April 5, 2021 – In a recent article, my colleague Peter Mehta discussed the Tax Court’s decision in Little Sandy Coal v. Commissioner and questioned the Court’s strict application of the process of experimentation substantially all test under section 41(d)(1)(c), excluding direct supervision and support from the numerator of the fraction to determine the 80% or more requirement, as well as imposing a more difficult substantiation requirement. In this article, I take a deeper dive into the applicable code and regulations to conclude the standard set forth by the Court in Little Sandy Coal contradicts the statute, regulations, and prior case law.
Under Treas. Regs. section 1.41-4(a)(6), 80% or more of the taxpayer’s research activities with respect to a business component must constitute elements of a process of experimentation for a purpose described in section 41(d)(3). The remaining 20% of the activities need not constitute a process of experimentation if they can be treated as expenses under section 174 and are not an excluded activity under section 41(d)(4). This test is applied separately to each business component. It has long been understood that the 80% or more is computed for each business component by taking the total of all qualified activities, including direct supervision and direct support related to the process of experimentation, divided by the total activities previously outlined.
Example: A taxpayer spends $2,000,000 in wages on activities to develop a new product meeting all the requirements of section 41(d). Of the total, $700,000 for engineers actually conducting the experiments, $200,000 for direct supervision of engineers conducting research, $900,000 for those directly supporting the engineers conducting research, and $200,000 for research activities not involving a process of experimentation, but deductible under section 174 and not specifically excluded.
Conclusion: Based on a long-time understanding of what is included in the computation to determine the 80% or more requirement, the numerator is $1,800,000 ($700,000 + $200,000 + $900,000) over the denominator of $2,000,000, arriving at 90% of the activities constituting a process of experimentation. Treas. Regs. section 1.41-4(a)(8), Example 4, supports this application of the 80% requirement.
However, the Court in Little Sandy Coal ruled that by definition direct supervision and direct support cannot be part of a process of experimentation because the activities are not part of the actual experiments. This means the numerator includes the wages only of those directly performing the experiments and the denominator incudes direct experimentation, supervision, support, and activities not constituting a process of experimentation, but which are deductible under section 174 and not specifically excluded.
Applying this interpretation to the example above, we would have to eliminate direct supervision and direct support from the numerator giving us a percentage of 35% – ($700,000/$2,000,000) for activities constituting a process of experimentation. This would mean that none of the expenses would qualify for the research credit.
This interpretation would essentially eliminate or drastically reduce the research credit taxpayers are allowed. The likelihood that 80% or more of total wages expended for a research project are for researchers doing direct research is slim.
Is the Court Correct in Their Decision?
I want to focus on four questions:
- What is the definition of “elements of a process of experimentation”?
- Does direct support and direct supervision constitute elements of a process of experimentation?
- Is the 80% or more test for a process of experimentation an activities-based calculation?
- Was a new standard used for substantiation?
Section 41(d) of the code and Treas. Regs. section 1.41-4 provide guidance for determining the 80% computation. Code section 41(b)(1) defines qualified research expenses and includes wages for qualified services under section 41(b)(2)(A)(i), which are defined under section 41(b)(2)(B) as services for engaging in qualified research, direct supervision, or direct support of research activities which constitute qualified research.
Next let us look at the 80% or more rule for process of experimentation outlined in Treas. Regs. section 1.41-4(a)(6) and described above. Throughout the section it refers to activities. Based on this, I believe the Court got it right to say when calculating the percent, you can only use activities.
When computing the numerator, clearly all direct research is included. The question is what else is included? It has long been accepted that direct supervision and support are also included. Not according to the Court in Little Sandy Coal. They argue that by definition a person providing supervision or support cannot actually be conducting research. Our question is, what is meant by the phrase, “constitute elements of a process of experimentation”? Code section 41(d)(1)(C) merely requires substantially all of the activities of which constitute elements of a process of experimentation for a qualified purpose. So, the definition of “constitute elements of…” a POE is unanswered.
Following Treas. Regs. section 1.41-4(a)(6) the denominator includes the amounts in the numerator plus any wages paid for activities on the research project that are not elements of a process of experimentation, which are deductible under section 174, and not specifically excluded under section 41(d)(4).
The answer to whether direct supervision and support are included in the numerator depends on what activities are included in the phrase “constitutes elements of a process of experimentation”?
By citing Treasury Regulation section 1.41-2(c) the Court determined that direct supervision and support by definition cannot be included because the individuals were not actually conducting experiments. I disagree. Code section 41(d)(1)(C) states substantially all of the activities of which constitutes elements of a process of experimentation and does not limit the definition to those conducting experiments.
In addition, for purposes of applying the substantial all test for the process of experimentation, the controlling regulations are Treas. Regs. section 1.41-4(a)(5), and Treas. Regs. section 1.41-4(a)(6). Treas. Regs. section 1.41-4(a)(5) defines a process of experimentation as:
A process of experimentation…involves the identification of uncertainty concerning the development or improvement of a business component, the identification of one or more alternatives intended to eliminate that uncertainty, and the identification and the conduct of a process of evaluating the alternatives (through, for example, modeling, simulation, or a systematic trial and error methodology).
Treas. Regs. section 1.41-4(a)(6) provides the substantially all test is met only if 80 percent or more of a taxpayer’s research activities constitute elements of a process of experimentation for a purpose described in section 41(d)(3).
In promulgating section 41(d)(1)(C) Congress chose to include the phrase “constitute elements of a process of experimentation.” Treasury incorporated the same wording in Treas. Regs. section 1.41-4(a)(6). Neither Congress nor Treasury limited the process of experimentation to simply conducting experiments. To interpret otherwise, simply ignores the plain meaning of the phrase “elements of a process…”
The Tax Court’s decisions in Union Carbide and Suder reject Little Sandy’s narrow interpretation of a what constitutes elements of a process of experimentation. In Union Carbide Corp. & Subs v. Commissioner, the court held the taxpayer’s broad range of activities on the Amoco anti-coking project satisfied the process of experimentation requirement: (1) review research of prior testing of available technology; (2) preparing a test plan; (3) designating reference and experimental cracking sets; (4) preparing for the test; (5) applying the pretreatment; (6) collecting test data; (7) analyzing the data; (8) forming a conclusion; (9) refining the hypothesis; and (10) repeating steps 4 through 8 for refined hypotheses.
Thus, Union Carbide included a broader range of activities constituting elements of a process of experimentation than Little Sandy Coal. Activities such as reviewing testing data for existing technology, prototype building activities, and data analysis (activities often conducted by employees who may not been involved in the actual conduct of experiments), were deemed to constitute elements of a process of experimentation.
Similarly, in Suder, the amount of QREs associated with direct support and supervision far exceeded the 20% threshold. Yet, the Court ruled the taxpayer had satisfied the process of experimentation test for all 12 projects evaluated. As discussed below, this result would have been impossible if direct support and direct supervision were excluded from the numerator.
Finally, to exclude direct supervision and support from the numerator and include the amounts in the denominator makes the 80% or more virtually unattainable. As just stated, the actual experiments are not where most of the costs are. Some will say shrink back; however, my question is to what? Shrinking back is to be appropriately applied, however, this does not address the question of what is meant by “constitutes elements of.”
It defies logic and the intent of the credit to limit the numerator to experiments only.
As to substantiation, Congress, Treasury, and previous court decisions have been reluctant to set burdensome substantiation requirements and taxpayers have been provided reasonable flexibility in substantiating their credit. Here, it seems the Court has indicated flexibility should not be given and has departed from previous decisions.
Application of the Little Sandy Coal Decision to Suder v. Commissioner
To take it a step further, let us apply the decision in Little Sandy Coal to the Tax Court decision in Suder. In Suder, the same court that decided Little Sandy Coal ruled for the taxpayer. I will use two key issues from Little Sandy in this comparison — substantiation and the application of the process of experimentation substantially all test.
First, the court in Suder accepted the taxpayer’s methodology of assigning a percentage of time each employee spent performing qualified services based on job titles and the testimony of senior management, and occasionally lower-level supervisors if Mr. Wende did not have direct knowledge of what the employees did. The result was wage QRE determined for the senior leadership team being Mr. Suder, CEO at 75%, Mr. Boyd, President and Chief Operating Officer at 25%, Mr. Wende, Senior VP of Product Development at 100%, and Mr. Hansen, Chief Technology Officer at 100%. The engineers, product managers and product testers were mostly qualified at 100% and the remaining employees at various percentages.
The second important difference is how the Court in Little Sandy Coal applied the substantially all test under section 41(d)(1)(c). In Suder, the taxpayer claimed wages, supplies, and contract labor as QREs. Focusing on wages, the following table shows wages allowed by the Court as QREs after reducing Mr. Suder’s wages under section 174(e) to a reasonable amount and allowing 75% as qualifying expenses.
|Wage QRE As Adjusted by the Tax Court|
|Year||Total Wage QRE Allowed by the Court||Total Wage QRE at 80% for Purposes of the POE Sub-all Test||Suder QRE @ 75% of Total Allowed by Court||Suder’s Wages as a % of Total Wage QRE||Boyd QRE @ 25%|
In Suder, the Court did not even raise the possibility that direct supervision or direct support should not be used in computing the substantially all requirement for the process of experimentation to qualify. To the contrary, the Court held, “we find that 80% or more of the activities with respect to each of the 12 projects constituted elements of a process of experimentation.”
The Court concluded Mr. Suder worked 20 to 30 hours per week, describing his activities:
“He spent most of his time brainstorming ideas for new products and ways to improve existing products. The inspiration for his ideas came from various sources, including reseller feedback, online research, trade publications, and alpha testing of ESl’s products, which he personally performed. His coworkers thought of him as the “chief idea guy” and the “product visionary.”
Suder’s wages alone exceed 20% of the total wage QRE and it is clear from the case he was not involved in actual experiments (or if he was it was limited). Based on this alone, if we excluded direct supervision and direct support as the Court did in Little Sandy Coal, it would be impossible for the taxpayer in Suder to meet the 80% substantially all requirement for the process of experimentation test. This is without considering the wages of Boyd, Wende, Henson, other supervision, and the required direct support involved in completing the development or improvement of a new product or process. In addition, the Court did not require a line-by-line analysis of the activities the employees performed, as the Court seems to suggest in Little Sandy Coal.
As I have previously written, it was a challenge as Director of Engineering at the IRS to examine research credit issues, due to the fact substantiation and practicable application is difficult for taxpayers. I welcome clarity from the Courts and the Little Sandy Coal case offers some.
However, in the areas of substantiation and the application of the POE 80% substantially-all test in Little Sandy Coal the Court has merely added to the confusion. What are the rules in these two areas? We have two opposing opinions within the Tax Court. I know from discussions with colleagues it would be appreciated if Treasury and IRS offers clarity. Taxpayers and practitioners calculate the research credit real time and we do not have the luxury of waiting for guidance. If there are different rules going forward taxpayers have to know it promptly.
Applying the substantially-all rule that excludes direct supervision and direct support from the numerator and leaves it in the denominator will reduce, if not eliminate, the amount of research credit a taxpayer can claim. This was not Congressional intent, nor Treasury’s intent and thwarts the purpose of the credit. I trust we will get guidance from the government soon.
 T.C. Memo 2021-15.
 T.C. Memo 2009-50.
 T.C. Memo. 2014-201.
 Under section 174(e), R&E expenses must be reasonable in amount to be deductible.
 T.C. Memo 2014-201, p. 21.
 Id. at 3.