The federal research credit under Internal Revenue Code (IRC) Section 41 encourages qualifying research conducted in the United States and other eligible U.S. jurisdictions. For architectural and engineering firms working under customer contracts, eligibility depends both on the activities performed and on the funded research exclusion. The architectural litigation examined in this study is Smith v. Commissioner, involving Adrian Smith + Gordon Gill Architecture, LLP (AS+GG).

The Tax Court’s December 18, 2024 order denying the IRS’s motion for summary judgment did not establish that the claimed research was unfunded or that the taxpayers were entitled to the credits. The subsequent opinion, Smith v. Commissioner, T.C. Memo. 2026-50, filed June 16, 2026, reached a mixed result: payment under all six sample contracts was not contingent on research success, but AS+GG retained substantial rights under four contracts. Any credits for those four projects remained subject to the regulatory funding allocation and further computation.

The Statutory Architecture of the Section 41 Research Credit

Innovation, complexity, or professional expertise alone does not establish eligibility. Activities must satisfy the statutory requirements for qualified research, avoid the applicable exclusions, and generate eligible expenses. The analysis applies separately to each business component.

The Four-Part Test

For the historical tax years considered in Smith, the expenditure requirement referred to Section 174. Under the amended statute applicable to tax years beginning after December 31, 2024, Section 41(d)(1)(A) refers to domestic research or experimental expenditures under Section 174A. Deduction treatment and credit eligibility are separate questions: meeting the research-expenditure requirement does not establish that every related cost qualifies for the credit.

Research must seek to resolve uncertainty about the capability, method, or appropriate design of a new or improved business component. The work must fundamentally rely on physical or biological sciences, engineering, or computer science. The intended improvement must concern function, performance, reliability, or quality; purely aesthetic or cosmetic objectives do not qualify.

A business component can be a product, process, computer software, technique, formula, or invention held for sale, lease, or license, or used in the taxpayer’s trade or business. The process of experimentation requirement calls for identifying uncertainty, identifying alternatives, and evaluating alternatives through an appropriate technical process, such as modeling, simulation, or systematic trial and error. Research need not advance the knowledge of an entire industry, and an unsuccessful research effort can qualify.

Under Treasury Regulation Section 1.41-4(a)(6), at least 80% of the relevant research activities must constitute elements of a process of experimentation for a qualified purpose, measured by cost or another consistently applied reasonable basis. This is an activities test, not a measure of how much of the finished product is new. The remaining activities must also satisfy the applicable expenditure requirement and avoid the statutory exclusions.

The Funded Research Exclusion under Section 41(d)(4)(H)

Section 41(d)(4)(H) excludes research to the extent funded by another person or governmental entity. Treasury Regulations Sections 1.41-4(c)(9) and 1.41-4A(d) require examination of payment contingency and retained substantial rights. All relevant agreements, including amendments and other arrangements affecting payment or ownership, must be considered.

Payments contingent on research success are treated differently from payments earned by performing services. Separately, a researcher that retains no substantial rights is treated as fully funded, even if it incurs unreimbursed costs. Incidental gains in experience do not establish substantial rights, but an enforceable nonexclusive right to use research results can suffice.

The rules are not simply an all-or-nothing allocation between customer and contractor. If the researcher retains substantial rights but receives noncontingent funding, the regulations provide for reducing research expenses by the applicable funding. Research expenses exceeding that funding may remain eligible, subject to the detailed allocation rules and all other credit requirements. A customer’s ability to claim contract research expenses requires its own analysis and does not automatically follow from disallowance of the contractor’s credit.

Analytical Overview of Smith v. Commissioner

AS+GG was a partnership whose research credits flowed through to its partners and their spouses. The litigation concerned credits generated in 2008, 2009, and 2010. The parties selected six sample architectural projects for trial: Atrium City Tower, Kingdom Tower, Masdar HQ, Atrium City Masterplan, Plot 14, and Plot R2.

The December 2024 summary judgment order left disputed funding issues for further proceedings. By the June 2026 opinion, the parties had resolved the four-part test and specified other qualification issues in the taxpayers’ favor. The remaining trial issues concerned funded research and the reasonableness of the partners’ 2008 compensation. Accordingly, the later opinion should not be described as leaving experimentation to a future trial.

Economic Risk and Payment Milestones

The taxpayers argued that design milestones and contractual performance obligations tied compensation to successful results. The IRS argued that payment was earned by providing architectural services under professional standards. These competing positions explained why the earlier summary judgment proceedings were significant, but surviving that motion was not a ruling on the ultimate merits.

In the June 2026 opinion, the Tax Court concluded that none of the payments under the six sample contracts was contingent on the success of the research. Milestones, fixed prices, professional obligations, and the possibility of reduced profits did not establish the required contingency under those agreements. The opinion therefore does not support a general proposition that milestone billing makes architectural research unfunded.

The practical question is what the researcher must accomplish to become entitled to payment, including whether unsuccessful research leaves the customer obligated to pay. A project can involve real technical uncertainty while still being funded under Section 41.

Substantial Rights and Foreign Copyright Arguments

The court found that AS+GG retained substantial rights for Atrium City Tower, Masdar HQ, Atrium City Masterplan, and Plot R2. It did not retain substantial rights for Kingdom Tower and Plot 14. The analysis depended on the particular contracts and relevant amendments or settlements, rather than a general assumption that architects retain all rights in their designs.

For Plot 14, the court rejected reliance on automatic copyright protection under UAE law. The taxpayers had not established that those protections overrode the contractual terms, and the court explained that the rights could be contracted away. Foreign copyright law therefore did not provide a general solution to restrictive ownership or consent provisions.

For the four projects with retained substantial rights, the court allowed the possibility of partial credits under Treasury Regulation Section 1.41-4A(d)(3), to the extent research expenses exceeded the applicable payments. It did not determine a dollar amount: the evidence was insufficient to calculate the credits, if any. The opinion also upheld the partners’ total 2008 compensation as reasonable under the independent investor test applicable in the Seventh Circuit and rejected the challenge to the funded research regulations based on Loper Bright.

Comparative Jurisprudence: Smith, MBJ, and Phoenix Design Group

These cases address distinct requirements. A favorable result on research qualification does not resolve funding, and retained intellectual property rights do not alone prove that payments depend on research success.

MBJ and General Commercial Risk

In Meyer, Borgman & Johnson, Inc. v. Commissioner, 100 F.4th 986 (8th Cir. 2024), the Eighth Circuit affirmed the denial of research credits to a structural engineering firm on funded research grounds. The court distinguished general commercial risk and obligations to provide competent services from contractual payment contingent on successful research.

Fixed pricing, code compliance, inspection, and acceptance provisions did not establish the necessary contingency on the record before the court. The holding does not require a particular phrase in every qualifying agreement: the substantive rights and obligations control. Smith’s later merits opinion similarly demonstrates why fixed-price or milestone-based billing cannot be treated as an automatic safe harbor.

Phoenix Design Group and the Process of Experimentation

In Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113, the Tax Court found that none of the three trial projects involved qualified research. The firm’s ordinary design procedures and project records did not establish the required technical uncertainty and evaluative experimentation for those projects.

Standard architectural design phases, engineering calculations, revisions, and successful deliverables do not by themselves prove experimentation. The evidence must explain what was uncertain and how alternatives were evaluated. This was a failure to establish qualification on the particular record, not a categorical exclusion of architectural or engineering services. Unlike Phoenix, Smith proceeded after the parties had resolved the four-part test in the taxpayers’ favor.

Substantiation and the Limits of Estimation

The cases illustrate the need to connect claimed expenses to qualifying activities. They do not establish that every taxpayer must maintain a laboratory notebook, use a prescribed timekeeping application, or produce one exclusive category of contemporaneous record.

The Cohan Doctrine and Evidentiary Foundations

The Cohan doctrine can permit estimation when a taxpayer establishes entitlement to a tax benefit but cannot prove the precise amount. It does not allow a court to assume that research qualified, that a particular employee performed qualifying work, or that an unsupported percentage is reliable. Little Sandy Coal Co. v. Commissioner, 62 F.4th 287 (7th Cir. 2023), illustrates the need for a factual basis before estimation can help.

Betz v. Commissioner, T.C. Memo. 2023-84, likewise illustrates why describing a difficult project and its eventual solution does not establish the investigative activities performed. Useful evidence can include dated design alternatives, calculations, models, test results, technical meeting notes, and credible explanations from participants. Contemporaneous records generally strengthen the connection between activities and costs, while later reconstructions require reliable support.

Form 6765 and Research-Expense Deduction Changes

For tax years beginning in 2022 through 2024, Section 174 generally required capitalization and amortization over five years for domestic research and 15 years for foreign research, using a midpoint convention. Public Law 119-21 subsequently added Section 174A, which generally permits immediate deduction of domestic research or experimental expenditures for tax years beginning after December 31, 2024. Taxpayers may instead elect amortization over at least 60 months. Foreign research remains subject to 15-year amortization under Section 174.

For a full 12-month tax year under the former five-year domestic amortization rule, research costs of X generally produced a first-year deduction of (X ÷ 5) × 0.5 = 0.1X. Immediate deduction would instead produce a deduction of X. Ignoring other tax effects, the corresponding tax reduction is the allowable deduction multiplied by the applicable tax rate. This historical comparison does not describe a continuing mandatory five-year rule for new domestic expenditures under Section 174A.

Transition rules address unamortized domestic costs from 2022–2024, including an election for recovery over one or two tax years beginning with the first tax year beginning after December 31, 2024. Separate retroactive relief was provided for eligible small businesses, subject to eligibility and filing deadlines. Revenue Procedure 2025-28 supplies procedural guidance. The Section 41 credit and Section 280C coordination rules must be considered separately from the deduction.

Under the December 2025 Instructions for Form 6765, Section G is optional for tax years beginning before 2026 and required for tax years beginning after 2025, subject to exceptions. These include qualifying small businesses electing the payroll tax credit, and specified original-return filers meeting both the $1.5 million QRE and $50 million average-gross-receipts thresholds, as defined in the instructions.

Where Section G applies, detailed entries generally follow the 80%/Top 50 business-component rule, with remaining components aggregated. Information sought in column 49(f) currently applies to amended returns. Requirements for valid refund claims must also be checked separately. The instructions do not support a claim that every component always requires a full experimentation narrative or that an automated system automatically denies claims before an examination.

Implications for Future R&D Tax Credit Applications

The practical lessons concern contract review, technical evidence, expense allocation, and the law applicable to the claim year. Taxpayers should assess the commercial substance of their arrangements rather than assuming that terminology determines eligibility.

Contract Terms and Risk Allocation

Payment contingency: Examine the conditions for earning fees, including acceptance criteria, termination payments, refunds, and the consequences of unsuccessful research. Technical milestones help only where they affect substantive payment rights. Any contract changes should accurately reflect the intended commercial arrangement.

Warranties and remedies: Warranty provisions and governing law can affect who bears the financial consequences of failure. System Technologies, Inc. v. Commissioner, Docket No. 12211-21, illustrates the relevance of Indiana-law remedies in the funding analysis. A warranty is not independently sufficient; its scope and interaction with payment and breach provisions matter.

Rights to results: Identify enforceable rights to use relevant research results in future business, including nonexclusive rights where appropriate. Review confidentiality terms, assignments, licenses, consent requirements, and subsequent settlements. Ownership of a deliverable and rights to underlying research results may differ.

Documentation and the Shrinking-Back Rule

Treasury Regulation Section 1.41-4(b)(2) requires applying the qualification tests first to the discrete business component. If it does not qualify, the analysis proceeds to the most significant subset of its elements and continues as necessary. It is not a rule requiring taxpayers to begin with the smallest possible component or to omit a qualifying overall component.

For example, if an overall building-related business component fails the requirements, a qualifying subsystem may be examined through shrinking back. Qualification must still be demonstrated at that level; merely calling a subsystem novel or assigning it a project code does not establish eligibility.

Activity and cost records: Organize records so that qualifying work and associated wages, supplies, and other eligible expenses can be identified by component. Time records, payroll data, project documents, and supported allocation methods should be consistent.

Technical narratives: Explain the uncertainty, technical principles, alternatives, and evaluation actually undertaken. Use the statutory requirements to structure an accurate account, without substituting legal terminology for evidence.

Direct supervision and support: Treasury Regulation Section 1.41-2(c) distinguishes immediate supervision of qualified research from higher-level management. Moore v. Commissioner, T.C. Memo. 2023-20, illustrates the evidentiary problems with unsupported executive wage claims. Job title alone is not decisive: an executive’s own qualifying research, dire