Documenting qualified research activities for the research tax credit

Editor: Lori Anne Johnston, CPA, J.D.

The credit for increasing research activities under Sec. 41, more commonly known as the research tax credit, has long been a widely used federal tax incentive for many businesses. The Joint Committee on Taxation estimates a total of $11.7 billion in federal tax expenditures for the research tax credit for fiscal year 2019, projected to increase to $16.4 billion in fiscal year 2023 (Estimates of Federal Tax Expenditures for Fiscal Years 2019-2023 (JCX-55-19) (12/18/19)).

Many taxpayers are now familiar with the Sec. 41(d) four-part test and understand that a business component must meet all four parts:

• A permitted purpose, which includes a new or improved function, performance, reliability, or quality, related to a product or process;
• An intent to discover information that is technological in nature by relying on the principles of a hard science;
• Technical uncertainty at the project’s outset regarding capability, method, or appropriate design; and
• A process of experimentation to attempt to resolve technical uncertainties (Sec. 41(d)).

As part of the process-of-experimentation requirement, substantially all (i.e., 80% or more) of the taxpayer’s research activities for a particular business component must constitute a process of experimentation (id.). The regulations provide that a process of experimentation is designed to evaluate one or more alternatives to achieve a result where the capability or the method of achieving that result is uncertain as of the beginning of the taxpayer’s research activities (Regs. Sec. 1.41-4(a)(5)). The term “business component” is defined as a product, process, software, technique, formula, or invention that is to be held for sale, lease, or license or used in the taxpayer’s trade or business (Sec. 41(d)(2)(B)).

The process-of-experimentation requirement has received increased focus by taxing authorities in examinations. The U.S. Tax Court’s decision in Siemer Milling Co., T.C. Memo. 2019-37, closely examines the taxpayer’s process of experimentation for each of the claimed projects. Several other cases are pending on the Tax Court docket that deal with similar issues. Additionally, the California Office of Tax Appeals (OTA) recently issued an opinion in the case In the Matter of the Appeals of Walden, No. 2018-OTA-209 (11/28/18), where the OTA also seemed to home in on the process-of-experimentation requirement of the four-part test.

Siemer Milling Co.

On April 15, 2019, the Tax Court, in Siemer, held that the taxpayer had not properly supported its R&D credit under Sec. 41 and disallowed the taxpayer’s R&D credit claim for the tax years ended May 31, 2011, and May 31, 2012. After applying the four-part test to each project, the Tax Court concluded that the taxpayer had failed to provide documentation to support its claim that substantially all of the research activities constituted a process of experimentation under Sec. 41. The court repeatedly referred to the Tax Court case Union Carbide, T.C. Memo. 2009-50, in making its decision.

In Siemer, thecourt found that the taxpayer did not establish it had 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 Tax Court also disallowed the credit for several projects based on the technological-in-nature test. Even though the taxpayer argued that it relied on principles of engineering and the physical and biological sciences, the court found that Siemer failed to establish the specific engineering principles on which the research activities relied.

Regarding the process of experimentation, the IRS argued the following position in its trial briefs:

The record is devoid of evidence that petitioner formulated or tested hypotheses, or engaged in modeling, simulation, or systematic trial and error during the credit years. Nor is there any evidence that petitioner evaluated alternatives during the credit years. … This failure attests to the conclusion that petitioner’s purported research during the credit years involved nothing more than simple trial and error.

The Tax Court’s holding suggests that the court is seeking to draw a line between “systematic” trial and error and “simple” trial and error regarding experimentation.

Additionally, Siemer shows that a research project must specify how the individual research project relied on the principle of science. For example, the taxpayer argued that the experimentation for its Pulsewave business component “relied on principles of engineering and the physical and biological sciences.” The court disagreed because it did not find that the taxpayer established the principles on which its research activities relied. It appears that the court believed that the taxpayer’s changes to the machine involved in the business component were slight adjustments, such as replacing bearings, and were not part of a process of experimentation to improve either the machine or the production method in which the taxpayer used the machine.

In the Matter of the Appeals of Walden

The OTA heard the appeal of the taxpayer in Walden regarding the taxpayer’s research credit claim related to the design and construction of modular buildings. The appeal arose from the California Franchise Tax Board’s (FTB’s) denial of the taxpayer’s refund claim for several years, stretching as far back as 2003. The FTB denied the refund claims for all years.

In its nonprecedential opinion, the OTA provided a relatively taxpayer-friendly determination regarding the elimination of uncertainty that was in line with Treasury regulations for Secs. 41 and 174. The OTA acknowledged that every research project begins with some existing knowledge, and the uncertainty requirement of Secs. 41 and 174 does not require that a taxpayer start “a project from a blank slate without any knowledge of engineering or design principles” and that the uncertainty requirement need not require that the taxpayer had “been seeking a scientific breakthrough.” The uncertainty requirement merely requires that the information available to the taxpayer does not “‘establish’ the method for developing the project or the appropriate design of the project.” The OTA determined that the taxpayer met the uncertainty requirement of Sec. 41.

The OTA, however, found that the majority of the examined business components did not meet the process-of-experimentation requirement of Sec. 41. The FTB relied on Trinity Industries, Inc., 691 F. Supp. 2d 688 (N.D. Tex. 2010), to argue that the taxpayer did not engage in an evaluative process involving the systematic testing of alternatives that resulted in 80% of the costs of each business component constituting a process of experimentation.

The taxpayer was unable to provide sufficient evidence supporting the design changes to five of the six examined business components. An example that the taxpayer offered to support its contention that substantially all of its activities constituted a process of experimentation was that the taxpayer made design changes because the customer wanted to hang something heavy on a wall, and the taxpayer had to order solid blocking because of wall stud and spacing issues. The OTA determined that rather than showing a process of experimentation that was the result of a “systematic process of considering and testing hypotheses that might be considered a process of experimentation,” the example showed that some of the claimed research activities were ordinary construction work. The fact that the taxpayer’s activities included ordinary construction work meant that some of its activities were not a process of experimentation, and the taxpayer had failed to prove that at least 80% of its research activities were a process of experimentation.

Lessons learned

These two cases, especially Siemer, show that documenting the research activities of a project is critically important to supporting the existence of qualified research activities. This has always been a well-understood concept for research credit claims. However, often the documentation was focused on demonstrating that a business component did have uncertainty concerning the development or improvement to the business component while describing process-of-experimentation activities in a manner that was not specifically descriptive of the scientific nature of the activities. Both Siemer and Walden show that taxpayers and tax practitioners must diligently document the process of experimentation and show how principles of science were applied during the experimentation.

Indeed, Siemer also shows that the documentation must very specifically show the process of experimentation undertaken by a taxpayer. Merely making general changes and adjustments, i.e., an unstructured iterative design process, does not appear to be sufficient to support the process-of-experimentation claim. Rather, the taxpayer must methodically undertake a process of experimentation.

Often, a taxpayer is more concerned with accomplishing the research projects it is undertaking than documenting a formal hypothesis and the testing to prove or disprove the idea. To incorporate the Siemer guidance while continuing to focus on its R&D processes, a taxpayer would be wise to develop a standard testing methodology specific to its technological field, which it applies for all projects. The standardized testing methodology will help show that the taxpayer executed 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, as the Siemer court mentioned. A standard testing methodology will also help ensure that the taxpayer does not inadvertently skip particular steps or tests it intended to conduct.

Siemer also demonstrates that a taxpayer’s research credit claim must fully document not just the outcome of the test, but also what precisely the test was intended to prove. It is unlikely that most taxpayers are formally writing down the tested hypothesis and whether the results supported the hypothesis. Nonetheless, a taxpayer should have documentation to support why a test occurred and what the test proved. Furthermore, in keeping with Walden, a taxpayer should be able to show that substantially all of the costs associated with a project were part of the process of experimentation or use the Regs. Sec. 1.41-4(b)(2) shrinkback rule to identify how the development of one or more subcomponents meets the four-part test.

Finally, when a taxpayer is going through the process of calculating and claiming the research credit, merely stating, without evidence or explanation, that the testing relied on a principle of science is vague and unhelpful for its research credit claim. The taxpayer is better off stating what scientific principle the process of experimentation applied. That will show how the particular research project met the technological-in-nature test, and it will further support the process of experimentation.

The R&D tax credit remains a widely used tax incentive. As taxing authorities continue to conduct examinations to ensure proper compliance with the credit’s various requirements, taxpayers should review the contemporaneous documentation they create in the conduct of their research activities to verify that they continue to support their research credit claims.

EditorNotes

Lori Anne Johnston, CPA, J.D., is a manager, Washington National Tax for RSM US LLP.

For additional information about these items, contact John Deininger (John.Deininger@rsmus.com); Jared Boucher (Jared.Boucher@rsmus.com); and Tom Windram (Tom.Windram@rsmus.com).

Unless otherwise noted, contributors are members of or associated with RSM US LLP.