OPINION
LYDON, Senior Judge:
The plaintiff in this litigation is an affiliated group of corporations which includes a parent corporation, Fairchild Industries, Inc., and various subsidiaries. For convenience, throughout this opinion the affiliated group of corporations shall be referred to as Fair-child. Fairchild seeks to recover deficiencies assessed by the Internal Revenue Service after the Service disallowed a portion of research tax credits Fairchild claimed on its 1983 and 1984 tax returns for work performed pursuant to a government contract. The issue before the court is whether the research Fairchild performed in its work on the contract was “funded” by the government. Having considered the parties’ cross-motions for summary judgment and having heard oral argument, the court concludes that Fairchild’s research was “funded” by the government and accordingly grants the government’s motion and dismisses the case.
I.
A. Fairchild’s Costs and Contract Payments
Beginning in 1977, the United States Department of the Air Force (Air Force) began a program designed to replace the Cessna T-37 plane as its primary trainer for use in the initial training of student pilots. During the initial development of the trainer over the next five years, Fairchild and competing contractors worked with the Air Force to develop the basic design of a “Next Generation Trainer.” Fairchild’s efforts included the construction and testing of a flying .62 scale model of the proposed aircraft, which it used in making its bid proposal to the Air Force. At the end of this process, on July 2, 1982, one of Fairchild’s divisions, the Fairchild Republic Company, entered into Contract Number F33657-82-C-2128 with the Air Force to design, develop, test, and produce the new •trainer aircraft, which was to be known as the T-46A, a plane designed to be a twin-engine jet aircraft with a top speed of about 400 knots. This project was generally known as “the T-46A contract.” The issue in this case is whether Fairchild is entitled to a tax credit for claimed research expenses generated under this contract.
1. Claimed Research Expenses
The following facts have been stipulated. The contract covered two phases, a full-scale development phase (FSD) and a production phase. The T-46A contract was a fixed-price incentive contract, which established a target cost and a target price. The original target cost for the FSD phase of the contract was $100,263,200, and the target price was set at $104,005,700, setting a target profit for Fair-child of $3,742,500.
With respect to the FSD phase of the contract, Fairchild’s books and records reflect that the claimed research expenses and the years in which they were claimed on Fairchild’s original tax returns are as follows:
Year Amount 1982 $ 3,963,298 1983 22,792,434 1984 42,118,036 1985 40,488,334 Total $109,362,102
On examination of Fairchild’s income tax returns for 1982 through 1985, the Internal Revenue Service disallowed approximately $19,608,497 of the claimed research expenses reproduced above for reasons unrelated the tax credit at issue in this case. But, the [841]*841remaining claimed research expenses, total-ling $89,753,605, related to the T-46A contract. The Service disallowed 55.8 percent of these remaining research expenses ($50,082,-512) on the ground that they were “funded” within the meaning of § 44F(d)(3) of the Internal Revenue Code of 1954. The $50,-082,512 of claimed research expenses that were disallowed as “funded” and the years in which these expenses were claimed to have been incurred are as follows:
Year Amount
1982 $ 2,198,837
1983 10,537,331
1984 17,101,020
1985 20,245,324
Total $50,082,512
The Service allowed the balance of Fair-child’s claimed research expenses, $39,671,-093. As a result of the disallowance of the 55.8 percent of the claimed research expenses, a correlative amount of research credits relating to the T-46A contract were also disallowed. Fairchild estimates that this amount is approximately $5.8 million.
2. Payments Received by Fairchild During the years 1982-1986, Fairchild received progress payments totalling at least $147 million in connection with the FSD and production phases of the T-46A contract. On its books, Fairchild summarized and allocated those progress payments as follows:
Full Scale Year Development Production 1982 $ 4,390,900 1983 32,894,892 1984 56,961,308 1985 5,965,708 $16,160,963 1986 13,631,689 17,954,938 Total 113,844,497 $34,115,901
During the period from April 27, 1984 through March 12, 1987, Fairchild invoiced the Air Force for work completed with respect to specific line items in the FSD and production phases of the T-46A contract. On March 13, 1987, the parties agreed to terminate the contract under an Agreement in Principle. Upon termination, the Air Force did not seek to recover any progress payments it made to Fairchild under the contract. After this Agreement had been entered into, during the remainder of 1987 and in early 1988 Fairchild invoiced the Air Force for additional payments that the Air Force had agreed to make in connection with the termination of the contract. These additional invoices show that Fairchild’s progress payments reproduced in the table above were liquidated in the following amounts in the following years:
Full Scale Year Invoice No. Development Production Termination 1984 484R1298 $17,379,222 1984 1184R1070 3,276,735 1985 985RCR11 759,918 1985 985R1028 14,502,049 1986 786R1592 13,758,859 1987 187R1375 $2,912,383 1987 387K1109 3,325,654 1987 987R1189 $88,306,463 $53,002,437 $2,912,383 $88,306,463
In total, between 1984 and 1988 Fairchild received a total of $159 million in connection with the contract. On its books, Fairchild summarized and allocated that amount as follows:
Full Scale Development Progress payment $113,844,497 Termination payment 6,755,503 Total $120,600,000
Production Progress payment $ 34,115,901 Termination payment 284,199 Subtotal $ 34,400,000 Total Contract Payments $155,000,000 Severance Payment 4,000,000 Grand Total $159,000,000
[842]*8423. Costs Incurred by Fairchild
In relation to the FSD phase of the contract, Fairchild claims it incurred direct costs (exclusive of overhead and profit) of $216,-056,000. On its books Fairchild allocated these costs as follows:
General and Administrative
Year Direct Costs Expenses
1982 $ 6,019,472
1983 32,876,666
1984 61,262,424
1985 57,718,593
1986 35,318,131
1987 (to 3/1) (approx.) 3,460,714 $19,400,000
Total $196,656,000 $19,400,000
4. The Service’s Determinations
As the tables demonstrate, Fairchild claims that the payments it received relative to the FSD phase of the contract, $120,600,-000, were exceeded by the direct costs it incurred, $216,056,000. The ratio of these two amounts, 0.558, is the figure the Service used to compute the disallowance of the research expenses claimed by Fairchild on the ground that this portion of the expenses was “funded” by the government; ie., of the $89,753,605 Fairchild claimed as research expenses, 55.8 percent was disallowed. Accordingly, the Service denied Fairchild the research tax credit that could be generated from this amount, about $5.8 million. Although these facts have been stipulated for the purposes of these motions, the parties do not agree that the summarization and allocation on Fairchild’s books or the costs Fair-child claims it incurred, as reproduced in the tables above, are accurate. Further, the parties do not agree that the “claimed research expenses” discussed above constitute “qualified research expenses” for tax purposes even if the court determines that the research was not funded.
On its returns filed for the tax years between 1982 and 1985, Fairchild claimed research tax credits under § 44F of the Internal Revenue Code of 1954.1 Because of carryovers and carrybacks, the parties believe that the research tax credits in issue here affect only the returns Fairchild filed for the 1983 and 1984 tax years. Upon audit, the Service disallowed a portion of the credits as described above and assessed deficiencies. Fairchild paid the deficiencies, waived disal-lowance of the refund claims it filed, and filed its complaint in this court.
II.
A Summary Judgment
Before the court are cross-motions for summary judgment. That there are cross-motions for summary judgment does not mean that the court must grant judgment as a matter of law for one side or another. Summary judgment is properly granted only when there is no genuine issue of material fact. Each party’s motion must be evaluated on its own merits and care must be taken to draw all reasonable inferences against the party whose motion is under consideration. Brown v. United States, 30 Fed.Cl. 23, 26 (1993). The parties in this matter have stipulated to most of the salient facts; each, however, presses upon the court facts it feels should compel the denial of the other party’s motion. Only a genuine issue concerning a material fact will cause this court to deny summary judgment. A material fact is one that will make a difference in the result of a case, and those facts that are material are identified by reference to the substantive law that governs the case. Id. The issue in this ease involves statutory construction. Thus, the matter at issue is one of law, and may be resolved by summary disposition. See P.J. Maffei Bldg. Wrecking Corp. v. United States, 732 F.2d 913, 916 (Fed.Cir. 1984).
B. Section MF
1. Rules of Construction
In considering a statute that features terms whose definitions are not facially apparent, the court must look for any express definitions in the statute. If there are none, the court must consider whether the statute has a plain meaning. If there is neither a [843]*843plain meaning to the statute nor any express definitions, then the court must look for a clear legislative intent. Johns-Manville Corp. v. United States, 855 F.2d 1556, 1560 (Fed.Cir.1988). Ideally, the legislative intent can be evinced from the plain meaning of the statutory language itself, but if it cannot the court must turn to extrinsic aids such as legislative history, rules of statutory construction, and the construction placed on the statute by the agency that administers it. J.M. Huber Corp. v. United States, 27 Fed.Cl. 659, 663 (1993).
The parties have drawn the court’s attention to regulations promulgated by the Service interpreting § 44F. The rules of statutory construction are appropriate tools to use to interpret a regulation, but in interpreting a regulation the issue to be decided is whether the Commissioner’s interpretation of the statute as expressed by the regulation is reasonable. The regulations will be sustained unless unreasonable and plainly inconsistent with the revenue statutes. Commissioner v. Portland Cement Co., 450 U.S. 156, 169, 101 S.Ct. 1037, 1045, 67 L.Ed.2d 140 (1981). It is axiomatic that statutes that confer special tax reduction benefits are to be construed strictly. Furthermore, “determinations of the Commissioner of Internal Revenue are presumptively correct, and the Commissioner’s understanding of the statute, as expressed in properly enacted regulations, unless unreasonable, is virtually conclusive.” Lima Surgical Assocs., Inc. v. United States, 944 F.2d 885, 888 (Fed.Cir.1991).
Also, the court notes that even if the statute and regulations are construed using the correct rules of statutory interpretation, further inquiry may be necessary — the court might have to examine the language of the contract as bargained for by the parties. Pure contract interpretation is a matter of law that may be resolved by summary judgment. P.J. Maffei Bldg. Wrecking Corp. v. United States, 732 F.2d 913, 916 (Fed.Cir. 1984). The contract must be read with an eye for determining what was the nature of the agreement entered into between the parties, and there is no surer way to find out what the parties meant than to see what they have done. See Insurance Co. v. Dutcher, 95 U.S. 269, 273, 24 L.Ed. 410 (1877); Crown Laundry and Dry Cleaners, Inc. v. United States, 29 Fed.Cl. 506, 516 (1993).
2. The Applicable Language
During the tax years in issue, 1983 and 1984,1.R.C. § 44F(a) allowed a taxpayer to take as a credit against the tax imposed for the taxable year an amount equal to 25 percent of the excess (if any) of the qualified research expenses for the taxable year over the base period research expenses. In this case, the parties do not seem to dispute that Fairchild incurred the “research expenses” it claimed in its returns. What is in dispute is whether these expenses are “qualified research expenses” as defined by § 44F(d). Section 44F(d) provides:
(d) Qualified Research — For purposes of this section the term “qualified research” has the same meaning as the term research or experimental has under section 174, except that such term shall not include—
(1) qualified research conducted outside the United States,
(2) qualified research in the social sciences or the humanities, and
(3) qualified research to the extent funded by any grant, contract, or otherwise by another person (or any governmental entity).
The term “funded” is not defined in the section or elsewhere in the subchapter. The regulations for this statute also do not describe when research is “funded.” Treasury Regulation § 1.41-5(d)2 governs qualified research funded by any contract for taxable years beginning before January 1, 1986. Subparagraph (1) of this regulation states that:
[844]*844Research does not constitute qualified research to the extent it is funded by any grant, contract, or otherwise by another person (including any governmental entity). All agreements (not only research contracts) entered into between the taxpayer performing the research and other persons shall be considered in determining the extent to which the research is funded. Amounts payable under any agreement that are contingent on the success of the research and thus considered to be paid for the product or result of the research ... are not treated as funding.
Treas.Reg. § 1.41-5(d)(l).
As the court will discuss below, the parties disagree whether the payments Fairchild received were “contingent on the success of the research” as used in this regulation. The parties do seem to agree, however, that Fair-child “retains substantial rights” in the research it performed as that term is used in § 1.41 — 5(d)(3), which provides: “If a taxpayer performing research for another person retains substantial rights in the research under the agreement providing for the research, the research is funded to the extent of the payments (and fair market value of any property) to which the taxpayer becomes entitled by performing the research.”
3. Legislative History of § 44F
Section 44F was added to the 1954 Code by the Economic Recovery Tax Act of 1981, Pub.L. 97-34, § 221(a), 95 Stat. 172, 241. As part of general discussion on tax credits for research, the House Committee on Ways and Means noted some of the reasons it felt § 44F should be adopted:
The committee believes that a substantial tax credit for incremental research and experimental expenditures will overcome the resistance of many businesses to bear the significant costs of staffing, supplies, and certain computer charges which must be incurred in initiating or expanding research programs. While such costs bear characteristics of investment activity, the relationships between the investment in research and the subsequent earnings often are less directly identifiable, and many businesses are reluctant to allocate scarce investment funds for uncertain rewards.
Research and experimentation are basic activities that must precede (1) the development and application of new techniques and equipment to production and (2) the development and manufacture of new products. The committee believes that the approach in the bill, designed to stimulate a higher rate of capital formation and to increase productivity, appropriately includes incentives for greater private activity, in research.
H.R.Rep. No. 201, 97th Cong., 1st Sess. 111. See also Joint Committee on Taxation, General Explanation of the Economic Recovery Tax Act of 1981, 97th Cong., 1st Sess. 119-120 [hereinafter 1981 Bluebook].3 But, as [845]*845noted above, § 44F(d)(3) specifically excluded from the credit expenditures funded by the government. A footnote to the report expresses the Committee’s thoughts on how § 44F might affect funding by government contracts:
In the case of government contracts, a research expenditure is considered to be funded from the grant, contract, or subcontract to the extent that the cost is required by, or allocable to, a particular contract or is incurred pursuant to a negotiated advance agreement for the payment of such costs (see, e.g. 32 C.F.R. 1115-205.35(a) (1980), relating to the establishment of a cost ceiling for availability of “independent research and development costs under the Defense Acquisition Regulations). However, the government contractor’s independent research generally would not be considered to be funded from a grant, contract, or subcontract. Nevertheless, the costs of such company-sponsored research generally would be treated as being funded from a government contract to the extent that those costs are reimbursed by a grant, contract, or subcontract in accordance with applicable government contracting accounting rules.
Id. at 116 n. 8.
C. Fairchild’s Argument
In its motion for summary judgment,4 Fairchild argues that the regulations establish a “two-prong test” that determines whether or not research is funded by another entity, namely, research is not funded by the other entity if (1) payment for the taxpayer’s work is contingent on the success of the research and (2) the taxpayer retains substantial rights in the research. In the context of Fairchild’s proposed two-prong test, the government appears to concede that Fairchild retained substantial rights in the research as that term is defined by Regulation § 1.41-5(d)(2) because the T-46A contract neither assigned the Air Force exclusive rights in the research nor denied Fair-child patent, copyright, and licensing rights with respect to inventions and technology resulting from the research. Accordingly, this “prong” does not need to be discussed any further.
Fairchild makes much of the language in § 1.41-5(d)(l) that states that “[ajmounts payable under any agreement that are contingent on the success of the research and thus considered to be paid for the product or result of the research are not treated as funding.” In its briefs, Fairchild notes that the first prong of the test it advances, the “contingent payment” test, limits the credit to research for which the taxpayer will bear the economic loss if the research is unsuccessful. Fairchild argues that the government was not contractually obligated to pay Fairchild any money unless Fairchild produced a useful product. In support of this argument, Fairchild urges the court to consider the demands placed on it by the T-46A contract.
The FSD phase target price was approximately $104,000,000, which was payable with respect to eleven Contract Line Items which Fairchild says required it to design, develop, test, and deliver within strict time frames two prototypes of the Next Generation Trainer (NGT), as well as all related support systems and documentation relating to the aircraft. In § H.16 of the contract, Fairchild agreed to bear “Total System Responsibility” for the project, which included responsibility for “the installation and integration of the NGT system elements ... including the re[846]*846sponsibility for undertaking any and all actions necessary to assure that the total system will meet all requirements as defined by the System Specification.” Section J of the contract incorporates by reference over one thousand pages of performance specifications and technical requirements that required Fairchild to meet extensive performance requirements. Section J also mandated what Fairchild describes as “[e]xhaustive inspection and testing procedures.”
By drawing the court’s attention to the exhaustive specifications and performance requirements that inhere in many defense contracts, Fairchild suggests the T-46A contract obligated Fairchild to engage in difficult, time-consuming aircraft development for which there were no guarantees of payment by the government. In further support of this assertion, Fairchild states the contract made any payment expressly contingent upon acceptance by the Air Force. Consider Defense Acquisition Regulation (DAR) 7-302.2:
The contractor shall be paid, upon submission of proper invoices or vouchers, the prices stipulated herein for work delivered or rendered and accepted, less deductions, if any, as herein provided. Unless otherwise provided, payment will be made upon acceptance of any portion of the work delivered or rendered for which a price is separately stated in the contract.
The T-46A contract also included the standard termination for default and termination for convenience of the government clauses. Fairchild notes that under DAR 7-302, amounts payable under a fixed-price contract that are terminated for convenience are subject to a downward adjustment if the contract is being performed at a loss at the time of termination, and thus even if the contract were terminated for convenience Fairchild’s payment rights would have depended on whether it was successfully performing the contract at the time of the termination. Fairchild also points to provisions such as DAR 7-302.4, incorporated into the contract, to show that the Air Force could reject any work it considered to be defective or otherwise not in conformity with the contract.
As a final argument for its “contingent payment” test, Fairchild states that the provisions in the T-46A contract entitling Fair-child to periodic progress payments demonstrate clearly that Fairchild would only be paid for its work in the FSD phase if the Air Force was finally and fully satisfied with the work performed. Progress payments, Fair-child says, are essentially a form of financing in which the payments made to the contractor finance the work the contractor is undertaking, although those payments are subject to repayment by the contractor to the government. As the contractor’s work is completed and accepted by the government, asserts Fairchild, the contractor's repayment obligations are “liquidated,” treated as offsets or deductions from final payments due to the contractor for that work. If a contract is terminated for convenience, “unliquidated” progress payments are treated as excess amounts repayable to the government on demand, or as offsets to amounts otherwise payable to the contractor under a settlement or determination. See DAR 7-103.21(b) and DAR 7-104.35(a), incorporated into the T-46A contract.
Referring to these provisions, Fairchild argues that it “had no right to retain any of its progress payments until they were treated as a deduction from a final payment due on work accepted by the Air Force,” and thus its progress payments were “contingent on the success of the research.” The Air Force, however, never sought return of any progress payments made to Fairchild. The payment provisions (such as the “progress payments” clause and the “inspection and acceptance” clause) which Fairchild uses to tie the contract to the regulation are essentially contract management and administrative controls. When a contractor submits an invoice for payment, these payment provisions ensure that false invoices shall not preclude the government from getting improper payments back from the contractor. Fairchild also draws the court’s attention to § H.37, which required Fairchild to satisfy various “demonstration milestones” showing the Air Force that the research and development efforts called for by the contract were progressing successfully. For example, one demonstration milestone appears to have been “First [847]*847Flight — 15 Apr 85,” which provides: “Description: This milestone is accomplished when the T-l aircraft completes its first scheduled flight in accordance with the test plan. Success Criteria: Government on-site verification.” Fairchild states that failure to meet these milestones would cause the Air Force to reduce or suspend its progress payments, forcing it to continue work without the benefit of government financing and demonstrating again that any money Fair-child received from the Air Force was “contingent” on delivering the “product or result of the research.”
D. The Government’s Argument
In its opposition to Fairchild’s motion for summary judgment and in support of its own motion, the government argues that § 44F and its regulations and legislative history allow a taxpayer to be entitled to this research credit only if the taxpayer bears the risk and the cost of undertaking the new research. The government directs the court’s attention to language in the congressional reports noting that the need for an incentive credit such as this is reflected by the fact that there are instances when the connection between investment research and subsequent earnings are not “directly identifiable,” and that a contract that explicitly provides for payment of research expenses evinces such a directly identifiable connection. In other words, Congress intended to provide the research credit to subsidize a taxpayer actually paying for the research, and intended to deny the credit to a taxpayer performing research at another’s expense.
In its submissions, the government draws distinctions between those cases in which a taxpayer conducts research for another but only will be paid if certain specified and difficult criteria can be met, and those cases in which it is all but certain that a taxpayer will be paid for the cost of its research. Of course, by drawing such distinctions the government is arguing that the statute generates the question: “who is actually bearing the cost of the research?” The government suggests that this question can be correctly answered only by examining “all agreements” between the taxpayer and the sponsor of the research (citing the second sentence of § 1.41-5(d)(l)). Fairchild, on the other hand, argues that the question should be answered by examining only the “four corners of the contract” (relying on the third sentence of this regulation).
Not content to state that Fairchild’s “contingent payment” test is the improper way to answer this fundamental question, the government goes further. It suggests that the proper test for whether contract research is eligible for the research credit is whether “payment is expected and likely to be made in the normal course of events.”
In addition, the government urges that Fairchild’s “contingent payment” test focuses inappropriately narrowly on regulation § 1.41-5(d)(l) at the expense of the statute and the legislative history of the statute. It says the test would severely restrict the meaning of funded research, and “as most contracts do not provide funds ‘carte blanche’ to the party conducting the research, nearly any such party may point to some condition with respect to its research that may affect its right to payment.” The government responds further to Fairchild’s interpretation of this regulation by suggesting that the regulation should be read narrowly (and in a way that supports its own “test”): payments are “contingent on the success” of the research only where payment conditions are sufficient that it is not expected and likely that payment will be made in the normal course of business. The government claims that Fairchild satisfies the standards of its test by noting that it was highly unlikely that Fairchild would not be paid for the research it performed pursuant to the contract, that in the event of termination for convenience the government would be obligated to pay the contractor for work completed, and that Fairchild was in fact paid $120.6 million for the FSD phase of the contract, $16 million more than the original target price.
III.
Initially, the court must examine the plain meaning of § 44F(d)(3). As read in the context of all of § 44F, the import of this sub-paragraph is that the tax credit should be awarded to whomever furnishes the money [848]*848that allows the research to be conducted. The tax credit can be awarded to a sponsoring party regardless of whether it furnishes the money by grant or contract or in any other way, but the credit will be awarded to this sponsor only to the extent that the sponsor was itself responsible for furnishing the means for the research to be performed. This is reasonable way to interpret this section, but this interpretation is not compelled by the language itself. As used in this section, the phrase “to the extent funded” does not carry with it a plain meaning; accordingly, the court must determine whether there was a clear legislative intent in support of this view.
As the court mentioned earlier, the various House, Senate, and Committee reports that preceded the enactment of this provision into the ERTA do not differ substantially. Each report states that the credit was given to stimulate research that wouldn’t have been undertaken otherwise. The 1981 Bluebook suggests that companies were reluctant to bear the costs that must be incurred to initiate or expand research programs because they did not want to allocate scarce investment funds for uncertain rewards. As a result, Congress wrote a law that sought to encourage companies to take risks and put some of their money into more risky research. If a company takes some of its money out of the bank and spends it on a new research initiative, it would seem to be entitled to the research tax credit. If the money comes from another company’s bank account, however, then this other company is the party allocating money for an uncertain reward, and it should be allowed to claim the credit. If the research is sponsored by someone other than the researcher, then to the extent the sponsor allocates its scarce investment funds it should be entitled to the credit.
The legislative history of this section suggests to whom a research credit should be allocated, and the Service’s interpretation of this section should be examined to see if it comports with the plain language of the statute, the plain meaning of the statute, and the intent of Congress in enacting the statute. Any regulation must be read as a whole and should not be read as to be internally inconsistent or effect an absurd result, ie., a result that runs counter to the plain meaning and legislative intent. As the dispute in this matter centers on the proper interpretation of § 1.41-5(d)(l), the sentences of the regulation should be parsed individually.
“Research does not constitute qualified research to the extent it is funded by any grant, contract, or otherwise by another person (including any governmental entity).”— This sentence states that a taxpayer may not take the credit if the government funds the research. Taken alone, this sentence states that the taxpayer/researcher cannot take the credit if the government is paying for the research. Generally, “to fund” is “to furnish a sum of money to be set aside for a particular purpose.” This interpretation comports with the plain meaning and legislative intent of the statute.
“All agreements (not only research contracts) entered into between the taxpayer performing the research and other persons shall be considered in determining the extent to which the research is funded.” — This sentence follows naturally from the sentence before it. It states that to determine the extent to which the government funded the research, the full scope of the relationship between the researcher and the sponsor must be considered.
“Amounts payable under any agreement that are contingent on the success of the research and thus considered to be paid for the product or result of the research ... are not treated as funding.” — That is to say, an amount payable that is considered to be paid for research isn’t funding. Money payable to the taxpayer isn’t considered funding if the parties agree that the money will only be paid if the research is successful or the product is successfully made, and it’s not funding if the costs the researcher incurred will be covered by the sponsor only if “the research is successful.”
This reading of the third sentence fits in with the two previous sentences of the regulation. If the sponsor pays the researcher over the course of the research, with payments made to the researcher before the “product” is finalized, then the sponsor has [849]*849funded the research. The sponsor assumed the risk that the money it allocated to the researcher would eventually bear fruit. But, if the sponsor will allocate research money only if the research is successful, then the research and development costs must be paid for by the researcher up front — the researcher is allocating its own funds and is assuming the risk that the sponsor will not pay for the finished product because the product is found to be unacceptable. In such an instance, the research would not be “funded” research because the researcher was betting its own money and therefore should be entitled to the credit.5
Fairchild argues that it assumed the risk under the contract and would be paid “contingent on the success” of its work. The facts state otherwise. The contract provided for bimonthly progress payments for Fair-child’s work during the FSD phase of the contract. Of course, Fairchild had numerable specifications to meet, and ran the risk that the Air Force might be dissatisfied with the results, but the periodic payments were paid. The court is not confined to the four corners of the contract — it must look to the relationship between the parties and see what they actually did as contract performance was on-going. See Insurance Co. v. Dutcher, 95 U.S. at 273. A target cost and target profit were set, and it doesn’t appear that the parties said that the government would only pay Fairchild the full target sum if a useful and satisfactory product was produced. A number of progress payments were made for the FSD phase before the project was cancelled by legislation, payments totalling $120 million.
It is apparent that this was also the determination of the Service. Fairchild’s total FSD costs were $216,056,000, of which $120,-600,000 were remitted to Fairchild by the Air force in the form of progress payments and payments made pursuant to the Agreement in Principle to end the contract in 1987. Thus, the Service found that $120,600,000 of the $216,056,000 was funded, and computed the credit accordingly. Fairchild’s independent research expenses, found to be not funded by the government ($39,671,093), were allowed by the Service to be used to compute a tax credit under § 44F. The court finds that the Service’s determinations reflect the most reasonable interpretation of its own regulations, an interpretation which harmonizes with the statutory language and intent of Congress. These determinations are presumptively correct, and because the Service’s understanding of the statute is reasonable, its determination is virtually conclusive. See Lima Surgical Assocs., Inc. v. United States, 944 F.2d at 888.
Fairchild has urged the court to adopt its two-prong “contingent payment” test, which Fairchild says will allocate the research credit to the taxpayer who will bear the economic loss if the research is unsuccessful. This test does appear to comport with the legislative intent; nevertheless, the court declines to approve of it for two basic reasons. First, a court must be careful not to approve of new “tests” where to do so would be inappropri[850]*850ate. After a new test is developed, litigants (as well as courts) might be tempted to squeeze the facts of a case into the new and tidy box, and as a result the clearest appreciation of all the facts might not be had. A test derived from agency regulations such as those featured in this case is unnecessary where the regulations themselves provide standards for determining what legal effect is to be given the salient facts.6 In the regulation at issue in this case, the Service has promulgated its method of determining when a government sponsor “funds” a contract. No new tests need to be announced here, especially when the new test does little more than restate the rule, emphasizing particular words at the expense of others. In addition, the regulation in this case suggests by its own terms that a new test would be inappropriate. Tests have a way of forcing to the forefront selected facts, and in this regulation, the Service has announced that “all agreements” shall be considered — the credit determination must be made on a case by case basis.
Second, even if the court were to approve of its test, it does not agree that Fairchild would have met its requirements. Fair-child’s right to payment was not “contingent” on success as that term is used in the regulation. This conclusion can be drawn from the contract and how the parties actually conducted their transactions during the FSD phase of the contract. True, Fairchild had to meet numerous technical specifications and the Air Force could reject that work it did not find to be in compliance. Fairchild did, however, receive payments under the contract to fund its research; at bottom it was spending the government’s money to conduct the T-46A research, not its own. Its right to full payment depended on the full success of its work, but the contract called for payments, targeted specifically for FSD, from the government to the contractor even if the government were to find that some equitable adjustment in target cost was necessary.
The court thus rejects Fairchild’s test, but it does not follow that the government’s own proposed test should be approved. The government asks the court to find that research is “funded” if “payment is expected and likely in the normal course of events.” The court finds no indication in the statute or the legislative history that this language should be used to determine when research is funded. The reason is clear: the government’s test is essentially tautological. It states nothing more than research is “funded” when it is, well, “funded.” For the reasons mentioned above why proposals for new tests should be scrutinized, and because the government’s proposition contributes nothing not already expressed in the regulations, the court rejects it.
The holding of the court and the Service’s position in this case is simple and direct. The same cannot be said of the parties’ positions. The amount of the research expenses at issue is $89,753,605. Fairchild is entitled to and has been allowed a tax research credit figured from the $39,671,093 that represents the portion of the $89,753,605 that was not funded for the T-46A contract by the government. Fairchild is not entitled to recover a tax credit figured from the remaining $50,-082,512 because that amount was funded by the government for Fairchild’s benefit.7 [851]*851This result constitutes a reasonable and rational construction and interpretation of § 44F(d) and its attendant regulations.
IV.
Accordingly, for the reasons given in this opinion, plaintiffs motion for summary judgment is denied and defendant’s motion for summary judgment is granted The Clerk of the Court is directed to enter judgment dismissing the complaint. No costs.