly or by clear implication condition payment on research success. Obligations to provide proper professional performance and compliant designs did not establish the necessary contingency.
A fixed price alone therefore does not settle the funding question. Conversely, the decision does not require particular magic words in every contract. Enforceable obligations, acceptance conditions, remedies, and the agreement as a whole matter. Cost-overrun exposure must be distinguished from the conditions governing entitlement to payment.
Smith: Procedural Developments and Partial Credits
The December 18, 2024 order in the consolidated Smith cases denied the IRS summary judgment on the funded-research issue. That procedural ruling did not itself award credits. The later merits opinion, Smith v. Commissioner, T.C. Memo. 2026-50, issued June 16, 2026, must also be considered.
The later court found that payments under the six sample contracts were not contingent on research success. It nevertheless found substantial rights retained under four contracts and permitted partial research credits subject to Treasury Regulation § 1.41-4A(d)(3). Two contracts failed the substantial-rights requirement. Accordingly, the case should not be reduced to a rule that milestone payments establish unfunded research or that all third-party funding necessarily eliminates every eligible expenditure.
Delivery of drawings to a client does not alone determine substantial rights. Contract provisions and applicable intellectual-property law must be examined. General experience gained from performing a project is only an incidental benefit and is not, by itself, a substantial right in its research results.
Substantiation and the Shrinking-Back Rule
Taxpayers generally bear the burden of establishing credit entitlement, subject to applicable burden-shifting rules and procedural exceptions. Financial-accounting classifications and consultant estimates must be reconciled with tax requirements. ASC 730 treatment is not automatically equivalent to § 41 eligibility; any applicable IRS administrative approach has its own conditions.
Sampling and the Scope of the Evidence
PDG’s three-project trial sample was expressly nonbinding. The opinion should not be described as mechanically disallowing all remaining projects because one sample failed. Sampling can be useful, but the selection method, supporting evidence, and any agreement governing extrapolation determine its significance. A favorable or unfavorable sample finding does not create a universal rule for every claim.
Applying the Shrinking-Back Rule
Under Treasury Regulation § 1.41-4(b)(2), if a business component fails the qualification tests, the analysis proceeds to its most significant subset of elements and then to smaller subsets as appropriate. This is an analysis of the component’s elements, not simply a selection of employees, accounting categories, or design phases.
PDG illustrates the need to explain the investigation at the relevant level. Identifying a problem and its final solution, without evidence of the intervening activities, may be insufficient. Records should connect actual work and expenditure to the component or subcomponent. The regulation does not prescribe one mandatory timekeeping application or make a particular form of contemporaneous log the sole permissible evidence.
| Substantiation Element | Typical Failure Point | Future Compliance Standard |
|---|---|---|
| Activity-Level Documentation | Using broad design phases without explaining investigative activities. | Maintain records explaining uncertainty, alternatives, evaluation, and results; these are practical recommendations, not a mandatory document format. |
| Direct Supervision | Equating general executive oversight with direct supervision of qualified research. | Identify the research directly supervised and the employee’s actual responsibilities. |
| Qualified Wage Allocation | Assigning percentages solely by job title without adequate factual support. | Use substantiated allocations connecting services to qualifying activities, applying any relevant wage rules. |
| Contract Language | Assuming a fixed price or milestone schedule resolves the funding analysis. | Evaluate payment contingencies, substantial rights, and any partial-funding allocation under the full agreements. |
The “In Connection With” and “Carrying On” Distinction
Snow’s interpretation of former § 174 differs from § 162’s requirement that ordinary business expenses arise while carrying on a trade or business. Section 41 has its own trade-or-business requirement and, in § 41(b)(4), a special rule for certain startup in-house research expenses intended for use in a future active business. Pre-revenue status therefore does not automatically defeat a credit, but an intention to sell a product is not by itself sufficient.
Startup expenses, organizational expenses, research deductions, and research credits require separate analyses. Section 481 addresses adjustments associated with accounting-method changes; it is not a general mechanism automatically recapturing disallowed startup research credits. The depreciation adjustment in Johnson should not be extrapolated into such a rule.
Federal Research Credits and State Constitutional Litigation
Magee v. Boyd, 175 So. 3d 79 (Ala. 2015), concerned constitutional challenges to the Alabama Accountability Act’s education-related provisions and tax credits. It did not decide federal R&D credit eligibility, engineering experimentation, or § 6662 penalties.
Its discussion of tax credits and appropriations arose in a particular state constitutional setting. It does not establish a universal classification of every credit or alter the federal four-part test. Federal credit eligibility follows the applicable federal statute and regulations, including their exceptions and rules allowing qualifying subsets.
Practical Implications for Future R&D Tax Credit Claims
Contract Review
Before entering research arrangements, identify the parties’ actual commercial allocation of risk and rights. Review when payment becomes enforceable, what happens if research fails, whether results may be reused, and whether funding allocations apply. Contract language must reflect the real transaction; inserting tax-oriented wording alone cannot establish eligibility. Professional standards and necessary client protections should be evaluated in context.
Documenting the Investigative Process
- Describe the specific uncertainty concerning capability, method, or design.
- Identify the business component and any relevant subcomponents.
- Explain the alternatives considered and how they were evaluated.
- Preserve relevant calculations, models, test results, design revisions, and technical explanations.
- Connect eligible wages and other expenditures to the substantiated activities.
These practices help establish the facts but do not replace the legal tests. Novelty, successful development, and technical difficulty are neither automatic qualification nor automatic disqualification.
The Role of Technical Experts and Tax Advisers
Technical staff should explain the engineering facts, while competent tax advisers evaluate the statutory requirements and calculations. Management should review the claim and resolve material inconsistencies. Signing a return does not create a special judicial certification under an unverified “Magee standard,” and a consultant’s involvement does not guarantee penalty protection.
