## Facts

- Applied Systems, Inc., an S corporation, developed and sold business software for independent insurance agencies.
- In the early 1990s, Applied Systems enhanced its software to add features (e.g., more ratings computations, insurer–agency transactions, multi-user file access without data corruption, and improved memory usage).
- The company replaced a licensed word-processing module with a simpler text editor designed to reduce memory demands while supporting form-letter generation.
- Shareholders, as pass-through owners, claimed the federal research credit under 26 U.S.C. § 41(a) based on the company’s software-development expenditures.
- The IRS disallowed the credit, disputing whether the activities met the statutory definition of “qualified research” under § 41(d), particularly the “discovery” and “process of experimentation” elements.
- The Tax Court held the expenditures were not for “qualified research,” and the shareholders appealed.

## Issues

1. Whether the company’s commercial software-development activities constituted “qualified research” under 26 U.S.C. § 41(d), so that related costs could generate a § 41(a) research credit.
2. Whether a more taxpayer-favorable standard stated in a proposed Treasury regulation could be applied to soften the governing “discovery” requirement.

## Decision

- The Seventh Circuit affirmed the Tax Court’s denial of the § 41 research credit.
- The court held the software work was normal, incremental commercial development and did not satisfy the regulatory “discovery” requirement for qualified research.
- The court upheld the finding that the work did not constitute a qualifying “process of experimentation” under § 41(d)(1)(C).
- The court rejected reliance on Proposed Treasury Regulation § 1.41-4, concluding proposed regulations have no legal effect and cannot displace binding regulations.

## Legal Principles

- For the § 41 credit, “qualified research” must satisfy § 41(d)’s requirements, including that it be undertaken to discover technological information and involve a process of experimentation.
- Under the applicable regulations, research does not qualify if it amounts to routine implementation or refinement using known techniques rather than generating information beyond existing knowledge in the relevant technological field.
- Iterative development work is not a statutory “process of experimentation” when it is not aimed at resolving genuine technological uncertainty about feasibility or method.
- Proposed Treasury regulations are not binding authority and do not override existing, effective regulations.

## Conclusion

The court affirmed the IRS’s disallowance of the research credit because the company’s commercial software enhancements were routine development using known methods, lacked a qualifying experimental process, and could not be recharacterized as qualified research through reliance on nonbinding proposed regulations.


Eustace v. Commissioner of Internal Revenue, 312 F.3d 905 (7th Cir. 2002)