Comments on proposed regulations relating to the credit for increasing research activities: March 6, 2002.

On March 6, 2002, Tax Executives Institute filed comments with the U.S. Treasury Department and the Internal Revenue Service on proposed regulations under section 41 of the Internal Revenue Code, relating to the credit for increasing research activities. The comments were prepared jointly by TEI' s Federal Tax and IRS Administrative Affairs Committees, whose chairs are Mitchell S. Trager of Georgia-Pacific Corporation and David L. Bernard of Kimberly-Clark Corporation, respectively. Contributing substantially to the development of TEI's comments were Eileen D. Frack of Kimberly-Clark Corporation and Loren M. Opper of Ford Motor Company. *

On December 14, 2001, the Treasury Department and the Internal Revenue Service released proposed regulations (REG-112991-01) on the definition of qualified research activities and other matters affecting the computation of the research tax credit under section 41 of the Internal Revenue Code. (1) The proposed regulations follow through on the Treasury Department's announcement (2) in Notice 2001-19 that, in response to concerns expressed by taxpayers about the rules set forth in T.D. 8930, (3) it would review those rules and issue re-proposed regulations. When final, the proposed regulations will modify the regulations under section 41 in T.D. 8930 that were issued on January 3, 2001, and subsequently suspended. The proposed regulations in REG-112991-01 were published in the Federal Register (66 Fed. Reg. 66362) and in the Internal Revenue Bulletin (2002-4 I.R.B. 404). (4) A public hearing on the proposed regulations is scheduled for March 27, 2002.

Background

Tax Executives Institute is the preeminent association of business tax executives in North America with more than 5,200 members representing 2,800 of the leading corporations in the United States, Canada, and Europe. TEI represents a cross-section of the business community, and is dedicated to the development and effective implementation of sound tax policy, to promoting the uniform and equitable enforcement of the tax laws, and to reducing the cost and burden of administration and compliance to the benefit of taxpayers and government alike. As a professional association, TEI is firmly committed to maintaining a tax system that works -- one that is administrable and with which taxpayers can comply.

Members of TEI are responsible for managing the tax affairs of their companies and must contend daily with the provisions of the tax law relating to the operation of business enterprises. We believe that the diversity and professional training of our members enable us to bring an important, balanced, and practical perspective to the proposed regulations on the research credit. Indeed, TEI has a vital interest in ensuring that the incentive effect of the research credit is properly implemented. Over the years, the Institute has submitted a plethora of comments on various aspects of the research credit as well as on the need for guidance under section 41(d), including pre-regulation comments (which were filed on June 9, 1997), comments on the December 1998 proposed regulations (which were filed on March 31, 1999), and comments in response to Notice 2001-19 (which were filed on May 7, 2001). We are pleased to submit the comments that follow.

Executive Summary

TEI commends the Treasury Department and IRS for issuing Notice 2001-19 and for undertaking a review of the statute, legislative history, court decisions, T.D. 8930, comments submitted in connection with the previously proposed regulations (REG-105170-97), and comments submitted in response to Notice 2001-19. The new proposed regulations represent a laudable and bold change in course. By eliminating the discovery test (or the "common knowledge" test) set forth in T.D. 8930 and aligning the requirements for qualified research more closely with section 174, the IRS and Treasury have taken significant steps to ensure that the regulations conform more closely with congressional intent to provide a salutary incentive for research activities and to ease the process for claiming the credit.

The proposed regulations have also beneficially clarified the process-of-experimentation tests set forth in T.D. 8930 so that a process of experimentation may include a process designed to evaluate one or more alternatives where capability or the method of achieving the result, or the appropriate design of that result is uncertain. Finally, the proposed regulations abandon the requirement in T.D 8930 that taxpayers create credit-specific documentation at the inception of a research project. Eliminating onerous recordkeeping requirements will significantly ease the administrative burden of claiming the credit. (5) TEI applauds these changes and believes that they address many of the concerns expressed by TEI and other commentators as well as those expressed by Congress in the Conference Report to the Tax Relief Extension Act of 1999. (6)

As a result, the new rules -- properly administered -- hold great promise for diminishing the scope, number, and degree of audit controversies about qualified research activities. Notwithstanding our enthusiasm for the many improvements in the proposed regulations, TEI has concerns and questions about some new provisions in the proposed rules as well as some provisions that are unchanged from T.D. 8930. Specific comments and requests for clarification are detailed below. A summary of the principal revisions that TEI recommends includes:

* Modifying the overbroad definition of "gross receipts" to adopt a consistent and easily administrable definition of gross receipts based on sales from a U.S. trade or business, which for most taxpayers is line 1(c) of Form 1120.

* Expanding the patent safe harbor in order to make it more effective.

* Clarifying the scope of excluded activities, especially research undertaken after commercial production to permit modifications to functional capabilities or design specifications to qualify for the research credit and to avoid creating per se exclusion rules for nebulous terms such as "debugging" and "trial production runs" that may involve a process of experimentation.

* Modifying and clarifying the rules relating to internal-use software, especially where the software is used in connection with the taxpayer's business of providing services to customers.

* Adopting a new procedure to permit taxpayers to obtain automatic consent for revocation of the alternative incremental credit computation under section 41(c)(4).

Gross Receipts

1. Definition. T.D. 8930 adopted a broad definition of the term gross receipts for purposes of computing the research credit. Gross receipts means the total amount derived by the taxpayer from all its activities and for all sources, reduced by certain enumerated items. (7) Regrettably, the proposed regulations retain the broad definition of gross receipts set forth in T.D. 8930. Many commentators, including TEI, said that the overbroad definition of gross receipts would impose a substantial administrative burden on taxpayers to recompute their base periods, especially where the data may not be available. The Preamble avers that the administrative burden is attributable to the incremental nature of the credit and the statutorily determined base period. (8) In TEI's view, the lack of guidance on the definition of gross receipts for more than 11 years contributed to the administrative burden. Indeed, in the absence of guidance, most taxpayers adopted a definition of gross receipts based on line 1(c) of Form 1120 (sales less returns and allowances) and excluded gross investment income and other receipts that have no direct relationship to the products or services for which research activities are conducted. (9)

TEI believes that taxpayers and the IRS would be better served by a definition of gross receipts directly related to the business activities that taxpayers' research activities are intended to support. Thus, where a taxpayer has consistently included in the computation of gross receipts subcategories of receipts under Prop. Reg. [section] 1.41-3(c)(2) for purposes of computing its research credit and where the continued use of such subcategories will not materially distort the computation of the credit, the taxpayer should be permitted to continue using that method notwithstanding Prop. Reg. [section] 1.41-3(c)(2). For example, where a taxpayer consistently used sales receipts less returns and allowances in its computation (even though it had receipts for other subcategories such as dividends, rents, royalties, and interest), then the continuing use of sales less returns should be permissible so long as it does not materially distort the taxpayer's credit computation. Evidence indicating that the use of such subcategories of gross receipts does not produce a material distortion should include the IRS's accepting the approach in prior examinations or a showing that the method has been used on prior returns.

In addition, the definition of gross receipts should be narrowed to enable U.S.-based multinational companies to be more competitive with foreign-based companies operating in the United States. The proposed regulations provide that foreign-based companies need only consider gross receipts that are connected with the conduct of a U.S. trade or business, whereas U.S.-based companies must include all gross receipts, including receipts from income from foreign subsidiaries or branches that are not connected with the conduct of the U.S. trade or business. This seems counterproductive to the purposes of the credit in encouraging U.S.-based research and ensuring that products made in the United States are competitive in the global marketplace. To ameliorate this effect, the regulations should permit a domestic corporation to compute gross receipts as though it were a foreign corporation with taxable income arising from income effectively connected with the conduct of a U.S. trade or business. In the event that solution is considered too complex, the definition of gross receipts for U.S.-based companies should minimally exclude interest (10) and all dividends, including Subpart F inclusions and section 78 gross-up amounts.

2. Consistency Rule and Effective Date. Under Prop. Reg. [section] 1.41-3(d), qualified research expenses and gross receipts that are taken into account in computing a taxpayer's "fixed-base percentage" and "base amount" must be determined on a basis consistent with the definition of qualified research expenses and gross receipts for the credit year. The consistency requirement requires taxpayers to restate base period gross receipts under the expanded definition of gross receipts and creates an unduly harsh administrative burden because many taxpayers may not have retained the necessary records to recompute the fixed-base percentage. (11) Moreover, since the regulations would be effective for tax years ending on or after December 26, 2001, they are in effect retroactive to January 1, 2001. The effective date may be beneficial to taxpayers in applying the definition of qualified research expenditures but detrimental in applying the gross-receipts rule. The harsh administrative burdens associated with the over-inclusive definition of gross receipts are best addressed by limiting the definition of gross receipts. If the definition of gross receipts remains unchanged, however, taxpayers should be permitted to elect to apply the regulations to tax years beginning after December 25, 2001.

Qualified Research Activities

1. Application of Discovering Information Requirement. Prop. Reg. [section] 1.41-4(a)(3)(ii) elaborates on the discovering information requirement by providing that --

   [a] determination that research is undertaken for the purpose of
   discovering information that is technological in nature does not require
   the taxpayer be seeking to obtain information that exceeds, expands or
   refines the common knowledge of skilled professionals in the particular
   field of science or engineering in which the taxpayer is performing the
   research. In addition, a determination that research is undertaken for the
   purpose of discovering information that is technological in nature does not
   require that the taxpayer succeed in developing a new or improved business
   component. (12)

TEI recommends that the following text be added at the end of the foregoing subparagraph:

   Moreover, a determination that research is undertaken for the purpose of
   discovering information that is technological in nature does not require
   that the taxpayer seek to develop an entirely new business component. The
   credit is available if the technological advance sought is evolutionary.
   Finally, a determination that research is undertaken for the purpose of
   discovering information that is technological in nature does not require
   that the taxpayer be the first to achieve the advance.

The text that TEI recommends be added to the regulation tracks the congressional committee reports confirming that evolutionary research activities are eligible for the credit. (13) Moreover, the added text confirms that eligibility for the research credit is not limited to the first taxpayer to discover information that is technological in nature.

2. Patent Safe Harbor. Prop. Reg. [section] 1.41-4(a)(3)(iii) provides that--

   [f]or purposes of section 41(d) and paragraph (a)(3)(i) of this section,
   the issuance of a patent by the Patent and Trademark Office under the
   provisions of 35 U.S.C. 151 (other than a patent for design issued under
   the provisions of 35 U.S.C. 171) is conclusive evidence that a taxpayer has
   discovered information that is technological in nature that is intended to
   eliminate uncertainty concerning the development or improvement of a
   business component. (14)

The issuance of a patent, however, is not a precondition for credit availability.

Where a taxpayer obtains a patent on a product or business component, TEI believes that fact should be presumptive evidence that all the qualification tests for research activities of section 41(d) and Prop. Reg. [section] 1.41-4(a)(2) are satisfied. That result may have been intended by the revision to the foregoing proposed regulation, but the rule's cross reference to "paragraph (a)(3)(i)," (15) -- a cross reference carried over from T.D. 8930 -- is confusing and potentially limits the scope and utility of the safe harbor. The reference in the safe harbor to "discovering information that is technological in nature" should, by definition, satisfy the requirements of Prop. Reg. [subsections] 1.41-4(a)(2)(ii) and (a)(4) in respect of discovering information that is technological in nature. In addition, we believe that the issuance of a patent is substantial evidence that the taxpayer has undertaken the "process of experimentation" required by Prop. Reg. [subsections] 1.41-4(a)(2)(iii) and (a)(5) and incurred expenditures that qualify under section 174 as well. In other words, research activities that lead to the issuance of a patent should satisfy most, if not all, the threshold requirements of section 41(d) and Prop. Reg. [section] 1.41(a)(2)(i) for qualified research expenditures.

In addition to clarifying the technical provisions of the safe harbor, the scope of property rights and know-how benefitting from the safe harbor should be expanded. Specifically, TEI believes that the safe harbor should be revised to expressly cover research activities that result in patentable interests in property. (16) At a minimum, the regulations should be expanded to include a safe harbor for copyrighted software. Indeed, since there are many legal means to protect the fruits of research activities, the IRS should consider extending the safe harbor to all items (trade secrets) that are protectable under common law or state or federal statutes.

The safe harbor regulation includes the beneficial statement that the issuance of a patent is not a precondition for availability of the credit. TEI agrees. Indeed, the prosecution of a patent claim can take many years and many patent claims may not be completed by the time of the taxpayer's examination, especially in light of ongoing efforts to increase audit currency. We do not believe the evaluation of taxpayers' research credit claims should be delayed pending the determination of a patent application. (17) In addition, taxpayers frequently refrain from seeking patents in order to retain maximum control of proprietary information and trade secrets. (18) Hence, as a practical matter, proprietary information and documents demonstrating the patentability of a product or business component should suffice to satisfy the safe harbor.

Finally, in addition to the statement currently in the proposed rule that a patent is not a precondition to eligibility of research activities, TEI recommends that the regulations include a statement that no adverse inference should be drawn regarding the eligibility of the research activities for the credit where a taxpayer abandons its efforts to obtain a patent or where the patent application is denied. (19) A patent may be denied because the U.S. Patent Office deems the component too similar to an item for which a patent has already been issued. In that case, the taxpayer's research activities are no less qualified simply because the taxpayer was not the first to achieve a particular technological advance.

3. Process of Experimentation -- "Readily Discernible" Standard. The Preamble to the proposed regulations states that --

   [a] taxpayer's activities do not constitute elements of a process of
   experimentation where the capability and method of achieving the desired
   new or improved business component and the appropriate design of the
   desired new or improved business component, are readily discernible and
   applicable as of the beginning of the taxpayer's research activities....
   (20)

In TEI's view, the determination whether the capability, method, or appropriate design of a new or improved business component is "readily discernible and applicable" as of the beginning of the research activity is an extremely vague standard that affords far too much discretion to revenue agents to challenge taxpayers' research activities. Indeed, TEI is concerned that, absent clarification and appropriate training, some revenue agents may -- as a means of perfunctorily disallowing taxpayers' research credit claims -- substitute the "readily discernible and applicable" standard of a process of experimentation for the discarded and misbegotten "discovery" test of T.D. 8930. To minimize this potential, TEI recommends inserting the phrase "to the taxpayer" immediately after the phrase "readily discernible" wherever it appears in the proposed regulations. (21) Inserting the phrase will help minimize disputes between the IRS and taxpayers about whether the taxpayer's research personnel "knew" or "should have known" that the capability, method, or design of a product could be "readily discerned." In other words, the IRS should not be permitted to use hindsight to question the skills or knowledge of the taxpayer's engineers or scientists. In TEI's view, the "readily discernible" standard should be based on a taxpayer's actual knowledge and the best evidence of a taxpayer's knowledge is its conduct, i.e., on the scope and extent of its research activities and process of experimentation, rather than on an undefinable measure of technological expertise that a taxpayer's engineers and scientists might be expected to possess. The key focus should be on what the taxpayer does, and where it engages in research activities, then the capability, method, or design of a new or improved business component is not "readily discernible" to the taxpayer.

To make this clear, the general standard of Prop. Reg. [section] 1.41-4(a)(5)(i) should be revised, as follows:

   For purposes of section 41(d), a process of experimentation is a process
   designed to evaluate one or more alternatives to achieve a result where the
   capability or the method of achieving that result, or the appropriate
   design of that result, is uncertain as of the beginning of the taxpayer's
   research activities. Thus, a taxpayer may undertake a process of
   experimentation if there is no uncertainty concerning the taxpayer's
   capability or method of achieving the desired result so long as the
   appropriate design of the desired result is uncertain as of the beginning
   of the taxpayer's research activities. However, a process of
   experimentation does not include the evaluation of alternatives to achieve
   the desired result if the capability and method of achieving the desired
   result, and the appropriate design of the desired result, are readily
   discernible to the taxpayer and applicable as of the beginning of the
   taxpayer's research activities. The capability or the method of achieving
   that result, or the appropriate design of that result is not readily
   discernible to the taxpayer if the taxpayer conducts tests to discover
   technological information that is not available to the taxpayer at the
   beginning of the taxpayer's research activities. A process of
   experimentation may include developing one or more hypotheses designed to
   achieve the desired result, designing and conducting an experiment to test
   and analyze those hypotheses, and refining or discarding the hypotheses as
   part of a design process to develop or improve the business component. For
   purposes of this paragraph (a)(5), factors that tend to indicate that the
   taxpayer has engaged in a process of experimentation are listed in
   paragraph (a)(5)(iv) of this section. A process of experimentation includes
   the testing of hypotheses to determine the cause of a design flaw even if,
   after the cause is determined, the correction to eliminate the flaw is
   readily apparent.

The additional italicized text emphasizes that the principal characteristic of a process of experimentation is the performance of testing. Rational taxpayers will not incur costs to (re)discover information of a technological nature that is "readily discernible and applicable." Thus, the taxpayer's testing activities should be prima facie evidence of the existence of a process of experimentation and the added italicized text will help clarify that result.

4. Process of Experimentation -- Qualified Purpose. Prop. Reg. [section] 1.41-4(a)(5)(iii) provides that "[f]or purposes of section 41(d) and this section, a process of experimentation is undertaken for a qualified purpose if it relates to a new or improved function, performance, reliability or quality of the business component. Research will not be treated as conducted for a qualified purpose if it relates to style, taste, cosmetic, or seasonal design factors." (22)

The list of qualifying purposes omits research activities undertaken in order to reduce the cost of components, products, or processes. TEI believes that the development of a new or improved product or process that has the same function, performance, reliability, or quality as an existing product or process but a lower cost of production should constitute a qualified purpose. Achieving a cost reduction without a degradation in function, performance, reliability, or quality requires the evaluation of alternatives that is inherent in a process of experimentation. Hence, TEI recommends the following sentence be inserted between the first and second sentences of Prop. Reg. [section] 1.41-4(a)(5)(iii):

   Research activity intended to achieve a decrease in the cost of a product,
   component, or process without a degradation in function, performance,
   reliability, or quality of the product or component is a qualified purpose.

5. Factors Indicating a Process of Experimentation. Prop. Reg. [section] 1.41-4(a)(5)(iv) sets forth a facts and circumstances test for determining whether a taxpayer has engaged in a process of experimentation. The proposed rule also includes a list of factors that "tend" to indicate that the taxpayer has engaged in a process of experimentation. No factor is dispositive in determining whether a taxpayer has undertaken a process of experimentation and other factors can be considered. The factors include testing "numerous alternative hypotheses," engaging in "extensive, comprehensive, intricate or complex scientific or laboratory testing," or evaluation of "numerous or complex specifications."

TEI believes the factors in the proposed rules may encourage IRS examiners to place undue weight on the number of tests that taxpayers conduct or to overemphasize the "complexity" or "intricacy" of the scientific or laboratory challenge as a threshold requirement to qualify research activity as credit-eligible. As important, Prop. Reg. [section] 1.41-4(a)(5)(i) only requires that a taxpayer "evaluate one or more alternatives" that "may include developing one or more hypotheses." Concededly, much research activity will require multiple testing of multiple variables, but only one test of one hypothesis should be required, as the operative rule of the proposed regulations (i.e., Prop. Reg. [section] 1.41-4(a)(5)(i)) specifies. Hence, clause (A) of Prop. Reg. [section] 1.41-4(a)(5)(iv) should be revised, as follows:

   The taxpayer tests and analyzes one or more alternative hypotheses to
   develop a new or improved business component.

In addition, evolutionary advances and improvements in a business component or the design of a business component may not require "extensive," "complex," or "intricate" testing. Conversely, the most creative, revolutionary scientific or engineering advances may result from relatively few hypotheses or the formulation of an elegantly simple test. The genius of an advance or the brilliance of the testing of a hypothesis should not be disqualified because of its inherent simplicity. To provide balance against the notion that "complex" or "intricate" testing is required, TEI recommends that another factor be added to the list of Prop. Reg. [section] 1.41-4(a)(5)(iv), as follows:

   The taxpayer employs an elegant, but simple experiment to validate its
   hypothesis.

Finally, the essence of experimental commercial engineering and product development involves computer simulations, models, and physical prototypes. Iterative testing and modifications of all three are usually performed until a product is deemed ready for commercial production. Indeed, the legislative history expressly refers to a process of experimentation through modeling or simulation as a hallmark of qualified research activities. (23) In order to balance the factors (and overcome any vestiges of bias in favor of "white labcoat" scientific research over field-based or factory-floor experimentation), the following factors should be added to the three listed in Prop. Reg. [section] 1.41-4(a)(5)(iv):

   D. The taxpayer develops product and process prototypes, which are tested
   in laboratory and field settings and are subsequently modified as a result
   of information obtained from the tests.

   E. The taxpayer engages in product and process prototyping for a period of
   time prior to the commercial sale or use of the product or process.

6. Substantially All Requirement. Prop. Reg. [section] 1.41-4(a)(6) states that --

   [t]he substantially all requirement of section 41(d)(1)(C) and paragraph
   (a)(2)(iii) of [the proposed regulations] is satisfied only if 80 percent
   or more of the research activities, measured on a cost or other
   consistently applied reasonable basis (and without regard to [section]
   1.41-2(d)(2)), constitute elements of a process of experimentation for a
   purpose described in section 41(d)(3). The substantially all requirement is
   applied separately to each business component. (24)

In the Preamble, the Treasury and the IRS invite comments on the application of the substantially all rule, (25) especially whether research expenses incurred for non-qualified purposes are includible in the credit computation where substantially all of the research expenses constitute elements of a process of experimentation.

TEI believes that, as long as the "substantially all" requirement is satisfied, expenditures for research activities that include style, taste, cosmetic, and seasonal design factors should constitute qualified research expenditures. Under section 41(d)(3)(B), style, taste, cosmetic, and seasonal design factors are disqualified "purposes" for determining whether a process of experimentation has been undertaken, but they are not excluded activities listed in section 41(d)(4) for which the credit may not be claimed. So long as the "substantially all" requirement is satisfied, taxpayers should not be required to segregate and eliminate from the qualified research expenditures the incidental costs related to style, taste, cosmetics, or seasonal design factors. In footnote 26 we provide an example to illustrate the application of the rule. (26) In addition to illustrating that non-qualified activities are included in the elements of a process of experimentation where they constitute no more than 20 percent of the activities (whether measured on a cost or time basis), the foregoing example could be used to provide guidance to IRS agents in order to focus the scope of examinations. Where a taxpayer can demonstrate that substantially more than 80 percent of the costs of a research project do not relate to style, taste, cosmetics, and seasonal design factors, agents should not feel compelled to scrutinize every dollar of expenditure. Statistical sampling techniques could also be beneficially applied to minimize information document requests requiring taxpayers to trace every dollar of expenditure to a "qualified" purpose.

Finally, TEI recommends modifying the "substantially all" rule to expressly state that taxpayers may satisfy the requirement by employing a time-based measure. In other words, taxpayers should be permitted to use "cost, time, or other consistently applied reasonable" measure to determine whether the "substantially all" rule is satisfied.

7. Shrinking-Back Rule. Prop. Reg. [section] 1.41-4(b)(2) states that the requirements of section 41(d) "are to be applied first at the level of the discrete business component, that is, the product, process, computer software, technique, formula, or invention to be held for sale, lease, or license, or used by the taxpayer in a trade or business of the taxpayer. If the requirements for credit eligibility are met at that first level, then some or all of the taxpayer's qualified research expenses are eligible for the credit." (27) believes that including the phrase "some or all of" in the foregoing sentence renders the regulation vague and misleading because it implies that there is a basis for disqualifying research expenditures apart from the statutory exclusions set forth in section 41(d)(4). Hence, we recommend deleting the phrase "some or all of" and inserting, after the word credit and before the period for that sentence, the phrase "except to the extent disqualified under section 41(d)(4) and paragraph (c) of this section." (28)

8. Activities Deemed to Occur after Commercial Production. Prop. Reg. [section] 1.41-4(c)(2)(ii) sets forth a list of activities that are deemed to occur after the beginning of commercial production of a business component, including preproduction planning for a finished business component; tooling-up for production; trial production runs; trouble shooting involving detecting faults in production equipment or processes; accumulating data relating to production processes; and debugging flaws in a business component.

By definition, the activities will never qualify as research activities regardless of when or why they occur. TEI has significant issues with this language.

First, the list of activities deemed to occur after commercial production in Prop. Reg. [section] 1.41-4(c)(2)(ii) is drawn from the Conference Committee Report to the 1986 Tax Reform Act. The regulations, however, misconstrue the report. The Committee Report states that "no expenditures relating to a business item are eligible for the credit after the item has been developed to the point where it either meets the basic functional and economic requirements of the taxpayer for such item, or is ready for commercial sale or use." The Committee Report then lists examples of expenditures for activities -- in the context of a finished business component -- for which the credit is not available. By stating categorically that the excluded activities "are deemed to occur after the beginning of commercial production," the proposed regulations establish a per se rule excluding the activities from credit eligibility. In the context of the Committee Report, however, the six activities are excluded because the component is "finished," i.e., it meets the functional and economic requirements of the taxpayer or the product is ready for commercial sale or use.

Second, the meaning and interpretation of the terms or activities listed in Prop. Reg. [subsections] 1.41-4(c)(2)(ii)(A) to (F) that are deemed to occur after production varies from industry to industry. The regulations seemingly assume that the six activities have a common definition or uniform interpretation to determine when research is completed and the business component is no longer a subject of qualified research. For example, one of the six listed activities is the accumulation of data relating to a production process. Accumulation of data relating to production processes, however, can be a preliminary step in developing a hypothesis for testing and, hence, can form a part of a process of experimentation.

Third, taxpayers rarely cease development activities on their product lines. In the high-tech area, products are subject to rapid obsolescence so the next product version is often being developed before the current version is ready for commercial production. Even in industries with longer product life cycles, the products are subject to a continuing improvements and upgrades. Accordingly, post-production improvements to the current product version may reflect improvements developed for the next version of the product. These improvements should be considered eligible for the credit if the activities satisfy the other requirements of section 41(d).

Fourth, a business component may be mistakenly placed in production before it is actually ready for commercial production. Such products may suffer a severe malfunction or failure because of a design flaw not detected when the product was initially placed in production. The product was actually not ready for commercial production, and the research activities to correct the malfunction or failure should be considered qualified research eligible for the credit, even though the product had been erroneously considered ready for commercial production and sale.

Fifth, product debugging activities should be eligible for the credit if they meet all the requirements for qualified research. Debugging flaws in a business component can cover a broad range of activities from routine maintenance to substantive changes in the design of a product or process. Software developers, for example, rely on alpha and beta testing by customers to identify critical flaws in a new or improved software program that would prevent the software from satisfying its performance requirements. Eliminating these bugs often requires significant changes in the program's design, thereby delaying its release for commercial production and sale. Alpha and beta testing, together with related remedial changes, are integral parts of the software development process and should not be excluded from the definition of qualified research by a "post-production" rule. (29) Accordingly, TEI recommends that the regulations acknowledge that the term "debugging flaws in a business component" may constitute qualified research activity.

To address the foregoing issues, TEI recommends easing the per se rule excluding the listed activities in Prop. Reg. [section] 1.41-4(c)(2)(ii). Taxpayers should be permitted to demonstrate that the listed activities satisfy the requirements of section 41(d) either because they constitute a significant improvement to a product already in commercial production or because they are necessary to complete or remedy a product released prematurely to commercial production. TEI's concern could be mitigated by revising the opening sentence of Prop. Reg. [section] 1.41-4(c)(ii), as follows:

   The following activities, if they do not otherwise satisfy the requirements
   of section 41(d)(1) and paragraph (a)(2) of this section, are deemed to
   occur after the beginning of commercial production of a business component.

Finally, although section 41(d)(2)(C) requires that any production process be treated as a separate business component, a unique production process may in certain circumstances be an integral part of the successful development of the product being produced. If the production process requirements are an integral part of the product development efforts and, hence, influence the research process objectives and the product design, the production process development should be considered a part of the product development activities for purposes of satisfying the requirements of section 41(b).

9. Adaptation of Existing Business Components. Prop. Reg. [section] 1.41-4(c)(3) restates the statutory exclusion of section 41(d)(4)(B): "Activities relating to adapting an existing business component to a particular customer's requirement or need are not qualified research." (30) The exclusion does not apply, however, merely because a component is intended for a particular customer.

Distinguishing between research activities for an adaptation of an existing business component for which no credit is allowed and research for an "improved" business component for which section 41(d)(1)(B)(ii) permits a credit likely requires an examination of all the facts and circumstances in individual cases. Nonetheless, the distinction between the two activities should be addressed in the regulations. The General Explanation of the Economic Recovery Tax Act of 1981, in which the research credit was first enacted, provides some guidance to distinguish the two activities. Specifically, "adaptation of an existing capability to a particular requirement or customer's need as part of a continuing commercial activity" does not qualify for the credit. (31) Accordingly, an adaptation might be defined as occurring where an existing capability of a business component is used for another purpose. An improved business component, on the other hand, requires a change in the capability of a business component. In other words, once the known design limitations of a business component are exceeded, a qualified research project (rather than an adaptation) is underway. Hence, an improvement or change in a business component's function will qualify for the credit, whereas adaptation of the same functionality to a different use will not qualify. In addition, an adaptation occurs in the conduct of continuing commercial activity, rather than being a separate business project for the improvement of a component.

The fundamental distinction between an adaptation and an improvement is that a qualifying "improvement" to a business component must satisfy all the prongs of section 41(d)(1). An improvement must involve discovering information to resolve uncertainty (in the sense required under section 174), be technological in nature, and involve a process of experimentation. Under Treas. Reg. [section] 1.174-2(a)(1), "[u]ncertainty exists if the information available to the taxpayer does not establish the capability or method for developing or improving the product or the appropriate design of the product." Since the section 174 test is incorporated in sections 41(d)(1)(A) and (B), an improvement necessarily involves resolving an uncertainty that is technological in nature that can alter the capability of a product or the process of producing it. Where a customer requests a change to a product that does not involve a new capability or function, the change should be deemed an adaptation. A change that requires a new capability or function but that can be achieved without resolving any technological uncertainty would also be an adaptation. A change in capability or function that requires the resolution of technological uncertainty and a process of experimentation, however, should qualify as a credit-eligible improvement rather than as an adaptation. TEI recommends that the regulations clarify that research activities that are directed to produce a change in capability (including factors such as performance, weight, speed, reliability, safety, or cost) or function of an existing component (including the ability to manufacture the component) that involves a process of experimentation constitutes an improvement to a business component rather than an adaptation.

10. Surveys, Studies, and Research Relating to Management Functions. Prop. Reg. [section] 1.41-4(c)(5) states the general rules that "[q]ualified research does not include activities related to (i) efficiency surveys; (ii) management functions or techniques, including such items as preparation of financial data and analysis, development of employee training programs and management organization plans, and management-based changes in production processes (such as rearranging work stations on an assembly line); ...." TEI recommends adding the phrase "that, do not require a process of experimentation," immediately after the parenthetical and before the semicolon. The regulation conclusively presumes that laying out a production line never requires or involves the discovery of engineering information through a process of experimentation. TEI believes that this should not be assumed; taxpayers should be afforded some flexibility to demonstrate that they have engaged in a process of experimentation that results in the discovery of information that is technological in nature.

Recordkeeping for the Research Credit

Prop. Reg. [section] 1.41-4(d) states that a taxpayer claiming a credit under section 41 must --

   retain records in sufficiently usable form and detail to substantiate that
   the expenditures claimed are eligible for the credit. For the rules
   governing record retention, see [section] 1.6001-1. To facilitate
   compliance and administration, the IRS and taxpayers may agree to
   guidelines for the keeping of specific records for purposes of
   substantiating research credits. (32)

The Preamble explains that upon re-evaluating whether a research credit-specific documentation requirement is warranted, the Treasury and IRS concluded that because of the "high degree of variability in the objectives and conduct of research activities conducted in the United States," taxpayers must be afforded reasonable flexibility in the manner in which they substantiate their research credits. (33) TEI agrees with the statement in the Preamble and commends the Treasury and IRS for eliminating the requirement that taxpayers keep credit-specific documentation and for adopting instead the general "books and records" requirement of Treas. Reg. [section] 1.6001-1.

The Preamble invites comments on recordkeeping "rules" that will facilitate compliance, including whether pre-filing agreements should extend to the qualification of particular cost centers or to the procedures established by the taxpayer for determining the expenditures qualifying for the credit. The Treasury and IRS also invite comments on the extent to which guidelines may be developed on an industry-by-industry basis. (34)

Prefatorily, even within the same industry, taxpayers conducting the same or substantially similar research activities may adopt different recordkeeping practices. Hence, we urge caution in attempting to provide recordkeeping "rules" of any form, especially rules that might require standardized or uniform practices. Indeed, in any formal guidance the IRS should confirm that taxpayers are not required to follow the specific recordkeeping practices set forth in such rules. With that caveat, we agree that it may be helpful to taxpayers and revenue agents for the IRS to provide examples of recordkeeping practices that it believes can be employed to support research credits. Examples of records that may be helpful in substantiating that a taxpayer engages in qualified research include:

1. Experimentation and testing tending to show that the taxpayer's research is based on the hard sciences and evaluates one or more hypotheses.

2. Reports of prototype testing tending to show that technological information is obtained from the testing.

3. Reports to management summarizing issues arising from the testing and experimental process.

4. Documents setting forth attributes and features intended for a new or improved business component.

5. Surveys or questionnaires completed by knowledgeable individuals providing details of the technological nature of the research, the taxpayer's process of experimentation, and other factors.

TEI believes that pre-filing agreements and the industry issue resolution program hold great promise for reducing the scope and degree of controversy in many areas, including research credit claims. Hence, we encourage the IRS to explore using both issue management tools as a means of improving compliance and administration. (35) We caution, however, that the key to achieving improved compliance and administration is voluntary participation. Where a taxpayer's books and records support its research credit claims, a decision not to adopt an industry guideline or participate in a pre-filing agreement should not trigger increased audit scrutiny.

Finally, we note that record retention limitation agreements (RRLA) represent an alternative means to document claimed research credits. Rev. Proc. 98-25 currently provides guidance on the use of RRLAs to define the scope of taxpayer obligations in respect of electronic recordkeeping. We recommend that the IRS consider adapting the guidance set forth there in order to apply it to the research credit area. Indeed, the IRS should consider broadly promoting the use of RRLAs -- including positive record retention agreements -- with taxpayers and, especially, with revenue agents as a means of clarifying the records to be retained in support of a taxpayer's tax liability generally and the research credit specifically.

Internal-Use Software

1. General. Under Prop. Reg. [section] 1.41-4(c)(6), software "developed by (or for the benefit of) the taxpayer primarily for internal use by the taxpayer is eligible for the credit only" (36) where the software:

* is used in an activity that constitutes qualified research;

* is developed for use in a production process;

* is developed for use in providing computer services to customers;

* satisfies a high threshold of innovation requirement; or

* is developed together with hardware as a single product that is used to provide services to customers in the taxpayer's trade or business.

Under Prop. Reg. [section] 1.41-4(c)(6)(iv) software is presumed to be primarily for internal use unless it "is developed to be commercially sold, leased, licensed, or otherwise marketed, for separately stated consideration to unrelated third parties ...." (37) The broad presumption that software is for internal use applies even where software developed primarily for internal use is subsequently sold, leased, licensed, or otherwise markets the computer software for separately stated consideration to unrelated third parties. (38) The proposed regulations cite computer software developed for "general and administrative" purposes and for providing "noncomputer services" as examples of internal use software.

In comments on the previous proposed regulations and T.D. 8930, companies in service industries have urged the Treasury and IRS to use the grant of authority in section 41(d)(4)(E) to write legislative regulations expanding the exceptions to internal-use software in order to afford an exception from internal-use software for self-developed software used in providing services to customers in the ordinary course of business. Since manufacturers are accorded a statutory exception under section 41(d)(4)(E)(ii) for software developed for "production processes" such an administrative exception would ensure parity in the treatment of self-developed software for service businesses. Regrettably, the Treasury Department and IRS have concluded that all software used in providing "noncomputer services" should be subject to the stringent rules relating to internal-use software. (39) Hence, the broad definition of internal-use software applies to many software applications developed by service industries for use in delivering services to customers. In rejecting proposals for an administrative exception, the Treasury and IRS seemingly ignore the most recent expressions of congressional intent in the extension of the research credit. Specifically, the Conference Report to the Tax Relief Extension Act of 1999 provides:

   The conferees further note the rapid pace of technological advance,
   especially in service-related industries, and urge the Secretary to
   consider carefully the comments he has and may receive in promulgating
   regulations in connection with what constitutes "internal use" with regard
   to software expenditures. The conferees also observe that software
   research, that otherwise satisfies the requirements of section 41, which is
   undertaken to support the provision of a service, should not be deemed
   "internal use" solely because the business component involves the provision
   of a service. (40)

This language reflects a clear congressional intent to place service businesses on an equal footing with other industries. We urge the Treasury Department and the IRS to reconsider the proposed regulations and provide an exception to internal-use software for software developed by service industries for the delivery of their services.

2. Primarily for Internal Use. Under Prop. Reg. [section] 1.41-4(c)(6)(iv),

   [u]nless computer software is developed to be commercially sold, leased,
   licensed, or otherwise marketed, for Separately stated consideration to
   unrelated third parties, computer software is presumed developed by (or for
   the benefit of) the taxpayer primarily for the taxpayer's internal use. For
   example, the computer software may serve general and administrative
   functions of the taxpayer or may be used in providing a noncomputer
   service. General and administrative functions include, but are not limited
   to, functions such as payroll, bookkeeping, financial management, financial
   reporting, personnel management, sales and marketing, fixed asset
   accounting, inventory management and cost accounting.

TEI has two specific comments about the definition of internal use.

a. The term "sales and marketing" as an example of "general and administrative" functions is vague. We believe the regulations are intended to address software related to internal customer service functions such as customer sales and contact tracking, billing, order entry, and tracking through delivery, as well as procurement. We recommend that the regulations be so clarified.

b. In contrast with software that is wholly internal to the taxpayer and essential to its internal functioning (and are discussed above), other software will permit customers and suppliers to interact directly with the taxpayer's systems in connection with the performance of a taxpayer's service for the customer. The Preamble notes that commentators

   urged that the definition of internal-use software exclude any software
   that includes an interface with customers or the public. An exclusion for
   software that includes an interface with customers or the public would
   entail substantial administrative difficulties and may inappropriately
   permit certain categories of costs (e.g., certain website development
   costs) to constitute qualified research expenses without having to satisfy
   the high threshold of innovation test. (41)

TEI believes that the exclusion in the proposed regulations is overbroad and, hence, may exclude from the credit some of the most highly innovative software developed and implemented over the past ten years that relates to customer package delivery, financial services, communications and messaging, and other services. (42) We urge the Treasury Department and IRS to consider inserting the following text immediately after the first sentence of Prop. Reg. [section] 1.41-4(c)(6)(iv):

   Computer software that is primarily used by customers, suppliers, service
   providers, or other third parties is not internal-use software. Software
   that enables customers to order merchandise, track package shipments, send
   messages, or otherwise enter directly into a market transaction in an
   automated system is not internal-use software.

3. Computer vs. Noncomputer Services. Under Prop. Reg. [section] 1.41-4(c)(6)(ii)(C)(3) research activities to develop computer software to provide computer services to customers is not required to satisfy the stringent requirements for internal-use software. The proposed regulations define a computer service as a "service offered by a taxpayer to customers who conduct business with the taxpayer primarily for the use of the taxpayer's computer or software technology. A taxpayer does not provide a computer service merely because the customers interact with the taxpayer's software." (43)

The proposed regulations require that a taxpayer determine a customer's primary motivation for entering into the business relationship. TEI believes this subjective and vague standard will prove unworkable and, thus, recommends that the regulations provide a better definition of computer services. One approach to revising the rule would be to define computer services as any service that creates, converts, distributes, manipulates, stores, or transmits machine-sensible electronic data. Another approach would be to employ the concept of "separately stated consideration" as used elsewhere in the proposed regulations. Thus, the first sentence of Prop. Reg. [section] 1.41-4(c)(6)(v) should be revised, as follows:

For purposes of this section, a computer service is a service offered for separately stated consideration by a taxpayer to customers who [begin strikethrough]conduct business with purchase such service[end strikethrough] from the taxpayer primarily for the use of the taxpayer's computer or software technology or know-how.

By employing the "separately stated consideration" concept, the added text clarifies that a taxpayer can develop software technology that is part of a computer service (and thus excepted from the requirements of internal-use software) and also develop software to be used in connection with noncomputer services that might be required to comply with the more rigorous internal-use software requirements.

TEI's recommended text also adds software "know-how" to the services that are encompassed in "computer services." In addition to developing their own software technology to provide services, many computer services companies apply "know-how," or expertise in the application of third-party software, when providing services to customers. The definition and qualification of a taxpayer's service as a "computer service" should not depend on the quantum of self-developed software included in the services but rather should be based upon the total computer or software technology component (including the application of computer or software know-how) embedded in the taxpayer's services.

4. High-Threshold-of-Innovation Test. Prop. Reg. [section] 1.41(c)(6)(vi) sets forth a three-prong test for determining whether internal-use software is eligible for the credit under the high-threshold-of-innovation test of Prop. Reg. [section] 1.41(c)(6)(ii)(4). Under the first prong of the test, software is considered innovative if it is intended to be "unique or novel" and is intended to differ in a "significant and inventive" way from prior software implementations or methods. None of the terms is defined so the ordinary dictionary meaning of the words should apply. As a result, TEI believes the standard is redundant because "unique or novel" software will nearly always differ in a "significant and inventive" way from prior software implementations or method. (44) TEI recommends, at a minimum, eliminating one part of the redundant first prong of the highthreshold-of-innovation standard and providing a definition and clarification of the remaining term. More important, it is unclear what the terms are intended to measure. Is uniqueness, novelty, significance, or inventiveness of software measured by the state of the art of software generally or within the taxpayer's industry? Assuming a benchmark can be established for one of the terms, how is a taxpayer's internal-use software to be compared against the benchmark? The legislative history of the 1986 Act (as well as the previous proposed regulations and T.D. 8930) defines innovation in terms of reductions in cost or improvements in speed, but the Preamble states that the Treasury and IRS rejected these factors in order to "update the definition of innovative, and to provide a more flexible definition with continuing application." (45) In the absence of a better measure of innovation, we recommend that the regulations retain the "reductions in cost" and "improvements in speed" criteria reflected in T.D. 8930 and the legislative history of the 1986 Tax Act. The "unique or novel" and "significant or inventive" test is highly subjective and, hence, likely not workable.

5. Effective Date of Internal-Use Software Regulations. Prop. Reg. [section] 1.41-4(c)(6)(ix) states that the internal-use software provisions apply to all tax years beginning after 1985. Depending on the number of open taxable years, the retroactive effective date will produce widely disparate treatment of similarly situated taxpayers. We urge the government to reconsider the requirement that all taxpayers apply the new internal-use software rules on a retroactive basis. We believe that taxpayers should be permitted, but not required, to apply the new rules retroactively.

Automatic Consent Procedure for Revocation of Alternative Incremental Research Credit

Section 41(c)(4) accords taxpayers an election to compute their research credit under an alternative method of computation. Under section 41(c)(4), an election to use the alternative incremental research credit (AIRC) computation applies to the taxable year for which it is first made and all subsequent taxable years unless revoked with the consent of the Secretary. Rather than require taxpayers to file an application for consent to the revocation, TEI recommends that the IRS issue a revenue procedure setting forth the requirements for granting taxpayers automatic consent to revoke their alternative incremental credit elections? The automatic consent procedure should permit taxpayers to make the change by computing their applicable credit under the "regular" method of section 41(a) and attaching a form in the taxpayer's timely filed (including extensions) original federal income tax return for the year of change. Should the National Office believe it necessary to monitor such changes, a copy of the application for automatic consent to the revocation should be filed with the National Office simultaneously with the return.

Examples in the Proposed Regulations

The proposed regulations include a number of examples illustrating the application of the definitions and operative rules. In a number of cases, the examples provide the only practical guidance for applying the vague concepts in the proposed regulations. Regrettably, some examples can be misapplied as a template for routinely and mechanically disallowing research credit claims. In order to provide better balance to the examples, we provide in the Appendix a number of comments, recommendations, revisions, and alternative examples that we believe should be considered for inclusion in the regulations.

Conclusion

Tax Executives Institute appreciates this opportunity to present our views on the proposed regulations relating to the computation of the research credit and the definition of qualified research activities. The comments were prepared under the aegis of TEI's Federal Tax Committee whose chair is Mitchell S. Trager. If you should have any questions, please do not hesitate to call either Mr. Trager at 404.652.2690, or Jeffery P. Rasmussen of the Institute's legal staff at 202.638.5601.