The federal research and development (R&D) tax credit under Internal Revenue Code (IRC) Section 41 can apply to qualifying work in engineering, architecture, manufacturing, software, and other industries. Eligibility depends on the activities performed, the expenses incurred, and the statutory exclusions. Technical complexity or a professional qualification alone does not establish entitlement to the credit.

The case name requires clarification. Johanson v. Commissioner, T.C. Memo. 2006-105, affirmed at 541 F.3d 973 (9th Cir. 2008), concerned alimony, not the R&D tax credit. It should not be presented as a research-credit decision or as the origin of a special “Johanson principle” governing funded research. This study distinguishes that case from the Section 41 authorities relevant to contractual risk, experimentation, and substantiation.

The Relationship Between Research Deductions and Section 41

Research expenditure deductions and research credits are related but distinct. Under the Tax Cuts and Jobs Act, research expenditures incurred in tax years beginning after December 31, 2021, generally became subject to five-year domestic or fifteen-year foreign amortization. That description is no longer a complete statement of current domestic treatment.

Public Law 119-21 added Section 174A, generally permitting immediate deductions for domestic research or experimental expenditures in tax years beginning after December 31, 2024, with an alternative capitalization election. Foreign research expenditures remain subject to fifteen-year amortization under Section 174. Transition provisions address previously capitalized domestic costs and eligible small businesses. IRS Revenue Procedure 2025-28 explains the applicable elections and accounting-method procedures.

The current Section 41(d)(1)(A) refers to domestic research or experimental expenditures under Section 174A. Earlier cases applied the statutory language governing their tax years. Deductibility does not, by itself, establish that an expenditure qualifies for the credit.

A failure at the overall business-component level does not necessarily eliminate all potential credit: qualifying subsets may remain after applying the shrinking-back rule. Separate exclusions, including funded research, must also be considered.

Johanson and the Role of State Law in Tax Disputes

In Johanson, the courts considered the federal tax treatment of support payments under the then-applicable alimony provisions. California law mattered to whether the payment obligation would terminate on the recipient’s death. The Ninth Circuit affirmed the Tax Court’s treatment of the payments as alimony.

This illustrates that state law can help determine underlying legal obligations relevant to a federal tax question. It does not establish that state law overrides federal credit requirements, or that Johanson controls later research-credit disputes. Funded-research analysis should be grounded in Section 41, the applicable Treasury regulations, and decisions addressing research agreements.

Funded Research and the Risk of Failure

Section 41(d)(4)(H) excludes research to the extent another person funds it. Treasury Regulation § 1.41-4A(d), incorporated into the modern qualified-research regulations, addresses both payment risk and rights to research results.

Payments contingent on successful research are not treated as funding under the regulation. If the researcher retains no substantial rights in the results, the research is treated as fully funded. Where substantial rights remain but payments constitute funding, otherwise eligible expenses must be reduced under the regulation’s project-specific rules; partial eligibility can therefore remain in appropriate circumstances.

A fixed price, milestone schedule, warranty, or potential cost overrun does not automatically establish that payment depends on research success. The agreements and legally enforceable obligations require substantive analysis.

The Meyer, Borgman & Johnson Decision

In Meyer, Borgman & Johnson, Inc. v. Commissioner, T.C. Memo. 2022-127, affirmed at 100 F.4th 986 (8th Cir. 2024), a structural engineering firm sought approximately $190,000 in research credits. The Eighth Circuit upheld the determination that the research was funded because payment was not contingent on research success.

Requirements to meet professional standards and produce competent designs did not establish the necessary contingency. The decision does not create a universal rule that a contract must contain a particular refund clause. Refund obligations, acceptance terms, and payment conditions are relevant to the full contractual analysis.

Smith and System Technologies: Procedural Outcomes and Later Developments

An earlier denial of summary judgment in Smith preserved disputed issues; it did not finally establish credit eligibility. The later merits decision, Smith v. Commissioner, T.C. Memo. 2026-50, filed June 16, 2026, found that payment under none of the six sample contracts was contingent on research success. AS+GG retained substantial rights in four projects, allowing potential partial credits after applying the funding rules. It lacked substantial rights in the other two. The court also found the disputed partner compensation reasonable. Milestone billing should therefore not be described as an automatic taxpayer victory.

In System Technologies, Inc. v. Commissioner, Docket No. 12211-21, an order dated January 3, 2025, denied the IRS’s motion for partial summary judgment. The court considered Indiana law, which would require repayment if the company failed to deliver the contracted product. The order rejected the IRS’s funded-research argument on that record. It did not award all claimed credits or resolve every remaining eligibility requirement.

The Process of Experimentation

Technical work must satisfy the experimentation requirement, even when it passes the funded-research analysis. Treasury Regulation § 1.41-4 describes a process that evaluates alternatives to resolve uncertainty about capability, method, or appropriate design. Modeling, simulation, and systematic trial and error can provide evidence of that process.

The 80% threshold concerns activities, not the proportion of a product that is novel or the percentage of all company spending labeled R&D. Ordinary quality control, adaptation, duplication, and other excluded activities must be distinguished from qualifying research. The applicable exclusions and business-component boundaries matter.

Phoenix Design Group and Technical Uncertainty

Phoenix Design Group, Inc. v. Commissioner is T.C. Memo. 2024-113, filed December 23, 2024, rather than a 2023 decision. It concerned research credits for 2013 through 2016 and examined six sample engineering projects.

The court found that the taxpayer had not established qualified research in those projects. Engineering calculations and design revisions did not, on the evidence presented, demonstrate the required research uncertainty and experimental process. The lesson is evidentiary: records should explain the technical question, the alternatives considered, and how evaluation addressed the uncertainty. The decision does not categorically exclude engineering or every iterative calculation.

Suder and Incremental Innovation

Suder v. Commissioner, T.C. Memo. 2014-201, recognized qualifying research in telephone-system development supported by extensive evidence. An improvement need not create a new scientific principle to qualify. Existing technologies can be used in research aimed at resolving uncertainty about a new or improved business component.

However, project duration, commercial importance, or novelty alone does not prove eligibility. Suder also addressed expense substantiation and reduced the chief executive’s compensation eligible for inclusion because part of it was unreasonable. It was not an unconditional allowance of every claimed expense.

Documentation and IRS Refund-Claim Review

Research-credit refund claims are subject to an initial review for required information as well as possible examination of their merits. Describing this process as a new automated “Classifier” that rejects claims without human involvement is unsupported. Filing sufficiency and substantive entitlement are different questions.

Moore and the Limits of Unsupported Time Allocations

In Moore v. Commissioner, T.C. Memo. 2023-20, affirmed by the Seventh Circuit on April 30, 2024, the taxpayers failed to substantiate a qualifying share of a company president and chief operating officer’s compensation. Payroll records established compensation but did not establish the time devoted to qualifying services. His executive position did not itself establish direct research, direct supervision, or direct support.

Reasonable estimates and credible testimony are not categorically prohibited. Their usefulness depends on an adequate factual foundation connecting the claimed amounts to qualifying activities. The Cohan doctrine is not a substitute for proving that qualified research occurred. Union Carbide litigation likewise should not be read as a general exemption from research-credit substantiation requirements.

Information Required for a Research-Credit Refund Claim

For the IRS’s research-credit refund-claim sufficiency rules, the required information currently includes:

• All business components to which the credit claim relates for the claim year.
• The research activities performed for each business component.
• Total qualified employee wage, supply, and contract research expenses for the claim year.

Effective June 18, 2024, the IRS waived the upfront requirements to identify each individual performing an activity and the specific information each individual sought to discover. The IRS may still request this information during examination. Under the extension announced in IRS news release IR-2025-99, the transition period providing 45 days to perfect a deficient claim runs through January 10, 2027.

These minimum filing items do not replace evidence supporting the four-part test, expense amounts, or contractual eligibility.

International Comparative Insights: Correcting the Johansen Attribution

The Norwegian return estimates in the source belong to Jarle Møen’s 2018 study, Corporate Returns to Subsidized R&D Projects: Direct Grants vs Tax Credit Financing. They should not be attributed to Klette and Johansen’s 1998 work as an evaluation of SkatteFUNN. The similarly spelled names do not connect this economic research with Johanson’s alimony litigation.

Møen estimated private returns of approximately 16% for tax-credit-financed R&D and 19% for self-financed R&D. The estimate for directly grant-funded projects was not statistically distinguishable from zero and had high variance. These are study-specific estimates, not guaranteed investment returns, credit rates, or measures of U.S. credit eligibility.

The Shrinking-Back Rule and Subcomponent Analysis

Treasury Regulation § 1.41-4(b)(2) provides for applying the qualification requirements to the most significant subset of elements when the overall business component fails them, continuing to smaller subsets as necessary. This is an eligibility analysis, not permission to assume an arbitrary qualifying percentage.

Little Sandy Coal Co. v. Commissioner, 62 F.4th 287 (7th Cir. 2023), involved a shipbuilder and two representative vessels, a tanker barge and a dry dock. The appellate court affirmed the disallowance because the taxpayer did not provide a principled basis for determining which activities constituted experimentation. It disagreed with some aspects of the Tax Court’s reasoning but found the evidentiary deficiencies decisive.

Novelty and broad wage allocations did not establish the substantially-all test, and the record did not provide an adequate basis for shrinking back to qualifying subsets. Maintaining evidence by component and, where appropriate, subcomponent can make that analysis possible.

Implications for R&D Tax Credit Claims

These authorities support separate reviews of contractual funding, technical qualification, and expense substantiation. Success on one issue does not resolve the others. A sound claim explains what was uncertain, how alternatives were evaluated, who performed qualifying services, and how the claimed amounts were calculated.

Strategic Contract Management

• Payment risk: Review acceptance criteria, payment entitlements, termination provisions, and remedies to determine who bears the consequences of unsuccessful research. Contract language should accurately reflect the commercial arrangement.
• Research rights: Examine whether the developer can use the research results in its business. Retaining general experience alone is insufficient.
• Applicable law: Assess governing law where it affects contractual obligations. Do not assume that a choice-of-law clause by itself establishes an unfunded arrangement.

Contemporaneous Documentation and Retrospective Studies

Contemporaneous records generally make substantiation easier. Useful evidence can include design versions, testing records, simulation results, technical meeting notes, project accounting, and supported employee time allocations. Recordkeeping should connect the activities to the relevant business components and expenses.

Retrospective studies have not been categorically prohibited. Their reliability depends on the underlying records and testimony, and unsupported recollections can be inadequate. There is no universal requirement to document every failed attempt or use a particular time-tracking system.

Application to Architecture, Engineering, and Construction

Firms in these industries should distinguish ordinary professional design work from activities that satisfy the research-credit requirements. Documentation should describe actual technical uncertainty and evaluation of alternatives, rather than merely restating statutory terminology. Separate qualifying development from routine compliance, production, and excluded work.

Final Thoughts

Johanson is an alimony authority, not an R&D tax-credit precedent. Relevant research-credit decisions demonstrate the importance of contract-specific risk analysis, retained research rights, evidence of experimentation, and supportable expense calculations. Procedural victories should not be presented as final credit allowances.

A defensible claim combines accurate descriptions of the work with the law applicable to the claim year. Current domestic research-deduction rules, refund-claim requirements, and later court decisions must be considered when applying lessons from older cases.