The federal Credit for Increasing Research Activities under Internal Revenue Code (IRC) Section 41 requires taxpayers to establish both qualifying activities and qualifying expenditures. Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113, illustrates the importance of evidence explaining what engineers actually did to resolve technical uncertainty. Hayden v. Commissioner, 112 T.C. 115 (1999), concerns a different provision: the Section 179 deduction. The two decisions should not be treated as successive stages of a single R&D eligibility test.

This study examines the limited relevance of Hayden, the substantive and evidentiary issues in Phoenix Design Group, the funded research exclusion, software requirements, and UK compliance developments. It also distinguishes the rules applicable to the historical years in the cases from the domestic research expensing provisions effective for tax years beginning after 2024.

Hayden v. Commissioner: The Scope of the Section 179 Decision

Hayden concerned the taxable-income limitation on the Section 179 election to expense qualifying property. It did not decide eligibility for the Section 41 research credit or research deductions under Section 174. Its entity-level reasoning must be understood within the specific statute and regulation at issue, rather than imported as a general restriction on research incentives.

Dennis and Sharon Hayden claimed a $17,500 Section 179 deduction allocated by Leddos Frozen Yogurt, LLC, which was treated as a partnership and had a loss for 1994. They argued that the partnership was not a taxpayer under Section 701. The Tax Court rejected the argument, and the Seventh Circuit affirmed in Hayden v. Commissioner, 204 F.3d 772 (2000). Treasury Regulation Section 1.179-2(c)(2) applies the Section 179 taxable-income limitation at both partnership and partner levels.

Entity-Level Restrictions on Research Incentives

That Section 179 limitation does not establish a taxable-income ceiling for generating or allocating Section 41 research credits. A business operating at a loss may conduct qualified research. Whether an owner can use an allocated credit currently is a separate question involving the applicable credit limitations, carryover provisions, and other relevant rules.

Research claimants should identify the entity conducting the work, its trade or business, and the expenses it actually incurs. The research credit has its own pass-through and controlled-group rules. Eligible qualified small businesses may elect the payroll tax credit under Section 41(h). Neither an immediate profit nor the characterization of an entity as a taxpayer in Hayden independently determines R&D eligibility.

Phoenix Design Group: Engineering Activities and Evidentiary Requirements

On December 23, 2024, the Tax Court issued Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113. The engineering firm designed mechanical, electrical, plumbing, and fire protection systems for hospitals and laboratories. The dispute concerned research from 2013 through 2016 and credit years 2015 through 2019. More than 200 projects were implicated, but the parties tried a nonbinding sample of three projects. The opinion held that those three projects did not involve qualified research; it expressly did not determine the overall deficiency amounts.

The decision illustrates that technical complexity alone does not establish qualified research. The court examined the claimed uncertainty and the evidence of investigative and experimental activities. It did not categorically disqualify professional engineering or establish a new statutory test for the industry.

Applying the Four-Part Test in Professional Services

Section 41(d) requires each relevant business component to satisfy the qualifying research requirements. The following table describes the historical tests considered in Phoenix Design Group. The current statute refers to Section 174A for the expenditure requirement for applicable post-2024 years; references to Section 174 remain appropriate when describing the historical litigation.

Deconstructing the Section 174 Uncertainty Threshold

Technical uncertainty concerns capability, method, or appropriate design in light of the information available when the research begins. Treasury Regulation Section 1.174-2 distinguishes experimental work from ordinary activities. The possibility that a design may later change does not, by itself, demonstrate that its appropriate design was technically uncertain. A later change may create a narrower issue that must be evaluated on its own facts.

Business uncertainty about profitability, customer preferences, or the timing of client instructions is different from technical uncertainty. However, use of established engineering principles, standard software, or commercially available tools does not automatically disqualify research. The relevant question is whether the actual activities meet the statutory and regulatory requirements.

The Rigor of the Process of Experimentation (PoE)

The process of experimentation requirement asks whether the taxpayer evaluated alternatives to resolve technical uncertainty for a permitted purpose. Under Treasury Regulation Section 1.41-4(a)(6), at least 80% of the research activities for the business component, measured by cost or another consistently applied reasonable basis, must constitute elements of that process. This is not a rule that 80% of all company activities must be R&D.

PDG described a six-stage workflow: Basis of Design, Schematic Design, Design Development, Construction Document, Bidding, and Construction Administration. Naming a workflow or showing design revisions did not establish the required investigative activities. The court required evidence of the work performed, rather than accepting the firm’s overall process description as sufficient proof.

Implications for Engineering and Software Development

A useful substantiation framework explains the reasons for technical decisions and connects them to actual research activities. Companies should retain evidence showing that they:

• Identified the relevant technical uncertainty when the research began.
• Identified one or more alternative approaches intended to resolve that uncertainty.
• Evaluated the alternatives through an appropriate process, such as modeling, simulation, or systematic trial and error.
• Recorded the results and explained resulting decisions, including unsuccessful attempts.

Section 41(d)(4)(B) excludes research related to adapting an existing business component to a particular customer’s requirement or need. Uncertainty alone does not automatically override this exclusion. Separately identifiable development or improvement work must satisfy the four-part test and avoid the applicable exclusions.

Documentation and Contemporaneous Record-Keeping

Phoenix Design Group and Little Sandy Coal Co. v. Commissioner illustrate the risks of unsupported activity allocations and insufficient evidence of experimentation. Hayden is not an R&D documentation decision. The Cohan estimation principle has not been categorically abolished for research credits, but estimates cannot substitute for proof that qualified research occurred or for a reasonable factual basis for quantifying qualifying expenditures.

In Phoenix Design Group, generic time descriptions and unexplained design materials did not adequately connect the engineers’ work with particular uncertainties and investigative activities. Testimony and project records should support one another. The practical lesson is to explain the activities behind recorded time, not simply to produce a large volume of drawings or a general project narrative.

Best Practices for Future Substantiation

Companies should maintain records that distinguish qualifying research, direct supervision, and direct support from routine production, administration, and other nonqualifying work. Useful evidence can include design histories, technical emails, test results, issue trackers, calculations, and credible time allocations. The classifications below are illustrative; the facts and the employee’s actual role determine treatment.

Treasury Regulation Section 1.41-4(d) requires records in sufficiently usable form and detail to substantiate eligibility and amount. It does not prescribe one universal activity-log format or categorically prohibit retrospective studies. Contemporaneous evidence is often stronger, but a later study can organize and explain reliable underlying records. The cited IRS research-credit audit guide dates to June 2005 and should not be presented as newly issued guidance or binding law.

Funded Research and the Allocation of Financial Risk

Section 41(d)(4)(H) excludes research to the extent another person or governmental entity funds it. Treasury Regulation Section 1.41-4A(d), incorporated into the funded research rules, addresses contingent payment and substantial rights. Customer payment does not automatically make research funded. The contract, governing law, and retained rights must be examined together.

Orders discussed in January 2025 commentary on System Technologies, Inc. v. Commissioner and Smith v. Commissioner addressed IRS summary-judgment motions; they should not be described as blanket final approvals of the claimed credits. In System Technologies, the analysis considered Indiana law and refund remedies following total breach. The Smith discussion concerned disputed contractual rights and the effect of governing foreign law. These procedural rulings underscore the importance of contract interpretation without resolving every qualification issue.

Rights Retention and Contractual Language

The taxpayer must retain substantial rights to the research results. Exclusive ownership or patent ownership is not invariably required, but having to pay to use the results can prevent the rights requirement from being satisfied. Financial risk and substantial rights are separate inquiries. Phoenix Design Group’s sampled-project holding should not be presented as a funded research ruling based on professional-standard contract language.

Contract reviews should address payment conditions, acceptance criteria, termination, warranties, remedies, and rights to use the results. A repair-or-replacement clause does not automatically eliminate financial risk: governing law may provide other remedies for total failure. Conversely, a fixed-price label alone does not prove that payment is contingent on successful research. Contract terms must reflect the actual arrangement.

The Shrinking-Back Rule: A Strategy for Salvaging Claims

A critical tool for taxpayers in the wake of Phoenix Design Group is the “shrinking-back rule” found in Treasury Regulation § 1.41-4(b)(2). If a business component—such as an entire building or a complex ship—fails the four-part test, the test may be applied to a subset of that component, such as a specific sub-system or part.

The court examined smaller systems and subcomponents in Phoenix Design Group, but the evidentiary deficiencies persisted. Taxpayers should identify the level at which the research occurred and support the related activities and costs. Shrinking back is not permission to select an unsupported percentage of an otherwise nonqualifying project.

Global Perspective: UK R&D Reform

The UK operates a separate statutory framework. HMRC guidance provides that the merged R&D expenditure credit scheme and enhanced R&D intensive support (ERIS) replace the former schemes for accounting periods beginning on or after 1 April 2024. This timing depends on the accounting period, rather than merely the date a claim is filed.

UK contracted-out R&D rules require analysis of who decided or contemplated that R&D would be undertaken, subject to detailed statutory conditions and exceptions. They should not be reduced to a US-style financial-risk test. Moore Kingston Smith has discussed how these changes affect manufacturers, with Thomas Hayden identified in its commentary; that professional commentary is not a judicial holding or HMRC rule.

Comparative Compliance Standards

The US 80% experimentation test and the UK ERIS intensity condition measure different things. The latter generally concerns whether relevant R&D expenditure reaches 30% of total expenditure, including connected companies and applicable grace provisions. Neither establishes a shared US–UK audit presumption that claims are incorrect.

Internal Use Software (IUS) and the High Threshold of Innovation

Software developed primarily for general and administrative functions may be internal-use software under Treasury Regulation Section 1.41-4(c)(6). Applicable internal-use software must meet the high-threshold-of-innovation requirements as well as the ordinary research tests. Software enabling interactions with third parties may fall outside that classification, and dual-function software has specific rules. Software used by a bank or consultancy is not automatically internal-use software merely because the business supplies services.

Where the high-threshold-of-innovation test applies, the software must:

• Be intended to produce a substantial and economically significant improvement, such as reduced cost or increased speed.
• Involve substantial resources and significant economic risk arising from technical uncertainty over their recovery within a reasonable period.
• Not be commercially available for the intended purpose without modifications that themselves satisfy the innovation and economic-risk requirements.

United Stationers addressed an older regulatory framework and should not be used to impose an industry-wide discovery requirement on current claims. Current regulations expressly allow reliance on existing science and technology. Customizing vendor software is neither automatically eligible nor automatically excluded; the particular development activities and applicable software rules control.

Economic Reality and the Substance-Over-Form Doctrine

Substance-over-form principles may matter when the tax characterization of an arrangement differs from its actual operation. However, research credit eligibility ordinarily requires application of Section 41 to the activities, expenses, and contractual arrangements established by the evidence. Economic-substance analysis should not be asserted as the basis of a case without support in the opinion.

Hayden applied the Section 179 income limitation; it did not hold that Leddos was a sham research partnership. Phoenix Design Group concerned failure to prove qualifying research on the trial projects. Neither holding supports a general claim that the court found an artificial research-credit shelter in those cases.

The Role of Accuracy-Related Penalties (Section 6662)

Section 6662 generally imposes a 20% accuracy-related penalty on the portion of an underpayment attributable to specified grounds, including negligence or substantial understatement. Disallowance of a research credit does not automatically establish a penalty. Reasonable cause and good faith under Section