On February 11, 2021, the Tax Court ruled in favor of the IRS in Little Sandy Coal Company v. Commissioner, finding that the taxpayer (petitioner) failed to meet the “substantially all” requirement with respect to the “process of experimentation” test under Internal Revenue Code (IRC) Section 41.
The dispute involved research tax credits claimed by Corn Island Shipyard, Inc. (CIS), a subsidiary of Little Sandy Coal Company, for the construction of several vessels for the tax year ending June 30, 2014. CIS is a corporation that designs, constructs and launches seafaring craft such as tankers and dry docks. The IRS took the position that CIS did not meet the “substantially all” requirement, i.e., that substantially all of the activities involved in the research constitute elements of a process of experimentation, which according to the relevant regulations means that 80% or more of the taxpayer’s research activities constitute elements of a process of experimentation for the costs related to the business component to be qualified. The IRS also determined that there was no identifiable sub-business component for the “Shrink Back” provision of Treas. Reg. §1.41-4(b)(2) to apply (under which a taxpayer will be permitted to claim certain improvement costs where the improvements made to an existing product account for less than 80% of the total product costs) and that CIS failed to produce sufficient records to substantiate its research tax credit. The Tax Court analyzed two selected projects: the “Apex Tanker” project and the “Dry Dock” project as representative samples of the types of activities undertaken by the taxpayer in the year at issue.
Although the Tax Court had “no doubt” that the iterative and detailed processes in designing the Apex Tanker and the Dry Dock evidenced processes of experimentation, the Court remained unconvinced that the taxpayer proved that the development activities constituted at least 80% of a process of experimentation.
For the Apex Tanker, Petitioner argued that many elements of the vessel—compared to previous models CIS had designed—were “redesigned and re-engineered during the development process.” CIS claimed that the tanker’s hull—“which makes up 90% of the vessel and was part of an extensive process of experimentation—demonstrated that the vast majority of the vessel, including every major system on the vessel, was re-engineered and redesigned meeting the substantially all requirement.” The court rejected this argument, ruling that the substantially all test must “be applied in reference to activities—not physical elements of the business component being developed or improved.”
Furthermore, although the Dry Dock project was a completely new design, the Court stated that, “accepting that the dry dock consisted entirely of novel elements that CIS had not previously designed and built would establish only that the design of those elements was uncertain at the start of the development process. Resolving that design uncertainty may have required, but did not necessarily require, a process of experimentation.” For both projects, the Court is clear that, “the novelty of an element of a business component does not establish that the work involved in developing that element involves a process of experimentation.”
In both examples, CIS asserted that substantially all of the time spent by its employees went toward a process of experimentation. The Tax Court responded that, while employee time or wages is a reasonable basis for determining if the substantially all test is satisfied, there must be a line-by-line analysis.
Significantly, the Court held that only the activities of the employees conducting direct research may be used to ascertain the numerator of the fraction that determines if a business component meets the 80% threshold and would not include supply costs or employees directly supporting or supervising research. According to the Court, the latter employees should be removed as they would not be directly involved in activities that constituted a process of experimentation. The Court further held that because supplies are not activities, when the fraction described in Treas. Reg. §1.41-4(a)(6) is computed using costs as a measure of activities, the costs of supplies used in the development of the product should not be taken into account. Correspondingly, the denominator is broad. In the view of the Tax Court, the sum of all other qualified activities and direct research should be included in the denominator. Here, it was mathematically impossible for either the Apex Tanker or the Dry Dock—analyzed at the project level—to meet the 80% threshold because the activities of directly supporting and supervising employees were excluded.
Even if a project did not meet the 80% threshold, the Tax Court signaled its acceptance that sub-components of these vessels could have likely satisfied a process of experimentation test, but given that CIS chose an “all or nothing strategy” to identify only the project at large as the business component without looking to the subcomponent level, the Tax Court was prevented from applying the “Shrink Back Rule.” This resulted in all activities related to the projects to be disqualified from inclusion in IRC Section 41.
The Tax Court focused on the taxpayer’s lack of substantiation with respect to its non-time- tracked employees. Although the Court recognized that CIS was burdened because there was an absence of any nontax reason to track the work of some of its employees by project and activity, the Tax Court highlights that Treas. Reg. §1.41-4(d) requires taxpayers to retain sufficient records to demonstrate they are eligible for the credit. Although the record-keeping requirement is liberal, the Court found that CIS’ choice “not to maintain detailed records of how its (non-production employees) spent their time placed on (CIS) the burden of demonstrating . . . the portion of those individuals’ work that did and did not involve a process of experimentation.” Thus, because CIS lacked detailed documentation, the burden fell on it to produce evidence of time allocations that it did not know. Consequently, CIS failed to meet its burden and all of the qualified activities associated with the projects were excluded.
The Tax Court’s decision rejects the U.S. District Court for the Northern District of Texas’ holding in Trinity Industries v. United States for its lack of detailed analysis around how two of the ships at issue, in that case, satisfied the substantially all requirement. The Tax Court’s decision reinforces the importance of analyzing business components and having a methodology to support a reasonable basis to quantify whether the substantially all test is met. The Court is particularly direct in requiring a “line by line” analysis of the costs associated with the processes of experimentation that occur for each business component.
Although Petitioner argued that analysis should be conducted based on assessing the new or improved nature of the product, the Court found that the process of experimentation test must be based on the activities of the employees. Without sufficient substantiation to support that substantially all of the activities constituted a process of experimentation, the taxpayer was unable to meet its burden and, due to the taxpayer’s all or nothing approach, its entire research tax credit was rejected. Taxpayers should carefully consider documenting employees’ time by type and by project.