delivered the opinion of the Court.

This case is a sequel to Merrion v. Jicarilla Apache Tribe, 455 U. S. 130 (1982), in which we held that the Jicarilla Apache Tribe (Tribe) has the power to impose a severance tax on the production of oil and gas by non-Indian lessees of wells located on the Tribe’s reservation. We must now decide whether the State of New Mexico can continue to impose its severance taxes on the same production of oil and gas.

I

All 742,135 acres of the Jicarilla Apache Reservation are located in northwestern New Mexico. Id., at 133. In 1887, President Cleveland issued an Executive Order setting aside this tract of public land “as a reservation for the use and occupation of the Jicarilla Apache Indians.” 1 C. Kappler, Indian Affairs, Laws and Treaties 875 (1904). The only qualification contained in the order was a proviso protecting bona fide settlers from defeasance of previously acquired federal rights.1 *167Ibid. The land is still owned by the United States and is held in trust for the Tribe.

The Tribe, which consists of approximately 2,500 enrolled members, is organized under the Indian Reorganization Act. 48 Stat. 984, 25 U. S. C. §461 et seq. The Indian Mineral Leasing Act of 1938 (1938 Act) grants the Tribe authority, subject to the approval of the Secretary of the Interior (Secretary), to execute mineral leases. 52 Stat. 347, 25 U. S. C. § 396a et seq. Since at least as early as 1953, the Tribe has been leasing reservation lands to nonmembers for the production of oil and gas. See Merrion, supra, at 135. Mineral leases now encompass a substantial portion of the reservation and constitute the primary source of the Tribe’s general operating revenues. In 1969, the Secretary approved an amendment to the Tribe’s Constitution authorizing it to enact ordinances, subject to his approval, imposing taxes on nonmembers doing business in the reservation. See Revised Constitution of the Jicarilla Apache Tribe, Art. XI, § 1(e) (Equity). The Tribe enacted such an ordinance in 1976, imposing a severance tax on “any oil and natural gas severed, saved and removed from Tribal lands.” Oil and Gas Severance Tax, Ordinance No. 77-0-02, Jicarilla Apache Tribal Code (hereinafter J. A. T. C.), Tit. 11, ch. 1 (1987) (Equity); see also Merrion, supra, at 135-136. The Secretary approved the ordinance later that year, and in. 1982 this Court upheld the Tribe’s power to impose a severance tax on pre-existing as well as future leases. See Merrion, supra. Subsequently, the Tribe enacted a privilege tax, which the *168Secretary also approved. See Oil and Gas Privilege Tax, Ordinance No. 85-0-434, J. A. T. C., Tit. 11, ch. 2 (1985).2

In 1976, Cotton Petroleum Corporation (Cotton), a non-Indian company in the business of extracting and marketing oil and gas, acquired five leases covering approximately 15,000 acres of the reservation. There were then 15 operating wells on the leased acreage and Cotton has since drilled another 50 wells. The leases were issued by the Tribe and the United States under the authority of the 1938 Act. Pursuant to the terms of the leases, Cotton pays the Tribe a rent of $125 per acre, plus a royalty of 1214 percent of the value of its production.3 In addition, Cotton pays the Tribe’s oil and gas severance and privilege taxes, which amount to approximately 6 percent of the value of its production. Thus, Cotton’s aggregate payment to the Tribe includes an acreage rent in excess of $1 million, plus royalties and taxes amounting to about 1814 percent of its production.

Prior to 1982, Cotton paid, without objection, five different oil and gas production taxes to the State of New Mexico.4 The state taxes amount to about 8 percent of the value of Cotton’s production. The same 8 percent is collected from producers throughout the State. Thus, on wells outside the *169reservation, the total tax burden is only 8 percent, while Cotton’s reservation wells are taxed at a total rate of 14 percent (8 percent by the State and 6 percent by the Tribe). No state tax is imposed on the royalties received by the Tribe.

At the end of our opinion in Merrion, 455 U. S., at 158-159, n. 26, we added a footnote rejecting the taxpayer’s argument that the tribal tax was invalid as a “multiple tax burden on interstate commerce” because imposed on the same activity already taxed by the State. One of the reasons the argument failed was that the taxpayer had made no attempt to show that the Tribe was “seek[ing] to seize more tax revenues than would be fairly related to the services provided by the Tribe.” Ibid. After making that point, the footnote suggested that the state tax might be invalid under the Commerce Clause if in excess of what “the State’s contact with the activity would justify.”5 Ibid, (emphasis in original).

*170In 1982, Cotton paid its state taxes under protest and then brought an action in the District Court for Santa Fe County challenging the taxes under the Indian Commerce, Interstate Commerce, Due Process, and Supremacy Clauses of the Federal Constitution. App. 2-15. Relying on the Merrion footnote, Cotton contended that state taxes imposed on reservation activity are only valid if related to actual expenditures by the State in relation to the activity being taxed. Record 421. In support of this theory, Cotton presented evidence at trial tending to prove that the amount of tax it paid to the State far exceeded the value of services that the State provided to it and that the taxes paid by all nonmember oil producers far exceeded the value of services provided to the reservation as a whole.6 Cotton did not, however, attempt to prove that the state taxes imposed any burden on the Tribe.

After trial, the Tribe sought, and was granted, leave to file a brief amicus curiae. Id., at 128. The Tribe argued that a decision upholding the state taxes would substantially interfere with the Tribe’s ability to raise its own tax rates and would diminish the desirability of on-reservation oil and gas leases. Id., at 124. The Tribe expressed a particular concern about what it characterized as a failure of the State “to provide services commensurate with the taxes collected.” Ibid.

*171After the Tribe filed its brief, the New Mexico District Court issued a decision upholding the state taxes. App. to Juris. Statement 14. The District Court found that “New Mexico provides substantial services to both the Jicarilla Tribe and Cotton,”7 and concluded that the State had a valid interest in imposing taxes on non-Indians on the reservation8 Squarely rejecting Cotton’s theory of the case, the court stated that “[t]he theory of public finance does not require expenditures equal to revenues.” Id., at 17. Turning to the question whether the state taxes were inconsistent with the federal interest in fostering the economic development of Indian tribes, the District Court found that the “economic and legal burden of paying the state taxes falls on Cotton or its buyers” and that “[n]o economic burden falls on the tribe by virtue of the state taxes.” Id., at 15. More specifically, it found that the state taxes had not affected the Tribe’s ability to collect its taxes or to impose a higher *172tax, and had “not in any way deterred production of oil and gas” on the reservation. Id., at 16-17. It concluded that the taxes had no adverse impact on tribal interests and that they were not pre-empted by federal law. Id., at 17-18. Finally, the District Court held that the taxes were fully consistent with the Commerce and Due Process Clauses of the Federal Constitution. Ibid.

The New Mexico Court of Appeals affirmed. 106 N. M. 517, 745 P. 2d 1170 (1987). Like the District Court, it was left unpersuaded by Cotton’s contention that the New Mexico taxes are invalid because the State’s expenditures on reservation activity do not equal the revenues collected. The Court of Appeals correctly noted that the Merrion footnote, 455 U. S., at 159, n. 26, “intimate[s] no opinion on the possibility of such a challenge,” but simply suggests that a state tax “might” be invalid if greater than the State’s “contact with the [on-reservation] activity would justify.” 106 N. M., at 520, 745 P. 2d, at 1173. Finding no support for Cotton’s position in Merrion, the Court of Appeals looked instead to our opinion in Commonwealth Edison Co. v. Montana, 453 U. S. 609 (1981), and concluded that a State’s power to tax an activity connected to interstate commerce is not limited to the value of the services provided in support of that activity. 106 N. M., at 521, 745 P. 2d, at 1174. Agreeing with the trial court that the New Mexico taxes were fairly related to the services provided to Cotton, the Court of Appeals rejected Cotton’s Commerce Clause challenge. Ibid.

The Tribe, again participating as an amicus curiae, urged a different approach to the case. Unlike Cotton, the Tribe argued that the state taxes could not withstand traditional pre-emption analysis. The Tribe conceded that state laws, to the extent they do not interfere with tribal self-government, may control the conduct of non-Indians on the reservation. It maintained, however, that the taxes at issue interfered with its ability to raise taxes and thus with its right to self-government. The Court of Appeals rejected *173this argument because the record contained no evidence of any adverse impact on the Tribe and, indeed, indicated that the Tribe could impose even higher taxes than it had without adverse effect.9

The New Mexico Supreme Court granted, but then quashed, a writ of certiorari. 106 N. M. 511, 745 P. 2d 1159 (1987). We then noted probable jurisdiction and invited the parties to brief and argue the following additional question:

“Does the Commerce Clause require that an Indian Tribe be treated as a State for purposes of determining whether a state tax on nontribal activities conducted on an Indian Reservation must be apportioned to account for taxes imposed on those same activities by the Indian Tribe?” 485 U. S. 1005 (1988).

We now affirm the judgment of the New Mexico Court of Appeals.

II

This Court’s approach to the question whether a State may tax on-reservation oil production by non-Indian lessees has varied over the course of the past century. At one time, such a tax was held invalid unless expressly authorized by Congress; more recently, such taxes have been upheld unless expressly or impliedly prohibited by Congress. The changed approach to these taxes is one aspect of the evolution of the doctrine of intergovernmental tax immunity that we recently discussed in detail in South Carolina v. Baker, 485 U. S. 505 (1988).

During the first third of this century, this Court frequently invalidated state taxes that arguably imposed an indirect eco*174nomic burden on the Federal Government or its instrumentalities by application of the “intergovernmental immunity” doctrine. That doctrine “was based on the rationale that any tax on income a party received under a contract with the government was a tax on the contract and thus a tax ‘on’ the government because it burdened the government’s power to enter into the contract.” Id., at 518. In a case decided in 1922, the Court applied the intergovernmental immunity doctrine to invalidate a state tax on income derived by a non-Indian lessee from the sale of his interest in oil produced on Indian land. See Gillespie v. Oklahoma, 257 U. S. 501. Consistently with the view of intergovernmental immunity that then prevailed, the Court stated that “a tax upon such profits is a direct hamper upon the effort of the United States to make the best terms that it can for its wards.” Id., at 506 (citing Weston v. Charleston, 2 Pet. 449, 468 (1829)). The same reasoning was used to invalidate a variety of other state taxes imposed on non-Indian lessees at that time.10

Shortly after reaching its zenith in the Gillespie decision, the doctrine of intergovernmental tax immunity started a long path in decline and has now been “thoroughly repudiated” by modern case law. South Carolina v. Baker, supra, at 520. In 1932, four Members of this Court argued that Gillespie was unsound and should be overruled. See Burnet v. Coronado Oil & Gas Co., 285 U. S. 393, 401 (Stone, J., dissenting); id., at 405 (Brandéis, J., dissenting). Five years later, the Court took a substantial step in that direction, rejecting the view that a nondiscriminatory state tax on a *175private party contracting with the Government is invalid because the economic burden of the tax may fall on the Government. See James v. Dravo Contracting Co., 302 U. S. 134 (1937). “With the rationale for conferring a tax immunity on parties dealing with another government rejected, the government contract immunities recognized under prior doctrine were, one by one, eliminated.” South Carolina v. Baker, supra, at 522. Specifically, in Helvering v. Mountain Producers Corp., 303 U. S. 376, 386-387 (1938), the Court squarely overruled Gillespie, supra. Thus, after Mountain Producers Corp., supra, was decided, oil and gas lessees operating on Indian reservations were subject to nondiscriminatory state taxation as long as Congress did not act affirmatively to pre-empt the state taxes. See ibid. See also Oklahoma Tax Comm’n v. Texas Co., 336 U. S. 342 (1949).

In sum, it is well settled that, absent express congressional authorization, a State cannot tax the United States directly. See McCulloch v. Maryland, 4 Wheat. 316 (1819). It is also clear that the tax immunity of the United States is shared by the Indian tribes for whose benefit the United States holds reservation lands in trust. See Montana v. Blackfeet Tribe, 471 U. S. 759, 764 (1985). Under current doctrine, however, a State can impose a nondiscriminatory tax on private parties with whom the United States or an Indian tribe does business, even though the financial burden of the tax may fall on the United States or tribe. See id., at 765; South Carolina v. Baker, supra, at 523. Although a lessee’s oil production on Indian lands is therefore not “automatically exempt from state taxation,” Congress does, of course, retain the power to grant such immunity. Mescalero Apache Tribe v. Jones, 411 U. S. 145, 150 (1973). Whether such immunity shall be granted is thus a question that “is essentially legislative in character.” Texas Co., supra, at 365-366.

The question for us to decide is whether Congress has acted to grant the Tribe such immunity, either expressly or *176by plain implication.11 In addition, we must consider Cotton’s argument that the “multiple burden” imposed by the state and tribal taxes is unconstitutional.

Ill

Although determining whether federal legislation has preempted state taxation of lessees of Indian land is primarily an exercise in examining congressional intent, the history of tribal sovereignty serves as a necessary “backdrop” to that process. Cf. Rice v. Rehner, 463 U. S. 713, 719 (1983) (quoting McClanahan v. Arizona State Tax Comm’n, 411 U. S. 164, 172 (1973)). As a result, questions of pre-emption in this area are not resolved by reference to standards of preemption that have developed in other areas of the law, and are not controlled by “mechanical or absolute conceptions of state or tribal sovereignty.” White Mountain Apache Tribe v. Bracker, 448 U. S. 136, 145 (1980). Instead, we have applied a flexible pre-emption analysis sensitive to the particular facts and legislation involved. Each case “requires a particularized examination of the relevant state, federal, and tribal interests.” Ramah Navajo School Bd., Inc. v. Bureau of Revenue of New Mexico, 458 U. S. 832, 838 (1982). Moreover, in examining the pre-emptive force of the relevant federal legislation, we are cognizant of both the broad policies that underlie the legislation and the history of tribal independence in the field at issue. See ibid. It bears emphasis that although congressional silence no longer entails a broad-based immunity from taxation for private parties doing business with Indian tribes, federal pre-emption is not limited to cases in which Congress has expressly — as compared to *177impliedly — pre-empted the state activity. Finally, we note that although state interests must be given weight and courts should be careful not to make legislative decisions in the absence of congressional action, ambiguities in federal law are, as a rule, resolved in favor of tribal independence. See ibid.

Against this background, Cotton argues that the New Mexico taxes are pre-empted by the “federal laws and policies which protect tribal self-government and strengthen impoverished reservation economies. ” Brief for Appellants 16. Most significantly, Cotton contends that the 1938 Act exhibits a strong federal interest in guaranteeing Indian tribes the maximum return on their oil and gas leases. Moreover, Cotton maintains that the Federal and Tribal Governments, acting pursuant to the 1938 Act, its accompanying regulations, and the Jicarilla Apache Tribal Code, exercise comprehensive regulatory control over Cotton’s on-reservation activity. Cotton describes New Mexico’s responsibilities, in contrast, as “significantly limited.” Brief for Appellants 21. Thus, weighing the respective state, federal, and tribal interests, Cotton concludes that the New Mexico taxes unduly interfere with the federal interest in promoting tribal economic self-sufficiency and are not justified by an adequate state interest. We disagree.

The 1938 Act neither expressly permits state taxation nor expressly precludes it, but rather simply provides that “unallotted lands within any Indian reservation or lands owned by any tribe . . . may, with the approval of the Secretary of the Interior, be leased for mining purposes, by authority of the tribal council ... , for terms not to exceed ten years and as long thereafter as minerals are produced in paying quantities.” 25 U. S. C. §396a. The Senate and House Reports that accompanied the Act, moreover — even when considered in their broadest possible terms — shed little light on congressional intent concerning state taxation of oil and gas produced on leased lands. See S. Rep. No. 985, 75th Cong., 1st Sess. (1937); H. R. Rep. No. 1872, 75th Cong., *1783d Sess. (1938). Both Reports reflect that the proposed legislation was suggested by the Secretary and considered by the appropriate committees, which recommended that it pass without amendment. Beyond this procedural summary, the Reports simply rely on the Secretary’s letter of transmittal to describe the purpose of the Act. That letter provides that the legislation was intended, in light of the disarray of federal law in the area, “to obtain uniformity so far as practicable of the law relating to the leasing of tribal lands for mining purposes,” and, in particular, was designed to “bring all mineral leasing matters in harmony with the Indian Reorganization Act.” Id., at 1, 3; S. Rep. No. 985, supra, at 2, 3. In addition, the letter contains the following passage:

“It is not believed that the present law is adequate to give the Indians the greatest return from their property. As stated, present law provides for locating and taking mineral leases in the same manner as mining locations are made on the public lands of the United States; but there are disadvantages in following this procedure on Indian lands that are not present in applying for a claim on the public domain. For instance, on the public domain the discoverer of a mineral deposit gets extralateral rights and can follow the ore beyond the side lines indefinitely, while on the Indian lands under the act of June 30, 1919, he is limited to the confines of the survey markers not to exceed 600 feet by 1,500 feet in any one claim. The draft of the bill herewith would permit the obtaining of sufficient acreage to remove the necessity for extralateral rights with all of its attending controversies.” Id., at 2; H. R. Rep. No. 1872, supra, at 2 (emphasis added).

Relying on the first sentence in this paragraph, Cotton argues that the 1938 Act embodies a broad congressional policy of maximizing revenues for Indian tribes. Cotton finds support for this proposition in Montana v. Blackfeet Tribe, 471 U. S. 759 (1985). That case raised the question *179whether the 1938 Act authorizes state taxation of a tribe’s royalty interests under oil and gas leases issued to nonmembers. Applying the settled rule that a tribe may only be directly taxed by a State if “Congress has made its intention to [lift the tribe’s exemption] unmistakably clear,” id., at 765, we concluded that “the State may not tax Indian royalty income from leases issued pursuant to the 1938 Act,” id., at 768. In a footnote we added the observation that direct state taxation of Indian revenues would frustrate the 1938 Act’s purpose of “ensur[ing] that Indians receive ‘the greatest return from their property,’ [S. Rep. No. 985, supra, at] 2; H. R. Rep. No. 1872, supra, at 2.” Id., at 767, n. 5.

To the extent Cotton seeks to give the Secretary’s reference to “the greatest return from their property” talismanic effect, arguing that these words demonstrate that Congress intended to guarantee Indian tribes the maximum profit available without regard to competing state interests, it overstates its case. There is nothing remarkable in the proposition that, in authorizing mineral leases, Congress sought to provide Indian tribes with a profitable source of revenue. It is however quite remarkable, indeed unfathomable in our view, to suggest that Congress intended to remove all state-imposed obstacles to profitability by attaching to the Senate and House Reports a letter from the Secretary that happened to include the phrase “the greatest return from their property.” Read in the broadest terms possible, the relevant paragraph suggests that Congress sought to remove “disadvantages in [leasing mineral rights] on Indian lands that are not present in applying for a claim on the public domain.” S. Rep. No. 985, supra, at 2; H. R. Rep. No. 1872, supra, at 2. By 1938, however, it was established that oil and gas lessees of public lands were subject to state taxation. See Mid-Northern Oil Co. v. Walker, 268 U. S. 45 (1925). It is thus apparent that Congress was not concerned with state taxation, but with matters such as the unavailability of extralateral mineral rights on Indian land. Nor do we *180read the Blackfeet footnote, 471 U. S., at 767, n. 5, to give the Secretary’s words greater effect. We think it clear that the footnote simply stands for the proposition that the Act’s purpose of creating a source of revenue for Indian tribes provides evidence that Congress did not intend to authorize direct state taxation of Indian royalties.

We thus agree that a purpose of the 1938 Act is to provide Indian tribes with badly needed revenue, but find no evidence for the further supposition that Congress intended to remove all barriers to profit maximization. The Secretary’s letter of transmittal, even when read permissively for broad policy goals and even when read to resolve ambiguities in favor of tribal independence, supports no more.

Our review of the legislation that preceded thé 1938 Act provides no additional support for Cotton’s expansive view of the Act’s purpose. This history is relevant in that it supplies both the legislative background against which Congress enacted the 1938 Act and the relevant “backdrop” of tribal independence. Congress first authorized mineral leasing on Indian lands in 1891. See Act of Feb. 28, 1891, §3, 26 Stat. 795, 25 U. S. C. §397 (1891 Act). That legislation, which empowered tribes to enter into grazing and mining leases, only applied to lands “occupied by Indians who have bought and paid for the same,” and was thus interpreted to be inapplicable to Executive Order reservations. See British-American Oil Producing Co. v. Board of Equalization of Montana, 299 U. S. 159, 161-162, 164 (1936). Mineral leasing on reservations created by Executive Order — like the Jicarilla Apache Reservation — was not authorized until almost four decades later. After years of debate concerning whether Indians had any right to share in royalties derived from oil and gas leases in Executive Order reservations,12 *181Congress finally enacted legislation in 1927 that authorized such leases. See Indian Oil Act of 1927, 44 Stat. (part 2) 1347, 25 U. S. C. §398a (1927 Act).

While both the 1891 and 1927 Acts were in effect, Gillespie was the prevailing law and, under its expansive view of intergovernmental tax immunity, States were powerless to impose severance taxes on oil produced on Indian reservations unless Congress expressly waived that immunity. Just two years after Gillespie was decided, Congress took such express action and authorized state taxation of oil and gas production in treaty reservations. See Indian Oil Leasing Act of 1924, 43 Stat. 244 (1924 Act), current version at 25 U. S. C. §398. See also British-American Oil Producing Co. v. Board of Equalization, supra (applying 1924 Act to uphold state tax imposed on the production of oil and gas in *182the Blackfeet Indian Reservation). More significantly for purposes of this case, when Congress first authorized oil and gas leasing on Executive Order reservations in the 1927 Act, it expressly waived immunity from state taxation of oil and gas lessees operating in those reservations. See 44 Stat. (part 2) 1347, 25 U. S. C. § 398c. Thus, at least as to Executive Order reservations, state taxation of nonmember oil and gas lessees was the norm from the very start. There is, accordingly, simply no history of tribal independence from state taxation of these lessees to form a “backdrop” against which the 1938 Act must be read.

We are also unconvinced that the contrast between the 1927 Act’s express waiver of immunity and the 1938 Act’s silence on the subject suggests that Congress intended to repeal the waiver in the 1938 Act and thus to diametrically change course by implicitly barring state taxation. The general repealer clause contained in thé 1938 Act provides that “[a]ll Act[s] or parts of Acts inconsistent herewith are hereby repealed.” 52 Stat. 348. Although one might infer from this clause that all preceding, nonconflicting legislation in the area, like the 1927 Act’s waiver provision, is implicitly incorporated, we need not go so far to simply conclude that the 1938 Act’s omission demonstrates no congressional purpose to close the door to state taxation. Moreover, the contrast between the 1927 and 1938 Acts is easily explained by the contemporaneous history of the doctrine of intergovernmental tax immunity. In 1927, Gillespie prevailed, and States were only permitted to tax lessees of Indian lands if Congress expressly so provided. By the time the 1938 Act was enacted, however, Gillespie had been overruled and replaced by the modern rule permitting such taxes absent congressional disapproval.13 Thus, Congress’ approaches to both the *1831927 and 1938 Acts were fully consistent with an intent to permit state taxation of nonmember lessees.14

Cotton nonetheless maintains that our decisions in White Mountain Apache Tribe v. Bracket, 448 U. S. 136 (1980), and Ramah Navajo School Bd., Inc. v. Bureau of Revenue of *184New Mexico, 458 U. S. 832 (1982), compel the conclusion that the New Mexico taxes are pre-empted by federal law. In pressing this argument, Cotton ignores the admonition included in both of those decisions that the relevant preemption test is a flexible one sensitive to the particular state, federal, and tribal interests involved. See id., at 838; Bracket, supra, at 145.

In Bracket, we addressed the question whether Arizona could impose its motor carrier license and use fuel taxes on a nonmember logging company’s use of roads located solely within an Indian reservation. Significantly, the roads at issue were “built, maintained, and policed exclusively by the Federal Government, the Tribe, and its contractors,” 448 U. S., at 150, and the State was “unable to identify any regulatory function or service [it] performed . . . that would justify the assessment of taxes for activities on Bureau and tribal roads within the reservation,” id., at 148-149. See also id., at 174 (Powell, J., concurring) (“The State has no interest in raising revenues from the use of Indian roads that cost it nothing and over which it exercises no control”). Moreover, it was undisputed in Bracket that the economic burden of the taxes ultimately fell on the Tribe. Id., at 151. Based on these facts and on our conclusion that collection of the taxes would undermine federal policy “in a context in which the Federal Government has undertaken to regulate the most minute details” of the Tribe’s timber operations, we held that the taxes were pre-empted. Id., at 149.

Ramah Navajo School Bd. involved a similar factual scenario. In the late 1960’s, New Mexico closed the only public high school that served the Ramah Navajo children. The State then sought to tax two nonmember construction firms hired by the Tribe to build a school in the reservation. As in Bracket, the State asserted no legitimate regulatory interest that might justify the tax. Ramah Navajo School Bd., supra, at 843-846. Also as in Bracket, the economic burden of the tax ultimately fell on the Tribe. And finally, again *185as in Bracket, we noted that federal law imposed a comprehensive regulatory scheme. Ramah Navajo School Bd., 458 U. S., at 839-842. We concluded: “Having declined to take any responsibility for the education of these Indian children, the State is precluded from imposing an additional burden on the comprehensive federal scheme intended to provide this education — a scheme which has ‘left the State with no duties or responsibilities.’” Id., at 843 (quoting Warren Trading Post Co. v. Arizona Tax Comm’n, 380 U. S. 685, 691 (1965)).

The factual findings of the New Mexico District Court clearly distinguish this case from both Bracket, supra, and Ramah Navajo School Bd., supra. After conducting a trial, that court found that “New Mexico provides substantial services to both the Jicarilla Tribe and Cotton,” costing the State approximately $3 million per year. App. to Juris. Statement 16. Indeed, Cotton concedes that from 1981 through 1985 New Mexico provided its operations with services costing $89,384, but argues that the cost of these services is disproportionate to the $2,293,953 in taxes the State collected from Cotton. Brief for Appellants 13-14. Neither Bracket, nor Ramah Navajo School Bd., however, imposes such a proportionality requirement on the States.15 Rather, both cases involved complete abdication or noninvolvement of the State in the on-reservation activity. The present case is also unlike Bracket and Ramah Navajo School Bd., in that the District Court found that “[n]o economic burden falls on the tribe by virtue of the state taxes,” App. to Juris. Statement 15, and that the Tribe could, in fact, increase its taxes without adversely affecting on-reservation oil and gas development, id., at 17. Finally, the District Court found that the *186State regulates the spacing and mechanical integrity of wells located on the reservation. Id., at 16. Thus, although the federal and tribal regulations in this case are extensive,16 they are not exclusive, as were the regulations in Bracket and Ramah Navajo School Bd.

We thus conclude that federal law, even when given the most generous construction, does not pre-empt New Mexico’s oil and gas severance taxes. This is not a case in which the State has had nothing to do with the on-reservation activity, save tax it. Nor is this a case in which an unusually large state tax has imposed a substantial burden on the Tribe.17 It is, of course, reasonable to infer that the New *187Mexico taxes have at least a marginal effect on the demand for on-reservation leases, the value to the Tribe of those leases, and the ability of the Tribe to increase its tax rate. Any impairment to the federal policy favoring the exploitation of on-reservation oil and gas resources by Indian tribes that might be caused by these effects, however, is simply too indirect and too insubstantial to support Cotton’s claim of pre-emption. To find pre-emption of state taxation in such indirect burdens on this broad congressional purpose, absent some special factor such as those present in Bracket and Ramah Navajo School Bd., would be to return to the pre-1937 doctrine of intergovernmental tax immunity.18 Any adverse effect on the Tribe’s finances caused by the taxation of a private party contracting with the Tribe would be ground to strike the state tax. Absent more explicit guidance from Congress, we decline to return to this long-discarded and thoroughly repudiated doctrine.

<J

Cotton also argues that New Mexico’s severance taxes — “insofar as they are imposed without allocation or apportionment on top of Jicarilla Apache tribal taxes” — impose “an un*188lawful multiple tax burden on interstate commerce.” Brief for Appellants 33. In support of this argument, Cotton relies on three facts: (1) that the State and the Tribe tax the same activity; (2) that the total tax burden on Cotton is higher than the burden on its off-reservation competitors who pay no tribal tax; and (3) that the state taxes generate revenues that far exceed the value of the services it provides on the reservation.

As we pointed out in the Merrion footnote, see n. 5, supra, a multiple taxation issue may arise when more than one State attempts to tax the same activity. If a unitary business derives income from several States, each State may only tax the portion of that income that is attributable to activity within its borders.19 See, e. g., Exxon Corp. v. Wisconsin Department of Revenue, 447 U. S. 207 (1980). Thus, in such a case, an apportionment formula is necessary in order to identify the scope of the taxpayer’s business that is within the taxing jurisdiction of each State. In this case, however, all of Cotton’s leases are located entirely within the borders of the State of New Mexico and also within the borders of the Jicarilla Apache Reservation. Indeed, they are also within the borders of the United States. There are, therefore, three different governmental entities, each of which has taxing jurisdiction over all of the non-Indian wells. Cf. Washington v. Confederated Tribes of Colville Indian Reservation, 447 U. S. 134 (1980) (Indian Tribe did not oust State of power to impose cigarette tax on on-reservation sales to non-Indian customers by imposing its own tax on transaction). *189The federal sovereign has the undoubted power to prohibit taxation of the Tribe’s lessees by the Tribe, by the State, or by both, but since it has not exercised that power, concurrent taxing jurisdiction over all of Cotton’s on-reservation leases exists. Cf. Commonwealth Edison Co. v. Montana, 453 U. S., at 617 (noting that because the taxed activity took place exclusively within Montana — although much of it on federal lands within the State — no nexus or apportionment problem existed). Unless and until Congress provides otherwise, each of the other two sovereigns has taxing jurisdiction over all of Cotton’s leases.

It is, of course, true that the total taxes paid by Cotton are higher than those paid by off-reservation producers. But neither the State nor the Tribe imposes a discriminatory tax. The burdensome consequence is entirely attributable to the fact that the leases are located in an area where two governmental entities share jurisdiction. As we noted in Merrion,. the tribal tax does “not treat minerals transported away from the reservation differently than it treats minerals that might be sold on the reservation.” 455 U. S., at 157-158. Similarly, the New Mexico taxes are administered in an evenhanded manner and are imposed at a uniform rate throughout the State — both on and off the reservation. See 106 N. M., at 521, 745 P. 2d, at 1174.

Cotton’s most persuasive argument is based on the evidence that tax payments by reservation lessees far exceed the value of services provided by the State to the lessees, or more generally, to the reservation as a whole. See n. 6, supra. There are, however, two sufficient reasons for rejecting this argument. First, the relevant services provided by the State include those that are available to the lessees and the members of the Tribe off the reservation as well as on it. The intangible value of citizenship in an organized society is not easily measured in dollars and cents; moreover, the District Court found that the actual per capita state expenditures for Jicarilla members are equal to or greater than *190the per capita expenditures for non-Indian citizens. See App. to Juris. Statement 16. Second, there is no constitutional requirement that the benefits received from a taxing authority by an ordinary commercial taxpayer — or by those living in the community where the taxpayer is located — must equal the amount of its tax obligations. See Keystone Bituminous Coal Assn. v. DeBenedictis, 480 U. S. 470, 491, n. 21 (1987). As we recently explained:

“[T]here is no requirement under the Due Process Clause that the amount of general revenue taxes collected from a particular activity must be reasonably related to the value of the services provided to the activity. Instead, our consistent rule has been:

“‘Nothing is more familiar in taxation than the imposition of a tax upon a class or upon individuals who enjoy no direct benefit from its expenditure, and who are not responsible for the condition to be remedied.

“‘A tax is not an assessment of benefits. It is, as we have said, a means of distributing the burden of the cost of government. The only benefit to which the taxpayer is constitutionally entitled is that derived from his enjoyment of the privileges of living in an organized society, established and safeguarded by the devotion of taxes to public purposes. Any other view would preclude the levying of taxes except as they are used to compensate for the burden on those who pay them, and would involve abandonment of the most fundamental principle of government — that it exists primarily to provide for the common good.’ Carmichael v. Southern Coal & Coke Co., 301 U. S. 495, 521-523 (1937) (citations and footnote omitted).

“There is no reason to suppose that this latitude afforded the States under the Due Process Clause is somehow divested by the Commerce Clause merely because the taxed activity has some connection to interstate com*191merce; particularly when the tax is levied on an activity conducted within the State.” Commonwealth Edison Co., supra, at 622-623.

Cotton, in effect, asks us to divest New Mexico of its normal latitude because its taxes have “some connection” to commerce with the Tribe. The connection, however, is by no means close enough. There is simply no evidence in the record that the tax has had an adverse effect on the Tribe’s ability to attract oil and gas lessees. It is, of course, reasonable to infer that the existence of the state tax imposes some limit on the profitability of Indian oil and gas leases — just as it no doubt imposes a limit on the profitability of off-reservation leasing arrangements — but that is precisely the same indirect burden that we rejected as a basis for granting non-Indian contractors an immunity from state taxation in Helvering v. Mountain Producers Corp., 303 U. S. 376 (1938); Oklahoma Tax Comm’n v. United States, 319 U. S. 598 (1943); Oklahoma Tax Comm’n v. Texas Co., 336 U. S. 342 (1949); Moe v. Confederated Salish and Kootenai Tribes of Flathead Reservation, 425 U. S. 463 (1976); and Washington v. Confederated Tribes of Colville Indian Reservation, 447 U. S. 134 (1980).

V

In our order noting probable jurisdiction we invited the parties to address the question whether the Tribe should be treated as a State for the purpose of determining whether New Mexico’s taxes must be apportioned. All of the Indian tribes that have filed amicus curiae briefs addressing this question — including the Jicarilla Apache Tribe — have uniformly taken the position that Indian tribes are not States within the meaning of the Commerce Clause. This position is supported by the text of the Clause itself. Article I, § 8, cl. 3, provides that the “Congress shall have Power ... To regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes.” Thus, the Commerce Clause draws a clear distinction between “States” and “In*192dian Tribes.” As Chief Justice Marshall observed in Cherokee Nation v. Georgia, 5 Pet. 1, 18 (1831): “The objects to which the power of regulating commerce might be directed, are divided into three distinct classes — foreign nations, the several states, and Indian Tribes. When forming this article, the convention considered them as entirely distinct.” In fact, the language of the Clause no more admits of treating Indian tribes as States than of treating foreign nations as States. See ibid.

It is also well established that the Interstate Commerce and Indian Commerce Clauses have very different applications. In particular, while the Interstate Commerce Clause is concerned with maintaining free trade among the States even in the absence of implementing federal legislation, see McLeod v. J. E. Dilworth Co., 322 U. S. 327, 330 (1944); Pike v. Bruce Church, Inc., 397 U. S. 137 (1970), the central function of the Indian Commerce Clause is to provide Congress with plenary power to legislate in the field of Indian affairs, see Morton v. Mancari, 417 U. S. 535, 551-552 (1974); F. Cohen, Handbook of Federal Indian Law 207-208, and nn. 2, 3 and 9-11 (1982). The extensive case law that has developed under the Interstate Commerce Clause, moreover, is premised on a structural understanding of the unique role of the States in our constitutional system that is not readily imported to cases involving the Indian Commerce Clause. Most notably, as our discussion of Cotton’s “multiple taxation” argument demonstrates, the fact that States and tribes have concurrent jurisdiction over the same territory makes it inappropriate to apply Commerce Clause doctrine developed in the context of commerce “among” States with mutually exclusive territorial jurisdiction to trade “with” Indian tribes.

Accordingly, we have no occasion to modify our comment on this question in the Bracker case:

“Tribal reservations are not States, and the differences in the form and nature of their sovereignty make it *193treacherous to import to one notions of pre-emption that are properly applied to the other.” 448 U. S., at 143.

The judgment of the New Mexico Court of Appeals is

Affirmed.

dissenting.

Although the Jicarilla Apache Tribe is not a party to the appeal, this case centrally concerns “the boundaries between state regulatory authority and [the Tribe’s] self-government.” White Mountain Apache Tribe v. Bracker, 448 U. S. 136, 141 (1980). The basic principles that define those boundaries are well established. The Court today, while faithfully reciting these principles, is less faithful in their application.

Pre-emption is essentially a matter of congressional intent. In this case, our goal should be to determine whether the State’s taxation of Cotton Petroleum’s reservation oil production is consistent with federal Indian policy as expressed in relevant statutes and regulations. First and foremost, we must look to the statutory scheme Congress has established to govern the activity the State seeks to tax in order to see whether the statute itself expresses Congress’ views on the question of state taxation. As the discussion in Part I below reveals, the statute most relevant to this case makes clear that Congress intended to foreclose the kind of tax New Mexico has imposed. Second, we must consider other indications of whether federal policy permits the tax in question. Part II below demonstrates that, under established principles, state taxation is pre-empted by federal and tribal interests in this case. Because the record is more than adequate to demonstrate the pre-emptive force of federal and tribal interests, I dissent.1

*194I

The most relevant statute is the Indian Mineral Leasing Act of 1938, 52 Stat. 347, 25 U. S. C. §396a et seq. (1938 Act), pursuant to which the Jicarilla Apache entered into mineral leases with appellant Cotton Petroleum. The 1938 Act is silent on the question of state taxation. But, as by this Court in Montana v. Blackfeet Tribe, 471 U. S. 759 (1985), the silence of the 1938 Act is eloquent and argues forcefully against the result reached by the majority.

In Montana, the State sought to tax the Blackfeet Tribe’s royalty interests under oil and gas leases held, pursuant to the 1938 Act, by non-Indian lessees operating on the The State sought to do so despite the fact that the 1938 Act contains no express authorization for any state tax on such leases. The State based its claim of taxation authority on a 1924 statute enacted to permit oil and gas leasing on reservations created by treaty.2 Act of May 29, 1924, ch. 210, 43 Stat. 244, 25 U. S. C. § 398 (1924 Act). The 1924 Act contained a proviso that “the production of oil and gas and other minerals on such lands may be taxed by the State in which said lands are located in all respects the same as on unrestricted lands, and the Secretary of the is authorized and directed to cause to be paid the tax so *195assessed against the royalty interests on said lands.” The State took the position that the 1938 Congress could not be presumed by mere silence to have abrogated the law permitting state taxation. 471 U. S., at 765-766.

In Montana, we squarely rejected the State’s argument. After noting that the 1938 Act was “comprehensive legislation,” id., at 763, containing a general repealer of all statutory provisions “‘inconsistent herewith,’” id., at 764, quoting § 7 of the 1938 Act, see note following 25 U. S. C. § 396a, we held that, under the canons of construction applicable to laws governing Indians, the general repealer clause could not be taken as implicitly incorporating consistent provisions of earlier laws. Rather, in the Indian context, clear congressional consent to state taxation was required and, on that point, we found no “indication that Congress intended to incorporate implicitly in the 1938 Act the taxing authority of the 1924 Act.” 471 U. S., at 767. Interpreting the 1938 Act as preserving the taxing authority of the 1924 Act, we held, would not “satisfy the rule requiring that statutes be construed liberally in favor of the Indians.” Ibid. In addition, we observed that such an interpretation would undermine the purposes of the 1938 Act as reflected in its legislative history: to achieve uniformity in tribal leasing, to harmonize tribal leasing with the goals of the Indian Reorganization Act, Act of June 18, 1934, ch. 576, § 16, 48 Stat. 987, codified at 25 U. S. C. § 476 et seq. (IRA), and “to ensure that Indians receive ‘the greatest return from their property.’” 471 U. S., at 767, n. 5.

The majority appropriately acknowledges that Congress knew when it enacted the 1938 Act that a statute governing tribal leases that failed expressly to authorize state taxation of Indian royalty interests would have the effect of leaving the States without the power to tax those interests. Ante, at 183, n. 14. Thus, the clear import of our decision in Montana is that Congress’ silence in 1938 expressed an intent substantially to narrow state taxing authority.

*196But the majority takes the position that the 1938 Act’s silence means something completely different when it comes to the kind of taxation at issue here, and expends considerable energy attempting to support that view. The majority argues that the same silence that reflected an intent to prohibit state taxation of Indian tribes’ royalty interests was “fully consistent with an intent to permit state taxation of nonmember lessees,” ante, at 183 (emphasis added). The majority notes that when the pre-1938 mineral-leasing statutes were enacted (including the 1927 Act, 44 Stat. 1347, 25 U. S. C. §398a et seq., which is of the greatest relevance here, see n. 2, supra), express congressional authorization was required not only for direct taxes on Indians (or other “sovereigns”), but also for taxes on those who contracted with Indians. See, e. g., Gillespie v. Oklahoma, 257 U. S. 501 (1922) (applying intergovernmental immunity doctrine to a tax on the net income of the non-Indian holder of a reservation mineral lease). In contrast, “[b]y the time the 1938 Act was enacted, . . . Gillespie had been overruled and replaced by the modern rule permitting such taxes absent congressional disapproval.” Ante, at 182. From this, the majority infers that because Congress knew in 1938 that it could maintain the pre-1938 status quo regarding lessee taxation simply by saying nothing, Congress’ silence is consistent with an intent to maintain that status quo.

The argument that the 1938 congressional silence regarding lessee taxation is consistent with an intent to permit such taxation cannot, for two reasons, withstand close scrutiny. First, even if the majority is correct in seeking the meaning of Congress’ silence in changes in this Court’s intergovernmental tax immunity jurisprudence, the facts defeat the majority’s theory. Second, and fundamentally, the majority’s court-centered approach fails to give due weight to a far more significant intervening event: the major change in federal Indian policy embodied in the IRA.

*197The ease which overruled Justice Holmes’ opinion for the Court in Gillespie was Helvering v. Mountain Producers Corp., 303 U. S. 376 (1938). Mountain Producers was decided on March 7, 1938. The majority, indeed, is correct that the 1938 Act was enacted on May 11, 1938, after that case was decided. But a review of the history of the 1938 Act reveals that it had assumed final form well before this Court’s decision in Mountain Producers. The majority’s chronology thus is somewhat misleading, at least if the realities of the legislative process are to have any relevance to the analysis of legislative intent.

The 1938 Act was drafted by the Department of the Interior and was submitted to the respective Committees on Indian Affairs of the House and Senate on June 17, 1937. See S. Rep. No. 985, 75th Cong., 1st Sess., 1 (1937) (Senate Report); H. R. Rep. No. 1872, 75th Cong., 3d Sess., 1 (1938) (House Report).3 The proposed bill was reported out of the Senate Committee in July 1937, with a recommendation that it be passed without amendment. Senate Report, at 1. The bill was passed by the Senate without debate on August 6, 1937. See 81 Cong. Rec. 8399. The bill was reported out of the House Committee on Indian Affairs on March 3, 1938, again with a recommendation that it pass without amendment. House Report, at 1. All this took place before the March 7, 1938, decision in Mountain Producers, during a period in which, the majority acknowledges, the proposed statute’s silence on the question of taxation would have meant that the States had no power to tax non-Indian lessees’ *198oil and gas production. The House passed the bill, also without debate, on May 2, 1938. See 83 Cong. Rec. 6057-6058.

Thus, although the majority is technically correct that the 1938 Act did not become law until after the announcement of this Court’s decision in Mountain Producers, the legislation was formulated, considered by the House and Senate Committees, referred out of the Committees without amendment, and passed by the Senate, all before Mountain Producers on March 7, 1938, changed the law of intergovernmental tax immunity. Up until that point, the clear meaning of the statute, as our decision in Montana makes clear, is that the State lacked power to impose the tax at issue in this case. There is no evidence that the change in the law wrought by Mountain Producers was brought to the attention of the House. It defies historical sense to make Mountain Producers the centerpiece of the interpretation of a statute which reached final form before Mountain Producers was decided.4

The Court in Montana put forward a more sensible explanation of the absence of state taxation authority in the 1938 Act. As the relevant House and Senate Reports explain, the 1938 Act was crafted, proposed, and enacted in light of the recently enacted IRA. The IRA worked a funda*199mental change in federal Indian law marked by two principal goals: “ ‘to rehabilitate the Indian’s economic life and to give him a chance to develop the initiative destroyed by a century of oppression and paternalism.’ ” Mescalero Apache Tribe v. Jones, 411 U. S. 145, 152 (1973), quoting H. R. Rep. No. 1804, 73d Cong., 2d Sess., 6 (1934). In reviewing pre-1934 Indian mineral-leasing statutes, the Interior Department found them wanting in both respects. The statutes not only gave the Indians no “voice” in the granting of leases, but also were not “adequate to give Indians the greatest return from their property.” House Report, at 2; Senate Report, at 2. The 1938 Act was proposed to “bring all mineral-leasing matters in harmony with the Indian Reorganization Act” in these respects. House Report, at 3; Senate Report, at 3. The Court observed in Montana that “these purposes would be undermined” by treating the 1938 Act as explicitly or implicitly leaving the taxation provisions of prior statutes in force. 471 U. S., at 767, n. 5.5

The majority’s observation, ante, at 182, that “[t]here is ... no history of tribal independence from state taxation of these lessees to form a ‘backdrop’ against which the 1938 Act must be read” cannot be dispositive. The IRA, enacted only a few years before the 1938 Act, is itself sufficient “backdrop” to inform our interpretation, for the IRA marked the rejection of all the assumptions upon which prior statutes providing for state taxation of reservation mineral production had been based.

The expectation that animated Indian policy under the General Allotment Act of 1887, ch. 119, 24 Stat. 388, was *200that at the expiration of a 25-year trust period, there would be no difference between Indians and other citizens: tribal life would come to an end, the Indians would be assimilated and fully subject to state governmental authority, Indian lands would be freely alienable to non-Indians and subject to state taxation, and surplus lands would be opened to private development. See generally F. Cohen, Handbook of Federal Indian Law 131-132 (1982); Readjustment of Indian Affairs: Hearings on H. R. 7902 before the House Committee on Indian Affairs (History of the Allotment Policy), 73d Cong., 2d Sess., pt. 9, pp. 428-489 (1934); Blackfeet Tribe v. Montana, 729 F. 2d 1192, 1195 (CA9 1984), aff’d, 471 U. S. 759 (1985).

With the passage of time, eventual state control remained the goal of the allotment policy, but delays in the full implementation of that policy became a matter of concern to the States. This was particularly evident in the area of mineral leasing. Such leasing for periods of up to 10 years had been authorized by statute in 1891, Act of Feb. 28, 1891, ch. 383, § 3, 26 Stat. 795, but it became increasingly clear that longer-term leases were an economic necessity. A pattern soon developed: in return for Congress’ extending the period during which mineral rights would be reserved to the Indian tribes, States were given the power to tax mineral production. See 3 Indians of the United States, Hearings before a Subcommittee of the House Committee on Indian Affairs 191-192, 281, 444-445 (1920). The taxation proviso in the 1924 Act, which was included at the insistence of members of the Subcommittee, was true to that pattern. See H. R. No. 386, 68th Cong., 1st Sess. (1924); see generally Brief for United States as Amicus Curiae in Montana v. Blackfeet Tribe, O. T. 1983, No. 83-2161, pp. 16-26.

By 1927, when Congress addressed the problem of oil and gas leasing on Executive Order reservations, the States were anxious to open those lands for mineral development and the debate in Congress squarely addressed the conflicting inter*201ests of States and Indian tribes. The Attorney General had issued a controversial opinion that the Mineral Lands Leasing Act of 1920, 41 Stat. 437, did not apply to Executive Order reservations, 34 Op. Atty. Gen. 181 (1924) (opinion of then-Attorney General Harlan F. Stone), and the matter was in litigation. See Development of Oil and Gas Mining Leases on Indian Reservations, Hearings on S. 1722 and S. 3159 before the Subcommittee of the Senate Committee on Indian Affairs, 69th Cong., 1st Sess., 29-30 (1926). If the Attorney General’s position did not prevail in the courts, the Indians would receive no income from mineral production on Executive Order reservations, and any claim of Indian entitlement to those reservations would be substantially undermined. In that uncertain legal climate, the 1927 Act is best viewed as a compromise: Indian interests acquiesced in the States’ claim that they had a right to increase their general revenues by sharing in the profits of reservation mining activities; in return, the Indians avoided legislation that would have obliterated any hope of obtaining recognition of their legal entitlement to Executive Order lands. See id., at 9, 55, 61, 63, 98-99. See also Hearings on S. 3159 and S. 4152 before the Senate Committee on Indian Affairs, 69th Cong., 1st Sess., 14, 24-25, 33-34 (1926).6

The political climate changed dramatically with the passage in 1934 of the IRA, in which “ ‘ [t]he policy of allotment and sale of surplus reservation land was repudiated’ ” as antithetical to tribal interests. Moe v. Confederated Salish and Kootenai Tribes of Flathead Reservation, 425 U. S. 463, 479 (1976), quoting Mattz v. Arnett, 412 U. S. 481, 496, n. 18 (1973). It would be a mistake to impute the political compromises of the allotment period into legislation enacted soon after the pas*202sage of the IRA, legislation expressly tailored to bring Indian mineral policy into line with a radically altered set of assumptions about the political and economic future of the Indians.

Furthermore, the IRA embodied an approach to tribal independence which would be undone by limiting Indian tribes to those powers they had been permitted to exercise in the past. The Department of the Interior, in interpreting the IRA at Congress’ request, realized that the assertions of Indian automony that the IRA sought to foster would seem novel, and would likely come at the expense of settled state expectations.

“It is a fact that State governments and administrative officials have frequently trespassed upon the realm of tribal autonomy, presuming to govern the Indian tribes through State law or departmental regulation or arbitrary administrative fiat, but these trespasses have not impaired the vested legal powers of local self-government which have been recognized again and again when these trespasses have been challenged by an Indian tribe. Tower and authority rightfully conferred do not necessarily cease to exist in consequence of long nonuser’. . . . The [IRA], by affording statutory recognition of these powers of local self-government and administrative assistance in developing adequate mechanisms for such government, may reasonably be expected to end the conditions that have in the past led the Interior Department and various State agencies to deal with matters that are properly within the legal competence of the Indian tribes themselves.” Powers of Indian Tribes, 55 I. D. 14, 28-29 (1934).

The Department noted: “Chief among the powers of sovereignty recognized as pertaining to an Indian tribe is the power of taxation.” Id., at 46. It would be entirely consistent with the spirit of the IRA for the Department, and for Congress, to have done away with the express authorization of state taxation in order to leave room for Indians to operate *203in the sphere of taxation unimpeded by the States. That Indians had never before asserted the right to freedom from state taxation was simply a product of the unfortunate state of affairs that the IRA sought to remedy.

In sum, we are given to choose between two possible interpretations of the silence of the 1938 Act. One, adopted by the majority, focuses on the change in this Court’s intergovernmental immunity doctrine which took place at the very end of the process leading to the 1938 Act. The other focuses on a fundamental change in congressional Indian policy which took place shortly before the process began, and was expressly noted as its motivating force. The latter interpretation is clearly the more compelling. I must conclude that, contrary to the majority’s view, the silence of the 1938 Act is not consistent with a congressional intent that non-Indian lessees of Indian mineral lands shall be subject to state taxation for their on-reservation activities.7 This conclusion does not constitute, as the majority says, a “return to [the] long-discarded and thoroughly repudiated doctrine” of constitutional intergovernmental tax immunity. Ante, at 187. Rather, it reflects a fuller understanding of the policies underlying federal Indian law in the mid- to late-1930’s and continuing, in relevant part, into the present time.

II

Even if we did not have such direct evidence of Congress’ intent to preclude state taxation of non-Indian oil production on Indian lands, that conclusion would be amply supported by a routine application of the traditional tools of Indian pre-emption analysis. The majority concludes other*204wise because it distorts the legal standard it purports to apply. Instead of engaging in a careful examination of state, tribal, and federal interests required by our precedents, see e. g., Ramah Navajo School Board, Inc. v. Bureau of Revenue of New Mexico, 458 U. S. 832, 838 (1982), the majority has adopted the principle of “the inexorable zero.” Teamsters v. United States, 431 U. S. 324, 342, n. 23 (1977). Under the majority’s approach, there is no pre-emption unless the States are entirely excluded from a sphere of activity and provide no services to the Indians or to the lessees they seek to tax. That extreme approach is hardly consistent with the “flexible” standard the majority purports to apply. Ante, at 184.

The Court has identified “two independent but related barriers to the assertion of state regulatory authority over tribal reservations.” White Mountain Apache Tribe v. Bracker, 448 U. S., at 142. The exercise of state authority may be impermissible solely on the ground that the state intervention would interfere with “the right of reservation Indians to make their own laws and be ruled by them.” Williams v. Lee, 358 U. S. 217, 220 (1959). Alternatively, state law may be pre-empted by the existence of a comprehensive federal regulatory scheme governing the subject matter. See, e. g., Warren Trading Post Co. v. Arizona Tax Comm’n, 380 U. S. 685, 688-690 (1965). These methods of analysis overlap. Indian sovereignty is not a “platonic” concept. McClanahan v. Arizona State Tax Comm’n, 411 U. S. 164, 172 (1973). It is a growing tradition, actively supported by federal legislation. Our pre-emption cases recognize that “federal law . . . reflects . . . related federal and tribal interests,” and that “the . . . encouragement of [Indian] sovereignty in congressional Acts promoting tribal independence and economic development” must inform our pre-emption analysis. Ramah Navajo, 458 U. S., at 838. In this case, all the elements that traditionally have resulted in a finding of federal pre-emption are present.

*205Federal regulation of leasing of Indian oil lands “is both comprehensive and pervasive.” Id., at 839. Provisions of the 1938 Act regulate all stages of the process of oil and gas leasing and production on Indian reservations. The auction or bidding process through which leases are acquired is supervised by the Department of the Interior. 25 U. S. C. §396b. Successful lessees must furnish a bond to secure compliance with lease terms, § 396c, and each lessee’s operations are in all respects subject to federal rules and regulations, § 396d. Longstanding regulations promulgated pursuant to the 1938 Act govern the minute details of the bidding process, 25 CFR §211.3 (1988), and give the Secretary of the Interior the power to reject bids that are not in the best interest of the Indian lessor, § 211.3(b). Federal law sets acreage limitations, §211.9, the term of each lease, §211.10, and royalty rates, methods, and times of payment, §§211.13 and 211.16. Turning to the regulation of the lessee’s operations, federal law controls when operations may start, §211.20, and federal supervisory personnel are empowered to ensure the conservation of resources and prevention of waste, §§211.19-211.21. Additional restrictions are placed on lessees by the Federal Oil and Gas Royalty Management Act of 1982, 96 Stat. 2447, 30 U. S. C. §1701 et seq., which further safeguards tribal interests by imposing additional inspection, collection, auditing, security, and conservation requirements on lessees.

In addition, the Jicarilla Apache, as expressly authorized by their Constitution, have enacted regulations of their own to supplement federal guidelines, and have created a tribal Oil and Gas Administration to exercise tribal authority in this area.8 See Jicarilla Apache Tribal Code, Tit. 18, ch. 1, §§ 1-7 (1987) and their Revised Constitution, Art. XI, § 1(a)(3). In*206deed, just as we earlier found of the Mescalero Apache: “The Tribe has engaged in a concerted and sustained undertaking to develop and manage the reservation’s . . . resources specifically for the benefit of its members.” New Mexico v. Mescalero Apache Tribe, 462 U. S. 324, 341 (1983).

The majority acknowledges that federal and tribal regulations in this case are extensive. But because the District Court found that the State regulates spacing and the mechanical integrity of wells, and that federal and tribal regulations are therefore not “exclusive,” the majority concludes without further ado that there is sufficient state activity to support the State’s claimed authority to tax.9 The majority’s reliance on the proposition that “[t]his is not a case in which the State has had nothing to do with the on-reservation activity, save tax it,” ante, at 186, reflects a mechanical and absolutist approach to the delicate issue of pre-emption that this Court expressly has repudiated. White Mountain Apache, 448 U. S., at 145. “[Cjomplete abdication or noninvolvement,” ante, at 185, has never been the applicable standard.

The taxes the State seeks to impose “would threaten the overriding federal objective of guaranteeing Indians that they will ‘receive . . . the benefit of whatever profit [their oil and gas reserves are] capable of yielding,”’ and would “reduc[e] tribal revenues and diminis[h] the profitability of the enterprise for potential contractors.” White Mountain Apache, 448 U. S., at 149. State taxes would also reduce the funds available to oil and gas producers to meet the financial obligations placed upon them by the extensive federal and tribal regulatory schemes. Ibid. Tribal and federal regulations clearly leave no room for these taxes. See id., at 151, n. 15.

*207Just as the majority errs by adopting a standard of “exclusivity,” it places undue significance on the fact that the State made some expenditures that benefited Cotton Petroleum’s on-reservation activities.10 Concededly, the State did spend some money on the reservation for purposes directly and indirectly related to oil and gas production. It is clear on this record, however, that the infrastructure which supports oil and gas production on the Jicarilla Apache Reservation is provided almost completely by the federal and tribal governments rather than by the State. Indeed, the majority appears to accept the fact that the state taxes are vastly disproportionate, ante, at 185, as well it must: $89,384 in services, as compared with $2,293,953 in taxes, speaks for itself.11 But the majority deems this fact legally irrelevant in order to *208avoid imposing a “proportionality requirement” that would be inconsistent with the notion that taxation is not based on a quid pro quo. Ante, at 185, n. 15. That notion, drawn from Due Process and Commerce Clause analysis, see ante, at 189-190, is inapposite in the pre-emption context. Preemption analysis calls for a close consideration of conflicting interests and of their potential impact on one another. Under the majority’s analysis, insignificant state expenditures, reflecting minimal state interests, are sufficient to support state interference with significant federal and tribal interests. The exclusion of all sense of proportion has led to a result that is antithetical to the concerns that animate our Indian pre-emption jurisprudence.

Finally, the majority sorely underestimates the degree to which state taxation of oil and gas production adversely affects the interests of the Jicarilla Apache. Assuming that the Tribe continues to tax oil and gas production at present levels, on-reservation taxes will remain 75% higher (14% as opposed to 8% of gross value) than off-reservation taxes within the State. The state trial court was not disturbed by this fact: it found that Cotton Petroleum had plans to dig new wells, and took that to be proof positive that the taxes imposed by the State did not deter drilling. But the court failed to recognize that Cotton Petroleum’s new wells were infield (or “infill”) wells, drilled between existing producing wells to increase the efficiency of drainage on lands already leased. Tr. 41-42, 68; see H. Williams & C. Meyers, Oil and Gas Terms 468 (7th ed. 1987). An infill well is essentially a no-risk proposition, in that there is little doubt that the well will be productive. Therefore, Cotton Petroleum’s willingness to drill infill wells does not reflect its willingness to develop new lands. Federal and tribal interests legitimately include long-term planning for development of lease revenues on new lands, where there is greater economic risk, see Tr. 450, and a greater probability that difference in tax rates will have an adverse effect on a producer’s willingness to drill new wells and on the competitiveness of Jicarilla *209leases. Id., at 68, 504. “[B]oth the rate at which mining companies acquire Indian land leases and the rate at which they develop them are dependent on the future balance between the deterrents to and the advantages of Indian land leasing. Where the balance will be struck cannot be predicted, for there are simply too many variables involved.” Federal Trade Commission, Staff Report on Mineral Leasing on Indian Lands 48 (1975) (FTC Report). Dual state and tribal taxation inevitably affects that balance.

In weighing the effect of state taxation on tribal interests, logic dictates that it is necessary to consider not only the size of the tax, but also the importance of the taxed activity to the tribal economy. See California v. Cabazon Band of Mission Indians, 480 U. S. 202, 218 (1987) (noting, in invalidating state regulation of tribal bingo operations, that bingo games constituted the sole source of tribal income). In this case, too, it is undisputed that oil and gas production is the Jicarilla Apache economy — a common pattern in reservations with substantial oil and gas reserves. See Tr. 159 (oil and gas royalties account for 90% of tribal income); FTC Report, at 10; Anders, Indians, Energy, and Economic Development, 9 J. Contemp. Business 57 (1980).

Furthermore, where, as here, the Tribe has made the decision to tax oil and gas producers, the long-term impact of state taxation on the Tribe’s freedom of action in the sphere of taxation must also be considered.12 Tribal taxation has been widely perceived as necessary to protect Indian interests.13 The fact that the Jicarilla Apache have seen fit to impose their own taxes renders the threat to tribal interests *210which is always inherent in state taxation all the more apparent. 14 The market can bear only so much taxation, and it is inevitable that a point will be reached at which the State’s taxes will impose a ceiling on tribal tax revenues. That the Jicarilla Apache have not yet raised their taxes to a level at which the combined effect of tribal and state taxation has been proved to diminish tribal revenues cannot be dis-positive. Our decisions have never required a case-specific showing that state taxation in fact has deterred tribal activity; the potential for conflict is sufficient.

The majority observes that this is not “a case in which an unusually large state tax has imposed a substantial burden on the Tribe,” ante, at 186, and deems the tribal interest “indirect and . . . insubstantial,” ante, at 187. But the majority does not explain why interferences with federal policy of only the dramatic magnitude of the tax at issue in Montana v. Crow Tribe, 484 U. S. 997 (1988), meet the pre-emption threshold. In Warren Trading Post Co. v. Arizona Tax Comm’n, 380 U. S., at 691, the Court rejected a 2% tax on the gross proceeds of a non-Indian trader on an Indian reservation because it put “financial burdens on [the trader] or' the Indians ... in addition to those Congress or the tribes have prescribed, and could thereby disturb and disarrange the statutory plan Congress set up in order to protect Indians.” Indeed, the dissenters in White Mountain Apache characterized the less-than-1% tax struck down in that case *211as “relatively trivial” and “unlikely to have a serious adverse impact on the tribal business,” 448 U. S., at 159 (Stevens, J., dissenting). That the tax burden was held sufficient to support a finding of pre-emption in White Mountain Apache and Warren Trading Post undermines the majority’s position here.

Ill

In sum, under established Indian pre-emption principles, the case before us should have been straightforward. We deal here with state taxes on oil producers engaged in the development of the natural resource upon which the economic future of the Jicarilla Apache depends. The federal statute governing the producers’ activities, unlike its historical predecessors, contains no express authorization of state taxation. That statute was enacted in a period in which concern with tribal sovereignty and tribal self-sufficiency was at the very core of federal Indian policy. Pursuant to that statute, the Federal Government regulates every aspect of the producers’ activities to advance the Indians’ best interests. The statute also encourages tribes to assert their own sovereign authority in this area; the Jicarilla Apache have done so through regulation and taxation. On the other side of the balance, New Mexico asserts little more than a desire to increase its general revenues at the expense of tribal economic development. That purpose “is insufficient to justify the additional burdens imposed by the tax on the comprehensive federal scheme . . . and on the express federal policy of encouraging Indian self-sufficiency in [this] area.” Ramah Navajo, 458 U. S., at 845.

I respectfully dissent.