Siemer Milling Co. v. Commissioner

The federal research tax credit, governed by Section 41 of the Internal Revenue Code, has long stood as a cornerstone of American industrial policy, designed to incentivize the private sector to undertake the technical risks inherent in innovation. However, the application and defense of this credit have entered an era of unprecedented scrutiny following the United States Tax Court decision in Siemer Milling Co. v. Commissioner, T.C. Memo. 2019-37. This case, alongside the subsequent appellate affirmation of Little Sandy Coal Co. v. Commissioner and the recent Phoenix Design Group v. Commissioner, signals a definitive shift in the judicial and administrative expectations for documentation, scientific rigor, and the "process of experimentation." For professional tax practitioners, corporate controllers, and R&D directors, the Siemer Milling decision is not merely a localized defeat for a wheat milling company; it is a foundational text that redefines the "four-part test" and mandates a transition from retrospective, interview-based credit studies toward contemporaneous, activity-level data capture.

The Siemer Milling Co. v. Commissioner Contextual Background

Siemer Milling Company, a family-owned enterprise with a history in the wheat milling industry dating back to the 1950s, operated multiple mills in the United States, focusing on the production and sale of wheat flour. For the tax years ending May 31, 2011, and May 31, 2012, the company claimed research tax credits of  and, respectively. These claims were supported by a credit study prepared by their long-standing accounting firm, CLA, which had been engaged since 2004 to conduct these studies and file amended returns.

The controversy began when the Commissioner of Internal Revenue disallowed the credits in their entirety, asserting that Siemer had failed to prove that its activities constituted "qualified research" under IRC Section 41. The dispute centered on seven specific projects that the company argued were aimed at improving products and processes. The Tax Court's subsequent analysis provided a granular look at how a failure to document a systematic, scientific approach can lead to total disallowance, even when the underlying work is technically complex and potentially innovative.

The Role of Accounting Advice and Penalty Mitigation

A significant aspect of the Siemer Milling case was the taxpayer's reliance on professional advice. While the court upheld the disallowance of the credits, it notably held that Siemer was not liable for accuracy-related penalties under IRC Section 6662. The court determined that the company had acted in good faith by relying on a competent adviser, providing that adviser with all necessary and accurate information, and demonstrating actual reliance on the firm’s expertise. This finding underscores the importance of the "reasonable cause" defense in R&D credit controversies. The court recognized that while the documentation was legally insufficient for the credit, the company’s effort to comply with the law through professional consultation shielded it from punitive assessments.

Exhaustive Project-Level Analysis and Technical Failures

The Tax Court’s decision was rooted in a project-by-project examination, where it found that none of the seven identified projects satisfied all elements of the four-part test. The projects spanned from product development to mechanical modifications of milling equipment, yet they all suffered from a common deficiency: a lack of documented, methodical experimentation in the scientific sense.

The Pulsewave Project: Routine Maintenance vs. Scientific Research

The Pulsewave project involved a machine operating on the principle of "resonance disintegration," which uses the physics of resonance, shock waves, and vortex-generated shearing forces to reduce particle size, rather than traditional crushing methods. Siemer's objective was to determine if it could increase the machine's operating speed from 3,600 rotations per minute (RPM) to 5,000 RPM.

The court's rejection of this project as qualified research was multi-faceted. First, it found that Siemer failed to establish technical uncertainty. Evidence showed that Pulsewave LLC, the manufacturer, had already provided data indicating the machine could run at speeds up to 5,000 RPM. The court characterized Siemer's activities—which included addressing mechanical problems that arose at higher speeds—as "more akin to mechanical maintenance" rather than research intended to discover technological information. Furthermore, the court found that the company did not establish reliance on principles of engineering or physical and biological sciences in a way that met the technological information test.

The Ozone Project: Evaluation vs. Experimentation

In the Ozone project, Siemer attempted to introduce ozone into various stages of the milling process—including cleaning, tempering, and crushing—to produce a low-microorganism flour for products that might not be baked. While the company was uncertain how ozone would affect the flour’s composition, flavor, and bacterial content, the court found that Siemer failed the process of experimentation test.

The court noted that although Siemer described steps designed to evaluate alternatives for inserting ozone, it did "not expand on what theory that it may have been testing or how it refined its process based on data collected". This distinction is vital; merely "running tests" or "evaluating alternatives" is insufficient if those activities do not constitute a systematic trial-and-error methodology where a hypothesis is formed, tested, analyzed, and refined.

The Flour Heat-Treatment and Littleford Day Projects

The Flour Heat-Treatment project sought to develop processes for producing cake flour without chlorine and low-microorganism flour without chemicals. This involved heating flour for various durations using different methods and testing for composition and bacterial levels. Similarly, the Littleford Day project involved working on a system to toast wheat and bran, experimenting with different times and temperatures.

Both projects failed the process of experimentation test because the court found the testimony provided by the Vice President of Production focused on the uncertainties faced rather than the process used to eliminate them. The documentation provided—which included undated recipes, lab results, and sample test orders—was found to be insufficient to demonstrate a methodical plan or scientific rigor.

The Wheat Hybrids Project: Development vs. Testing

Perhaps the most clear-cut failure involved the Wheat Hybrids project, where Siemer tested new varieties of wheat from breeders to see if they met customer specifications. The court ruled that this failed the business component test because Siemer was not developing a new process or product; it was simply "determining what was available from wheat breeders and growers". The court likened this to a taxpayer evaluating software already on the market, which is a non-qualifying activity.

The Process of Experimentation: Refining the "Scientific Sense"

A central takeaway from Siemer Milling is the court's strict adherence to the definition of a process of experimentation as outlined in Union Carbide Corp. & Subs v. Commissioner. This requires a "methodical plan involving a series of trials to test a hypothesis, analyze the data, refine the hypothesis, and retest the hypothesis so that it constitutes experimentation in the scientific sense".

The differentiation between "routine trial and error" and a "qualified process of experimentation" has become the primary battleground in IRS audits. While many companies naturally follow an iterative process—particularly in food science, engineering, and software—the Siemer ruling emphasizes that this process must be documented to reflect scientific methodology rather than just an eventual solution to an issue.

Substantiation Standards and the Shrinking-Back Rule

The Siemer Milling decision serves as a "barometer" for the level of detail now required for R&D credit claims. The court’s rejection of undated, unauthored, and vague documents suggests that the "traditional models" of collecting R&D data—relying on high-level employee interviews and retrospective summaries—are no longer adequate.

Contemporaneous Documentation Requirements

Taxpayers must now ensure that their records are "in sufficiently usable form and detail to substantiate that the expenditures claimed are eligible for the credit". This includes:

• Project lists with clearly defined technical objectives and unknowns.
• Design iterations, test results, and engineering notes captured in real-time.
• Employee activity logs or time-tracking systems that link labor specifically to research activities.
• Technical correspondence (emails) that discuss specific challenges and hypothesis testing.

Application of the Shrinking-Back Rule

A critical concept reaffirmed in recent litigation, including Phoenix Design Group v. Commissioner (2024), is the "shrinking-back rule" under Treas. Reg. § 1.41-4(b)(2). This rule dictates that if an entire business component (the overall product) fails to meet the four-part test, the test should be applied to the most significant subcomponent. The "shrinking" continues until a subset is found that satisfies the test or until the most basic element is reached and fails.

In Phoenix Design Group, the court acknowledged the shrinking-back rule but could not apply it because the taxpayer’s documentation was inadequate even at the subcomponent level. This highlights a vital strategy for future claims: taxpayers should document their research at the modular or subcomponent level to ensure that even if a large-scale project is deemed too "routine" as a whole, specific technical breakthroughs within that project can still qualify.

Comparative Jurisprudence: Little Sandy Coal and Phoenix Design Group

The judicial trend toward stricter substantiation is further evidenced by two other landmark cases that have shaped the post-Siemer landscape.

Little Sandy Coal Co. v. Commissioner (2021, Affirmed 2023)

In Little Sandy Coal, a shipbuilder claimed credits for first-in-class vessels. The case is significant for the "novelty argument" rejection. The taxpayer argued that because the entire ship was a "novel" prototype, all activities associated with it should be considered part of a process of experimentation. The court disagreed, stating that the "substantially all" test applies to activities, not the physical elements of the product.

The Seventh Circuit’s affirmation of this case highlighted that "shortcut estimates" and departmental allocations are insufficient; the taxpayer must provide a principled breakdown of employee time by experimentation activity. However, the appellate court did provide a more favorable view for taxpayers by stating that "direct support" and "direct supervision" can be included in the numerator of the PoE fraction if those individuals were actually engaged in tasks that supported experimentation, diverging from the Tax Court’s narrower "silver bullet" exclusion.

Phoenix Design Group v. Commissioner (2024)

This case involved a mechanical, electrical, and plumbing (MEP) engineering firm. The court disallowed credits across three sample projects (hospitals and university buildings), finding that the firm's design process was "linear" and focused on code compliance rather than "iterative". The court emphasized that performing "routine calculations on available data" does not constitute experimentation because the information was already effectively known. This case is a stark reminder that even highly complex engineering work may not qualify if it does not involve the resolution of true technological uncertainty through a systematic evaluative process.

Administrative Shifts: The Overhaul of Form 6765

In response to these judicial outcomes, the IRS has implemented significant changes to Form 6765, Credit for Increasing Research Activities, effectively "front-loading" the audit process by requiring detailed qualitative information at the time of filing.

New Disclosure Requirements

• Section E – Other Information (Effective 2024): Requires disclosure of the number of business components generating QREs and the total amount of officer wages included in the claim. The emphasis on officer wages stems from cases like Moore v. Commissioner, where a lack of documentation for a CEO's time led to disallowance.
• Section G – Business Component Information (Mandatory 2026): Taxpayers will be required to report information on a business-component basis for their top 50 projects (or those accounting for 80% of QREs). This includes specifying the type of business component and the information sought for discovery.

The goal of these changes is to enhance transparency and allow the IRS to analyze potential deficiencies in a claim—such as a failure to satisfy the "substantially all" test—before an examination even begins.

Strategic Legislative Context: Section 174 and the OBBBA

The legislative environment surrounding R&D has also seen dramatic shifts. The Tax Cuts and Jobs Act (TCJA) of 2017 introduced a requirement to capitalize and amortize research expenditures over five years (domestic) or fifteen years (foreign) under Section 174, beginning in 2022.

However, the "One Big Beautiful Bill Act" (OBBBA), signed on July 4, 2025, has largely reversed this for domestic research. The OBBBA reinstated immediate expensing for domestic research expenditures under a new Section 174A. Crucially, the law provides "transition rules" allowing taxpayers to deduct remaining unamortized costs from 2022-2024 and even allows "eligible small businesses" to apply Section 174A retroactively by amending prior returns.

These changes optimize the relationship between R&D deductions and credits, but they also increase the stakes for correct classification. Because Section 280C requires that R&D deductions be reduced by the amount of the credit taken (unless a reduced credit is elected), the precision of the underlying R&D study is paramount to maximize the overall tax benefit.

Implications for Future R&D Tax Credit Applications

The Siemer Milling case and the ensuing judicial and administrative trends create a new "playbook" for taxpayers seeking to claim the research credit. The traditional reliance on "conclusory statements" and "after-the-fact interviews" is a high-risk strategy that is likely to fail in the current environment.

Redefining Technical Uncertainty

Future applications must move beyond "general design challenges" and clearly define the "scientific or technological questions" that the research seeks to answer. In Siemer, the failure to show that information was not already available (as in the Pulsewave speed issue) was fatal. Documentation must now specifically "map" technical unknowns to the investigative activities undertaken to resolve them.

The Activity-Based Accounting Requirement

The Little Sandy Coal ruling emphasizes that R&D must be tracked by activity, not by the physical object being created. This necessitates a project-level cost accounting system that can break down labor into categories of "direct performance," "direct supervision," and "direct support". Taxpayers should avoid "shortcut estimates" and instead implement time-tracking systems that capture these distinctions contemporaneously.

Proactive Litigation and Audit Preparation

The IRS is increasingly using automated review systems (like the "Classifier" system mentioned in Meyer, Borgman & Johnson) to deny claims that lack specificity at the time of filing. Consequently, a "procedurally weak" claim at filing—one that does not identify business components or expenses with the required detail—may be rejected before it even reaches an examiner.

Conclusion: A New Paradigm for American R&D Incentives

The decision in Siemer Milling Co. v. Commissioner represents a permanent shift in the topography of the federal research tax credit. It marks the conclusion of the era where the mere existence of complex technical activity was sufficient to justify the credit. Instead, the "Siemer Standard" demands that taxpayers function as scientists as much as businessmen, documenting their innovations through a structured, iterative, and hypothesis-driven process.

The implications for future applications are clear: precision is mandatory. As the IRS moves toward requiring business-component-level disclosures on the face of the tax return, and as the judiciary continues to refine the definition of "experimentation" in a way that excludes linear design and routine adaptation, the burden of proof has never been higher. However, for those companies that invest in contemporaneous documentation and rigorous project management, the R&D credit remains a vital and defensible incentive, further strengthened by the legislative restoration of immediate expensing under the OBBBA. The "win" for the IRS in Siemer Milling is a "warning" for all other taxpayers—one that mandates a transformation in how American industry captures, documents, and claims its investment in the future.