delivered the opinion of the Court.
In these consolidated cases, we are asked by 26 private telephone companies and the United States to sustain the holding of the Court of Appeals for the Fourth Circuit that orders of the Federal Communications Commission (FCC or Commission) respecting the depreciation of telephone plant and equipment pre-empt inconsistent state regulation. They are opposed by the Public Service Commissions of 23 States, backed by 30 amici curiae, who argue that the Communications Act of 1934 (Act), 48 Stat. 1064, as amended, 47 U. S. C. § 151 et seq., expressly denied the FCC authority to establish depreciation practices and charges insofar as they relate to the setting of rates for intrastate telephone service.
Respondents suggest that the heart of the cases is whether the revolution in telecommunications occasioned by the federal policy of increasing competition in the industry will be thwarted by state regulators who have yet to recognize or *359accept this national policy and who thus refuse to permit telephone companies to employ accurate accounting methods designed to reflect, in part, the effects of competition. We are told that already there may be as much as $26 billion worth of “reserve deficiencies” on the books of the Nation’s local telephone companies, a reserve which, it is insisted, represents inadequate depreciation of a magnitude that threatens the financial ability of the industry to achieve the technological progress and provide the quality of service that the Act was passed to promote. Petitioners answer that the Act clearly establishes a system of dual state and federal authority over telephone service. They contend that the Act vests in the States exclusive power over intrastate rate-making, which power, petitioners argue, includes final authority over how depreciation shall be calculated for the purpose of setting those intrastate rates. Petitioners note also that the Due Process Clause of the Fourteenth Amendment necessarily represents a check on the power of the States to set depreciation rates at what would amount to confiscatory levels, and that respondents therefore overstate the danger of the States crippling the financial vitality of phone companies.
In deciding these cases, it goes without saying that we do not assess the wisdom of the asserted federal policy of encouraging competition within the telecommunications industry. Nor do we consider whether the FCC should have the authority to enforce, as it sees fit, practices which it believes would best effectuate this purpose. Important as these issues may be, our task is simply to determine where Congress has placed the responsibility for prescribing depreciation methods to be used by state commissions in setting rates for intrastate telephone service. In our view, the language, structure, and legislative history of the Act best support petitioners’ position that the Act denies the FCC the power to dictate to the States as it has in these cases, and accordingly, we reverse.
*360I
The Act establishes, among other things, a system of dual state and federal regulation over telephone service, and it is the nature of that division of authority that these cases are about. In broad terms, the Act grants to the FCC the authority to regulate “interstate and foreign commerce in wire and radio communication,” 47 U. S. C. § 151, while expressly denying that agency “jurisdiction with respect to . . . intrastate communication service . . . .” 47 U. S. C. § 152(b). However, while the Act would seem to divide the world of domestic telephone service neatly into two hemispheres — one comprised of interstate service, over which the FCC would have plenary authority, and the other made up of intrastate service, over which the States would retain exclusive jurisdiction — in practice, the realities of technology and economics belie such a clean parceling of responsibility. This is so because virtually all telephone plant that is used to provide intrastate service is also used to provide interstate service, and is thus conceivably within the jurisdiction of both state and federal authorities. Moreover, because the same carriers provide both interstate and intrastate service, actions taken by federal and state regulators within their respective domains necessarily affect the general financial health of those carriers, and hence their ability to provide service, in the other “hemisphere.”
In 1980 and 1981, the FCC issued two orders that ultimately sparked this litigation. In the 1980 order the FCC changed two depreciation practices affecting telephone plant. Property Depreciation, 83 F. C. C. 2d 267, reconsideration denied, 87 F. C. C. 2d 916 (1981). First, the order altered how carriers could group property subject to depreciation. Because carriers employ so many individual items of equipment in providing service, it would be impossible to depreciate each item individually, and property is therefore classified and depreciated in groups. The order permitted companies the option of grouping plant for depreciation pur*361poses based on its estimated service life (the “equal life” approach). This replaced the FCC’s prior practice of requiring companies to classify and depreciate property according to its year of installation (the “vintage year” method). This change was made to allow depreciation to be based on smaller and more homogeneous groupings, which, the FCC concluded, would result in more accurate matching of capital recovery with capital consumption.
The 1980 order further sought to promote improved accounting accuracy by replacing “whole life” depreciation with the “remaining life” method. Under remaining life, and unlike the treatment under a whole life regime, if estimates upon which depreciation schedules are premised prove erroneous, they may be corrected in midcourse in a way that assures that the full cost of the asset will ultimately be recovered.
The third FCC-mandated change in plant depreciation was announced in a 1981 order, and involved the cost of labor and material associated with the installation of wire inside the premises of a business or residence. The new rule provided that this so-called “inside wiring” no longer be treated as a capital investment to be depreciated over time, but rather as a cost to be “expensed” in the year incurred. Uniform System of Accounts, 85 F. C. C. 2d 818.
Later in 1981, the National Association of Regulatory Utility Commissioners (NARUC) petitioned the FCC for a “clarification” of its order respecting inside wiring. Specifically, NARUC sought a declaration that the FCC’s order did not restrict the discretion of state commissions to follow different depreciation practices in computing revenue requirements and rates for intrastate services.
On April 27, 1982, the FCC issued a memorandum opinion and order in which it agreed with NARUC that its order respecting the depreciation of inside wiring did not preclude state regulators “from using their own accounting and depreciation procedures for intrastate ratemaking pur*362pose[s] . . . Uniform System of Accounts, 89 F. C. C. 2d 1094, 1095. In reaching this conclusion, the FCC declared that it had not intended the 1981 order to “have any preemptive effect that does not arise by operation of law,” and added that “[n]o policy of this Commission would be furthered by requiring state commissions to adhere to the rules we have adopted for the purposes of computing the interstate revenue requirement.” Id., at 1097. The FCC then examined the language and legislative history of sections of the Act dealing with jurisdiction and depreciation and found that they did not support the position that unwilling state commissions either were required by operation of law or could be required in the discretion of the FCC to follow all accounting and depreciation methods prescribed by the Commission. Two commissioners issued a written dissent in which they argued that the FCC had, in its 1981 order, intended to pre-empt inconsistent state depreciation practices, and that deference to the States was especially inappropriate where an important federal policy — that of maturing a “brave new world” of competition in the industry — was at stake.
Respondents petitioned for reconsideration of the order, and the FCC reversed itself and held that § 220 of the Act, which deals expressly with depreciation, does operate automatically to pre-empt inconsistent state action where the Commission has acted to prescribe depreciation rates for a carrier. Amendment of Part 31, 92 F. C. C. 2d 864 (1983). As an alternative ground in support of pre-emption, the FCC asserted that federal displacement of state regulation is justifiable under the Act when necessary “to avoid frustration of validly adopted federal policies.” Id., at 875. Applying this standard to the facts before it, the FCC then found preemption appropriate. It noted that “adequate capital recovery is important to ‘make available, so far as possible, to all the people of the United States a rapid, efficient, Nationwide, world-wide wire and radio communication service with *363adequate facilities at reasonable charges . . .’ 47 U. S. C. 151,” and that “[s]tate depreciation rate prescriptions that do not adequately provide for capital recovery in the competitive environment, which constitutes this Commission’s policy in those markets found capable of supporting competition, would frustrate the accomplishment of that policy and are preemptable by this Commission.” 92 F. C. C. 2d, at 876.
The Fourth Circuit affirmed. Virginia State Corporation Comm’n v. FCC, 737 F. 2d 388 (1984).1 It acknowledged that the Act “does reserve to the states the authority to prescribe rates for intrastate telephone service,” but determined that “reservation [of authority] is not to be read as preserving the states’ sphere of intrastate jurisdiction at the expense of an efficient, viable interstate telecommunications network.” Id., at 392. The court then noted that the FCC had intended to pre-empt state practices, held that the authority to do so was statutorily entrusted to the FCC, and found that the regulations at issue were reasonably designed to ensure that federal objectives would not be frustrated. The Court of Appeals did not reach the Commission’s holding that § 220 of the Act automatically operates to pre-empt state-prescribed depreciation at odds with depreciation ordered by the FCC. We granted certiorari to review the decision of the Court of Appeals. 472 U. S. 1025 (1985).2
*364II
Both petitioners and respondents characterize this litigation as one in which two different persons seek to drive one car, a condition the parties agree is unsatisfactory.3 Where the parties disagree is with respect to who ought to be displaced from the controls. In order to address the contentions, it is appropriate to consider not only the structure of the Act and how it divides authority, but also the nature and function of depreciation as a component of utility regulation.
Depreciation is defined as the loss in service value of a capital asset over time. In the context of public utility accounting and regulation, it is a process of charging the cost of depreciable property, adjusted for net salvage, to operating expense accounts over the useful life of the asset. Thus, accounting practices significantly affect, among other things, the rates that customers pay for service. This is so because a regulated carrier is entitled to recover its reasonable expenses and a fair return on its investment through the rates *365it charges its customers, and because depreciation practices contribute importantly to the calculation of both the carrier’s investment and its expenses. See Knoxville v. Knoxville Water Co., 212 U. S. 1, 13-14 (1909). See generally, 1 A. Priest, Principles of Public Utility Regulation (1969); P. Garfield & W. Lovejoy, Public Utility Economics (1964); 1 A. Kahn, Economics of Regulation (1970).
The total amount that a carrier is entitled to charge for services, its “revenue requirement,” is the sum of its current operating expenses, including taxes and depreciation expenses, and a return on its investment “rate base.” The original cost of a given item of equipment enters the rate base when that item enters service. As it depreciates over time — as a function of wear and tear or technological obsolescence — the rate base is reduced according to a depreciation schedule that is based on an estimate of the item’s expected useful life. Each year the amount that is removed from the rate base is included as an operating expense. In the telephone industry, which is extremely capital intensive, depreciation charges constitute a significant portion of the annual revenue requirement recovered in rates; the parties agree that depreciation charges amount to somewhere between 10% to 15% of the intrastate revenue requirement.
In essence, petitioners’ argument is that the plain and unambiguous language of § 152(b) denies the FCC power to compel the States to employ FCC-set depreciation practices and schedules in connection with the setting of intrastate rates. In part, that section provides:
“[N]othing in this chapter shall be construed to apply or to give the Commission jurisdiction with respect to (1) charges, classifications, practices, services, facilities, or regulations for or in connection with intrastate communication service by wire or radio of any carrier . . . .”
Petitioners maintain that “charges,” “classifications,” and “practices” are “terms of art” which denote depreciation and accounting, and thus that the question presented by these *366cases is expressly answered by the statute. They argue also that the legislative history shows on a more general level that § 152(b) was intended to reserve to the States exclusive regulatory jurisdiction over intrastate service, especially intrastate ratemaking, and that given the importance of depreciation to ratemaking, to require state regulators to follow FCC depreciation practices would frustrate the statutory design of preserving the States’ ratemaking authority over intrastate service. Petitioners maintain that to confer this power on the FCC would be, in effect, to write the jurisdictional limitation of § 152(b) out of the Act.
Where petitioners focus on § 152(b), respondents’ principal argument is that this litigation turns on §220 of the Act, which they insist constitutes an unambiguous grant of power to the FCC exclusively to regulate depreciation. Their argument is that once the FCC has acted pursuant to that section, States are automatically precluded from prescribing different depreciation practices or rates. Section 220(b) states:
“The Commission shall, as soon as practicable, prescribe for such carriers the classes of property for which depreciation charges may be properly included under operating expenses, and the percentages of depreciation which shall be charged with respect to each of such classes of property, classifying the carriers as it may deem proper for this purpose. The Commission may, when it deems necessary, modify the classes and percentages so prescribed. Such carriers shall not, after the Commission has prescribed the [classes] of property for which depreciation charges may be included, charge to operating expenses any depreciation charges on classes of property other than those prescribed by the Commission, or after the Commission has prescribed percentages of depreciation, charge with respect to any class of property a percentage of depreciation other than that prescribed therefor by the Commission. No such carrier shall in any case include in any form under its *367operating or other expenses any depreciation or other charge or expenditure included elsewhere as a depreciation charge or otherwise under its operating or other expenses.”
Respondents assert that their understanding of § 220(b) is bolstered by other substantive provisions of §220. They note, for example, that under § 220(g), once the FCC has prescribed the “forms and manner of keeping accounts,” it is “unlawful... to keep any other accounts . . . than those so prescribed ... or to keep the accounts in any other manner than that prescribed or approved by the Commission,” and that subsections (d) and (e) of § 220 provide for civil and criminal penalties for failing to keep accounts as determined by the Commission. Moreover, § 220(h) permits the FCC in its discretion, if it finds such action to be “consistent with the public interest,” to “except the carriers of any particular class or classes in any State from any of the requirements” under the section “in cases where such carriers are subject to State commission regulation with respect to matters to which this section relates.” Respondents argue that this provision strongly suggests that unless the FCC affirmatively acts to waive or delegate its authority, i. e., to “except” carriers from its regulation, then under § 220(h) the States impliedly cannot adopt inconsistent regulations. Respondents also assert that §220(i) makes clear that the role of the States in depreciation is essentially advisory only. That section provides that the FCC, before exercising its authority, “shall notify” the state commissions and provide an opportunity to the States to “present [their] views” and also instructs the FCC to “consider such views and recommendations.” According to respondents, “Congress gave the states an opportunity to present their views because it expected them to be bound by the resulting prescriptions.” Joint Brief for Listed Private Respondents 14 (Joint Brief). In sum, the position of respondents is that “Congress clearly intended that there be one regime — rather than multiple regimes — of deprecia*368tion for each subject carrier. The FCC was given responsibility for establishing such a regime, and its depreciation decisions have to be respected unless and until it relinquishes authority to the states in individual instances. The states’ interest is recognized but their role is confined to providing their Views and recommendations.’” Ibid.
Although respondents rely primarily on §220 to support pre-emption, they also urge as an alternative and independent ground the reasoning relied on by the Court of Appeals, namely that the FCC is entitled to pre-empt inconsistent state regulation which frustrates federal policy. It is in the context of this argument that respondents most forcefully contend that state regulators must not be permitted to jeopardize the continued viability of the telecommunications industry by refusing to permit carriers to depreciate plant in a way that allows for accurate and timely recapturing of capital. This argument, which is pressed especially by the Solicitor General, relies largely on §151, which in broad terms directs the FCC to develop a rapid and efficient national telephone network.
Ill
The Supremacy Clause of Art. VI of the Constitution provides Congress with the power to pre-empt state law. Preemption occurs when Congress, in enacting a federal statute, expresses a clear intent to pre-empt state law, Jones v. Rath Packing Co., 430 U. S. 519 (1977), when there is outright or actual conflict between federal and state law, e. g., Free v. Bland, 369 U. S. 663 (1962), where compliance with both federal and state law is in effect physically impossible, Florida Lime & Avocado Growers, Inc. v. Paul, 373 U. S. 132 (1963), where there is implicit in federal law a barrier to state regulation, Shaw v. Delta Air Lines, Inc., 463 U. S. 85 (1983), where Congress has legislated comprehensively, thus occupying an entire field of regulation and leaving no room for the States to supplement federal law, Rice v. Santa Fe Elevator Corp., 331 U. S. 218 (1947), or where the state law stands as *369an obstacle to the accomplishment and execution of the full objectives of Congress. Hines v. Davidowitz, 312 U. S. 52 (1941). Pre-emption may result not only from action taken by Congress itself; a federal agency acting within the scope of its congressionally delegated authority may pre-empt state regulation. Fidelity Federal Savings & Loan Assn. v. De la Cuesta, 458 U. S. 141 (1982); Capital Cities Cable, Inc. v. Crisp, 467 U. S. 691 (1984).
In the present cases, two of these “varieties” of preemption are alleged. As noted above, respondents argue that §220 by its terms confers exclusive regulatory power over depreciation on the FCC, thus raising a claim that Congress has expressly manifested a clear intent to displace state law. In addition, respondents maintain that the refusal of the States to accept the FCC-set depreciation schedules and rules will frustrate the federal policy of increasing competition in the industry, and thus that state regulation “stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.” In our view, the jurisdictional limitations placed on the FCC by § 152(b), coupled with the fact that the Act provides for a “separations” proceeding to determine the portions of a single asset that are used for interstate and intrastate service, 47 U. S. C. § 410(c), answer both pre-emption theories.
The critical question in any pre-emption analysis is always whether Congress intended that federal regulation supersede state law. Rice v. Santa Fe Elevator Corp., supra. The Act itself declares that its purpose is “regulating interstate and foreign commerce in communication by wire and radio so as to make available, so far as possible, to all the people of the United States a rapid, efficient, Nation-wide, and world-wide wire and radio communication service with adequate facilities at reasonable charges . . . .” 47 U. S. C. §151. In order to accomplish this goal, Congress created the FCC to centralize and consolidate the regulatory responsibility that had previously been the province of the Interstate Commerce Commis*370sion and the Federal Radio Commission under predecessor statutes. See generally McKenna, Pre-Emption Under the Communications Act, 37 Fed. Comm. L. J. 1, 12-18 (1985). To this degree, § 151 may be read as lending some support to respondents’ position that state regulation which frustrates the ability of the FCC to perform its statutory function of ensuring efficient, nationwide phone service may be impliedly barred by the Act.
We might be inclined to accept this broad reading of § 151 were it not for the express jurisdictional limitations on FCC power contained in § 152(b). Again, that section asserts that “nothing in this chapter shall be construed to apply or to give the Commission jurisdiction with respect to (1) charges, classifications, practices, services, facilities, or regulations for or in connection with intrastate communication service . . . .” By its terms, this provision fences off from FCC reach or regulation intrastate matters — indeed, including matters “in connection with” intrastate service. Moreover, the language with which it does so is certainly as sweeping as the wording of the provision declaring the purpose of the Act and the role of the FCC.
In interpreting §§ 151 and 152(b), we are guided by the familiar rule of construction that, where possible, provisions of a statute should be read so as not to create a conflict. Washington Market Co. v. Hoffman, 101 U. S. 112 (1879). We agree with petitioners that the sections are naturally reconciled to define a national goal of the creation of a rapid and efficient phone service, and to enact a dual regulatory system to achieve that goal. Moreover, were we to find the sections to be in conflict, we would be disinclined to favor the provision declaring a general statutory purpose, as opposed to the provision which defines the jurisdictional reach of the agency formed to implement that purpose.
Respondents advance a number of arguments to counter the view that § 152(b) forbids the FCC to prescribe deprecia*371tion practices and charges in the context of ratemaking for intrastate service. We address each in turn.
A
Respondents assert that the legislative history of § 152(b), as well as the structure of the Act, shows that “charges” and “classifications” refer only to “customer charges,” not depreciation charges, and thus that § 152(b) does not purport to limit the FCC power to regulate depreciation. They seek to support this narrow reading of § 152(b) by noting that the words “charges,” “classifications,” “practices,” and “regulations” appear throughout the Act in contexts where it is clear that what is meant is charges which relate directly to carriers’ rate and service relationships with their customers, rather than depreciation or accounting charges. See §§201-205. Reading the sections in pari materia, we are told, makes it apparent that Congress was concerned in § 152(b) with preserving state autonomy over the rates charged by carriers for specific services, not over depreciation. According to respondents, this reading is bolstered by the legislative history of the section, which reveals that the provision was proposed by state regulators in reaction to this Court’s decision in the so-called Shreveport Rate Case, Houston, E. & W. T. R. Co. v. United States, 234 U. S. 342 (1914), which held, among other things, that the Interstate Commerce Commission had the power to order an increase in specific intrastate railroad rates charged to customers in order to avoid discrimination against interstate commerce. “In other words, Section 2(b)(1) was from the outset concerned with protection against federal preemption of the states’ setting of individual customer charges for specific intrastate services.” Joint Brief 34.
We reject this narrow reading of § 152(b). “Charges,” “classifications,” and “practices” are terms often used by accountants, regulators, courts, and commentators to denote depreciation treatment, see, e. g., United Railways & Elec*372tric Co v. West, 280 U. S. 234, 262 (1930); Smith v. Illinois Bell Telephone Co., 282 U. S. 133, 158 (1930); Wheat, The Regulation of Interstate Telephone Rates, 51 Harv. L. Rev. 846, 859 (1938); A. Kahn, Economics of Regulation (1970), and in accordance with the rule of construction that technical terms of art should be interpreted by reference to the trade or industry to which they apply, Corning Glass Works v. Brennan, 417 U. S. 188 (1974), we find that they do embrace depreciation. It is worth noting that the FCC itself, in the very orders underlying this litigation, used “charges” to mean “depreciation charges.” E. g., Property Depreciation, 83 F. C. C. 2d, at 275.
Nor does the Shreveport Rate Case carry the load that respondents ask of it. In that case, this Court interpreted the constitutional and statutory authority of the Interstate Commerce Commission to include the power to regulate, indeed, set, intrastate rates in order to prevent discrimination against interstate traffic. It is certainly true, as respondents assert, that when Congress was drafting the Communications Act, § 152(b) was proposed and supported by the state commissions in reaction to what they perceived to be the evil of excessive federal regulation of intrastate service such as was sanctioned by the Shreveport Rate Case; but we find no authority in the legislative history to support respondents’ position that the sole concern of the state commissioners was with “protection against federal preemption of the states’ setting of individual customer charges for specific intrastate services.” Joint Brief 34. Rather, the legislative history reveals that representatives from the industry and the States were fully aware that what was at stake in the Act were broad powers to regulate, including, but not limited to, the setting of individual rates, and that “[t]he question of an appropriate division between federal and state regulatory power was a dominating controversy in 1934.” McKenna, 37 Fed. Comm. L. J., at 2. In other words, while we agree that provisions in both the Senate and House bills were de*373signed to overrule the Shreveport Rate Case, we are not persuaded that it was anyone’s understanding that this “overruling” could or should be accomplished by merely including in the Act one section which forbade the FCC to establish specific rates for certain intrastate services; had this been the intention, it would hardly have been necessary to deny the FCC the jurisdiction over “charges, classifications, practices, services, facilities, or regulations for or in connection with intrastate communication service . . . Presumably, it would have sufficed simply to deny the FCC jurisdiction over “rates.” In sum, given the breadth of the language of § 152(b), and the fact that it contains not only a substantive jurisdictional limitation on the FCC’s power, but also a rule of statutory construction (“[N]othing in this chapter shall be construed to apply or to give the Commission jurisdiction with respect to . . . intrastate communication service . . .”), we decline to accept the narrow view urged by respondents, and hold instead that it denies the FCC the power to pre-empt state regulation of depreciation for intrastate ratemaking purposes.
B
Accordingly, we cannot accept respondents’ argument that § 152(b) does not control because the plant involved in this case is used interchangeably to provide both interstate and intrastate service, and that even if § 152(b) does reserve to the state commissions some authority over “certain aspects” of intrastate communication, it should be “confined to intrastate matters which are ‘separable from and do not substantially affect’ interstate communication.” Joint Brief 36. With respect to the present cases, respondents insist that the refusal of the States to employ accurate measures of depreciation will have a severe impact on the interstate communications network because investment in plant will be recovered too slowly or not at all, with the result that new investment will be discouraged to the detriment of the entire network. Numerous decisions of the Courts of Appeals are cited as au*374thority for the proposition that § 152(b) applies as a jurisdictional bar to FCC pre-emptive action only when two factors are present; first, when the matter to be regulated is purely local and second, when interstate communication is not affected by the state regulation which the FCC would seek to pre-empt. E. g., North Carolina Utilities Comm’n v. FCC, 537 F. 2d 787 (CA4), cert. denied, 429 U. S. 1027 (1976); North Carolina Utilities Comm’n v. FCC, 552 F. 2d 1036 (CA4), cert. denied, 434 U. S. 874 (1977); Puerto Rico Telephone Co. v. FCC, 553 F. 2d 694 (CA1 1977); New York Telephone Co. v. FCC, 631 F. 2d 1059 (CA2 1980).
The short answer to this argument is that it misrepresents the statutory scheme and the basis and test for pre-emption. While it is certainly true, and a basic underpinning of our federal system, that state regulation will be displaced to the extent that it stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress, Hines, 312 U. S., at 67, it is also true that a federal agency may pre-empt state law only when and if it is acting within the scope of its congressionally delegated authority. This is true for at least two reasons. First, an agency literally has no power to act, let alone pre-empt the validly enacted legislation of a sovereign State, unless and until Congress confers power upon it. Second, the best way of determining whether Congress intended the regulations of an administrative agency to displace state law is to examine the nature and scope of the authority granted by Congress to the agency. Section 152(b) constitutes, as we have explained above, a congressional denial of power to the FCC to require state commissions to follow FCC depreciation practices for intrastate ratemaking purposes. Thus, we simply cannot accept an argument that the FCC may nevertheless take action which it thinks will best effectuate a federal policy. An agency may not confer power upon itself. To permit an agency to expand its power in the face of a congressional limitation on its jurisdiction would be to grant to the agency *375power to override Congress. This we are both unwilling and unable to do.
Moreover, we reject the intimation — the position is not strongly pressed — that the FCC cannot help but pre-empt state depreciation regulation of joint plant if it is to fulfill its statutory obligation and determine depreciation for plant used to provide interstate service, i. e., that it makes no sense within the context of the Act to depreciate one piece of property two ways. The Communications Act not only establishes dual state and federal regulation of telephone service; it also recognizes that jurisdictional tensions may arise as a result of the fact that interstate and intrastate service are provided by a single integrated system. Thus, the Act itself establishes a process designed to resolve what is known as “jurisdictional separations” matters, by which process it may be determined what portion of an asset is employed to produce or deliver interstate as opposed to intrastate service. 47 U. S. C. §§ 221(c), 410(c). Because the separations process literally separates costs such as taxes and operating expenses between interstate and intrastate service, it facilitates the creation or recognition of distinct spheres of regulation. See Smith v. Illinois Bell Telephone Co., 282 U. S. 133 (1930). As respondents concede, and as the Court of Appeals itself acknowledged, 737 F. 2d, at 396, it is certainly possible to apply different rates and methods of depreciation to plant once the correct allocation between interstate and intrastate use has been made,4 Brief for Respondent *376GTE 36, just as it is possible to determine that, for example, 75% of an employee’s time is devoted to the production of intrastate service, and only one quarter to interstate service, and to allocate the cost of that employee accordingly. Respondents maintain that if the FCC and the States apply different depreciation practices to the same property, then the “whole purpose of depreciation, which is to match depreciation charges of the equipment with the revenues generated by its use,” will be frustrated. Ibid. But this is true and a concern only to the degree that the principles, judgments, and considerations that underlie depreciation rules reflect only “real world” facts, rather than choices made by regulators partially on the basis of fact and partially on the basis of such factors as the perceived need to improve the industry’s cash flow, spur investment, subsidize one class of customer, or any other policy factor. What is really troubling respondents, of course, is their sense that state regulators will not allow them sufficient revenues. While we do not deprecate this concern, § 152(b) precludes both the FCC and this Court from providing the relief sought. As we so often admonish, only Congress can rewrite this statute.
C
We also reject respondents’ argument that §220, which deals specifically5 and expressly with depreciation, requires *377automatic pre-emption of all state regulation respecting depreciation. As noted above, § 220 directs the FCC to prescribe the classes of property for which depreciation charges may be included under operating expenses, and prohibits carriers from departing from FCC-set regulations respecting depreciation. While it is, no doubt, possible to find some support in the broad language of the section for respondents’ position, we do not find the meaning of the section so unambiguous or straightforward as to override the command of § 152(b) that “nothing in this chapter shall be construed to apply or to give the Commission jurisdiction” over intrastate service. We note, for example, that a very strict reading of §220 — which is what respondents urge and upon which they ultimately rely — is simply untenable. There can be no dispute, for example, regarding the fact that taxing authorities of the Federal Government are entitled to require the carriers to employ, for tax purposes, depreciation practices and schedules different from those which might be ordered by the FCC for interstate ratemaking purposes. We are advised by petitioners that carriers do, as a routine matter, keep “separate” books in this connection. Were respondents’ reading of § 220 correct, this practice would violate the Act, and taxing authorities would be compelled to compute taxation on the basis of depreciation schedules employed by the FCC for ratemaking purposes. Moreover, despite the sweeping language of § 220, nowhere does it even allude to, let alone expressly refer to, depreciation as a component of state ratemaking. Nor is the word “pre-emption” used.
It is thus at least possible, as some petitioners argue, that the section was intended to do no more than spell out the authority of the FCC over depreciation in the context of interstate regulation. It is similarly plausible, as other *378petitioners contend, that the section, which is captioned “Accounts, records, and memoranda,” was addressed to the plenary authority of the FCC to dictate how the carriers’ books would be kept for the purposes of financial reporting, in order to ensure that investors and regulators would be presented with an accurate picture of the financial health of the carriers. In any event, we need not, in order to decide these cases, define fully the scope of the section, and we hold only that § 220 does not operate to pre-empt state depreciation regulation for intrastate ratemaking purposes.6
*379Like many statutes, the Act contains some internal inconsistencies, vague language, and areas of uncertainty. It is not a perfect puzzle into which all the pieces fit. Thus, it is with the recognition that there are not crisp answers to all of the contentions of either party that we conclude that § 152(b) represents a bar to federal pre-emption of state regulation over depreciation of dual jurisdiction property for intrastate ratemaking purposes.
For the reasons stated above, the judgment of the Court of Appeals for the Fourth Circuit is reversed, and the cases are remanded for further proceedings consistent with this opinion.
It is so ordered.
The Chief Justice and Justice Blackmun dissent. Justice Powell and Justice O’Connor took no part in the consideration or decision of these cases.Exclusive jurisdiction over final FCC orders lies with the courts of appeals. 28 U. S. C. §2342(1).
We originally postponed jurisdiction in No. 84-871, which came to us by way of appeal, rather than certiorari. A potential jurisdictional issue in that case arose as a result of the contention of the Government and the telephone companies that an appeal did not lie under 28 U. S. C. § 1254(2) because the decision of the Court of Appeals did not expressly strike down any particular state ratemaking order.
We need not address or resolve whether an appeal is proper in No. 84-871. The Louisiana Public Service Commission has asked that its jurisdictional statement be treated as a petition for a writ of certiorari, and we clearly have certiorari jurisdiction under 28 U. S. C. § 1254(1) to decide the case. We have, moreover, granted the petitions for certiorari in *364Nos. 84-889, 84-1054, and 84-1069, 472 U. S. 1025 (1985). In accordance with our customary practice, see, e. g., Renton v. Playtime Theatres, Inc., 475 U. S. 41, 43-44, n. 1 (1986), we dismiss the appeal in No. 84-871 and, treating the papers as a petition for certiorari, grant the writ of certiorari.
Petitioners suggest that overreaching by the FCC has resulted in a situation where one person has a foot on the accelerator of a car while another person is attempting to steer. Tr. of Oral Arg. 9, 21. Although it is not evident from the metaphor whether petitioners’ position is that the hand or the foot belongs to the FCC — whether, in other words, the FCC has stepped on the States’ authority, or, heavy-handedly grabbed the wheel — the notion is that it is the States’ responsibility under the Act to value property and to ascertain a rate base, and that it is inconsistent to confer depreciation authority — which is, according to the States, integrally bound up with valuation considerations and the determination of the rate base — on the FCC.
Respondents assert that this is “the case of two hands on the steering wheel,” id., at 40, by which, presumably, they mean to suggest that the hands belong to two different entities. Their position is that it makes no sense to have both the FCC and the state regulators depreciating the same piece of plant in two different ways.
Thus, these eases are readily distinguishable from those in which FCC pre-emption of state regulation was upheld where it was not possible to separate the interstate and the intrastate components of the asserted FCC regulation. See, e. g., North Carolina Utilities Comm’n v. FCC, 537 F. 2d 787 (CA4), cert. denied, 429 U. S. 1027 (1976), and North Carolina Utilities Comm’n v. FCC, 552 F. 2d 1036 (CA4), cert. denied, 434 U. S. 874 (1977) (Where FCC acted within its authority to permit subscribers to provide their own telephones, pre-emption of inconsistent state regulation prohibiting subscribers from connecting their own phones unless used ex-*376elusively in interstate service upheld since state regulation would negate the federal tariff).
Respondents maintain that since “[slpecific terms prevail over the general,” Fourco Glass Co. v. Transmirra Products Corp., 353 U. S. 222, 228 (1957), and § 220 deals specifically with depreciation, the general language of § 152(b) should not be read to bar FCC regulation of depreciation. The rule of construction cited by respondents is simply inapplicable in the context of these cases. First, § 152(b) deals with jurisdiction, and thus addresses a different subject than §220, which respondents correctly characterize as involving depreciation. Thus, while § 152(b) may be more “general’ than § 220, the sections are not general or specific with respect to each other. Second, § 152(b) not only imposes jurisdictional limits on the power of a federal agency, but also, by stating that nothing in the Act shall *377be construed to extend FCC jurisdiction to intrastate service, provides its own rule of statutory construction. In other words, the Act itself, in § 152(b), presents its own specific instructions regarding the correct approach to the statute which applies to how we should read § 220.
Respondents insist that the legislative history of the section proves that it was intended to provide the FCC with power to pre-empt state regulation over depreciation practices. They rely in particular on the fact that Congress, in drafting § 220, reenacted, almost verbatim, § 20(5) of the Interstate Commerce Act, 49 U. S. C. App. §20(5), which, respondents contend, had already been construed to require the ICC to prescribe depreciation for both telephone companies and railroads. Telephone and Railroad Depreciation Charges, 118 I. C. C. 295 (1926). Respondents note further that during hearings on an early version of the Act, state commissioners testifying before Congress argued that reenactment of § 20(5) and other provisions would permit the FCC to usurp “[a]ll matters of depreciation. . . without regard to the action upon the same subject by the State Commission.” Hearings on S. 6 before the Senate Committee on Interstate Commerce, 71st Cong., 2d Sess., pt. 15, p. 2243 (1930) (resolution of Montana Commission). They also note that state regulators did manage — initially—to persuade the drafters of the Act to add a new §220(j), which expressly permitted the States to prescribe their own depreciation practices for the purposes of determining intrastate rates. The fact that this section was rejected by the Conference Committee, despite the strong support of the States, we are told, is strong evidence that Congress intended to preserve in the FCC the broad power over depreciation that had been conferred on the ICC.
We are not persuaded. First, § 20(5) of the Interstate Commerce Act had never been interpreted to prohibit state commissioners from requiring carriers to keep additional records for the purposes of intrastate ratemaking. As the FCC itself noted in its 1982 order denying preemption, Uniform System of Accounts, 89 F. C. C. 2d 1094, 1101, it was only in dictum that the ICC suggested that it possessed authority under § 20(5) to prescribe depreciation for all property that might be used in interstate commerce, and that dictum did not even purport to address *379whether federal prescription of depreciation would pre-empt the States from prescribing additional depreciation practices for its regulatory purposes. Moreover, that is how this Court read the ICC order. Smith v. Illinois Bell Telephone Co., 282 U. S. 133, 159 (1930). And in Northwestern Bell Telephone Co. v. Nebraska State Railway Comm’n, 297 U. S. 471, 478 (1936), we expressly left open whether an ICC prescription, if issued, would be pre-emptive of state regulation.
Moreover, while it is true that Congress rejected the state-proposed §220(j), again, as the FCC noted in its order denying pre-emption, respondents make too much of too little. “The record of the Congressional hearings indicates little more than that the supporters of original section 220(j) believed that the provision was desirable to resolve a previous unsettled point of law under the predecessor provision of the Interstate Commerce Act. ... At most, this legislative history indicates that the 1934 Congress was not sure whether reenactment of the Interstate Commerce Act language would or would not preempt state accounting and depreciation rules and did not choose to resolve the question at that time.” 89 F. C. C. 2d, at 1103, 1106.