Computer Software and the Research Credit

TAX CASES

Congress enacted the research credit to encourage businesses to engage in additional, high-risk research activities. Whether computer software development qualifies for the credit is an ongoing question. Recently the Tenth Circuit Court of Appeals, in a case of first impression, reversed the district court that had allowed a taxpayer to claim the credit.

Tax and Accounting Software Corp. (TAASC) develops and sells software for use by tax and accounting professionals. It created four programs that provided unique functions at the time they were developed. TAASC claimed research credits of approximately $124,000 in 1993 and $192,500 in 1994 for the costs of these programs. The IRS denied the credits. The district court ruled in TAASC’s favor. The IRS appealed.

Result. For the IRS. The question before the court was whether the expenditures met the definition of qualified research. IRC section 41 defines such research as having five requirements:

1. The research must qualify as an expense under IRC section 174.
2. The research must be “undertaken for the purpose of discovering information.”
3. The information discovered must be “technical in nature.”
4. The information discovered must be intended for “the development of a new or improved business component.”
5. The research must “constitute elements of a process of experimentation.”

There was no dispute over the taxpayer’s research having met the first, third and fourth conditions. The parties disagreed about the second and fifth requirements.

The taxpayer argued that if research discovers technological information that leads to the development of an innovative product it meets the “discovering information” requirement. The government argued Congress intended the research credit to be more restrictive than the research deduction. Therefore, in order to qualify, the research must expand or refine a principle of science.

The Tenth Circuit rejected the taxpayer’s argument because it read the definition to require that the research develop new information used to develop a new product. Simply showing the development of a new product was not sufficient. The court also rejected the government’s position that the research must expand or refine a scientific principle. Information is not limited to expanding or refining physical or computer science. The Tenth Circuit concluded the research must result in new information. Then a company must take the information and apply it to create a new product or component.

The second area of dispute was the requirement that the research constitute a process of experimentation. The taxpayer argued that applying known methods with uncertainty as to which would solve the problem met the definition. The government argued that to meet the requirement the possibility of success must be uncertain. The Tenth Circuit found the phrase “process of experimentation” ambiguous. Therefore it turned to congressional intent.

The legislative intent was to encourage businesses to undertake high-risk research with an uncertain outcome. On this basis the court rejected the taxpayer’s argument and accepted the government’s position. In other words, a successful outcome from the research cannot be reasonably certain at the start of the process for a company to claim the credit.

The Tenth Circuit remanded the case to the district court for it to determine whether the questioned research met these clarified definitions. It’s doubtful this or any software development that has a high probability of success before research begins will meet the established requirements for the credit.

Tax and Accounting Software Corp. v. United States, CA 10.

Prepared by Edward J. Schnee, CPA, PhD, Culverhouse Professor of Accounting and director, MTA program, Culverhouse School of Accountancy, University of Alabama, Tuscaloosa.