Shami v. Commissioner

The federal research and development tax credit, established under Section 41 of the Internal Revenue Code, represents a cornerstone of the American fiscal strategy to foster technological innovation and maintain global economic competitiveness. However, the practical application of this credit is frequently characterized by high-stakes litigation, particularly concerning the substantiation of qualified research expenses (QREs) and the eligibility of executive compensation within research-intensive organizations. The landmark case of Shami v. Commissioner (T.C. Memo 2012-78 and 2012-103), subsequently affirmed by the United States Court of Appeals for the Fifth Circuit (741 F.3d 560), provides a definitive framework for understanding the evidentiary burdens placed upon taxpayers. This analysis explores the technical nuances of the Shami litigation, the statutory intersection of Sections 41 and 174, the judicial limitations of the Cohan rule, and the profound implications for the current landscape of IRS enforcement and the 2025 regulatory updates following the One Big Beautiful Bill Act (OBBBA).

The Statutory Architecture of the Section 41 Credit

To comprehend the failure of the taxpayers in Shami, one must first examine the multi-layered statutory requirements governing the research credit. The credit is calculated as a percentage of the amount by which a taxpayer's QREs for the taxable year exceed a defined base amount. The code defines QREs as the sum of in-house research expenses and contract research expenses incurred in carrying on a trade or business. In-house research expenses encompass wages paid to employees for "qualified services," supplies used in the conduct of qualified research, and amounts paid for the use of computers.

For any activity to be deemed "qualified research," it must satisfy a rigorous four-part test. First, the expenditure must be of a type that may be treated as an expense under Section 174. This requires the activity to be "research and development in the experimental or laboratory sense," specifically intended to eliminate uncertainty concerning the development or improvement of a business component. Second, the research must be undertaken for the purpose of discovering information which is technological in nature, meaning it relies on principles of physical or biological sciences, engineering, or computer science. Third, the application of that information must be intended to be useful in the development of a new or improved business component. Finally, substantially all of the activities must constitute elements of a process of experimentation, which involves the evaluation of alternative designs, the testing of hypotheses, and the systematic refinement of a product or process.

A critical mechanism within the wage component of the credit is the "80-percent rule" or the flush language of Section 41(b)(2)(B). If at least 80% of an employee's services performed during a taxable year constitute "qualified services"—defined as engaging in qualified research, direct supervision, or direct support of such research—then 100% of the wages paid to that employee may be included in the QRE calculation. This rule was the primary lever used by Farouk Systems, Inc. (FSI) to claim millions in tax credits, and its misapplication became the focal point of the judicial rebuke in Shami.

Factual Context and the Operation of Farouk Systems, Inc.

Farouk Systems, Inc. is a prominent manufacturer and distributor of professional hair care and cosmetic products, widely recognized for technological advancements in hair styling tools and chemical formulations, such as the CHI and BioSilk brands. As a Subchapter S corporation, FSI’s income, deductions, and tax credits flowed through to its shareholders, which included the petitioners in the Shami case. During the years in question (2003, 2004, and 2005), the company employed several hundred people, including an R&D staff ranging from 18 to 27 employees.

The company engaged a third-party consultant, alliantgroup, to conduct a research and development tax credit study to identify eligible expenses. The results of these studies led FSI to claim massive QREs, the vast majority of which were derived from the salaries of the company’s highest-level executives: Farouk Shami, the CEO and President, and John McCall, the Co-Chairman.

In the aggregate, the wages of these two individuals accounted for over 80% of the total wage QREs claimed by FSI across the three-year period. The IRS challenged the entirety of the credits, asserting that the company had failed to substantiate that the executives were performing qualified services and that the magnitude of the wages was unreasonable under Section 174(e).

The Section 174(e) Reasonableness Requirement and the Driggs Precedent

A pivotal element of the IRS's argument in Shami was the "reasonableness" limitation. While Section 162 generally allows for the deduction of reasonable compensation as a business expense, Section 174(e) specifically applies a reasonableness standard to research or experimental expenditures. This provision was added to the Internal Revenue Code by the Omnibus Budget Reconciliation Act of 1989 as a direct legislative response to the district court decision in Driggs v. United States (706 F. Supp. 20).

In Driggs, the court concluded that the then-existing version of Section 174 did not impose a reasonableness requirement on research deductions. If an expense met the definition of R&D, it was deductible regardless of the amount. Congress, concerned that this would allow taxpayers to hide unreasonable compensation or other unrelated costs within research claims, amended the statute to clarify that Section 174 only applies "to the extent that the amount thereof is reasonable under the circumstances".

In Shami, the IRS determined that classifying $8 million to $9.5 million in annual salary as a "research expense" was inherently unreasonable given the lack of evidence of the executives' day-to-day involvement in technical experimentation. The Tax Court agreed, noting that while the executives were clearly successful business leaders, their remuneration was vastly disproportionate to the value of any actual research services they may have performed. This established a critical precedent: the IRS can use Section 174(e) to bifurcate a salary into a "reasonable research component" and a "non-research business component," effectively disqualifying the latter from the credit calculation.

Evidentiary Failures and the Limitation of Judicial Discretion

The Shami litigation is perhaps best known for its focus on the quality of documentation required to satisfy the burden of proof. In the Tax Court, the Commissioner’s determination is presumptively correct, and the taxpayer bears the burden of proving it erroneous. FSI attempted to meet this burden by introducing over 4,500 pages of laboratory records. However, the Tax Court, exercising its discretion under Federal Rule of Evidence 403, limited the petitioners to a sample of dozens of representative records.

The court reasoned that the volume of the records was irrelevant if they did not establish a nexus between the activities and the specific individuals at issue. Most of the laboratory logs showed chemical tests performed by staff chemists or technicians; Farouk Shami’s name appeared only occasionally as having "approved" the documents, and John McCall’s name was largely absent. The Fifth Circuit affirmed this exclusion, stating that admitting the full 4,500 pages would have resulted in "needless delay, wasted time, and unnecessary cumulative evidence" while providing almost no probative value regarding the executives' time allocations.

The Credibility Gap and the Failure of Testimony

In the absence of contemporaneous time-tracking records, FSI relied heavily on the testimony of Farouk Shami and John McCall. Both testified that they spent 80% to 100% of their time on research and development activities. The Tax Court, however, found this testimony to be "contradictory, self-serving, and noncredible". The court noted that neither Shami nor McCall possessed formal education or training in chemistry or engineering, which cast doubt on their ability to perform the technical, hands-on experimentation they claimed to lead.

The Tax Court’s assessment of credibility is given high deference by appellate courts. Because the court did not believe the witnesses, and because the laboratory records did not corroborate their active participation, the taxpayers were found to have failed the "substantiality" requirement of the four-part test.

The Cohan Rule and the Threshold of Estimation

One of the most significant legal arguments raised by the petitioners was the application of the Cohan rule. Derived from Cohan v. Commissioner (39 F.2d 540), the rule provides that if a taxpayer can prove they incurred a deductible expense but cannot substantiate the exact amount, the court should make a "reasonable estimate" rather than disallowing the deduction entirely.

FSI argued that since research was clearly happening at the company, the court was legally required to estimate a portion of the executives' time that qualified for the credit. They pointed to United States v. McFerrin (570 F.3d 672), where the Fifth Circuit had remanded a case for the district court to consider estimates under the Cohan rule despite incomplete records.

However, the Fifth Circuit in Shami distinguished the two cases. The court clarified that the Cohan rule is only implicated once the taxpayer establishes they are entitled to some benefit—that is, they must first prove that some amount of qualified research services were actually performed by the specific employees in question. Because the Tax Court found the testimony regarding Shami and McCall to be entirely non-credible and the documentation insufficient to show they performed any qualified research, there was no "reasonable basis" upon which an estimate could be founded. The court noted that providing relief under such circumstances would be "unguided largesse".

Direct Supervision vs. General Administrative Oversight

The Shami decision provided essential clarity on the definition of "direct supervision" under Section 41(b)(2)(B). Qualified services include the direct supervision of qualified research activities, but this does not encompass general high-level management or administrative oversight of a department.

Evidence at trial indicated that FSI had a formal hierarchy: Farouk Shami supervised an R&D Manager (Ali Ghannad), who in turn supervised the laboratory chemists and assistants. The court found that this "second-tier supervision" (supervising the supervisor) does not qualify as "direct supervision" for purposes of the R&D credit. To qualify as direct supervision, the manager must be involved in the technical details of the research—reviewing lab notebooks, providing hands-on guidance for experimental design, and directly managing the personnel conducting the tests. Shami’s role was characterized as that of an upper-level manager involved in strategic decisions and business operations, which are not qualified research activities.

Judicial Precedents and the Evolving Definition of "Process of Experimentation"

The failure of FSI in Shami must be viewed alongside subsequent cases that have further tightened the "substantially all" requirement. In Little Sandy Coal Co., Inc. v. Commissioner (T.C. Memo 2021-15), the court addressed a claim involving the development of a tank barge and dry dock. Like FSI, the taxpayer in Little Sandy Coal failed to provide enough evidence to show that 80% or more of the activities for each business component involved a "process of experimentation". The court rejected the argument that "general engineering" automatically equates to a process of experimentation, emphasizing that the taxpayer must document the systematic evaluation of alternatives to resolve technical uncertainty.

Similarly, in Phoenix Design Group, Inc. v. Commissioner (2024), the Tax Court ruled that routine professional services, even those involving complex engineering calculations, do not qualify if the technological uncertainty is not clearly defined at the outset. These cases collectively affirm the Shami principle: the IRS and the courts will no longer accept broad, retrospective estimates of time or general descriptions of research. They require specific, contemporaneous data linking costs to the scientific process.

Administrative Reactions: CCA 20214101F and Form 6765 Revisions

The lessons learned from the IRS's victory in Shami have been formalized in recent administrative guidance. On October 15, 2021, the IRS Office of Chief Counsel released CCA 20214101F, which significantly altered the requirements for filing a valid research credit claim for refund. The memorandum established that for a refund claim to be considered "valid" (and thus to confer jurisdiction on a court for later review), the taxpayer must provide specific, foundational information at the time of the claim.

Taxpayers who fail to provide this information are given a 45-day transition period to "perfect" their claim, after which the IRS may summarily reject the claim as procedurally deficient. This procedural hurdle prevents taxpayers from filing "placeholder" claims and then attempting to find substantiation through the litigation process.

Revisions to Form 6765 (2024-2025)

The IRS has further operationalized these requirements through the 2024 and 2025 revisions to Form 6765, Credit for Increasing Research Activities. The new draft form includes Section G, which requires taxpayers to report their QREs at the business component level. While the final version of the form was slightly streamlined to allow taxpayers to report 80% of their total QREs (capped at 50 business components), the reporting burden remains significant. This shift from "cost-center" accounting to "project-based" accounting is a direct result of the judicial insistence on nexus seen in Shami.

Legislative Interplay: TCJA, OBBBA, and Section 174 Amortization

The landscape of R&D tax benefits has been further complicated by the Tax Cuts and Jobs Act of 2017 (TCJA) and the subsequent One Big Beautiful Bill Act (OBBBA) of 2025. Under the TCJA, beginning in 2022, taxpayers were no longer permitted to immediately deduct Section 174 expenditures; instead, they were required to capitalize and amortize them over 5 years (for domestic research) or 15 years (for foreign research).

This change created a massive tax burden for research-intensive companies, but it also inadvertently changed the "reasonableness" landscape. Since Section 41 credits are only available for expenses that would be deductible under Section 174, the amortization requirement meant that the credit was being calculated on expenses that were no longer currently deductible.

The One Big Beautiful Bill Act (OBBBA), enacted in July 2025, permanently restored the immediate expensing of domestic research and experimentation expenditures, effectively reversing the TCJA’s amortization requirement for domestic R&D. However, the OBBBA also reintroduced the requirement to "tax-effect" all research credits—meaning taxpayers must reduce their Section 174 deduction by the amount of the Section 41 credit claimed, or elect a reduced credit under Section 280C.

For future R&D applications, the restoration of immediate expensing makes the Section 174(e) reasonableness requirement even more critical. Because the tax benefit is now realized in the current year, the IRS is expected to increase its focus on high-wage QREs to prevent companies from over-claiming credits to offset current taxable income.

The Strategic Importance of Contemporaneous Documentation

The Shami case stands as the definitive warning against "look-back" studies and retrospective allocations. In a look-back study, a consultant interviews employees several years after the research occurred and asks them to estimate their time. The courts have increasingly found these estimates to be unreliable.

To survive an audit in the current environment, organizations must implement real-time tracking systems. These systems should not only track hours but also link those hours to specific technical business components. The use of "labor codes" or "activity codes" within an ERP system (like Oracle or SAP) is now considered the gold standard for substantiation.

Furthermore, the Shami decision highlights the evidentiary weight of patents. While FSI argued that Farouk Shami’s status as a named inventor on several patents proved he was conducting qualified research, the court ruled that a patent is not conclusive evidence of Section 41 compliance. A patent proves that information was discovered, but it does not prove how many hours an executive spent on that discovery, nor does it prove that the expenditures were reasonable under Section 174(e).

Binding Concessions and Procedural Integrity

Despite the overall loss on wages, the petitioners in Shami achieved a tactical victory regarding supply costs. During the trial, the Commissioner’s attorney had made several statements and stipulations indicating that the IRS was not challenging the company’s supply-cost QREs, focusing exclusively on the wage claims for the four highest-paid employees.

The Tax Court, however, had initially adopted the Commissioner’s original calculation of the deficiency, which did not account for these concessions. The Fifth Circuit vacated this portion of the judgment, ruling that stipulations and concessions made during trial are binding on the government. This served as a reminder of the "Primary Purpose" test and the importance of procedural rigor: once the government concedes an issue to narrow the scope of a trial, it cannot later revert to its original position to increase the tax deficiency.

Implications for High-Level Employee Participation

The Shami decision fundamentally altered how "C-Suite" involvement is viewed in R&D audits. The IRS now often "categorically excludes" managers above the first line of supervision unless the taxpayer provides overwhelming evidence of their technical contribution.

For technically-degreed executives (e.g., a CTO with a PhD in Physics), the burden is slightly lower, but for non-technical executives like Farouk Shami and John McCall, the burden is nearly insurmountable without time sheets. Organizations must carefully evaluate whether to include executive wages at all; a high-profile, poorly substantiated executive claim can often trigger a more intensive audit of the entire R&D department.

Future Outlook: The Role of AI and Classifier Systems in R&D Audits

As the IRS manages an increasing volume of research credit claims, it has begun deploying the "Classifier" review system to automate the initial screening of R&D refund claims. This system uses algorithms to identify claims that lack the "Five Essential Pieces of Information" required by CCA 20214101F. Claims that fall below a certain threshold of information density are flagged for summary denial, a trend observed in Meyer, Borgman & Johnson, Inc. v. Commissioner (2024).

For future applications, this means that the "narrative" portion of the claim is more important than ever. Taxpayers must provide a clear, technological description of their experimentation process that the Classifier system can recognize as a valid scientific methodology.

Conclusion

The legacy of Shami v. Commissioner is one of increased rigor and the end of the "informal" R&D tax credit claim. By affirming the Tax Court’s refusal to apply the Cohan rule in the face of non-credible testimony, the Fifth Circuit sent a clear signal that the research credit is a data-driven benefit. The case defined the boundaries of "direct supervision," clarified the power of the Section 174(e) reasonableness standard, and established the binding nature of IRS trial concessions.

In the wake of the 2025 One Big Beautiful Bill Act, and with the implementation of the new project-based Form 6765, the principles established in Shami have been fully integrated into the administrative fabric of the United States tax system. For future R&D credit applications, the lesson is clear: innovation must be matched by documentation. Success in claiming the credit now requires a granular, contemporaneous record of who performed the research, what technical uncertainty they sought to resolve, and a reasonable nexus between their compensation and the value of their scientific contribution. Organizations that continue to rely on retrospective estimates and executive prestige do so at significant peril in an era of automated screening and heightened judicial skepticism.