Opinion for the Court by Circuit Judge EDWARDS.
HARRY T. EDWARDS, Circuit Judge:Under the Nuclear Waste Policy Act of 1982 (“the Waste Act”),1 the Secretary of Energy is directed to prescribe fees for the disposal of spent nuclear fuel (“SNF”). At issue in this case is a Department of Energy (“DOE”) rule setting forth the basis upon which the agency will compute a onetime fee for fuel spent to generate electricity prior to April 7, 1983.
Section 302(a)(3) of the Waste Act prescribes a “1 time fee per kilogram of heavy metal in spent nuclear fuel ... in an amount equivalent to an average charge of 1.0 mil per kilowatt-hour for electricity generated by such spent nuclear fuel.” The Secretary interpreted this language to permit the adoption of a fee schedule which takes into account the variation in the efficiency of the electric fuel burnup. Accordingly, under DOE’s rule, the one-time “fee is designed to recover, on an industry-wide basis, a fee per kilogram of heavy metal which will bring in an average one mill [sic] for each kilowatt hour generated by the fuel discharged from reactors prior to April 7, 1983.” DOE’s Brief at 31-32. Under this approach, “the relatively increased cost of disposal of the spent fuel per unit of electricity generated from low burnup fuel is taken into account.” Id. at 6.2
In this case, General Electric Uranium Management Corporation (“GEUMCO”) claims that the one-time fee established by DOE is contrary to the fee prescribed by Congress in the Waste Act. In particular, GEUMCO argues that Congress intended the one-time fee “to bear a direct relationship to the amount of electricity generated, regardless of any discrepancies in fuel efficiency or in volume.” GEUMCO Brief at 22-23. Thus, GEUMCO complains about the “unjustifiable inequities” produced by the DOE formula, and asserts that it “must pay approximately three mills per kilowatt-hour simply because its spent fuel generated less electricity per volume than fuel with a higher burnup.” Id. at 23.
On October 16, 1983, GEUMCO filed for judicial review in the District Court. On December 13, 1983, DOE moved to dismiss on two grounds: (1) that the District Court lacked subject matter jurisdiction because the Waste Act vests exclusive jurisdiction over review of agency action in the courts of appeals, and (2) that the Waste Act permits the one-time fee rule adopted by DOE. GEUMCO moved for summary judgment. The District Court rejected DOE’s subject matter jurisdiction arguments but granted DOE’s motion to dismiss for failure to state a claim upon which relief could be granted, finding DOE’s rule to be consistent with the language, legislative history and purpose of the Waste Act.3
*265GEUMCO then petitioned this court for direct review of DOE’s rule and also appealed from the District Court’s dismissal. By order of May 29, 1984, we consolidated these cases. Since we hold that we have original and exclusive jurisdiction over the subject matter of this case, and therefore that the District Court did not have proper jurisdiction to decide the case, we vacate the District Court’s opinion.
On direct review, we affirm DOE’s rule as a reasonable exercise of its discretion under the Waste Act. The statute clearly delegates to the DOE the authority to prescribe a one-time fee rule and does not specify the precise method of computing that fee. Furthermore, the legislative history is at best ambiguous on this question. Since DOE has acted within an area of its expertise, reached a result that is not contrary to any expressed intent of Congress, and adopted an approach that is a reasonable compromise between conflicting policies, we defer to the agency. This result, we believe, comports with the principles enunciated by the Supreme Court in Chevron.4
I. Background
Through the Waste Act, Congress conceived a comprehensive scheme for the disposal of SNF and other high-level radioactive waste generated by civilian nuclear power reactors. That scheme included the establishment of waste repository facilities under federal auspices to be fully funded by those generating the nuclear wastes. Section 302 of the Waste Act, which provides for the establishment of a Nuclear Waste Fund, authorizes the Secretary to
enter into contracts with any person who generates or holds title to high-level radioactive waste, or spent nuclear fuel, of domestic origin for the acceptance of title, subsequent transportation, and disposal of such waste or spent fuel. Such contracts shall provide for payment to the Secretary of fees ... sufficient to offset expenditures described in subsection (d) of this section [to identify, acquire, develop, license, construct, operate, maintain and monitor nuclear waste storage facilities and to pay the administrative costs of the disposal program].5
Congress expressly contemplated that utilities would be charged two different types of fees for DOE’s disposal services: an ongoing fee for SNF used to generate electricity on and after April 7, 1983, and a one-time fee for SNF used to generate electricity before that date. Under the statute, Congress prescribed a relatively straight-forward cost-rule for the ongoing fee.6 However, the task of promulgating the one-time fee was delegated to the Secretary:
For spent nuclear fuel ... used to generate electricity in a civilian nuclear power reactor prior to the application of the [ongoing] fee ... to such reactor, the Secretary shall, not later than 90 days after January 7, 1983, establish a 1 time fee per kilogram of heavy metal in spent nuclear fuel, or in solidified high-level radioactive waste. Such fee shall be in an amount equivalent to an average charge of 1.0 mil per kilowatt-hour for electricity generated by such spent nuclear fuel, or such solidified high-level waste derived therefrom____7
DOE commenced rulemaking proceedings by publishing a notice on February 4, 1983,8 soliciting comments on a proposed standard form contract to be used for its SNF disposal services. Article VIII.A.2 of the proposed contract specified that the *266one-time fee would “be assessed by applying average dollar per kilogram charges to three (3) distinct ranges of fuel burnup reflecting actual disposal costs for those ranges so that the integrated cost across all discharged fuel is equivalent to 1.0 mill [sic] per kilowatt-hour.”9
DOE received over eighty written comments on its proposed rule, many of which addressed the one-time fee rule. GEUMCO submitted comments on the proposed rule and argued then, as here, that the only lawful method of assessing charges was the first alternative method to that in the proposed contract, which would compute the one-time fee at the rate of one mil per kilowatt-hour of electricity generated by each individual fuel assembly. GEUMCO contended that
[w]hile the [statutory] language for the one-time fee is phrased differently than that for the ongoing fee, it appears to be merely a recognition of the different stages the fuel is at rather than an attempt to have one category of fuel pay fees on a radically different basis than another category. After April 7, 1983, the fee will be collected on a periodic basis as electricity is generated. The fee is related to an ongoing stream of electricity that can be read at the meter. Electricity generation is not ongoing for the fuel prior to April 7, however, and Congress apparently perceived a need to allocate the fee to the physical commodity that remained so as to achieve an equivalent result.10
GEUMCO presented similar testimony at the public hearing that the DOE held on March 3, 1983.
DOE published its final rule on April 18, 1983, essentially adopting the contract set forth earlier in its notice of proposed rule-making. The rule prescribed three different sets of fees: a one-time fee for SNF discharged before April 7, 1983; an in-core fee for fuel which was in the core of an operating nuclear reactor on April 7, 1983; and an ongoing fee for SNF generated on or after April 7, 1983, set, as prescribed by the Waste Act, in the amount of 1.0 mil per kilowatt-hour on electricity generated by a purchaser’s nuclear power reactors. The one-time fee, which is the only fee at issue in this case, was described as follows:
For SNF, or solidified high-level radioactive waste derived from SNF, which fuel was used to generate electricity in a civilian nuclear power reactor prior to April 7, 1983, a one-time fee will be assessed by applying industry-wide average dollar per kilogram charges to four (4) distinct ranges of fuel burnup so that the integrated cost across all discharged (i.e. spent) fuel is equivalent to an industry-wide average charge of 1.0 mill per kilowatt-hour. For purposes of this contract, discharged nuclear fuel is that fuel removed from the reactor core with no plans for reinsertion. In the event that any such fuel withdrawn with plans for reinsertion is not reinserted, then the applicable fee for such fuel shall be calculated as set forth in this paragraph 2. The categories of spent nuclear fuel burnup and the fee schedule are listed below:
*267(In 1982 dollars)
Nuclear spent Dollars
fuel burnup per
_range_:_kilogram
0 to 5,000 MWDT/MTU $ 80.00
5.000 to 10,000 MWDT/MTU 142.00
10.000 to 20,000 MWDT/MTU 162.00
OVER 20.000 MWDT/MTU_184.0011
In explaining its reasons for adopting the burnup range approach, DOE noted that over 80 percent of those commenting on the one-time fee calculation favored this type of approach, including several utilities,
because it avoids extremely low and extremely high fees. It sustains the principle of equity and fairness by requiring all generators and owners of the spent nuclear fuel to bear a reasonable share of disposal costs. Further it corresponds more closely to program costs.12
Although originally proposed as a three burnup range, a four-tier approach was finally adopted as “more appropriate and equitable as it is more representative of the actual industrywide distribution of spent nuclear fuel encompassing low, average and high burnup fuel.” 13
DOE explained that it did not adopt the first suggested alternative approach, the individual assembly approach based on a fee of one mil per kilowatt-hour of electricity generated, because that fee rule would unduly penalize those possessing high burnup spent fuel while benefitting those with low burnup fuel.14 The second alternative formula approach, which was also based on an individual assembly basis, and garnered the support of many commenters who saw it as a straightforward approach, was rejected by DOE because it “weakens the fairness principle by unduly benefiting owners of low-burnup fuel while penalizing those with high burnup fuel [and] ... conflicts with the [congressional intent to have those responsible for generating nuclear waste and spent fuel bear a reasonable share of disposal costs.” 15
. On June 30, 1983, DOE entered into a standard contract with GEUMCO, including the one-time fee provision as it was articulated in the rule.16 At the same time, DOE and GEUMCO entered into a collateral agreement to preserve GEUMCO’s right to seek judicial review of the one-time fee.17
II. Discussion
A. Subject Matter Jurisdiction
As an initial matter, this case raises a jurisdictional question. The Department, relying on sections 119(a)(1)(A) and 111(b)(4) of the Waste Act, contends that the District Court lacked subject matter jurisdiction to review the one-time fee rule. Section 119(a)(1)(A) of the Waste Act provides that
(1) [e]xcept for review in the Supreme Court of the United States, the United States courts of appeals shall have exclusive and original jurisdiction over any civil action—
(A) for review of any final decision or action of the Secretary, the President, or the Commission under this part.18
This review provision is located in Part A, “Repositories for Disposal of High-Level Radioactive Waste and Spent Nuclear Fuel,” of Subchapter I, “Disposal and Storage of High-Level Radioactive Waste, Spent Nuclear Fuel and Low-Level Radioactive Waste.” Section 111(b)(4), also contained within Part A of Subchapter I, describes the purpose of the Waste Act as being
to establish a Nuclear Waste Fund, composed of payments made by the genera*268tors and owners of such waste and spent fuel, that will ensure that the costs of carrying out activities relating to the disposal of such waste and spent fuel will be borne by the persons responsible for generating such waste and spent fuel.19
DOE reasons that, because the goal of setting up the Nuclear Waste Fund is found in Part A, which vests the courts of appeals with original and exclusive jurisdiction over matters arising under that part, exclusive jurisdiction includes all matters associated with the creation of the fund.20
The District Court rejected DOE's jurisdictional argument, noting that section 302(a)(3) of the Waste Act, the provision addressing the one-time fee rule, does not fall within the same part as section 119, the judicial review provision. The District Court also was not persuaded by the argument that section 11(b)(4), which defines the establishment of the Nuclear Waste Fund as one of the Waste Act’s purposes, brings this litigation within the purview of the judicial review prescription of section 119(a)(1)(A). On this latter point, the trial court was particularly influenced by the fact that GEUMCO’s challenge to the onetime fee rule rested solely on section 302(a)(3).21
We conclude that DOE’s rule setting the one-time fee to be charged under section 302(a)(3) of the Waste Act is well within the class of agency actions reviewable under section 119(a)(1)(A), and that the reference to “under this part” does not remove this case from the exclusive jurisdiction of the court of appeals. First, despite the statute’s lack of clarity, we find every indication from those sections of the statute relied on by DOE, as well as from related sections, and from the structure of the statute as a whole, that Congress intended that the court of appeals would have original and exclusive jurisdiction in cases of this sort. Second, the legislative history, although not illuminating on this question, certainly does not compel a result contrary to the one we reach here. Third, the policy considerations underlying this case, and articulated in other judicial decisions covering analogous statutes, overwhelmingly support our conclusions.
The only statutory review provisions that exist in the Waste Act vest the courts of appeals with original and exclusive jurisdiction. There are two such provisions, section 119(a)(1)(A), which provides for our original and exclusive jurisdiction over review of any final decision or action “under this part,”22 and section 221, which is contained in Subchapter II, “Research, Development, and Demonstration Regarding Disposal of High-Level Radioactive Waste and Spent Nuclear Fuel,” and which refers back to section 119(a)(1)(A) by providing that
[¡judicial review of research and development activities under this subchapter shall be in accordance with the provisions of section [119] of this title.23
Section 302, which provides for the contracts to be made between private parties and the Government to fund the Government’s waste disposal activities and responsibilities under the Waste Act, including the one-time fee provision at issue in the instant case, is part of Subchapter III, “Other Provisions Relating to Radioactive Waste,” which has no specific review provision. Thus, the only two review provisions in the Waste Act, in Subchapters I and II, vest original and exclusive jurisdiction in the courts of appeals. We find it inconceivable that Congress intended to have review of all actions concerning waste disposal in the court of appeals — including the choice, *269characterization, approval of, and authorization for construction of candidate sites;24 federal agency actions such as coordination and environmental review;25 research and development questions relating to the disposal of high-level wastes and SNF,26 including funding and payments to the states and Indian tribes under the Act27 — except for questions concerning the composition of the Nuclear Waste Fund and a few other matters located in Subchapter III.
The structure of the Waste Act supports our view that it would be singularly incongruous for fee questions to be subject to judicial review in both the district courts and the courts of appeals, while all other matters would go directly to the courts of appeals. The Waste Act has a limited and integral purpose and structure. It focuses only on the disposal of nuclear wastes. Subchapter I addresses permanent repositories for the disposal of high-level wastes and SNF in Part A, an interim storage program in Part B, monitored retrievable storage in Part C, and low-level waste disposal in Part D. Subchapter II provides for research, development and demonstration, and Subchapter III is a catch-all for other necessary provisions. In Part A of Subchapter I, Congress expressed its goal to set up a Nuclear Waste Fund,28 and its intention that generators and owners of nuclear wastes and SNF should bear the costs of disposal.29 The mere fact that Congress granted the Secretary the authority to enter into contracts to provide for the payment of those disposal costs in a different subchapter does not compel the conclusion that Congress thereby intended that there would be entirely different judicial review procedures for those contracts.30
Finally, as we look at the statute as a whole, we note that section 302(a)(1) gives the Secretary the authority to enter into contracts to provide the fees for disposal “[i]n the performance of his functions under this chapter____”31 In this context, the word “chapter” refers to Chapter 108 of Title 42 which, of course, encompasses the entire Waste Act. One of the important “functions” of the Secretary under Chapter 108 is to set up the Nuclear Waste Fund as described in Part A of Subchapter I, wherein we find the provision according original and exclusive jurisdiction in the courts of appeals. Therefore, we find that section 302 incorporates by reference the functional responsibilities of the Secretary under Part A of Subchapter I. By the same token, the purpose to establish the Waste Fund, as explained above, is found in Part A of Subchapter I. This purpose incorporates by reference the contractual provisions of section 302. It thus appears that, from whatever vantage point we approach the statutory construction problem that we address here, we inevitably conclude that Congress intended to vest original and exclusive jurisdiction with the courts of appeals.32
*270The legislative history does not expressly aid us in our construction here. However, there is nothing in that history that compels any different conclusion. Indeed, the evolution of the placement of section 302 in the Waste Act strongly suggests that its physical separation from the judicial review provision in section 119 is pure happenstance and in no way indicates a congressional intent that review under the different subchapters be governed by different standards.33 There simply is no explanation for the separation in the legislative history.
Finally, we believe that there are certain policy considerations that militate in favor of the judgment that we reach on the jurisdictional question. While we do not rely on factors of policy in lieu of statutory construction, we do consider them relevant in a case such as this where the pertinent statutory provisions admit of ambiguity.
First, it frequently has been noted that, in administrative appeals, “where it is unclear whether review jurisdiction is in the district court or the court of appeals the ambiguity is resolved in favor of the latter.”34 A principal reason for this guideline is that exclusive jurisdiction in the court of appeals avoids duplicative review and the attendant delay and expense involved.35 While it is true that the delay associated with district court review already has occurred in this case, this in no way diminishes the legitimacy of the policy favoring exclusive jurisdiction in the court of appeals.
Second, there is no need for an evidentiary hearing in order for judicial review to occur in this case. The DOE rulemaking involved public notice, written comments from eighty-five sources, a hearing and a transcript of the proceedings. This information was fully considered by the Department precedent to the adoption of a final rule.36 Since there are no factual findings to be made, the court of appeals is fully able to consider this administrative record, without input from the district court, in reviewing the merits of GEUMCO’s claim.37 Indeed, even without an extensive record or hearing before the agency, it is rare that the factfinding capacity of the district court is required in judicial review of administrative agency actions.38
*271Third, original and exclusive jurisdiction in the courts of appeals promotes the congressional goals of efficiency and predictability. Although review of the one-time fee for disposal of spent fuel may not require further judicial consideration beyond the disposition of this case, it is but one of many like questions that may arise under the Waste Act. We find it hard to believe that Congress intended to have any of the basic issues surrounding disposal of SNF to be tied up in duplicative litigation for years on end. Nuclear reactors must be refuelled periodically at which time the SNF must be removed and stored. There is already a tremendous accumulation of SNF in short-term storage facilities with limited capacity. Consequently, nuclear facilities may well have to be shut down in the near future, and the matter of establishing safe and adequate disposal facilities is an acknowledged global problem of dramatic urgency.39 In this context, we cannot assume that Congress intended both the District Court and this court to perform the identical review function of the onetime fee rule based on an identical administrative record.40
For the foregoing reasons, we hold that the court of appeals has original and exclusive jurisdiction to review DOE’s one-time fee rule and that the District Court lacked subject matter jurisdiction to consider the instant case. Therefore, we vacate the opinion and judgment of the District Court and proceed to review DOE’s determination directly. For the reasons set forth below, we deny the petition for review.
B. The One-Time Fee Rule
GEUMCO avers that DOE’s interpretation of the Waste Act violates the intent of Congress, as expressed both in the statutory language and its legislative history, to recover fees based on the amount of electricity generated by the SNF rather than by the volume of SNF deposit*272ed in the federal waste repository. We disagree with these contentions, and find that DOE’s interpretation is a reasonable one which should be upheld. We find the statute and the legislative history to be, at best, ambiguous concerning the one-time fee and, given this lack of clarity, we are guided by recent Supreme Court precedent to defer to the agency’s implementation of the statute.
In Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc.,41 the Supreme Court held that “if the statute is silent or ambiguous with respect to [a] specific issue [within the general compass of the statute], the question for the court is whether the agency’s answer is based on a permissible construction of the statute.”42 Recently, we have set forth certain guidelines for adhering to the mandate of Chevron:
In determining, whether an agency’s interpretation represents a reasonable accommodation of conflicting statutory purposes, a reviewing court must determine both whether the interpretation is arguably consistent with the underlying statutory scheme in a substantive sense and whether “the agency considered the matter in a detailed and reasoned fashion.” 43
Applying these principles to the present case, we find that the statute is ambiguous with respect to the specific method of calculating the one-time fee, and that DOE’s rule “represents a reasonable accommodation of conflicting policies” served by the Act.44 Furthermore, we find that this is not a case where “it appears from the statute or its legislative history that the accommodation is not one that Congress would have sanctioned.”45 Finally, we are assured, on the record of this case, that DOE fully and fairly considered the matter in issue “in a detailed and reasoned fashion.”
We find, first, that DOE is indubitably entrusted with the administration of the Waste Act. We find, second, that nothing in the language of the statute explicitly states that the one-time fee must be based upon the amount of electricity generated rather than upon the volume of SNF in the repository. Section 302(a)(3) provides in relevant part that, as concerns the one-time fee,
the Secretary shall ... establish a 1 time fee per kilogram of heavy metal in spent nuclear fuel____ Such fee shall be in an amount equivalent to an average charge of 1.0 mil per kilowatt-hour for electricity generated by such spent nuclear fuel, or such solidified high-level waste derived therefrom, to be collected from any person delivering such spent nuclear fuel ... to the Federal Government.46
GEUMCO urges us to find that the statutory prescription in the words “in an amount equivalent to an average charge of 1.0 mil per kilowatt-hour” compels the Secretary to set the fee at a rate of one mil per kilowatt-hour for electricity generated. However, we can find no such clarity in the language because the Secretary is given the discretion to fix a fee based upon an “average charge.” Under GEUMCO’s construction of the statute, the Secretary would be denied all discretion in establishing the method of calculating the one-time fee rule. If that is what Congress intended, then the delegation of authority to the Secretary to fix a fee within the range of an “average charge” would appear superfluous. We decline to submit to such a tortured interpretation of the statute. Rather, it appears to us that Congress, in section 302(a)(3) of the Waste Act, gave DOE the authority to regulate the method *273of calculating the one-time fee for SNF, while stipulating that the method should yield the same result as “an amount equivalent to an average charge of 1 mil per kilowatt-hour.” This does not mean that the method itself must be a charge of 1 mil per kilowatt-hour.
The language of section 302(a)(2), which sets the ongoing fee, confirms this construction. In that section, Congress mandated a fee “equal to 1.0 mil per kilowatt-hour” rather than “equivalent to an average charge of 1 mil per kilowatt-hour.” We believe that Congress well perceived the distinction in its phraseology of these parallel sections of the same statute. Congress intended that the one-time fee would average out to a fee equivalent to one mil per kilowatt-hour, not be equal to one mil per kilowatt-hour. Thus, we are bound to conclude that DOE’s four-tier approach, relating the one-time fee to the fuel bumup ratio but yielding a final amount equivalent to an industry-wide average of 1.0 mil per kilowatt-hour reasonably accords with the language of the statute and should be affirmed.47
GEUMCO devotes a considerable portion of its brief to discussing the legislative history of section 302(a)(2), the ongoing fee rule. That history, although it reflects congressional resolution of the conflicting goals and policies related to the ongoing fee, in no sense compels a certain result in this case, which concerns the one-time fee, not the ongoing fee. While we take notice of the policy accommodation reached by Congress concerning the ongoing fee, we hold that it is not determinative of congressional intention with regard to the one-time fee. Indeed, it would have made no good sense for Congress to adopt two separate and distinct statutory provisions for the ongoing and one-time fees if there was a legislative intention to prescribe precisely the same formula for both fees.
So far as we have been able to determine, the legislative history nowhere clearly indicates that Congress intended that the one-time fee be calculated according to the amount of electricity generated. The fee provisions of section 302 are largely derived from the Senate version S. 1662, which originally contained no provision at all concerning the one-time fee. Following Committee hearings during which this deficiency was identified, the Senate Committee on Environmental and Public Works added the one-time fee provision. Thus, the legislative history relevant to the onetime fee rule is mostly limited to that Committee’s hearings. Noting that earlier versions of the bill had failed to impose a cost for previously generated SNF, but had instead required that “future consumers of nuclear power would be required to bear the cost of storage and disposal for existing waste as well as for future wastes,”48 the Committee added what was later section 302(a)(3) “to cover the cost of long-term storage and disposal of nuclear waste and spent fuel generated as of the date of enactment.”49 While, the Committee intended the two fees to be “comparable” and “equivalent,” these terms do not com*274pel an identical method for computing the ongoing and one-time fees.
We believe that the legislative history of section 302(a)(3) reflects congressional intent to delegate authority to the Secretary to regulate a problem that had been largely neglected during the development of the bill. GEUMCO is correct that the legislative history of the ongoing fee reflects an intent to assess costs on the basis of electricity generated, not on volume of SNF, thus passing on to the consumer the costs of disposal. But Congress was not content to lump the one-time fee with the ongoing fee. At the eleventh hour it created a separate statutory provision delegating the problem to the Secretary. In so doing, Congress did establish the parameters — the final average one-time fee was to be “equivalent ” to the bottom-line ongoing fee. Congress, however, did not establish the method for that one-time fee, either in the statute or the legislative history.
Our review of the statute and its legislative history leaves us unconvinced that the one-time fee rule is an illegal, or an arbitrary and capricious, exercise of the Secretary’s authority under section 302(a)(3) of the Waste Act. Accordingly, under the principles of Chevron, we uphold the Secretary’s determination.
III. Conclusion
For the foregoing reasons, we hold that subject matter jurisdiction over GEUMCO’s claim lies exclusively in the courts of appeals and that the Department’s rule setting the method for calculating the onetime fee due under section 302(a)(3) is neither unlawful nor arbitrary or capricious. Accordingly, we deny GEUMCO’s petition challenging the one-time fee rule.
. 42 U.S.C. §§ 10101 et seq. (1982).
. DOE describes the "low burnup" factor as follows:
To generate electricity in nuclear plants, rods of fissionable uranium fuel are inserted into the core of the reactor. In proximity to other rods, collectively referred to as an assembly, a controlled chain reaction creates heat, which produces steam, which turns the electric turbines. Many factors may influence the decision of when to withdraw a given fuel assembly, including the type of fuel and the type of reactor. The fission process is irreversible: as the uranium atoms are split into radioactive transuranic elements, at some point the rods are eventually "spent” and are discharged from the reactor. The amount of energy a utility will extract prior to discharging an assembly may vary considerably. This variation in the efficiency of energy extraction — which is reflected in the degree of fuel burnup — is the practical source of the instant litigation.
Utilities with low burnup ratios produce, per unit of electricity (kilowatt hour), greater quantities of spent fuel. Therefore, the cost for disposal, per unit of electricity, will also be greater.
DOE brief at 3.
. General Electric Uranium Management Corp. v. DOE, 584 F.Supp. 234 (D.D.C.1984).
. Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., — U.S.-, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984).
. 42 U.S.C. § 10222(a)(1) (1982).
. The ongoing fee is set forth in section 302(a)(2), which provides that
[f]or electricity generated by a civilian nuclear power reactor and sold on or after the date 90 days after January 7, 1983, the fee ... shall be equal to 1.0 mil per kilowatt-hour.
42 U.S.C. § 10222(a)(2) (1982).
. 42 U.S.C. § 10222(a)(3) (1982).
. Notice of Proposed Rulemaking, 48 Fed.Reg. 5458 et seq. (1983).
. Id. at 5464. In the same notice, DOE also identified three other possible approaches for the one time fee. The three alternative approaches DOE suggested included (1) determining the average burnup of all fuel discharged prior to April 7, 1983, calculating the dollar per kilogram for that average burnup equivalent to one mil per kilowatt-hour, and applying that dollar per kilogram charge to all discharged fuel; (2) using a prescribed formula which takes into account fuel burnup data in order to determine a charge per kilogram of heavy metal that is equivalent to one mil per kilowatt-hour, but making the calculation on an individual assembly basis; and (3) categorizing discharged fuel with less than a certain burnup level as abnormal fuel and subjecting the cost for disposal of that fuel to the 1 mil per kilowatt-hour equivalent charge, but subjecting the residual discharged fuel to an average dollar per kilogram charge such that total fees collected are equivalent to 1 mil per kilowatt-hour. Id. at 5459-60.
. Joint Appendix ("J.A.") 163-64.
. Art. VIH, A.2. 10 C.F.R. § 961.11 (1984). "MWDT/MTU" expresses megawatt-days (thermal) divided by metric tons of uranium. Megawatt-days (thermal) measures the heat generated by the fuel; metric tons of uranium is a volume measure.
. 48 Fed.Reg. 16590, 16593 (1983).
. Id.
. id.
. See Affidavit of Burton F. Judson, J.A. 18.
. GEUMCO s SNF has low fuel burnup, and has generated low amounts of electricity per volume. The financial impact on GEUMCO of the one-time fee adopted is therefore substantial.
.42 U.S.C. § 10139(a)(1)(A) (1982).
. 42 U.S.C. § 10131(b)(4) (1982).
. GEUMCO, although contending that the District Court had jurisdiction to review DOE's rule, also filed a protective petition for review in this court. GEUMCO claimed general federal question jurisdiction in the District Court, 28 U.S.C. § 1331 (1982), and also cited the Administrative Procedure Act, 5 U.S.C. § 702 (1982).
. 584 F.Supp. at 236-37.
. 42 U.S.C. § 10139(a)(1)(A) (1982).
. 42 U.S.C. § 10201 (1982).
. 42 U.S.C. §§ 10132-35, 10193 (1982). 42 U.S.C. §§ 10132-35 are all contained in Part A of Subchapter I; 42 U.S.C. § 10193 is found in Subchapter II.
. 42 U.S.C. § 10196 (1982) (Subchapter II).
. 42 U.S.C. §§ 10197, 10198 (1982) (Subchapter II).
. 42 U.S.C. §§ 10138, 10199 (1982) (Subchapter I, Part A, and Subchapter II respectively).
. 42 U.S.C. § 10131(b)(4) (1982).
. 42 U.S.C. § 10131(a)(4) (1982).
. Furthermore, there is nothing in the legislative history to evidence such an intention on the part of Congress.
. 42 U.S.C. § 10222(a)(1) (1982) (emphasis added).
. In Wisconsin Electric Power Co. v. Hodel, No. 83-2281 (D.D.C. July 18, 1984), appeal docketed, Nos. 83-2066 & 84-5671 (D.C.Cir. Sept. 21, 1984), District Judge Oberdorfer declined to take jurisdiction under section 302(a)(2) of the Waste Act, 42 U.S.C. § 10222(a)(2), the sister provision to that at issue here, which specifies the method for the ongoing fee. We find his reasoning persuasive. See id., slip op. at 5-6 (literal reliance on "part” is misplaced here, and there are strong policy and practical reasons for exclusive jurisdiction in the courts of appeals).
. Present section 119, the judicial review provision, was originally proposed in the same section of the Act that described the contract authority to collect fees. See H.R. 3809, 97th Cong., 2d Sess. § 118 (1981), U.S.Code Cong. & Admin.News 1982, pp. 3792, 3822. The judicial review and fee contract provisions remained in close proximity through various metamorphoses of the bill, but ultimately were physically separated from each other, apparently without design. The House later separated the Nuclear Waste Fund provisions and the review provisions. See, e.g., H.R. 6598, 97th Cong., 2d Sess. (1982); H.R. 5016, 97th Cong., 2d Sess. (1981); S. 1662, 97th Cong., 1st Sess. (1981); see also H.R. 7187, 97th Cong., 2d Sess. (1982).
. Denberg v. United States R. Retirement Board, 696 F.2d 1193, 1197 (7th Cir.1983), cert. denied, — U.S. -, 104 S.Ct. 1706, 80 L.Ed.2d 180 (1984).
. See, e.g., Harrison v. PPG Indus., 446 U.S. 578, 593, 100 S.Ct. 1889, 1898, 64 L.Ed.2d 525 (1980) ("The most obvious advantage of direct review by a court of appeals [of the new source performance standards under the Clean Air Act] is the time saved compared to review by a district court, followed by a second review on appeal.”); Foti v. Immigration and Naturalization Serv., 375 U.S. 217, 232, 84 S.Ct. 306, 315, 11 L.Ed.2d 281 (1963) (Harlan, J., concurring) (Congress vested exclusive jurisdiction in the courts of appeals to avoid delays in the deportation process under the Immigration and Nationality Act); City of Rochester v. Bond, 603 F.2d 927, 936 (D.C.Cir.1979) (Congress espoused policy against duplication and inconsistency in its judicial review provisions under the Federal Aviation Act and the Communications Act).
. 48 Fed.Reg. 16590, 16593-98 (1983).
. See, e.g., Investment Co. Inst. v. Board of Governors of the Fed. Reserve Sys., 551 F.2d 1270, 1276 (D.C.Cir.1977) (finding exclusive jurisdiction in court of appeals under the Bank Holding Company Act and that a factual hearing in the District Court was unnecessary since appellate review was based on the administrative record); see also Oljato Chapter of Navajo Tribe v. Train, 515 F.2d 654, 660 (D.C.Cir.1975) ("[I]t is well settled that bifurcated jurisdiction between District Court and Court of Appeals over identical litigation is not favored.”).
. See, e.g., Florida Power & Light Company v. Lorion, — U.S.-, 105 S.Ct. 1598, 1607, 84 L.Ed.2d 643 (1985) (citations omitted) (finding that Congress intended to locate jurisdiction of the subject-matter of Hobbs Act review in the
*271court of appeals whether or not a hearing occurred or could have occurred at the agency level). The Supreme Court stated:
Perhaps the only plausible justification for linking initial review in the court of appeals to the occurrence of a hearing before the agency would be that, absent a hearing, the reviewing court would lack an adequate agency-compiled factual basis to evaluate the agency action and a district court with factfinding powers could make up that deficiency. Such a justification cannot, however, be squared with fundamental principles of judicial review of agency action. "[T]he focal point for judicial review should be the administrative record already in existence, not some new record made initially in the reviewing court"____
If the record before the agency does not support the agency action ... or if the reviewing court simply cannot evaluate the challenged agency action on the basis of the record before it, the proper course, except in rare circumstances, is to remand to the agency for additional investigation or explanation. The reviewing court is not generally empowered to conduct a de novo inquiry into the matter being reviewed and to reach its own conclusions based on such an inquiry____ Moreover, a formal hearing before the agency is in no way necessary to the compilation of an agency record____
The factfinding capacity of the district court is thus typically unnecessary to judicial review of an agency decisionmaking. Placing initial review in the district court does have the negative effect, however, of requiring duplication of the identical task in the district court and in the court of appeals; both courts are to decide, on the basis of the record the agency provides, whether the action passes muster under the appropriate APA standard of review.
. See, e.g., Pacific Gas & Elec. Co. v. State Energy Resources Conservation and Dev. Comm’n, 461 U.S. 190, 195 & nn. 1, 2, 103 S.Ct. 1713, 1717-1718 & nn. 1. 2, 75 L.Ed.2d 752 (1982) (citing H.R.Rep. No. 785, pt. 1, 97th Cong., 2d Sess. 47 (1982) and U.S. Congress, Office of Technology Assessment. Managing Commercial High-Level Radioactive Waste 9 (April 1982)); see also Joint Hearings on S. 637 and S. 1662 Before The Senate Committee on Energy and Natural Resources and The Subcommittee on Nuclear Regulation of The Committee on Environmental and Public Works, 97th Cong., 1st Sess. 259, 265 (1981).
. For a discussion of the policy considerations supporting avoidance of duplicative review, see generally Schorr, The Forum for Judicial Review of Administrative Action: Interpreting Special Review Statutes, 63 B.U.L.Rev. 763 (1983); Currie & Goodman, Judicial Review of Administrative Action: Quest for the Optimum Forum, 75 Colum.L.Rev. 1 (1985); Note, Jurisdiction to Review Administrative Action: District Court or Court of Appeals, 88 Harv.L.Rev. 980 (1975).
. — U.S.-, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984).
. Id. at 2782 (footnotes omitted).
. Rettig v. Pension Benefit Guaranty Corp., 744 F.2d 133, 151 (D.C.Cir.1984) (quoting Chevron, 104 S.Ct. at 2793).
. Chevron, 104 S.Ct. at 2783 (citing United States v. Shimer, 367 U.S. 374, 81 S.Ct. 1554, 6 L.Ed.2d 908 (1961)).
. Id.
. 42 U.S.C. § 10222(a)(3) (1982).
. See, e.g., Defense Logistics Agency v. Federal Labor Relations Auth., 754 F.2d 1003, 1014 (D.C.Cir.1985); see also Eagle-Picher Industries, Inc. v. Environmental Protection Agency, 759 F.2d 905, 920 (D.C.Cir.1985).
GEUMCO urges that we read the word “average" in section 302(a)(3) to refer to an intra -corporation rather than to an inter -industry average. It is GEUMCO's belief that, because the statute states that the fee shall be computed by "an average charge of 1.0 mil per kilowatt-hour for electricity generated by such nuclear fuel," 42 U.S.C. § 10222(a)(3) (1982) (emphasis added), the statute stipulates a narrow reading for the word “average." We find absolutely no support for such a narrow reading of the word “average,” either in the statute or in its legislative history.
GEUMCO also argues that, because DOE's rule assessed in-core fuel costs on an individual assembly basis, and because in-core fuel consists in part of pre-April 7, 1983, SNF, it is inconsistent for DOE to assess the one-time fee on a different basis. We reject this contention. In-core fuel represents a distinct and unique regulatory and policy problem. It was not at all unreasonable for DOE to have arrived at a different accommodation of conflicting policy goals concerning that problem.
. S.Rep. No. 282, 97th Cong., 1st Sess. 26 (1981).
. Id.