Boeing Comments on Proposed Regs on Research and Experimental Expenses

On behalf of The Boeing Company ("Boeing"), Baker & McKenzie, LLP, is pleased to submit these comments on the proposed regulations REG-124148-05 (Research Expenditures) ("Proposed Regulations") issued by the U.S. Department of the Treasury ("Treasury Department") and the Internal Revenue Service ("Service") on September 6, 2013, regarding the amendment to the definition of research and experimental expenditures under section 174 of the Internal Revenue Code ("Code").¹ In general, Boeing commends the Service and Treasury drafters for their efforts in this important and evolving area. For the reasons set forth below, however, Boeing believes that the Service and Treasury should consider a number of changes to the proposed regulations.

Boeing also requests the opportunity to have its representatives testify on its behalf at the hearing scheduled for January 8, 2014, at 10 a.m., regarding the topics discussed below, as well as any topics that other commentators may raise.

Boeing is one of the world's major aerospace firms. Boeing's principal business segments include Boeing Commercial Airplanes ("Commercial Airplanes") and Boeing Defense, Space & Security ("Defense"). The Commercial Airplanes segment is engaged in the research, development, production, and marketing of commercial jet aircraft, and also provides related support services. Commercial Airplanes' customers generally include the commercial airline industry worldwide. The family of Commercial Airplane products includes the 737 narrow-body model and the 747, 767, 777, and 787 wide-body models. For its part, the Defense segment principally conducts research, development, production, modification, and support of products and related systems for military and government purposes. Over time, Boeing's aerospace research and engineering have been responsible for many of the most fundamentally transformative breakthroughs in the history of flight -- from the creation of commercial passenger planes to the mass production of warplanes to the development of transcontinental jetliners to the invention of the first revolutionary, composite-based Dreamliner. Boeing's breakthroughs likewise have contributed to the exploration of outer space, including research and development contributions that were essential to build and sustain the International Space Station, produce and deploy a fleet of Space Shuttles to commute to and from the station and conduct space-based research, and place American astronauts on the Moon and bring them safely home again.

All of these innovations, so familiar to all of us, which have brought us from the Wright Brothers' bi-plane to what would have been considered science fiction less than a century ago, are the product of research and development, particularly Boeing's research and development. That research and development, encouraged and supported by our tax laws since the 1950s, has made America's aircraft not only the most innovative, most efficient, and fastest in the world -- it has made America's aircraft safe. The comments below are about not losing sight of the most important goal: to keep America's aircraft, and the people who fly in them, safe.

I. Executive Summary.

In many respects, the Proposed Regulations provide helpful guidance regarding the application of section 174 to pilot models and have many positive aspects that will likely reduce the confusion and the unnecessary conflicts that have increased in frequency between taxpayers and the Service's exam teams over the past several years. In that regard, we commend the Treasury Department and the Service for seeking to bring some quietude to these disputes, which often have required further administrative review or even litigation to resolve. However, in other respects, the Proposed Regulations conflict directly with section 174 and its established regulatory authority and judicial precedents. And, perhaps most important to Boeing, the Proposed Regulations do not reflect the reality of the research and experimental efforts that Boeing and the rest of the aerospace industry must undertake to develop complex, highly integrated, and safe aircraft. It is these aspects of the Proposed Regulations that we believe should be revised to reflect the realistic considerations inherent in the production of aircraft in the aerospace industry.

The primary changes that the Treasury Department and the Service should make to the Proposed Regulations concern: (1) the "shrinking-back" rule; (2) the use of multiple pilot models; and (3) the expenses paid to a third party for a pilot model.

(1) The Proposed Regulations' shrinking-back rule, as applied to aerospace pilot models, fails in its goal of minimizing potential disputes because it disregards the interrelationship of an aircraft's components and subcomponents and the relationship of those components to the entirety of the aircraft. Under section 174 and the Treasury regulations thereunder, expenditures made to eliminate uncertainty concerning the development or improvement of a product are research or experimental expenditures eligible for deduction. The shrinking-back rule -- despite its name -- does not "shrink back" to expenditures that satisfy section 174 to the exclusion of nonqualified expenditures. Instead, the rule creates a new class of expenditures: expenses that satisfy the requirements of section 174 but that, nonetheless, may no longer be deductible because those expenses relate to a large tangible aircraft made up of numerous individual components. This distinction based on the type of the expense, rather than whether the expense eliminates uncertainty, is directly contrary to the Code, Treasury regulations, many Proposed Regulations examples, case law, and administrative guidance.

In our view, the Treasury Department and the Service should remove the shrinking-back rule in its entirety, including the example, from the Proposed Regulations. See Prop. Treas. Reg. § 1.174-2(a)(5) and Prop. Treas. Reg. § 1.174-2(a)(11), Example 8. In the alternative, the Proposed Regulations, to be consistent with section 174, should make explicit that elimination of uncertainty with respect to the integration of components is sufficient to avoid application of the shrinking-back rule, In other words, so long as the expenses incurred to integrate components are necessary to eliminate uncertainty, then the expenses will not be subject to the shrinking-back rule. In Boeing's case in particular, and in the aerospace industry in general, flight testing of the entire aircraft is indispensable and is a non-discretionary part of its research and experimentation to ensure that each aircraft design is safe. The shrinking-back rule as it is currently written, then, simply cannot be applied to aircraft because of their highly integrated components and subcomponents. In fact, the Federal Aviation Administration ("FAA") certification process and requirements, Boeing's flight testing, and incidents involving aircraft design show that all aircraft pilot models qualify under section 174 at the product level.

(2) Under an example in the Proposed Regulations, testing for multiple pilot models is permitted if each pilot model is tested for a discrete purpose. See Prop, Treas. Reg. § 1.174-2(a)(11), Example 5. In the aerospace industry, however, multiple pilot models are necessary for testing sometimes without such a discretely identified purpose. Without multiple pilot models, a failure may not be correctly identified as a design problem or a unique problem encountered by the pilot model because of, for example, a defect in materials. Many different types of flight testing must be repeated on different pilot models to identify the root cause of a problem or to confirm the absence of a problem. The example should thus make clear that the question is whether the expenditures for all of the multiple pilot models are necessary to eliminate uncertainty, regardless of whether the pilot models are intended for a specific testing purpose, to be qualified for section 174 treatment.

(3) The Proposed Regulations should clarify that all expenses paid to a third party to produce a pilot model qualify as research or experimental expenses under section 174. An example in the current Treasury regulations concludes that heat, light, and power, incurred in connection with a special project are research and development expenditures and, as such, are deductible under section 174. See Treas. Reg. § 1.174-4(c). The example in the Proposed Regulations leaves this point ambiguous. See Prop. Treas. Reg. § 1.174-2(b)(5), Example 1. The Proposed Regulations should clarify that all expenses attributable to the production of the pilot model constructed by the third party qualify under section 174.

II. Background.

The term "research and experimental expenditures" includes costs incurred in a trade or business for "research and development . . . in the experimental or laboratory sense." Treas. Reg. § 1.174-2(a)(1). Expenses are incurred for research and development if the expenditures are for activities that are "intended to discover information that would eliminate uncertainty concerning the development or improvement of a product." Id. "Uncertainty 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." Id. The term "product" includes any "pilot model, process, formula, invention, technique, patent, or similar property." Treas. Reg. § 1.174-2(a)(2). The product may be developed either for use in the taxpayer's business or for sale, lease, or license. See id. Qualification under section 174 "depends on the nature of the activity to which the expenditures relate, not the nature of the product or improvement being developed or the level of technological advancement the product or improvement represents." Id. (Emphasis added.) Generally, "research or experimental expenditures" include "all such costs incident to the development or improvement of a product." Treas. Reg. § 1.174-2(a)(1).

Many different types of expenses may be deducted under section 174. An example in the Treasury regulations identifies a variety of expenses as deductible under section 174, including: salaries; heat, light, and power; drawings; models; laboratory materials; attorneys' fees; and allocable depreciation on a building attributable to the research project. See Treas. Reg. § 1.174-4(c), Example.

The Proposed Regulations' definition of a pilot model contemplates application of a shrinking-back rule. The shrinking-back rule addresses situations in which the requirements of Treas. Reg. § 1.174-2(a)(1) are met with respect to only a component part of a larger product and are not met with respect to the overall product itself. See Prop. Treas. Reg. § 1.174-2(a)(5). This portion of the Proposed Regulations states that "[t]he presence of uncertainty concerning the development or improvement of certain components of a product does not necessarily indicate the presence of uncertainty concerning the development or improvement of other components of the product or the product as a whole." Id.

The shrinking-back rule of the Proposed Regulations is illustrated by an example that concerns a manufacturer of aircraft engines. See Prop. Treas. Reg. § 1.174-2(a)(11), Example 8. In this example, an aircraft engine manufacturer researches and develops a new type of compressor blade. A compressor blade is a component of an aircraft engine. The new compressor blade design improves the performance of the aircraft engine. The engine manufacturer produces and installs the compressor blade on an aircraft engine produced by the company. The example concludes that the "costs of producing and installing the compressor blade component that [the aircraft engine manufacturer] incurred represent research and development costs in the experimental or laboratory sense." Id. The costs incurred to produce and install the component qualify as research or experimental expenditures. "However, section 174 does not apply to [the manufacturer's] costs of producing the aircraft engine on which the component was installed." Id.

In addition to the example described above, the preamble to the proposed regulations discusses the shrinking-back rule. See Notice of Proposed Rulemaking and Notice of Public Hearing regarding "Research Expenditures," 78 Fed. Reg. 54,796, 54,798 (Sept. 6, 2013). The preamble states that qualifying expenditures "may relate only to one or more components of a larger product." Id. The preamble notes this scenario may occur in the case of "a large tangible asset made up of numerous individual components." Id. The preamble provides an example of an automobile. The design of the automobile may be certain except for the appropriateness of the braking system design. The preamble states that the Service and the Treasury Department believe that it is inappropriate to deny section 174 eligibility with respect to the braking system because there is not uncertainty with respect to the automobile's general design. The preamble also states that the shrinking-back rule applies only if the requirements of section 174 are not met with respect to an overall product, and the rule is not applied to exclude research or experimental expenditures from section 174 eligibility. See id.

The Proposed Regulations also provide that if expenditures qualify as research or experimental expenditures, then the disposition of that property, including the sale or use of the pilot model in the taxpayer's trade or business, is irrelevant to a determination of eligibility under section 174. See Prop. Treas. Reg. § 1.174-2(a)(1). These regulations also amend Treas. Reg. § 1.174-2(b)(4) to provide that the depreciable property rule is an application of the general definition of research or experimental expenditures described in Treas. Reg. § 1.174-2(a)(1) and should not be applied to exclude otherwise eligible expenditures. Finally, the Proposed Regulations state the general rule that the costs of producing a product after uncertainty concerning the development or improvement of a product is eliminated are not eligible for treatment under section 174, because the costs are, at that point, not for research or experimentation.

Confronted with the task of designing many complicated, highly integrated components, an engineer's ability to predict the behavior of one, or many, aircraft systems before an aircraft first takes flight is limited by the simulation and testing that can occur on the ground. Boeing uses laboratory tests and simulations to predict various aircraft characteristics including loading, takeoff, handling in engine-out and heavy crosswind conditions, stall speed, and landing. These predictions, however, must be validated during flight testing of an entire aircraft to demonstrate the capabilities, and the safety, of full-aircraft configurations in a range of environmental conditions. While Boeing uses the most advanced on-ground technology available to make these predictions, the predictions must be confirmed, or disproved, by aircraft behavior during flight testing. And Boeing cannot flight test only an aircraft's "components" or "subcomponents."

Boeing conducts extensive flight testing to make certain its aircraft satisfy the company's stringent safety and airworthiness requirements. This testing is crucial to ensure that all aircraft systems work and communicate per design, to understand what happens if systems fail and the consequences of failure, and to determine whether aircraft will perform successfully during flight conditions required by regulatory agencies, including the FAA.

Boeing has developed its flight test program requirements based on the company's nearly 100-year experience testing aircraft to enhance safety and performance. During flight testing of an aircraft, the company's engineers and flight test pilots will discover how the aircraft operates in its service environment. As these discoveries are made, Boeing engineers research and develop solutions to resolve and incorporate new designs into the tested aircraft and, eventually, future aircraft.

During flight testing, Boeing engineers receive feedback from the intricate monitoring and testing systems that are installed in each pilot model, as well as from the test pilots themselves, regarding the aircraft's behavior. The engineers review the aircraft's design based on that accumulated data and on the pilots' feedback to ensure that all safety issues are addressed before flight testing progresses. These design changes include modifications to ensure that the commercial airline pilots who will ultimately fly the aircraft with hundreds of passengers on board will be able to safely control the final product even in the most adverse conditions.

Some of the aircraft developed by Boeing are completely new and without precedent. Boeing's 787 aircraft represents, for example, a step change in the design and production of modern aircraft. Other aircraft, however, are based on, but differ significantly in design from, previous aircraft. This latter type of aircraft is commonly referred to as a "variant" (or "derivative") aircraft. As with a completely new and original aircraft design, designing variant aircraft and ensuring their safety is no different as compared to designing an entirely new aircraft. In this respect, a variant aircraft is a brand new aircraft. In fact, the difficulty of predicting how the integrated components and subcomponents of a new or variant aircraft will behave in flight is greatly complicated by the reality that a change to one component invariably will affect many others -- some foreseen and others not:

Ensuring that the designs of new and variant aircraft are safe and airworthy requires extensive testing both on the ground and in the sky. For example, flight testing for operational safety and performance capabilities is critical. Simply lengthening the fuselage of an aircraft, which may constitute the design of a "variant," will materially and permanently change the physics of the aircraft altogether. Each aspect of the aircraft's design and of its components and subcomponents must be tested to measure the effects that the "simple" act of adding another section to the fuselage (and thus perhaps adding another several dozen seats) has had on the existing aircraft design, such as the effects on flight control surfaces (e.g., the aircraft's rudder and elevator), secondary structures (e.g., structures that are not subject to primary loads), and other important components. Lengthening the fuselage of an aircraft will mean that the aircraft has less rotation capability and tail clearance. Performance during takeoff will be affected because the longer aircraft must take-off at a lower angle as compared to the original, shorter version. Too great a take-off angle will damage the aircraft tail by causing it to strike the ground. Additionally, a lengthened fuselage changes the aircraft's gross weight, trimming capability, and fuel-burn profiles. These types of effects, and almost innumerable others, require Boeing to perform exhaustive ground and flight testing to gather data to ensure the safety and performance of variant aircraft. Testing individual "components" and "subcomponents" alone, independently of the aircraft as a whole, is insufficient to eliminate uncertainty regarding the whole aircraft's design with the new components or subcomponents included.

The FAA Administrator regularly issues regulations and minimum standards governing aircraft safety. See 49 U.S.C. § 44701 (2000). The FAA regulations are found in Title 14 of the Code of Federal Regulations ("14 C.F.R."). In general, the statutory and regulatory scheme is designed to regulate the airworthiness of aircraft.

The FAA's Administrator issues "type certificates." A type certificate is issued to an applicant to signify the FAA's belief that the design of an aircraft is safe and complies with all applicable regulations.

FAA regulations provide the technical requirements, such as load conditions and other flight characteristics, that must be satisfied to obtain type certification for transport aircraft. See 14 C.F.R. § 25.1-1801.

An applicant, such as Boeing, receives a type certificate for an aircraft design if the design meets the applicable regulations and the applicant performs all testing and computations necessary to show that the design meets such regulations. See 14 C.F.R. § 21.21. Once the applicant's internal testing and computations are completed, the applicant presents the aircraft to the FAA together with all necessary documentation so that the FAA can perform its own testing. See 14 C.F.R. § 21.33.

To ensure the safety of every certificated aircraft type design, the FAA requires each applicant to perform flight testing. See 14 C.F.R. § 21.35. Before conducting FAA flight tests, the applicant must show: compliance with applicable structural requirements; completion of necessary ground inspections and tests; that the flight test aircraft conforms with the type design; and that the Administrator received a flight test report from the applicant containing the results of his tests. See 14 C.F.R. § 21.35(a)(1)-(4). Upon demonstrating satisfaction of the applicable requirements, the applicant "must make all flight tests that the Administrator finds necessary" to determine compliance with the applicable requirements and to determine whether there is reasonable assurance that the aircraft and its components are reliable and function property. 14 C.F.R. § 21.35(b)(1)-(2); 14 C.F.R. § 21.33(a).

After complying with all of the FAA's requirements, the FAA Administrator issues a type certificate to an applicant that includes the type design, operating limitations provided in the product's manual, certificate data sheet, the applicable regulations with which the FAA requires compliance, and any other conditions or limitations prescribed for the aircraft. See 14 C.F.R. § 21.41. A product's type design consists of: the drawings and specifications necessary to define the configuration and the design features of the product; information on dimensions, materials, and processes necessary to define the structural strength of the product; and any other data necessary to determine the aircraft design's airworthiness. See 14 C.F.R. § 21.31(a)-(e).

Certain design changes to a certificated aircraft type design require a new type certificate. "[I]f the FAA finds that the proposed change in design, power, thrust, or weight is so extensive that a substantially complete investigation of compliance with the applicable regulations is required," then an application for a new type certificate must be made. 14 C.F.R. § 21.19.

In general, changes in type design are classified as "major" or "minor." A "minor change" has no appreciable effect on the weight, balance, structural strength, reliability, operational characteristics, or other characteristics affecting the airworthiness of the product. 14 C.F.R. § 21.93(a). Minor changes may be approved "under a method acceptable to the FAA before submitting to the FAA any substantiating or descriptive data." 14 C.F.R. § 21.95. Aircraft undergoing a minor change generally do not require an amended type certificate or the accompanying flight testing. All other changes are considered "major changes." Id. The process for approval of major changes is extensive and requires an applicant to:

Typically, significant product level changes are major changes that require an amendment to the original type certificate. See Establishing the Certification Basis of Changed Aeronautical Products, Advisory Circular No. 21.101-1 A, (Sept. 3, 2010); see also 14 C.F.R. § 21.93. Because variant aircraft designs typically include many different major changes, an applicant generally must apply for an "amended" type certificate. The process for qualifying a variant aircraft design for an amended type certificate mirrors the process for receiving an entirely new type certificate, which requires pilot model testing of the entire aircraft. See 14 C.F.R. § 21.101. The Boeing 777-300ER is an example of a variant aircraft design that includes many major changes as compared to a previous type design and required Boeing to apply for an "amended" type certificate.

In addition to the FAA's requirements, Boeing's commercial aircraft must satisfy the requirements of the European Aviation Safety Agency and other foreign national aviation authorities around the world. These authorities perform functions similar to the FAA by conducting safety analysis and research, authorizing foreign operators, implementing and monitoring safety rules, and granting type certification for aircraft designs and components. These organizations all require Boeing to comply with stringent safety standards.

III. Flight Testing the Boeing 777-300ER.

The Boeing 777 is an extremely advanced long-range wide-body twin-engine jet aircraft that performed its first flight on February 24, 2003. The 777 is the world's largest twin-engine aircraft. Boeing designed the 777 to replace earlier wide-body aircraft. Boeing has developed a total of six unique 777 models. The most recently developed model, the 777-300ER, is also the largest and most advanced 777 passenger model. The 777-300ER is based on the design of previous 777 models and powered by General Electric GE90-115B turbofan engines, the largest and most powerful engine in the history of aviation.

Although technically a "variant" of the 777, the 777-300ER is essentially a new, more complex aircraft comprised of more than 3 million different parts. The General Electric GE90-115B turbofan engines that power the 777-300ER produce substantially more thrust than the Mercury-Redstone Launch Vehicle rocket that carried Alan Shepard when he became the first American to enter space. In light of the significant design changes to the 777-300ER that were necessary for the aircraft to achieve its new "extended range" (i.e., the "ER") and to handle the thrust of the new engines, Boeing engineers had to develop and conduct ground and flight testing on pilot models to eliminate many 777-300ER design uncertainties. Based on the company's extensive experience in the research and design of modern aircraft, the company decided that it needed three pilot models for this purpose. The three 777-300ER pilot models accumulated hundreds of ground test hours and over 1,500 flight test hours during Boeing's test program. These pilot model tests on the ground and in the sky satisfied many hundreds of development and certification test items and enabled Boeing's engineers to correct issues discovered during testing. The ultimate 777-300ER design would have been, however, impossible without flight testing. Flight test analysis simply could not be completed on the ground or on a computer. With the 777-300ER design, the critical data required to design the aircraft could not be obtained through simulation. Boeing needed to conduct flight tests to collect critical data and test all of the theories and simulated predictions for the aircraft.

Boeing conducted extensive flight testing on the 777-300ER to discover and correct any and all of the aircraft's design uncertainties. For example, Boeing engineers were uncertain about the optimal design of an oil pump located within the 777-300ER's engine. The oil pump enabled the engine blades to "windmill" at reduced speeds when the engines shutdown inflight. During inflight engine shutdown, the airflow through the engine from the aircraft's forward momentum caused the engines to continue spinning and generate limited amounts of power. During flight tests of the pilot model, the engineers discovered that the engine lost oil at low windmilling speeds. During low speeds, oil gathered in the engines' drains and blew out of the back of the engine. This resulted in unacceptable losses of oil from the engine. Boeing engineers were uncertain of the cause of the problem and the appropriate design of the oil pump to enable it to operate properly at low windmilling speeds.

To evaluate the oil loss characteristics of the pump experienced at low windmilling speeds, Boeing engineers created a plan to evaluate the oil pump. Boeing engineers configured the flight test aircraft's right-hand engine with a newly designed oil pump. The new oil pump design had shown, during earlier testing, superior low speed performance relative to the previous oil pump design. The flight test plan specified eleven test conditions. For all conditions, the crew would shut down the engine and maintain flight speed for five to ten minutes. The crew would adjust the aircraft's speed and altitude to achieve different airflow pressure and windmilling speeds. After each condition, engineers recorded oil quantity for the engine to identify any amount of oil loss. Boeing engineers also monitored the engines for any oil loss.

Boeing conducted the flight tests to evaluate the new oil pump design, first completing seven conditions where the engine was shut down and windmilled at low speed for a total of 70 minutes. Boeing could not perform the tests on the engine alone. During all of the conditions Boeing engineers did not detect or visually witness any oil loss from the engine. Visual inspection of the drain after the flight showed no oil loss. The engineers concluded that the new oil pump installed on the right-hand engine for this test demonstrated satisfactory operation during low speed windmilling.

IV. Application of the Proposed Regulations to the Aerospace Industry and Boeing.

The shrinking-back rule in the Proposed Regulations inappropriately disallows a deduction for a type of expense that qualifies under section 174, however. This treatment of expenses under section 174 is unprecedented. All other legal authority, including examples in the Proposed Regulations themselves, determines whether a research and experimental expense qualifies under section 174 based on when the taxpayer incurred the expense, namely whether the taxpayer incurred the expense before or after the elimination of uncertainty.

Section 174 permits taxpayers to treat research or experimental expenditures paid or incurred during the taxable year in connection with their trade or business as expenses that are not chargeable to a capital account. These expenditures are allowed as a section 174 deduction. The Treasury regulations under section 174 define qualifying expenses as expenditures for activities intended to discover information that would eliminate uncertainty concerning the development or improvement of a product. See Treas. Reg. § 1.174-2(a)(1). Indeed, the Treasury regulations explicitly state that qualification for section 174 "depends on the nature of the activity to which the expenditures relate, not the nature of the product or improvement being developed or the level of technological advancement the product or improvement represents." Id. (Emphasis added.) An example in the Treasury regulations permits many different types of expenses including: salaries; heat, light and power; drawings; models; laboratory materials; attorneys' fees; and allocable depreciation on a building attributable to the research project. See Treas. Reg. § 1.174-4(c), Example. The qualification of expenses as research or experimental expenditures is thus based on when the expenses were incurred, before or after the elimination of uncertainty. See I.R.C. § 174(a)(1) and Treas. Reg. § 1.174-2(a)(1).² Consistent with this definition based on timing, all of the examples in the Proposed Regulations that precede the shrinking-back rule example base the determination of whether an expense qualifies under section 174 on the expenditure's timing, specifically before the resolution of uncertainty; the examples do not otherwise disallow research and experimental expenditures based on their type. See Prop. Treas. Reg. § 1.174-2(a)(11), Examples 3, 4, 5, 6, and 7.

Consistent with the Code and Treasury regulations, the case law under section 174 supports the inclusion of virtually any type of expenditure made for research and development. The court in Magee v. Commissioner, 32 T.C.M. (CCH) 1277 (1973), determined whether the costs incurred by a painter in the development of a portable scaffold invention qualified under the trade or business test in section 174. See id. Among the types of expenditures for which the painter claimed a section 174 deduction were those incurred in the purchase of: materials and parts; small tools and supplies; labor (sheet metal work, machine shop, welding, electrical work), photography and blueprints; patent attorneys fees; travel expense; telephone expense; sales tax exemption certificate; a house trailer including the remodeling (labor and materials); gas and heat; and a hitch for a utility trailer. See Id. at 1278. The Tax Court, ruling that the painter satisfied the trade or business test under section 174, upheld research or experimental expenditure treatment for all of the listed expenses incurred in connection with the taxpayer's trade or business. See id. at 1280.

In Kilroy v. Commissioner, 41 T.C.M. (CCH) 292 (1980), the taxpayer invented a new hydraulic mining system. See id. The Service challenged whether "miscellaneous office expenses" related to this invention were deductible under section 174. Id. at 296. The court observed that the taxpayer used his office in connection with his inventing business for such purposes as sending and receiving correspondence relating to the business, coordinating his research, and attempting to market his inventions. See id. The court allowed the taxpayer to deduct the miscellaneous office expenses for the amount related to inventing activities. See id. Similarly, in Best Universal Lock Co., Inc. v. Commissioner, 45 T.C. 1 (1965), the Service conceded on brief that, in the event the court found that all of the costs for experimental projects qualify under section 174, the taxpayer would be entitled to deduct the expenditures that, though concededly experimental costs, were merely designated as "shop payroll," "office payroll," "cash," and similar unidentified categories. Id. at 12. The court held hat the taxpayer could deduct the expenses under section 174. See id.

The Service also has allowed the deduction of a vast spectrum of expense types under section 174 in its own administrative guidance. In Rev. Rul. 73-275, 1973-1 C.B. 134, the Service considered whether expenditures connected with a product engineering department, "including salaries and overhead," were research and development expenditures within the meaning of section 174. Id. The sole function of the department was to design and develop, through research and experimentation, the concept of the products manufactured and sold by the corporation. See id. The Service ruled that the types of expenditures at issue connected with the product engineering development, including indirect costs, qualified under section 174. See id.

The Service addressed the question of whether indirect costs qualified under section 174 in the context of a joint research venture in which the taxpayer participated. Rev. Rul. 73-20, 1973-1 C.B. 133. Two organizations carried out the joint venture: one to promote and collect funds from the taxpayer (and additional contributors) for a research and development project, and the other to build and operate the project. See id. Monies for the project were funneled through the promoting organization to the organization operating the project. See id. The Service ruled that the type of costs incurred by the taxpayer in connection with the research and development project and paid to the promoting organization were section 174 expenditures. See id. at 134. The deductible expenditures included the amounts related to the "operating and administrative expenses" of the promoting organization. Id.

The Proposed Regulation's shrinking-back rule creates an unprecedented distinction in section 174 between research and experimental expenditures that are deductible and those that are not. Despite its name, the shrinking-back rule does not distinguish between expenditures that satisfy, and do not satisfy, the requirements of section 174. The rule does not "shrink back" to only those expenses incurred for qualified activities. Under the Code, Treasury regulations, Proposed Regulations examples, case law, and administrative guidance, virtually any expenditure made for an activity that otherwise meets all the statutory requirements is deductible under section 174. Instead, the shrinking-back rule in the Proposed Regulations creates a new class of expenditures: expenses that qualify as section 174 expenditures but that, nonetheless, cannot be deducted because those expenses relate to "a large tangible asset made up of numerous individual components." Notice of Proposed Rulemaking and Notice of Public Hearing regarding "Research Expenditures," 78 Fed. Reg. at 54,798. This standard is potentially subject to many different interpretations and could apply to any tangible asset with component parts. For example, what constitutes a "large" tangible asset? How many individual components are "numerous"? This broad language creates significant ambiguity. Further, the Proposed Regulations attempt to rewrite section 174(a)(1) to read:

Aircraft must be ground and flight tested using complete pilot models because the integration of an aircraft's components and subcomponents, as well as the relationship of all components to the whole aircraft, cannot be tested to determine the aircraft's safety in any other manner. Boeing cannot flight test aircraft "components" and "subcomponents" to ensure an aircraft's safety. The shrinking-back rule applied to aircraft pilot models disregards the factual interrelationship of an aircraft's components and subcomponents, as well as the relationship of such components to the entire aircraft. Boeing's experience with the 777-300ER, the company's safety requirements, and the FAA's requirements demonstrate the necessity for testing an entire aircraft, whether the aircraft is entirely new or a variant based on a previous model.

How an aircraft behaves while it is flying simply cannot be simulated on a computer, on a runway, or in a warehouse; a pilot model aircraft must become airborne. Uncertainty will always exist for all components and all subcomponents until the entire aircraft is flight tested with all components and subcomponents operating together in flight. As a result, Boeing, and all aerospace companies, must test every new and variant aircraft in its entirety before certifying the aircraft as safe. The aircraft and its components simply are not severable for this purpose.

Boeing's experience developing the 777-300ER, a variant aircraft based on previously designed models that included many major changes as defined by the FAA, shows the uncertainty created in designing and testing every new aircraft model. During flight test, Boeing's engineers corrected an oil pump design, among many other problems, which Boeing could only identify by flight testing the entire 777-300ER. No amount of laboratory testing on the ground could identify these problems. The company needed to assemble three complete 777-300ER pilot models and fly the aircraft to identify and correct these problems, eliminating uncertainty regarding the aircraft's design. Nothing less than complete models would do, because the highly integrated components and subcomponents of the 777-300ER can, and during flight testing did, behave in unanticipated ways. Boeing cannot eliminate all uncertainty in the design of its aircraft, such as the 777-300ER, and all underlying components and subcomponents without the use of pilot-model aircrafts.

The FAA's requirement that all aircraft be tested in their entirety to receive a type certificate or amended type certificate shows the need to test the entire aircraft as a pilot model. No other form of testing can provide the necessary assurance that the aircraft design is safe and airworthy. Pilot model testing is indispensable to eliminating uncertainty and ensuring the safety of an aircraft. In fact, despite all of the testing conducted by aerospace companies, and in compliance with the FAA requirements, aircraft incidents still occur. These incidents show that uncertainty can exist at the aircraft level even after exhaustive pilot model testing. Flight testing is essential to minimize or completely eradicate these incidents and identify as many aircraft-level design uncertainties as possible prior to an aircraft entering into service.

The Treasury Department and the Service should make explicit in the Proposed Regulations that uncertainty with respect to the integration and interoperability of an aircraft's components and subcomponents is sufficient to avoid application of the shrinking-back rule. Therefore, pilot models for all aircraft, including variant aircraft based on previous designs that include major changes such as the 777-300ER, qualify in their entirety, and the shrinking-back rule is inapplicable. The FAA certification process and requirements, Boeing's flight testing, and incidents involving aircraft design show that all aircraft pilot models qualify at the product level.

To show that the shrinking-back rule is not applicable to aircraft pilot models, the preamble discussion of products to which the shrinking-back rule is applicable should clarify that the rule is not applicable to tangible assets made up of individual components if testing the product in its entirety is necessary to eliminate uncertainty.

This example is unrealistic in practice. Modern commercial turbofan engines are highly complicated and integrated products composed of a fan, compressor, combustor, turbine, mixer and nozzle. The fan, located at the front of the engine, sucks air into the engine and splits the air into two separate portions. One portion of the air bypasses the center of the engine while the second portion passes through the engine's center. Within the center of the engine, the compressor squeezes air into progressively smaller areas resulting in an increase in the energy potential of the air. This compressed air passes into the combustors where fuel is injected and the mixture is ignited. The ignited airflow coming out of the combustor goes into the turbine, causing the turbine blades to rotate. The turbines are fixed on shafts that rotate all of the blades that spin during operation. Finally, at the very back of the aircraft engine, the mixer combines the bypass air with the engine exhaust escaping over the nozzle.

The example in the regulations ignores the interrelationship between all of the complex components and subcomponents within an aircraft engine, a highly sophisticated and integrated product. Similar to an entire aircraft, isolation of aircraft engine components is impossible due to the complexity of the engine's highly integrated components and subcomponents. "Components" and "subcomponents" of aircraft engines cannot be individually tested to eliminate uncertainty. The fan, compressor, combustor, turbine, mixer and nozzle are all under immense stress managing the tremendous forces produced during operation. Some of the highly integrated components spin thousands of times per minute, and the combustion chamber reaches temperatures well in excess of a thousand degrees Fahrenheit.

An aircraft engine manufacturer developing a new type of compressor blade would need to conduct extensive, comprehensive testing involving the entire aircraft engine. The critically integrated components of an aircraft engine operate together under extreme conditions. A dramatic change to one of the subcomponents, such as an entirely new type of compressor blade, would require changes to many other related components of the engine. An aircraft engine manufacturer designing a new compressor blade, in fact, would likely redesign the entire engine rather than redesign a critical individual part of the engine due to the difficulty in predicting how the new component would affect the engine's operation.

The same interrelationship is true for the aircraft itself. Individual, seemingly discrete, changes to components or subcomponents can have profound changes on the overall structure, operation, and safety of the aircraft. See John C. Dalton, Advances in Safety Assessment in New Airplane Design, American Institute of Aeronautics and Astronautics, Inc., Sept. 28, 1998, at 985606; Mohammad H. Sadraey, AIRCRAFT DESIGN: A SYSTEMS ENGINEERING APPROACH, 69-70 (2013). For example, as described, Boeing engineers discovered during flight testing of the 777-300ER that an oil pump design did not work as intended. Boeing could only discover this uncertainty through testing the aircraft, and the interrelationship of all components, in its entirety. Testing the oil pump alone simply was not an option.

Example 8 under Prop. Treas. Reg. § 1.174-2(a)(11) should be removed or revised to provide that an aircraft engine manufacturer that changes the design of its compressor blade is entitled to all expenditures for both the compressor blade and the engine. In the alternative, this example should be replaced with an example that makes clear that the application of the shrinking-back rule is impermissible where the overall product is composed of integrated components and subcomponents that affect the whole.

On its face, the example is sound and seems to be an appropriate application of section 174. The problem, however, is that it implies that only products that achieve that level of "new" and experimental -- those that perhaps reflect a step change in the industry -- would have expenses eligible for section 174 treatment. This example should, therefore, state explicitly that any pilot model aircraft used for flight testing qualifies under section 174. So long as the research and experimental expenditures incurred with regard to the pilot model are to eliminate uncertainty, then they should be deductible. The example's current use of "take off and land vertically" to describe the aircraft implies that an aircraft must possess a design feature never before accomplished in the aerospace industry to qualify under section 174. Interpreted in this manner, this excessively heightened standard of uncertainty would effectively preclude virtually every taxpayer from satisfying section 174.

This heightened standard is not applicable to section 174 expenses, which need only eliminate uncertainty concerning the development or improvement of a product. See Treas. Reg. § 1.174-2(a)(1). The design of all aircraft, including the 777-300ER in Boeing's case, are uncertain until flight testing pilot models of the aircraft has been completed. Therefore, this example should be changed to clarify that a pilot model of any aircraft qualifies under section 174. The first two sentences of the example should be changed to read:

This example is certainly a useful step to resolve one aspect of the disputes between taxpayers and the Service's exam teams regarding multiple pilot models. Under the example, however, the deductibility of the section 174 expenses for multiple pilot models is permitted only if each pilot model is tested for a discrete purpose. In the aerospace industry, companies such as Boeing that have built tens of thousands of aircraft through the years know from experience that they need multiple pilot models for testing. That extensive experience informs judgments as to how many pilot models and for what purpose each model will be used, sometimes using two or three models, sometimes as many as six or eight. Sometimes they use multiple pilot models that have discretely identified purposes, and sometimes they build repetitive testing into the aircraft test plans. Indeed, without multiple pilot models, a failure may not be correctly identified as a design problem or a unique problem encountered by the pilot model because of, for example, a defect in materials. Many different types of testing must be repetitive to identify the root cause of a problem. Therefore, provided that each pilot model is used to eliminate uncertainty regardless of any individually assigned type of testing, the expenditures for all pilot models should be qualified for treatment under section 174. There should be no limitation on the judgment of aerospace industry companies regarding how multiple pilot models are used. The Code and Treasury regulations already limit taxpayers to claim only amounts that are reasonable under the circumstances. See I.R.C. § 174(e) and Treas. Reg. § 1.174-2(a)(6). The fourth sentence of the example, therefore, should omit reference to separate environment testing and instead read:

Multiple models are necessary to test the design and eliminate uncertainty.

The taxpayer's payment to the third party for engineering and design labor in addition to materials and supplies used to develop the machine's design must necessarily include other expenses borne by the third party attributable to its carrying out the outsourced research and development project. The third party would be entitled to compensation from the taxpayer for all such expenses and such compensation to the third party should be deductible to the taxpayer under section 174. Example 1 should make clear that the expenditures made to the third party include all attributable costs and that those expenses are deductible unless the design uncertainty is eliminated before that cost is incurred.

V. Conclusion.

Aerospace industry products must be safe. To ensure the flying public's safety, Boeing cannot flight test "components" and "subcomponents." Boeing must undertake extensive ground and flight testing of entire pilot models to develop the complex, highly integrated, and safest aircraft in the world. Flight testing of the entire aircraft is an indispensable part of Boeing's research and experimentation to ensure that each aircraft design is safe. No other form of simulation or investigation is adequate. This form of testing is equally critical for both entirely new aircraft and variant aircraft to ensure that the aircraft design is as safe as possible. To reflect the unique considerations of the aerospace industry, the Treasury Department and Service should change the Proposed Regulations consistent with the suggestions herein.

Emily McMahon, Deputy Assistant Secretary (Tax Policy), Treasury

Lisa Zarlenga, Tax Legislative Counsel, Treasury

Alexa Claybon, Attorney-Advisor, TLC, Treasury

Scott Mackay, Taxation Specialist, TLC, Treasury

Curt Wilson, Associate Chief Counsel (Pass-Throughs & Special Industries), IRS

David McDonnell, Attorney-Advisor, Branch 5, ACC (PSI), IRS

Nicole Cimino, Senior Technical Reviewer, Branch 5, ACC (PSI), IRS

² The Treasury regulations specify activities that do not satisfy the requirements of section 174 as a threshold matter. See Treas. Reg. § 1.174-2(a)(3) (e.g., literary research) and Treas. Reg. § 1.174-2(c) (e.g., oil exploration). The Treasury regulations do not, however, identify any research or experimental expenditures under section 174 that taxpayers must nonetheless not deduct because of the expenditure type.