A need for reformation: r&d theory and financial accounting treatment

By Jeff Burnison '93

Intermediate Accounting 1

Writing Objective: Engage in critical reading, to gain insight to form your own ideas on a chosen topic relevant to the course. Papers should reflect independent thinking based on evidence.

ABSTRACT: This paper focuses on existing R&D theory and practice as stated In FASB No. 2. Certain factors which led to the Board’s decision to expense R&D costs will be presented and criticized where poor theory, inconsistency, or faulty reasoning was used. Illustration of the negative impact FASB No. 2 has had on business profitability and management decisions will show that expensing R&D emphasizes the short-run and distorts long term goals. Finally, a look at some alternative ways of treating R&D will be explored and a new hybrid method of selective capitalization is endorsed.

In the accounting profession, a call for change sometimes arises when bad theory makes good practice or when good theory makes bad practice. After reviewing FASB No. 2,1 have found both bad theory and bad practice exist in treatment of R&D expenditures. Major flaws such as ignorance of the definition of an asset and disregard for the matching principle in present practice affirm the need to reform treatment of R&D.

Prior to 1975, methods of accounting for research and development were chosen by firms on an individual basis. As a result, numerous accounting and reporting practices existed, allowing for little consistency and comparability. It was this lack of R&D conceptual and methodological framework which led to the pronouncement of FASB No.2: Accounting for Research and Development Costs. Qualifying definitions of research and development are given:

Research is planned search or critical investigation aimed at discovery of new knowledge with the hope that such knowledge will be useful in developing a new product or service or a new process or technique or in bringing about a significant improvement to an existing product or process.

Development is the translation of research findings or other knowledge into a plan or design for a new product or process or significant improvement…It does not include routine or periodic alterations…and it does not include market research or market testing activities (FASB 2, par. 8).

Furthermore, FASB No. 2 does not apply to firms in the extractive industry unless the costs are actual R&D pertaining to new technology in methods, etc. Other exceptions have since been made in 1981 for the industries of record and music (FASB 50), cable TV (FASB 51), and motion pictures (FASB 53). Also, beginning in December of 1985, some capitalization of R&D is allowed in the development of computer software after a point of technological feasibility is reached (FASB 86). Except in these industries and under special circumstances (e.g., R&D conducted under contract), the general rule adopted by the FASB is that all research and development expenditures are to be expensed when incurred.

In reaching this decision, the underlying GAAP principle of conservatism was a pervasive element. By expensing all R&D, assets, net income, and owners’ equity are not overstated and presented in the most conservative fashion in the period costs are incurred. In future periods, however, net income would be “overstated” because no amortization would exist if R&D is not capitalized. Other factors presented by the Board in support of their decision were: a. Uncertainty of future benefits, b. Lack of causal relationship between expenditures and benefits, c. Accounting recognition of economic resources, d. Expense recognition and matching, and e. Usefulness of resulting information (FASB 2, par. 39-50). Next, a description of these factors according to the FASB and criticisms of the Board’s logic and theory will be explored.


The Board states that the, “…high degree of uncertainty about the future benefits of individual research and development projects…” was significant in reaching their conclusion (FASB 2, par. 39). Bierman and Dukes argue that these risks and uncertainties are blown out of proportion stating:

…business firms do not generally begin new product or process development projects until the principal technical uncertainties have been resolved through inexpensive research, conducted by their own personnel or by outsiders (1975. 49).

Managers are not likely to follow through with ridiculously high risk R&D projects. This is especially true in smaller firms when routine R&D losses cannot be afforded. Bierman and Dukes also contend that the Board erroneously based its decision on risks of individual R&D projects which are naturally high instead of considering the entire portfolio of projects. They illustrate this point with a simple example which I will summarize: Consider a firm with 100 independent R&D projects costing $10,000 each and the probability of success for each project is 10%. A successful project will result in $200,000 worth of present value benefits. The probability that there will be one or more successes is .99997. By applying expected present value probabilities, total expected future benefits are $2 million for the entire portfolio (Bierman and Dukes, 1975, 50). Clearly, the Board needs to more explicitly define uncertainty and risk and how these terms allow for expensing R&D. Many assets on the balance sheet have uncertain future economic benefit. Consider a machine with a long useful life in an environment of rapidly changing technology. The future economic benefits associated with this asset can’t be much more valuable than the future benefits associated with R&D costs going into a new machine that will potentially make the present machine obsolete. If both have uncertain benefits, how can it be that one is an asset and the other an expense? As Professor Robert Maurer claims, the R&D controversy may hinge on the definition of an asset. As defined by the FASB, “Assets are probable future economic benefits obtained or controlled by a particular entity as a result of past transactions or events” (Concept 6, par. 25). Following in a footnote, the FASB goes on to state that the term probable in this definition is, “…used with its usual general meaning, rather than in a specific accounting or technical sense…” (footnote 18). According to this definition of an asset, the portfolio (or a percentage of the portfolio) of R&D costs mentioned in the previous example are surety eligible for capitalization given a .99997 probability of future benefit.


The Board declares, “A direct relationship between research and development costs and specific future revenue generally has not been demonstrated…” therefore R&D must be expensed (FASB 2, par. 41). In the real world, seldom can one measure a causal relationship between costs and resulting benefits. There are external factors present in every situation. The environment in which the benefits are seen is an integral component in relation to the benefits. For example in the development of the cartoon “The Simpsons”, certain costs went into the production and marketing of the show. Obviously, the show has been a tremendous success. But, can one say that the costs of making the show caused the show to reap the kinds of profits it has? I contend that the watchers (external factor) made the program successful. To say that specific expenditures cause resulting benefits is to short-change the processes involved in our free market system. Certain marketing studies can predict consumer behaviors and then use this to predict probable future sales and profits. R&D costs can provide this same type of probable cost-benefit relationship if the pool of projects and their respective probabilities of success are large enough as previously shown. Also, in the production of “The Simpsons” you cannot point out a specific expenditure and show the specific benefit it caused. The entire pool of production costs is needed to produce an asset (the show) that has future probable economic benefit. R&D should be treated the same way. If the entire pool of R&D has a high probability of providing future economic benefit, then at least a portion of R&D should be capitalized.


The FASB argues here that R&D doesn’t meet the definition of an asset because the future economic benefits cannot be measured:

The criterion of measurability would recognize that a resource not be recognized as an asset…unless at the time it is acquired or developed its future economic benefit can be identified and objectively measured (FASB 2, par. 44).

The Board goes on to say, “…at the time most R&D costs are incurred the future benefits are at best uncertain” (FASB 2, par. 45). I disagree. After an R&D project reaches a point of feasibility, estimates can be made as to the types and amounts of expected resulting revenues and profits. A form of cost-benefit analysis could be used in application of R&D, Just as it is used when making purchase decisions concerning equipment, patents, or other assets that are capitalized and then subject to cost recovery. For example. Imagine MARS Corporation has determined that it would be feasible to produce and sell a sugar-free candy bar after some preliminary testing of the product and analysis of the market has been completed. Their staffs (e.g. accounting, management, sales, production, and marketing) conservatively estimate that $1 million worth of future profits could be realized if the candy bar was available to consumers. A portion of these benefits (profits) could be attributed to the R&D costs incurred in developing the candy bar to make It ready for mass production. As illustrated, after R&D reaches a point of feasibility, future economic benefits can be estimated and “measured” with the same accuracy (or lack of accuracy) that they are measured with a number of other assets.


In their conclusion to expense R&D, the Board maintains that FASB No. 2 adheres to “…matching in its most limited sense…” because no cause and effect relationship exists and there is no way to systematically and rationally allocate these expenses over time (FASB 2, par. 47). Clearly, the FASB has manipulated a definition to fit its cause. To say that the matching principle is adhered to in FASB No.2 is ludicrous. The purpose of R&D is to provide benefit in the future by generating new revenues through utilization of the products and processes developed in current R&D. It takes time for a project or process in the R&D stage to reach the market and actually begin to realize revenues. Expensing R&D when incurred is anything but good matching as Ellis and Mc Donald point out:

Even though one of accounting’s main principles is the matching of revenues to expenses of those revenues, the current (R&D) system fosters a bad match…with today’s revenues matched to spending for tomorrow’s products, and the costs of developing today’s revenues matched to yesterday’s revenues, the accounting mismatch is shifted one whole cycle (1990, 31).


Here, the FASB suggests that there is a lack of evidence demonstrating that capitalization would facilitate the assessment of future returns and risk: “…capitalization of research and development costs is not useful in assessing the earnings potential of the enterprise” (FASB 2, par. 50). Conversely, where disclosure is discussed, the FASB says that, “Disclosure shall be made in the financial statements of the total research and development costs charged to expense…” (FASB 2, par. 13) because, “The Board recognizes that disclosure of additional information about an enterprise’s research and development activities might be useful to some financial statement users” (FASB 2, par. 62). The FASB seems to be saying two different things. R&D expenditures are disclosed in the income statement because this information is useful in assessing future cash flows and growth. A firm with considerable investment in R&D will have a more favorable earnings potential compared to a similar firm with little invested in R&D. Still, by only disclosing total expenditures, investors are given only a very general idea of R&D costs as opposed to the quality and probabilities associated with those costs. If the portion of R&D costs which have been incurred after a point of technological feasibility is capitalized, a much better picture of R&D and its future impact would be painted for the external users of the statements.

Given the FASB’s overemphasis of conservatism and neglect of the matching principle, a new alternative method and theory needs to be sought. The present, simplistic method of expensing has caused many firms problems and will continue to do so unless a change is made. A study conducted by Nix and Peters (1988) investigated the negative effects of expensing all R&D.

Nix and Peters surveyed 200 R&D directors in 1983 at a time when profits were tight. They found that many corporate managers reduced current R&D expenditures in order to boost short term profits. “Seventy percent replied that when current profits are low, R&D expenditures for long term projects are reduced” (1988, 39). Unfortunately, this quick fix of reducing R&D to cushion ailing profits is going to cause greater harm to the firm in the long run. If a portion of R&D was capitalized and amortized over time, management would not be as tempted to use R&D as a manipulative tool in increasing or decreasing current profits. By taking this pressure off managers, more emphasis would be placed on the future health of the firm when budgeting for R&D.

In Appendix 2 of FASB No.2, the Board discusses the alternatives from which they chose to expense R&D when incurred. These alternatives were to capitalize all R&D when incurred, selectively capitalize, and accumulate these costs in a special category. Not all R&D costs provide future benefit so they should not all be capitalized for this would overstate the assets of the firm. But, some of the costs do provide future benefit; therefore, some of these costs should be capitalized and written-off over a scheduled time period.

Accumulating costs in a special category is supported by David and Paul Nix (1991). Their proposal is to put all of each year’s R&D costs (unless it’s obvious that no future benefit exists) into a contra stockholders’ equity account. At year end, a fixed percentage of the remaining balance in the contra account from prior years would be expensed for the current year. This cures the lump sum expensing problem and provides better matching, but these costs are never capitalized as assets. Instead, “capitalization” is accomplished Indirectly through the stockholders’ equity. The problem of understated assets still remains.

Ideal treatment of R&D should utilize a method that would neither overstate nor understate assets, expenses, net income, or stockholders’ equity. In addition to conservatism, proper matching would also be part of the practice and theory involved. As Gellein and Newman express:

GAAP requires revenue to be recognized when realized and costs to be associated with related revenue. Thus, costs incurred in one period to produce expected benefit in a future period are expenses of the future period and therefore should be deferred (1973, 73).

The method I am proposing is a hybrid sort of selective capitalization method. Selective capitalization is already used in the computer industry to account for R&D. FASB No. 86 allows for software R&D costs to be capitalized after a point of technological feasibility is determined. In other industries, the same ideology should apply. First of all, accounting treatment for R&D should be determined by considering the entire portfolio of costs and activities of a firm pertaining to R&D. Expected future revenues given the pool’s aggregate probability should be figured. For a small firm, if the portfolio’s probability of success is greater than 50% and the expected future value of revenues is greater than the sum of accumulated costs in the portfolio, an appropriate percentage of R&D costs should be capitalized. Capitalization would be figured on an item by item basis and capitalized over respective expected useful lives. For larger firms, the portfolio’s probability of success would need to be much higher (e.g., 85% or so). The reason larger firms are assigned a higher probability is two-fold. First, larger firms would be expected to be involved in a greater number of projects because they generally have more resources to do so. Each project has its own probability of success, therefore, the portfolio’s probability of success would naturally be higher for firms involved in more projects. Second, smaller firms are often more efficient R&D spenders as Nixon and Lonie point out:

In the U.S., the Joint Hearings (1978) established that firms with 100 or less employees produced 24 times more major inventions per research dollar expended than did larger firms (1991, 91).

A lower percentage would make the capitalization method more easily available to these smaller, more effective R&D spending firms.

Clearly, this approach is more complicated than expensing everything. Simple accounting for such a complex concept is inadequate considering the impact R&D treatment has on financial statements. Perhaps the accountant would have to call on other professionals, such as actuaries, to aid with the computations this method demands. Accountants would also have to work closely with management and forecasters to gather the necessary data. It should not be too much to ask of these professionals to give their professional judgements about the technological feasibility, the likelihood of a particular project succeeding, and the expected revenues that would be realized if the project was successful.

The theory and practice endorsed in FASB No. 2 are both in desperate need of reformation. Better matching and more useful financial statements are needed, especially when one considers the large amounts of R&D expenditures incurred by competitive and successful firms. In the future some form of selective capitalization must be practiced in all industries that spend material amounts on R&D. The computer, extractive, record and music, cable TV, and motion picture industries should not be the only ones allowed to use this superior methodology.