Accounting for research and development costs
The matching principle tells us to expense costs in the same period that those costs provide some benefit to the company. Interpretation of the matching principle gets a bit fuzzy when dealing with research and development.
1. Conceptual problem
Generally, a company can expect to gain substantial benefit from research and development activities. Innovations can lead to cost reductions or higher market share in the future. On the other hand, research projects might turn out to be fruitless endeavors. How should a company account for such uncertainty in light of the matching principle?
2. General accounting treatment
The Financial Accounting Standards Board (FASB) in the USA decided to err on the side of conservatism when it required the immediate expensing of most research and development costs. The costs to be expensed include professional salaries, research supplies, expert fees associated with projects, and certain indirect costs. It’s a fairly straightforward requirement - the only potential difficulty involves depreciable equipment.
Equipment should be separated into two categories. The first category is equipment that has no other potential uses in the future other than various research projects. The entire cost should be expensed regardless of useful life or salvage value. The second category is equipment that can eventually be used for some other purpose besides research. This equipment should be capitalized as an asset and depreciated over its useful life. While being used for research purposes, research and development expense should be debited. Once it is put into general service, depreciation expense should be debited. In both circumstances, accumulated depreciation is credited as usual.
3. Software development
Research and development costs related to retail software (software for sale) are expensed under different rules. Once a project reaches technological feasibility, development costs can be capitalized in a manner similar to inventory production costs. As the software is sold, the capitalized costs are amortized to expenses. Similarly, costs incurred to develop internal software are expensed until technological feasibility is reached. Costs to further develop the software are capitalized, and then amortized like other short-lived intangibles.
4. Amortization example
Let’s assume that Friends Company, a fictitious entity, develops commercial software for various governmental units and agencies throughout the United States. In January, 20X3, the company spends $400,000 researching and designing the initial code for a software program. Later that year, the program reaches technical feasibility, and Friends spends an additional $1 million bringing the program up to commercial standards and specifications. In 20X4, the company has revenues of $3 million related to the program.
For our example, $400,000 would be expensed as research and development costs in 20X3, and $1 million would be capitalized as an asset. In 20X4, the portion of the $1 million asset amortized to expense is the greater of two possible methods - straight line or percentage of revenue. If the economic life of the software is 5 years, the amortization under the straight line method would be $200,000. If the company expects to bring in $30 million of revenue for the program, the amortization under percentage of revenue would be $100,000. Therefore, the company will amortize $200,000 of the asset to expense for 20X4.