The federal research and development (R&D) tax credit under Internal Revenue Code Section 41 depends on both qualifying activities and adequate evidence of the associated costs. Recent decisions illustrate the risks of unsupported allocations, generalized project descriptions, and assumptions that novel products automatically qualify. They do not establish a universal prohibition on reasonable estimates or a new requirement that every claimant maintain a particular kind of timesheet.
This study examines substantiation, business-component analysis, customer-funded research, and changes to research-expense deductions and Form 6765. The applicable tax year matters: the treatment of domestic research costs changed substantially in 2025.
The Judicial Foundation of Research Credit Substantiation
Section 41 and its regulations require taxpayers to establish eligibility rather than merely characterize their work as innovative. The relevant evidence should explain the research performed, the business components involved, and the basis for including wages, supplies, and contract research costs. A retrospective credit study can organize this evidence, but cannot substitute for proof that qualifying activities occurred.
In Moore v. Commissioner, T.C. Memo. 2023-20, affirmed by the Seventh Circuit in 2024, the disputed costs concerned the compensation of Nevco’s president and chief operating officer. The courts rejected the claimed allocation because the evidence did not establish the extent of his qualifying work. The result does not exclude officers categorically or make contemporaneous time logs the only permissible evidence.
Comparative Legal Precedents and Judicial Outcomes
| Case Citation | Core Dispute | Judicial Finding | Implication for Future Applications |
|---|---|---|---|
| Little Sandy Coal Co. v. Commissioner, 62 F.4th 287 (7th Cir. 2023) | Whether vessel-development activities met the substantially-all experimentation test. | Disallowance affirmed because the taxpayer lacked a principled allocation of qualifying activities. | Novelty and arbitrary percentages do not establish the experimentation ratio. |
| Moore v. Commissioner, T.C. Memo. 2023-20, affirmed, No. 23-2681 (7th Cir. Apr. 30, 2024) | Substantiation of an officer’s qualified services. | The disputed compensation allocation was not adequately supported. | Document actual research, direct supervision, or direct support rather than relying on job title. |
| Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113 | Technical uncertainty and experimentation in engineering projects. | The taxpayer failed to establish qualifying research for the projects adjudicated; accuracy-related penalties were sustained. | Design phases and project complexity need evidence connecting activities to statutory requirements. |
| System Technologies, Inc. v. Commissioner, Docket No. 12211-21, summary-judgment order | Whether customer payments constituted research funding. | The court denied the IRS motion for partial summary judgment after considering contractual obligations and Indiana law. | A ruling on funding at summary judgment is not a final allowance of every claimed credit. |
Applying the Four-Part Test
Section 41 requires eligible research expenditures, information that is technological in nature, a qualifying business purpose, and substantially all research activities to constitute elements of a process of experimentation. Apply the requirements separately to each business component. Current Section 41(d)(1)(A) refers to Section 174A; earlier-year cases applied the Section 174 language then in effect.
A permitted purpose concerns a new or improved function, performance, reliability, or quality. Changes relating only to style, taste, cosmetic appearance, or seasonal design do not meet that requirement. The credit also has exclusions, including foreign research and research after commercial production, subject to the governing rules.
Research Expenditures and Technical Uncertainty
The uncertainty analysis asks whether the information available at the outset establishes the capability, method, or appropriate design for developing or improving the product. A project’s expense, commercial importance, or unfamiliarity to a particular employee is not sufficient by itself. Conversely, research need not produce knowledge that is new to the entire industry.
Phoenix Design Group illustrates why ordinary design work and the application of known engineering information may fail to establish qualifying uncertainty. Its outcome should not be read as excluding all engineering calculations, code-related development, or architectural projects. The actual investigative activities and supporting evidence determine the result.
Technological in Nature
The experimentation process must fundamentally rely on engineering, computer science, or physical or biological science. Existing scientific principles can support qualified research. An aesthetic objective alone is insufficient, but a software or design project may contain technical development activities that warrant separate analysis.
The Process of Experimentation and the 80% Rule
The substantially-all threshold is at least 80%, measured by cost or another consistently applied reasonable basis. The analysis concerns research activities for the business component, not the percentage of physical parts that are new.
Little Sandy Coal rejected reliance on first-in-class vessels and unsupported estimates. The court did not categorically exclude production or support activities from experimentation; their connection to the experimental process must be established. A qualifying prototype does not automatically make all construction costs eligible. The experimentation ratio and the separate rules for qualified employee services should not be conflated.
Administrative Changes: Form 6765
Section E: Entity-Level Disclosures
Section E asks for information including business-component counts and officer wages. Complete it when required by the applicable instructions. Disclosure of officer compensation is not itself a finding that the wages are ineligible.
Section G: Business Component Information
| Form 6765 Update | Description | Implementation Status | Strategic Impact |
|---|---|---|---|
| Section E | Additional QRE information. | Required when QREs appear on line 48. | Reconcile disclosures to supporting records. |
| Section G | Component-level cost details. | Optional before 2026; generally required for tax years beginning after 2025. | Prepare component allocations. |
| Section G exceptions | Qualifying payroll-credit small businesses; or original-return filers with controlled-group QREs of $1.5 million or less and the specified three-year average gross receipts of $50 million or less. | Subject to the instructions. | Exceptions do not waive substantiation. |
| Business Component Detail | 80% of QREs or 50 components, with remaining components aggregated. | Applies when Section G is required. | Follow ordering and group rules. |
The Form 6765 instructions continue to recognize permitted statistical sampling. Section G does not abolish sampling. Its information-sought field currently applies to amended returns; it is not a universal employee-by-employee narrative requirement.
Research-Expense Deductions and the Financial Implications of Innovation
The Tax Cuts and Jobs Act required five-year amortization for domestic research expenditures and fifteen-year amortization for foreign research expenditures for tax years beginning after December 31, 2021. That historical rule no longer describes the general treatment of new domestic expenditures.
Public Law 119-21, enacted July 4, 2025, added Section 174A. Domestic research or experimental expenditures generally may be deducted currently for tax years beginning after December 31, 2024. Taxpayers may instead elect capitalization and amortization over at least 60 months under the statutory conditions. Foreign research remains subject to fifteen-year amortization under Section 174.
Research Expenditure Treatment by Period
| Type of Research Expenditure | Amortization Period | Effective Date | Impact on Credit Value |
|---|---|---|---|
| Domestic R&D incurred in 2022–2024 tax years | Five years under the original TCJA rule, subject to subsequent transition provisions. | Tax years beginning after 2021 and before 2025. | Deduction timing is separate from credit eligibility. |
| Domestic R&D incurred in 2025 and later tax years | Current deduction generally available; optional amortization of at least 60 months. | Tax years beginning after 2024. | Section 174A restores current deductibility; Section 280C coordination remains relevant. |
| Foreign R&D | Fifteen years, using the statutory midpoint convention. | Tax years beginning after 2021. | Foreign research is generally excluded from the federal Section 41 credit. |
| Software development | Domestic or foreign treatment, as applicable. | Apply the rules for the expenditure’s location and tax year. | Software-expense treatment does not establish Section 41 eligibility automatically. |
Transition provisions address remaining unamortized domestic costs from 2022–2024, including accelerated recovery options and a retroactive election for eligible small businesses. Revenue Procedure 2025-28 provides implementation procedures. Eligibility, election deadlines, and accounting-method requirements must be evaluated separately; historical costs should not simply be deducted again.
The Funded Research Exclusion and Economic Risk
Research performed under a customer contract is not automatically ineligible. The funded-research rules examine whether payment depends on successful research and whether the researcher retains substantial rights to the results. Payments and expenses may require allocation. Exclusive intellectual-property ownership is not invariably required, but retaining merely incidental experience is insufficient.
In System Technologies, the court considered Indiana-law remedies for a complete failure to deliver and concluded that the warranty provisions did not eliminate relevant customer remedies. This supported denial of the IRS funding motion. The decision was not the unspecified “T.C. Memo. 2025” merits opinion described in the original draft.
Contract analysis should encompass the full agreement, actual performance, payment obligations, and governing law. In Meyer, Borgman & Johnson, Inc. v. Commissioner, No. 23-1523 (8th Cir. May 6, 2024), the court affirmed a funded-research determination. Ordinary professional obligations and financial exposure do not necessarily make customer payments contingent on research success.
Procedural Requirements for Research Credit Refund Claims
The IRS reviews whether amended-return refund claims contain sufficient information to identify their basis. This procedural review is distinct from deciding whether the underlying research qualifies. Public guidance does not support describing it as a universal automated “Classifier” that denies credits without human review.
Since June 18, 2024, the IRS has waived the initial requirement to supply individual researchers’ names and the information each individual sought. Claims must still identify the relevant business components, describe research activities for each, and state total qualified employee wage, supply, and contract research expenses. Additional details may be requested during examination.
The IRS extended the transition period for perfecting deficient research-credit refund claims through January 10, 2027. During that period, applicable notices generally provide 45 days to supply missing information. These procedures do not dispense with substantive qualification or applicable refund limitation periods.
Administrative Law and Agency Authority
Corner Post, Inc. v. Board of Governors of the Federal Reserve System addresses accrual of an Administrative Procedure Act claim under the general six-year federal limitations statute. It does not invalidate the research-credit regulations or remove tax-specific jurisdictional restrictions. Its possible application to a particular tax challenge requires a separate legal analysis.
Litigation about assessment of international information-return penalties should not be extrapolated into immunity from R&D-related accuracy penalties. Sections 6662 and 6664 govern relevant underpayment penalties and reasonable-cause relief. Credit disallowance does not, by itself, resolve every penalty issue.
Practical Implications for R&D Credit Claims
The Shrinking-Back Rule and Business Components
When the requirements are not met for an entire business component, apply the shrinking-back rule to the most significant subset of its elements, continuing as appropriate. It does not permit selecting an arbitrary group of employees or favorable hours. Evidence must still establish qualification at the narrower level.
Documentation Practices
- Project narratives: Record the uncertainty, alternatives, evaluations, and outcomes while the work is fresh. Preserve unsuccessful trials where relevant, without treating failure as a prerequisite or proof of eligibility.
- Expense connections: Reconcile payroll and purchasing records to the activities claimed. Explain allocation assumptions and avoid blanket departmental percentages.
- Engineering phases: Use existing design-stage records as an organizational aid. Phase labels alone do not demonstrate experimentation.
- Contract review: Examine payment contingencies, rights, remedies, and governing law. Contract language should reflect commercial reality rather than manufacture a tax position.
- Officer activities: Separate actual research, immediate technical supervision, and direct support from general management. Under Treasury Regulation Section 1.41-2, higher-level supervision and general administration do not qualify merely because they benefit a research department.
- Evidence quality: Preserve drawings, source-control history, test results, technical correspondence, and reliable time allocations. Interview evidence can explain the record; unsupported recollections can leave material gaps.
Final Thoughts
Research-credit claims are strongest when technical evidence and financial calculations support one another. The verified decisions illustrate fact-specific failures of proof, not a new doctrine of absolute substantiation or the end of all estimation. Good records improve defensibility without guaranteeing allowance.
Businesses should apply the rules for the relevant tax year, distinguish Section 41 credit eligibility from Sections 174 and 174A expense treatment, and follow the applicable Form 6765 and refund-claim procedures. Integrating these tasks into normal project administration can make the resulting study more accurate and easier to substantiate.
