Home »
Research Tax Credit »
Federal »
Case Law »
Caveney v. Commissioner
Caveney v. Bower (2003) is an Illinois Supreme Court decision concerning state-law R&D credit eligibility for S-corporation shareholders and statutory retroactivity. It did not abolish the federal discovery test or establish the federal consistency rule. Federal R&D tax credit standards, including technical uncertainty and process of experimentation, remain governed strictly by Section 41 and Treasury regulations.
The federal credit for increasing research activities under Section 41 of the Internal Revenue Code has evolved through legislation, Treasury regulations, and judicial decisions since its introduction in 1981. This study examines that development, distinguishing the Illinois decision in Caveney v. Bower, 207 Ill. 2d 82 (2003), from the federal rules governing technical uncertainty, experimentation, and consistent credit calculations. Caveney concerned state-law eligibility for S-corporation shareholders and statutory retroactivity; it did not abolish the federal discovery test or establish the federal consistency rule. Those federal standards arise from Section 41 and its implementing regulations.
Historical Context and the Legislative Intent of Section 41
The R&D tax credit was initially introduced as a temporary incentive to combat the stagnation of private-sector research investment during the late 1970s. The primary goal was to provide an immediate reduction in the after-tax cost of research, thereby encouraging firms to undertake higher-risk projects that might otherwise be economically unfeasible. Unlike the Section 174 deduction, which had been part of the code since 1954 and allowed for the expensing of research costs, Section 41 was intended as an incremental credit—rewarding only the increase in research spending over a determined base period.
The Tax Reform Act of 1986 narrowed the definition of qualified research by introducing the statutory requirements commonly described as the four-part test. Disputes subsequently arose over whether discovering technological information required an advance beyond the knowledge of skilled professionals. The statute and applicable regulations, including their effective dates, must be distinguished from later descriptions of those historical disputes.
The 2001 final regulations required information that exceeded, expanded, or refined the common knowledge of skilled professionals in the relevant field. Treasury later abandoned that requirement. Although applied engineering can qualify, neither the use of scientific principles nor the novelty of a project alone establishes eligibility. The eventual regulatory change should not be attributed to Caveney.
The Discovery Test Conflict and Regulatory Change
In Eustace v. Commissioner, 312 F.3d 905 (7th Cir. 2002), the Seventh Circuit upheld denial of credits for software-development activities under the standards applicable to the years at issue. The court also addressed inadequate support for employee wage allocations. Its treatment of the discovery requirement must be read in its historical regulatory context; proposed regulations did not themselves change the governing law.

Regulatory Era
Key Interpretation
Primary Objective

1981–1985
Research credit introduced with a broader research definition.
Encourage increased research investment.
1986–2001
Four-part statutory test; later regulations applied an industry-knowledge discovery standard.
Define qualifying technological research.
December 2001 proposal
Proposed removal of the industry-knowledge discovery requirement.
Refocus on technical uncertainty.
January 2004 final regulations onward
Uncertainty and process-of-experimentation requirements; subsequent amendments also apply.
Evaluate qualifying activities and substantiation.
Tax & Accounting Software Corp. v. United States likewise formed part of the historical dispute over software research and the discovery requirement. Treasury issued final regulations in January 2001, proposed their revision in December 2001, and adopted revised final regulations in January 2004 through Treasury Decision 9104. The revised rule focuses on eliminating uncertainty about developing or improving a business component rather than expanding industry-wide knowledge.
Analyzing Caveney v. Bower: Facts and Findings
Jack and Margaret Caveney were shareholders in Panduit Corporation, which elected S-corporation treatment for 1993, 1994, and 1995. They claimed Illinois research credits based on expenditures incurred by Panduit. Illinois disallowed the credits and assessed $1,091,131. in tax and interest, which they paid under protest. The litigation concerned shareholder entitlement under Illinois law and whether a 1999 statutory amendment applied to those earlier years.
The Trade or Business Requirement for Pass-Through Entities
In Caveney v. Bower, the Illinois Supreme Court held that the pre-amendment statute did not authorize the shareholders to claim credits for Panduit’s expenditures. The 1999 amendment creating shareholder eligibility was substantive and did not apply retroactively. The court also rejected the uniformity-clause challenge and directed judgment for the State. This state-law outcome does not override federal pass-through credit provisions or establish a general requirement that shareholders personally conduct corporate research.
The Demise of the Federal Discovery Test
Treasury Regulation Section 1.41-4(a)(3) provides the relevant federal standard. Research need not produce information beyond the knowledge of skilled professionals. Instead, it must seek to eliminate uncertainty concerning the capability, method, or appropriate design of a business component. The taxpayer must also meet the remaining statutory tests and avoid applicable exclusions.
This framework accommodates incremental industrial research even when competing products already exist. However, a product being new to the taxpayer is insufficient by itself. Routine adaptation, duplication, and ordinary quality-control activities remain subject to statutory exclusions. The taxpayer’s actual technical uncertainty and evaluation of alternatives determine whether its development work qualifies.
The Consistency Rule: Maintaining the Incremental Baseline
Section 41(c)(5)(A) and Treasury Regulation Section 1.41-3(d) require consistency between the credit-year and base-period determinations. For the regular credit, the taxpayer compares current qualified research expenses (QREs) with a base amount. This requirement is statutory and regulatory, rather than a holding of Caveney.
The Mechanics of Consistency
The consistency rule provides that the QREs and gross receipts taken into account in computing the fixed-base percentage must be determined on a basis consistent with the determination of QREs for the credit year. This ensures that the credit accurately measures the relative increase in research spending compared to what the taxpayer “typically” spent relative to its gross receipts.
If a taxpayer newly identifies a qualifying expense category in the credit year, it must account for comparable qualifying base-period expenses when determining its fixed-base percentage. Conversely, costs that do not qualify under the credit-year standard must be excluded from the base-period calculation. Matching job titles is not required: the relevant question is whether the underlying activities and expenses receive consistent treatment.

Component of Calculation
Consistency Requirement
Operational Impact

Wage Categories
Apply consistent qualifying-activity and expense standards; identical job titles are unnecessary.
Avoid distortions from inconsistent expense classifications.
Supply Definitions
Uniform treatment of “used in the conduct of” research.
Adjusts for shifts in inventory vs. R&D accounting.
Business Components
Apply consistent research-qualification criteria to relevant activities.
Support comparable classifications across periods.
Consistency operates in both directions. It prevents selective classification from overstating the increase in research spending, while requiring comparable treatment when expenses are excluded. Expiration of the refund limitations period for a base year does not eliminate the need to determine that year’s inputs correctly. These principles follow from the statute and regulation, not an independent judicial rule attributed to Caveney.
The 1984-1988 Base Period and Substantiation Challenges
For established taxpayers using the regular credit, the fixed-base percentage commonly draws on 1984–1988 data, subject to the statutory rules for start-up companies and other adjustments. The base amount also uses average gross receipts for the four preceding tax years and is subject to a minimum-base limitation. Historical substantiation can therefore be important. The IRS consistency audit guide discusses Research, Inc. v. United States as an example of inadequate base-period support. The alternative simplified credit generally uses the preceding three years of QREs instead of the 1984–1988 fixed-base calculation.
Modern Jurisprudence: The Shift from Discovery to Process
After Treasury removed the industry-wide discovery requirement, disputes continued over whether taxpayers demonstrated qualifying experimentation. Little Sandy Coal Co. v. Commissioner, 62 F.4th 287 (7th Cir. 2023), and Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113, illustrate the importance of proving the activities actually performed. These cases apply statutory requirements; they should not be described as developments caused by Caveney or as creating a universal new recordkeeping format.
The Four-Part Test in Detail
The qualification tests apply separately to each business component: a product, process, computer software, technique, formula, or invention held for sale, lease, or license, or used in the taxpayer’s trade or business. Product research and research on the associated commercial production process are tested separately.
Research-expenditure test: The research must satisfy the expenditure requirement in Section 41(d)(1)(A). The current statutory cross-reference is Section 174A; earlier tax years used Section 174. The relevant concept concerns experimental or laboratory work intended to resolve technical uncertainty. Deduction eligibility alone does not establish credit eligibility.
Technological in Nature Test: The research must fundamentally rely on principles of engineering, physics, chemistry, biology, or computer science.
Business Component Test: The research must be for a “permitted purpose,” such as developing a new or improved function, performance, reliability, or quality.
Process of experimentation test: At least 80% of the relevant research activities must constitute elements of a process of experimentation for a permitted purpose. The activities may be measured by cost or another consistently applied reasonable basis. This is an activity test, not a test based on the percentage of a finished product that is new.
Little Sandy Coal and the “Substantially All” Fraction
Little Sandy Coal involved credits associated with 11 first-in-class vessels, with two vessels selected for trial. The taxpayer relied heavily on novelty and estimated employee allocations. The Seventh Circuit affirmed the disallowance because the taxpayer did not substantiate the required proportion of qualifying research activities.
The denominator consists of research activities whose expenses satisfied the applicable Section 174 requirement and that were not excluded under Section 41(d)(4). The numerator consists of those activities that were elements of experimentation. The appellate court rejected the Tax Court’s categorical exclusion of direct support and supervision from the numerator: qualifying activities can belong in both parts of the fraction.
Prototype fabrication can be part of experimentation when it helps evaluate alternatives and resolve technical uncertainty. Little Sandy Coal did not categorically exclude pilot-model production. However, novelty and arbitrary time estimates did not establish the necessary activity allocation. This distinction makes the decision both an evidentiary warning and a correction of an overly restrictive interpretation.
The Shrink-Back Rule
Treasury Regulation Section 1.41-4(b)(2) requires testing the overall business component first and, if it fails, proceeding to its most significant subset of elements. The process continues until a qualifying subset is identified or the most basic element fails. For example, a qualifying propulsion subsystem may support a credit even if an entire vessel does not. The taxpayer must substantiate the qualifying activities and related expenses; shrinking back is not automatic approval of any selected cost percentage. Contemporaneous evidence is useful, but the regulation does not mandate a single form of project-time record.
Funded Research: Risk and Rights in Contractual Arrangements
Section 41(d)(4)(H) excludes research to the extent funded by another person or governmental entity. Customer contracts require careful analysis: the existence of a contract or a payment does not automatically mean all research is funded. Treasury Regulation Section 1.41-4A(d) addresses payment contingencies and rights in the research results.
The Two-Pronged Funding Test
To determine whether research is funded, the courts analyze two specific factors: the retention of “substantial rights” and the bearing of “economic risk”.
Economic risk: Consider whether payment is contingent on successful research. A right to reimbursement regardless of success generally indicates funding to that extent. Contract labels, such as fixed-price or time-and-materials, are relevant but do not replace examination of the operative terms and governing law.
Substantial rights: The contractor must retain substantial rights to use the research results. Exclusive ownership is not required, and incidental experience alone may be insufficient. An obligation to pay for use of the results can prevent the contractor from retaining substantial rights under the regulation.
In Meyer, Borgman & Johnson, Inc. v. Commissioner (8th Cir. 2024), the court affirmed that the engineering firm’s research was funded. General professional-performance obligations, inspection provisions, and the possibility of contractual liability did not establish that payment depended on successful research. Commercial risk must be evaluated in relation to the specific payment obligation, rather than assumed to establish the required contingency.
Milestone Payments and Local Law Remedies
The summary-judgment decisions discussed in Smith v. Commissioner and System Technologies, Inc. v. Commissioner illustrate the significance of governing law. In Smith, unresolved issues concerning contractual rights and foreign law prevented summary judgment for the IRS. In System Technologies, Indiana-law remedies supported the conclusion that payments for the projects at issue depended on successful performance, and the IRS’s partial-summary-judgment motion was denied. These rulings should not be presented as blanket approval of the taxpayers’ entire research credits or as proof that any refund remedy establishes eligibility.

Case
Result
Key Factor

Meyer, Borgman & Johnson
Funded; appellate affirmance in 2024.
Payment was not shown to depend on research success.
Geosyntec Consultants
Capped contracts on appeal were funded; fixed-price contracts had a different result below.
Cost-overrun exposure alone did not make the capped-contract payments contingent on success.
System Technologies
IRS partial-summary-judgment motion denied in the discussed 2025 ruling.
Indiana-law remedies informed the payment-contingency analysis; other credit requirements remained separate.
Smith v. Commissioner
IRS summary-judgment motion denied in the discussed ruling.
Unresolved contractual and foreign-law issues; this procedural ruling was not final allowance of the credit.
The Impact of Sections 174 and 174A on Research Expenditures
The Tax Cuts and Jobs Act of 2017 required capitalization of specified research expenditures for tax years beginning after December 31, 2021. Domestic expenditures were amortized over five years and foreign expenditures over 15 years, beginning at the midpoint of the year incurred. Subsequent legislation changed domestic treatment for tax years beginning after 2024.
Broader Definition of Section 174 Expenditures
The research-expenditure pool can be broader than Section 41 QREs. Properly allocable overhead and depreciation may enter the research-expenditure calculation, while the credit uses specifically defined categories, including qualifying wages, supplies, computer-use costs, and the allowable share of contract research. Acquisition costs of depreciable research equipment are not themselves qualifying supplies. Taxpayers should reconcile the two calculations rather than assume that all deductible research costs earn a credit.
Public Law 119-21, enacted July 4, 2025, added Section 174A and restored immediate deductions for domestic research expenditures in tax years beginning after December 31, 2024, with an elective capitalization alternative. Foreign research remains subject to 15-year amortization. Transition provisions allow accelerated recovery of remaining domestic 2022–2024 balances, and eligible small businesses could elect retroactive domestic-expensing treatment subject to statutory and procedural deadlines. Revenue Procedure 2025-28 provides implementing procedures. Section 280C coordination also remains relevant.
The Future of Documentation: Audit Trends and the Gatekeeper Review
The IRS announced additional information requirements for research-credit refund claims in 2021, with implementation beginning in January 2022. These filing requirements must be distinguished from proving substantive eligibility during examination. The IRS subsequently reduced the required upfront information and extended the transition period for correcting deficient claims.
Mandatory Elements of a Valid Claim
Under the IRS requirements applicable to research-credit refund claims postmarked on or after June 18, 2024, the claim must provide:
Identification of all business components to which the claim relates for the claim year.
Identification of the research activities performed for each business component.
Total qualified employee wage, supply, and contract research expenses for the claim year; Form 6765 may be used for these totals.
The IRS waived the upfront requirements to identify each individual and the information each individual sought to discover. That information may still be requested during examination.
The transition period allowing taxpayers 45 days to perfect deficient research-credit refund claims has been extended through January 10, 2027. Filing compliance does not by itself establish that the claimed activities and costs qualify.
Treasury Regulation Section 1.41-4(d) requires records in sufficiently usable form and detail to substantiate eligibility and amount. Contemporaneous engineering records, test results, payroll support, and reliable allocations can help. The rules do not universally require a particular timekeeping system or categorically bar credible testimony and reconstructed evidence, but unsupported estimates remain vulnerable. Eustace illustrates the evidentiary problems created by inadequate support for employee allocations.
Statistical Sampling and Discovery
Statistical sampling can help evaluate large populations of projects, but a taxpayer’s own sample does not automatically limit IRS discovery. In Kapur v. Commissioner, T.C. Memo. 2024-28, the Tax Court declined to confine discovery to the taxpayer’s preferred projects. Information about the larger population was relevant to selecting a representative sample. Early agreement on scope and methodology can reduce disputes, but sampling does not eliminate the burden of substantiating credit eligibility.
Final Thoughts
The development of the federal research credit reflects changes in regulations and continuing judicial scrutiny of qualification and proof. Treasury’s removal of the industry-wide discovery requirement allowed the focus to rest on the taxpayer’s technical uncertainty and experimentation. Caveney v. Bower remains a separate Illinois decision about shareholder eligibility and retroactivity.
The consistency requirement protects the integrity of the incremental calculation by applying comparable classifications across relevant periods. Historical records, current expense support, and the selected calculation method all matter. Sections 174, 174A, and 280C also require coordination with the credit calculation.
Manufacturers and service firms should connect their claimed costs to identifiable qualifying activities. Useful records explain the uncertainty, alternatives evaluated, work performed, and expense allocation. Contract research also requires analysis of payment contingencies and retained rights. A proper review of these issues is essential.