The Historical Foundations of Section 174 and the Driggs Controversy

Section 174, enacted in 1954, allowed taxpayers to elect current deductions for qualifying research and experimental expenditures connected with a trade or business. It reduced uncertainty about the tax treatment of research costs, including costs that otherwise could require capitalization. The statute initially contained no express requirement that the amount of an expenditure be reasonable. That omission became central to Driggs v. United States.

Factual Background of Driggs v. United States

Guy K. Driggs, a physician, was a partner in Typesetting Systems Research Joint Venture (TSR). TSR hired Typography Systems International, Inc. (TSI), in which Driggs was a shareholder, to develop typesetting software. TSR agreed to pay TSI $895,000 for its best efforts to produce two software packages. Evidence showed approximately $1.3 million of research spending over four years. The refund litigation concerned Driggs and his wife’s 1981 return, on which he deducted his pro rata share of TSR’s loss against medical-practice income.

The IRS disallowed the loss on the grounds that TSI was a sham and the research agreement was not entered into for profit. The taxpayers paid the resulting assessment and sought a refund. The jury found that TSI and the transactions were legitimate, but found that only $895,000 of the $1.3 million had been reasonably incurred. The court entered judgment for a refund of $102,735.12 plus statutory interest. The government’s post-verdict motion sought, among other relief, a reduction based on the jury’s reasonableness finding.

The Judicial Ruling on the Reasonableness Standard

The legal core of the Driggs case rested on whether the District Court could limit the taxpayer’s deduction to the “reasonable” amount found by the jury. The government filed a motion for judgment notwithstanding the verdict, arguing that a reasonableness limitation should be read into Section 174, similar to the “ordinary and necessary” requirements of Section 162. The court, however, rejected this invitation to engage in judicial legislation.

The court emphasized that where a statute is unfettered by higher court interpretation, the intent of Congress must be derived from the plain language of the legislative branch. Section 174(a)(1), as it existed in 1981, stated that a taxpayer “may treat research or experimental expenditures which are paid or incurred by him during the taxable year in connection with his trade or business as expenses which are not chargeable to capital account”. Unlike Section 162(a), which explicitly limits deductions for salaries to a “reasonable allowance,” Section 174 contained no such quantitative restriction.

The district court treated the statute’s limitation as definitional rather than quantitative and refused to add an amount limitation Congress had not enacted. It therefore rejected the government’s proposed reduction of the judgment. The court suggested that Congress could have considered even unsuccessful research useful. Its holding concerned the version of Section 174 applicable to the 1981 tax year; it did not exempt purported research payments from every other tax-law requirement or establish eligibility for the separate Section 41 credit.

The Legislative Response: Enactment of Section 174(e)

The Omnibus Budget Reconciliation Act of 1989 added former Section 174(e), limiting the provision’s application to expenditures reasonable under the circumstances. The legislative history addressed the concern illustrated by Driggs that qualifying research costs could otherwise escape an express amount limitation. This historical rule should not be described as an unchanged current statutory requirement: the Tax Cuts and Jobs Act rewrote Section 174 for amounts paid or incurred in taxable years beginning after December 31, 2021, and removed former subsection (e).

The Scope and Timing of the Reasonableness Requirement

Under former Section 174(e) and Treasury Regulation Section 1.174-2(a)(9), reasonableness generally concerned the amount ordinarily paid for comparable research activities by comparable enterprises in comparable circumstances. The historical standard did not authorize rejection merely because the research was unusual or unlikely to succeed. New Section 174A does not expressly reenact former Section 174(e). The continuing text of older regulations must be evaluated against the amended statute and the tax year at issue, rather than presented as proof that former subsection (e) still exists.

For years governed by former Section 174(e), compensation and related-party research payments require an appropriate reasonableness analysis. For later years, taxpayers should separately assess the governing statute, applicable regulations, wage definitions, actual services, and other relevant tax rules. Removal of the former statutory language does not mean that fabricated payments, nonresearch services, or unsupported allocations qualify for a research credit. Activity qualification and expenditure measurement remain separate inquiries.

Theoretical Convergence and the Four-Part Test of Section 41

The first element of the Section 41 qualified-research definition historically referred to Section 174. For amounts paid or incurred in taxable years beginning after December 31, 2024, Section 41(d)(1)(A) instead refers to expenditures treated as domestic research or experimental expenditures under Section 174A. Historical cases must be read under the version applicable to their tax years. Research-expense treatment alone is insufficient: the activities must meet the other Section 41 tests and exclusions, and the claimed costs must fit the expense categories in Section 41(b).

The Four-Part Test Framework

The identification of qualified research requires the cumulative satisfaction of the following tests:

The Research Expenditure Test: Research must satisfy the applicable research-expenditure provision: historically Section 174 and, under the amended statute, Section 174A for domestic research. The uncertainty inquiry concerns capability, method, or appropriate design in developing or improving a product, process, software, technique, formula, or invention. Ordinary business or commercial uncertainty alone is insufficient.

The Technological in Nature Test: The process of experimentation must fundamentally rely on principles of the physical or biological sciences, engineering, or computer science.

The Process of Experimentation Test: At least 80% of the relevant research activities, measured on a reasonable and consistently applied basis such as cost, must constitute elements of a process of experimentation for a permitted purpose. The process must identify uncertainty, identify one or more alternatives intended to resolve it, and evaluate those alternatives through methods such as modeling, simulation, or systematic trial and error. The denominator is relevant research activities, not every business activity or every dollar spent producing a product.

The Business Component and Permitted Purpose Test: The information sought must be intended to help develop a new or improved business component held for sale, lease, or license, or used in the taxpayer’s trade or business. Experimentation must relate to function, performance, reliability, or quality, rather than style, taste, cosmetic, or seasonal design factors. The tests apply separately to each business component, subject to the shrinking-back rule and statutory exclusions.

The Mathematics of the Research Credit

The regular research-credit method includes a credit of 20% of current-year qualified research expenses (QREs) exceeding the base amount. This is not the only calculation method, and it is not a 20% credit on all research spending. The regular QRE calculation, before Section 280C coordination and other limitations, can be expressed as follows:

Regular QRE credit = 20% × max(0, current-year QREs − base amount).

The base amount generally equals the fixed-base percentage multiplied by average annual gross receipts for the preceding four tax years, but it cannot be less than 50% of current-year QREs. The fixed-base percentage is subject to a 16% ceiling and special start-up rules; the historical 1984–1988 ratio does not apply uniformly to every taxpayer. QRE definitions must be applied consistently across the relevant years. An elective alternative simplified credit generally equals 14% of current QREs exceeding 50% of average QREs for the preceding three years; a 6% rule applies if the taxpayer had no QREs in any one of those three years. Section 41 also contains separate rules for basic research and energy research, which are outside this formula.

Contemporary Jurisprudence: The Shift to Activity-Level Substantiation

Siemer Milling Company v. Commissioner and Phoenix Design Group, Inc. v. Commissioner illustrate the need for evidence showing how particular activities satisfy Section 41. Their holdings do not categorically exclude oral testimony or studies prepared after the research occurred. The issue is whether the evidence, considered as a whole, establishes qualifying research and supports the expenses claimed.

Siemer Milling and the Failure of Conclusory Evidence

In Siemer Milling Company v. Commissioner, T.C. Memo. 2019-37, the Tax Court disallowed the research credits at issue for the flour miller’s 2011 and 2012 tax years because it failed to establish a qualifying process of experimentation. Descriptions of product development, milling improvements, and trial and error did not adequately show the methodical experimental process required for the claimed projects. The decision is a failure-of-proof case, not a ruling that milling research or retrospective testimony is inherently ineligible.

The practical lesson is to explain the uncertainty, alternatives, evaluation methods, and results for each claimed business component. General labels such as iterative development do not substitute for evidence of actual experimentation. Contemporaneous technical records can be persuasive, and knowledgeable witnesses can explain them. Neither a specialized scientific degree nor a universally prescribed laboratory-log format is a statutory prerequisite to the credit.

Little Sandy Coal and the “Substantially All” Fraction

In Little Sandy Coal Co., Inc. v. Commissioner, T.C. Memo. 2021-15, affirmed at 62 F.4th 287 (7th Cir. 2023), a shipbuilding business claimed research credits involving eleven vessels. The litigation used two vessels as representative projects. The taxpayer relied heavily on their novelty and asserted pilot-model status, but failed to provide sufficient evidence to apply the substantially-all test.

The substantially-all threshold concerns research activities, rather than the percentage of a vessel’s physical features that are new. Under Treasury Regulation Section 1.41-4(a)(6), the 80% determination must use a reasonable basis applied consistently. The Seventh Circuit also rejected a categorical exclusion of direct support, direct supervision, and pilot-model production from the numerator: such activities can count when they themselves constitute elements of a process of experimentation.

The Seventh Circuit’s affirmation clarified the structure of the 80% fraction:

Experimentation fraction = research activities constituting elements of a process of experimentation ÷ total relevant research activities under the applicable research-expenditure test, using a consistent measurement basis.

The evidence must permit a principled allocation between relevant research activities that constitute elements of experimentation and those that do not. Unsupported percentages and assumptions based solely on novelty failed in Little Sandy Coal. The Seventh Circuit expressly recognized that expense estimates may be available once the taxpayer proves qualifying research occurred. It did not impose universal employee timesheets or prohibit every reasonable estimate.

Technical Uncertainty vs. Routine Engineering: The Phoenix Design Group Precedent

In Phoenix Design Group, Inc. v. Commissioner, T.C. Memo. 2024-113, the Tax Court examined three agreed trial projects involving building engineering systems. It found that the taxpayer failed to establish the required research uncertainty and process of experimentation on the record presented. Professional engineering work, complexity, and a multistage design workflow did not by themselves prove qualified research.

The Routine Engineering Trap

The relevant distinction is between using available information to resolve routine design questions and conducting research to resolve uncertainty about capability, method, or appropriate design. Applying established engineering principles can be part of qualified research; new-to-the-world scientific knowledge is not required. However, standard calculations and project revisions do not automatically establish an experimental process. The taxpayer must demonstrate the qualifying uncertainty and the actual evaluation performed.

Phoenix Design Group also sustained accuracy-related penalties under Section 6662. A 20% penalty is not an automatic consequence of a denied research credit: the applicable statutory grounds, procedural requirements, and reasonable-cause and good-faith defense must be considered. The case does not by itself prove a new IRS enforcement policy or that missing contemporaneous records always produce penalties.

The Funded Research Exclusion: Risk and Rights

Section 41(d)(4)(H) excludes research to the extent funded by a grant, contract, or otherwise by another person. Treasury Regulation Section 1.41-4A(d), incorporated into the later research-credit framework, distinguishes payment contingent on research success from amounts payable for performing research. Retaining substantial rights is a separate requirement. If the taxpayer retains no substantial rights, the research is fully funded for this purpose. If substantial rights are retained but payments are not contingent on success, payments generally reduce otherwise eligible research expenses; qualifying excess costs may remain eligible.

The Contingency of Success Standard

In System Technologies, Inc. v. Commissioner, Docket No. 12211-21, an order dated January 3, 2025, denied the IRS’s motion for partial summary judgment concerning funded research. The court considered Indiana law as well as the contract language. Warranty provisions limited repair and replacement remedies but did not eliminate a refund remedy for total failure to deliver. The court concluded that payments we