Historical Context: The Fokker Nexus in United States Tax and Regulatory Law

Estate of Fokker v. Commissioner, 10 T.C. 1225 (1948), addressed whether Anthony Fokker was domiciled in the United States at death and the valuation of certain foreign assets. The Tax Court found U.S. domicile after considering his circumstances. This individual estate-tax decision did not establish a general rule placing all foreign aerospace companies within the full scope of U.S. taxation or determine research-credit eligibility.

United States v. Fokker Services B.V., 818 F.3d 733 (D.C. Cir. 2016), addressed the district court’s refusal to exclude time under the Speedy Trial Act for a deferred prosecution agreement because it considered the prosecution too lenient. The D.C. Circuit held that this provision did not authorize the court to second-guess the Executive’s charging decisions on that basis.

Fokker’s investigation and cooperation formed part of its criminal-resolution context. They did not create a documentation threshold for civil tax claims. A comparison between corporate compliance records and research-credit records is an organizational analogy only; the tax requirements arise from the Internal Revenue Code, Treasury regulations, and relevant tax decisions.

Theoretical Foundation: The Section 41 Four-Part Test

Research must satisfy all four requirements for the relevant business component, subject to statutory exclusions. Failure for one component does not automatically invalidate every other component in a claim. Separately eligible parts may require analysis under the shrinking-back rule.

The Research-Expenditure Test: The Elimination of Uncertainty

The cases discussed here applied the Section 174 reference in the law governing their historical tax years. For taxable years beginning after December 31, 2024, the amended Section 41(d)(1)(A) refers to expenditures eligible for treatment under Section 174A. The research-expenditure inquiry concerns experimental development or improvement: uncertainty can relate to capability, method, or appropriate design.

Routine calculations do not automatically demonstrate qualifying uncertainty or experimentation. Conversely, the law does not require a patentable invention, industry-wide novelty, or a technically unprecedented problem. Engineering work must be assessed from the taxpayer’s available information and the activities actually performed, rather than from the project’s description alone.

The Technological Nature and Permitted Purpose Tests

The work must fundamentally rely on physical or biological science, engineering, or computer science, and seek improvement in function, performance, reliability, or quality of a business component. A proposed aerospace example could involve evaluating alternative materials to improve structural performance. The example would still require its own factual qualification analysis; the age or reputation of an aircraft design establishes nothing by itself.

The Process of Experimentation (PoE) Test

The process must evaluate alternatives to resolve relevant uncertainty. Modeling, simulation, and systematic trial and error can provide evidence of that process. A formal document labeled “hypothesis” is not universally required, and a standard design workflow is not sufficient merely because it can be described using scientific terminology.

This comparison describes a change in the discovery standard, not an exemption from the other qualification rules. Meeting the four-part test and avoiding an exclusion are separate requirements. Technical uncertainty alone does not override an applicable exclusion.

The “Substantially All” Threshold and the Little Sandy Coal Precedent

Little Sandy Coal Co. v. Commissioner, 62 F.4th 287 (7th Cir. 2023), affirmed disallowance where the taxpayer did not substantiate that at least 80% of the relevant research activities constituted elements of experimentation. The company claimed expenses for 11 vessels; two vessels were selected for trial. Newness and unsupported allocations did not establish qualification.

The Quantitative Fraction

The numerator consists of research activities constituting elements of experimentation. The denominator consists of the relevant research activities eligible under the historical Section 174 standard and not excluded under Section 41(d)(4). Measurement must use cost or another consistently applied reasonable basis.

Experimentation percentage = (activities constituting elements of experimentation ÷ relevant non-excluded research activities) × 100; required threshold: at least 80%.

The Seventh Circuit rejected a categorical exclusion of direct supervision and direct support from the numerator. These activities may count when they constitute elements of experimentation; they should not be added automatically or counted twice. Construction is likewise not excluded merely because it is construction, but building a new vessel does not make every production activity experimental.

The Cohan Rule and the Failure of Estimates

The court distinguished proving qualified research under Section 41(d) from estimating eligible expenses under Section 41(b). Expense estimation may be available after qualification is established. The Cohan rule cannot supply the missing proof that the taxpayer performed qualified research.

Accordingly, this decision does not abolish every reasonable estimate or require one particular timesheet system. The practical issue is whether evidence supports the allocation. Contemporaneous project records can strengthen that evidence, but unsupported percentages remain vulnerable regardless of how polished the final credit study appears.

Documentation Lessons from Phoenix Design Group (2024)

Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113, examined three trial projects involving mechanical, electrical, plumbing, and fire-protection engineering. The court found no qualified research in those projects and sustained accuracy-related penalties associated with them. The opinion did not automatically resolve the remaining 235 projects; the trial-project determinations were nonbinding as to those projects.

Narrative Deficiencies and AIA Design Phases

Broad design phases, including schematic design and design development, did not prove qualifying activity. The court examined the claimed uncertainty and the actual work, including inconsistencies between activity descriptions and the asserted process. The possibility that a design might later change did not itself establish technical uncertainty.

Moore v. Commissioner, No. 23-2681 (7th Cir. 2024), affirmed T.C. Memo. 2023-20 concerning the compensation of Nevco’s president and COO, Gary Robert. The appellate court emphasized the inability to establish what fraction of his activity was qualified research. The decision did not announce that credible testimony is categorically inadmissible or that every executive’s wages are ineligible.

Scientific Substantiation: The Fokker F28 Case Study

NASA and FAA research involving a full-scale Fokker F28 crash test in 2019 illustrates the relationship between physical testing and computational models. It does not establish a private taxpayer’s right to a research credit. Government funding, contractual rights, the claimant’s expenses, and the other statutory requirements would require separate examination.

The Scientific Methodology in Fokker F28 Testing

NASA’s study, Simulation of a Full-Scale Crash Test of a Fokker F28 Fellowship Aircraft, describes finite-element modeling using LS-DYNA and comparison with measured crash-test responses. The aircraft impacted a soil bed at approximately 65.3 feet per second horizontally and 31.8 feet per second vertically. The analysis included quantitative test-to-model comparison using ISO/TR 16250 methodology.

• Model development: Establish a computational representation and identify assumptions affecting predicted structural response.
• Physical testing: Collect measurements under defined impact conditions.
• Comparison: Evaluate agreement and differences between predicted and measured responses.

These are useful examples of engineering evidence. Section 41 does not require LS-DYNA, a particular crash-test standard, or an equivalent level of project expense. A commercial claimant would need to explain how its own evidence establishes eligibility and supports the expenses claimed.

The Fokker-Planck Equation and Technological Nature

The source’s discussion of Fokker-Planck applications does not establish a connection between the cited scientific work, Fokker Services, and a research-credit claim. Advanced mathematical terminology should be included only where it describes the actual project. A credit study should explain what the taxpayer did and why, without implying eligibility from unrelated research or a shared name.

The Adaptation and Funded Research Exclusions

The exclusions require independent attention even when a project involves technical work. They can be particularly relevant where an engineering business customizes existing products or performs research for customers.

Adaptation of Existing Components

Section 41(d)(4)(B) excludes research related to adapting an existing business component to a particular customer’s requirement or need. Whether work is an excluded adaptation or development of a qualifying new or improved component depends on the facts. Identifying a new uncertainty does not automatically defeat this exclusion. The source’s connection between C-390 modification services and the Fokker sanctions agreement is not substantiated and is not a basis for tax qualification.

Economic Risk and Substantial Rights

For contracted research, Treasury Regulation Section 1.41-4A(d) requires analysis of funding and retained rights. Payment that is contingent on successful research can place the relevant risk on the performer; an unconditional entitlement to payment points toward funding. The performer must also retain substantial rights. Exclusive ownership of all resulting intellectual property is not invariably necessary.

In Meyer, Borgman & Johnson, Inc. v. Commissioner, 100 F.4th 986 (8th Cir. 2024), ordinary professional obligations and contractual quality provisions did not make payment contingent on successful research. The court affirmed disallowance under the funded research exclusion. This was not a decision about inadequate business-component descriptions in a refund claim.

Contract reviews should examine the actual payment conditions, remedies, and rights under all relevant agreements. The source’s generalized reference to “the Smith case” does not justify treating every milestone or fixed-price arrangement as unfunded. Contract labels and a potential risk of commercial loss do not by themselves decide the tax treatment.

Implications for Future R&D Tax Credit Applications

The tax cases support careful substantiation and contract review. They do not demonstrate that the Fokker criminal settlement caused a new Section 41 regime or that the IRS universally preapproves research credits.

The IRS Classifier Review System

IRS procedures include validity review of research-credit refund claims on amended returns. The IRS’s research-credit refund-claim FAQs describe review by employees and specialists; they do not establish automatic AI rejection of any claim considered weak.

For claims subject to these procedures, the IRS requires identification of relevant business components, the research activities for each, and total qualified wage, supply, and contract-research expenses. Since June 18, 2024, the IRS has waived the initial requirement to identify each individual and the information each sought to discover, although it may request that information in an examination. The published transition period through January 10, 2027 provides 45 days to perfect deficient claims covered by the procedures. Refund-claim validity and substantive entitlement are separate issues.

Best Practices for 2025–2026

The following are practical recommendations, rather than universally mandatory record formats:

• Project tracking: Connect employee activities and expenses to the relevant business components, with enough detail to assess separately eligible parts.
• Uncertainty records: Explain the capability, method, or design questions using the information available when the work began.
• Alternative evaluation: Retain relevant test results, simulations, design changes, and explanations of rejected approaches.
• Contract review: Identify who bears research-performance risk and what rights each party retains.
• Reconciliation: Connect the credit calculation to payroll, accounting records, and applicable tax-year rules.

Causal Relationships and Second-Order Insights

Organized records can support both corporate compliance and tax substantiation, but no causal relationship between the Fokker deferred prosecution agreement and the cited research-credit holdings has been established. Each legal regime has its own purpose and evidentiary questions.

Technical staff and tax professionals can work together to explain projects accurately. The cases do not require companies to hire a particular class of specialist or prevent a suitably competent accountant from preparing a research-credit claim.

The shrinking-back rule in Treasury Regulation Section 1.41-4(b)(2) calls for examining the most significant subset of elements when the requirements are not met for the entire business component, continuing as appropriate. For example, an aircraft subsystem may warrant separate analysis. Shrinking back does not guarantee a credit or excuse missing evidence.

The following table retains the original comparison headings. Its entries distinguish the historical TCJA regime from the federal changes enacted in 2025, including Section 174A.

Future Outlook: AI and Automation in R&D Substantiation

As a potential workflow aid, software can help organize engineering notes, issue-tracking records, and cost data. Any AI-generated description shoul