Read the Fifth Circuit’s decision [PDF 252 KB]
The taxpayer filed amended tax returns for 1994 and 1995 and claimed that it was entitled to a research tax credit for its expenses of developing certain vessels. The IRS denied the research credit claims, and the taxpayer filed a tax refund action in federal district court.
In a two-phase trial, the district court determined that only three of the eight ships developed in the claim years involved enough experimental work for their expenses to be counted as qualified research expenses (QREs). The taxpayer had lost much of its detailed records in Hurricane Katrina, and the court applied the statutory test for whether a project was qualified research on an all-or-nothing basis: If substantially all of the research activities in developing the project (80%) were part of a process of experimentation, the court would accept all of the costs as QREs; if less than 80% was experimental, none of the costs of that project would be QREs.
During the second phase, the district court considered how the research credit consistency rule would be applied. The research credit is 20% of the credit year QREs in excess of a base amount; the base amount is determined in part by reference to QREs in an earlier period.
The consistency rule requires that QREs in the credit year be determined on a basis consistent with their determination in the base period. The district court counted all 10 ships the taxpayer developed in the base period as being qualified, even though the taxpayer argued that the activities for four of them were less than 80% experimental. As a result, the base amount was higher than the taxpayer had computed, and it was allowed a research credit of only $135,788, rather than the $4.5 million it claimed.
Fifth Circuit’s decision
On appeal, the Fifth Circuit agreed with the taxpayer that the district court should have applied the same all-or-nothing test for the base period ships as it did for the credit year, and that vessels that did not satisfy the 80% experimentation standard should be excluded. The case was remanded, however, so that the district court could assess the credibility of testimony about how much experimentation took place on these four ships, and whether it would be appropriate to exclude their QREs from the base period amount, as the taxpayer argued.
The taxpayer had also appealed the district court’s determination that the work on one of the credit year ships did not meet the 80% test. The district court had made detailed findings about whether numerous specific activities in the ship’s development were experimental. The Fifth Circuit found the record of the district court’s analysis adequately documented and reasonable to support the overall finding that the ship in question was not sufficiently experimental to be counted.
The federal district court’s opinion included numerous holdings, largely in the taxpayer’s favor, about the standard for an activity to be qualified research. The district court also held that even when the materials used to resolve the developmental uncertainties of the project are the same materials that compose the finished ship that the taxpayer sells to its customer, those materials can be considered good supply expenses that are eligible for the research credit. The government did not challenge those holdings in the appeal of this case to the Fifth Circuit, and they may be useful to other taxpayers that wish to treat material costs as QREs.