delivered the opinion of the Court.

Section 310(b)(1) of the Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA), Pub. L. 97-248, 96 Stat. 596, 26 U. S. C. § 103(j)(1), removes the federal income tax exemption for interest earned on publicly offered long-term bonds issued by state and local governments unless those bonds are *508issued in registered form.1 This original jurisdiction case presents the issues whether § 310(b)(1) of TEFRA either (1) violates the Tenth Amendment and constitutional principles of federalism by compelling States to issue bonds in registered form or (2) violates the doctrine of intergovernmental tax immunity by taxing the interest earned on unregistered state bonds.

I

Historically, bonds have been issued as either registered bonds or bearer bonds. These two types of bonds differ in the mechanisms used for transferring ownership and making payments. Ownership of a registered bond is recorded on a central list, and a transfer of record ownership requires entering the change on that list.2 The record owner automatically receives interest payments by check or electronic transfer of funds from the issuer’s paying agent. Ownership of a bearer bond, in contrast, is presumed from possession and is transferred by physically handing over the bond. The bondowner obtains interest payments by presenting bond coupons to a bank that in turn presents the coupons to the issuer’s paying agent.

In 1982, Congress enacted TEFRA, which contains a variety of provisions, including § 310, designed to reduce the federal deficit by promoting compliance with the tax laws. Congress had become concerned about the growing magnitude of tax evasion; Internal Revenue Service (IRS) studies indicated that unreported income had grown from an estimated range of $31.1 billion to $32.2 billion in 1973 to a range of $93.3 billion to $97 billion in 1981. Compliance Gap: Hearing before the Subcommittee on Oversight of the Internal *509Revenue Service of the Senate Committee on Finance, 97th Cong., 2d Sess., 126 (1982). Unregistered bonds apparently became a focus of attention because they left no paper trail and thus facilitated tax evasion. Then Assistant Secretary of the Treasury for Tax Policy John Chapoton testified before the House Ways and Means Committee that a registration requirement would help prevent tax evasion because bearer bonds often represent unreported and untaxed income that, without a system of recorded ownership, the IRS has difficulty reconstructing. Hearings on H. R. 6300 before the House Committee on Ways and Means, 97th Cong., 2d Sess., 35 (1982). He also expressed concern that bearer bonds were being used to avoid estate and gift taxes and as a medium of exchange in the illegal sector. Ibid. In reporting out the bill containing the provision that eventually became § 310 of TEFRA, the Senate Finance Committee Report expressed the same concerns:

“The committee believes that a fair and efficient system of information reporting and withholding cannot be achieved with respect to interest-bearing obligations as long as a significant volume of long-term bearer instruments is issued. A system of book-entry registration will preserve the liquidity of obligations while requiring the creation of ownership records that can produce useful information reports with respect to both the payment of interest and the sale of obligations prior to maturity through brokers. Furthermore, registration will reduce the ability of noncompliant taxpayers to conceal income and property from the reach of the income, estate, and gift taxes. Finally, the registration requirement may reduce the volume of readily negotiable substitutes for cash available to persons engaged in illegal activities.” S. Rep. No. 97-494, Vol. 1, p. 242 (1982).

Section 310 was designed to meet these concerns by providing powerful incentives to issue bonds in registered form.

*510Because §310 aims to address the tax evasion concerns posed generally by unregistered bonds, it covers not only state bonds but also bonds issued by the United States and private corporations. Section 310(a) requires the United States to issue publicly offered bonds with a maturity of more than one year in registered form.3 With respect to similar bonds issued by private corporations, §§ 310(b)(2) — (6) impose a series of tax penalties on nonregistration. Corporations declining to issue the covered bonds in registered form lose tax deductions and adjustments for interest paid on the bonds, §§ 310(b)(2) and (3), and must pay a special excise tax on the bond principal, § 310(b)(4). Holders of these unregistered corporate bonds generally cannot deduct capital losses or claim capital-gain treatment for any losses or gains sustained on the bonds. §§ 310(b)(5) and (6). Section 310 (b)(1) completes this statutory scheme by denying the federal income tax. exemption for interest earned on state bonds to owners of long-term publicly offered state bonds that are not issued in registered form.

South Carolina invoked the original jurisdiction of this Court, contending that § 310(b)(1) is constitutionally invalid under the Tenth Amendment and the doctrine of intergovernmental tax immunity. We granted South Carolina leave to file the instant complaint against the Secretary of the Treasury of the United States, South Carolina v. Regan, 465 U. S. 367 (1984), and appointed as Special Master the Honorable Samuel J. Roberts, 466 U. S. 948 (1984). The National Governors’ Association (NGA) intervened.4 After conducting hearings and taking evidence, the Special Master concluded that § 310(b)(1) was constitutional and recommended *511entering judgment for the defendant. South Carolina and the NGA filed exceptions to various factual findings of the Special Master and to the Master’s legal conclusions concerning their constitutional challenges.

I — I hH

We address the claim that § 310(b)(1) violates the Tenth Amendment first.5 South Carolina and the NGA contend, and the Master found, that § 310 effectively requires States to issue bonds in registered form, noting that if States issued bonds in unregistered form, competition from other nonexempt bonds would force States to increase the interest paid on state bonds by 28-35%, and that even though almost all state bonds were issued in bearer form before § 310 became effective, since then no State has issued a bearer bond. Report of Special Master 2, 23-24. South Carolina and the NGA thus argue that, for purposes of Tenth Amendment analysis, we must treat § 310 as if it simply banned bearer bonds altogether without giving States the option to issue nonexempt bearer bonds. The Secretary does not dispute the finding that §310 effectively requires registration, see Brief for Defendant 19 (urging the Court to adopt all the Master’s findings), preferring to argue that § 310 survives Tenth Amendment scrutiny because a blanket prohibition by Congress on the issuance of bearer bonds can apply to States without violating the Tenth Amendment. For the purposes of Tenth Amendment analysis, then, we treat § 310 as if it directly regulated States by prohibiting outright the issuance of bearer bonds.6

*512A

The Tenth Amendment limits on Congress’ authority to regulate state activities are set out in Garcia v. San Antonio Metropolitan Transit Authority, 469 U. S. 528 (1985). Garcia holds that the limits are structural, not substantive— i. e., that States must find their protection from congressional regulation through the national political process, not through judicially defined spheres of unregulable state activity. Id., at 537-554. South Carolina contends that the political process failed here because Congress had no concrete evidence quantifying the tax evasion attributable to unregistered state bonds and relied instead on anecdotal evidence that taxpayers have concealed taxable income using bearer bonds. It also argues that Congress chose an ineffective remedy by requiring registration because most bond sales are handled by brokers who must file information reports regardless of the form of the bond and because beneficial ownership of registered bonds need not necessarily be recorded.

Although Garcia left open the possibility that some extraordinary defects in the national political process might render congressional regulation of state activities invalid under the Tenth Amendment, the Court in Garcia had no occasion to identify or define the defects that might lead to such invalidation. See id., at 556. Nor do we attempt any definitive articulation here. It suffices to observe that South *513Carolina has not even alleged that it was deprived of any right to participate in the national political process or that it was singled out in a way that left it politically isolated and powerless. Cf. United States v. Carotene Products Co., 304 U. S. 144, 152, n. 4 (1938). Rather, South Carolina argues that the political process failed here because § 310(b)(1) was “imposed by the vote of an uninformed Congress relying upon incomplete information.” Brief for Plaintiff 101.7 But nothing in Garcia or the Tenth Amendment authorizes courts to second-guess the substantive basis for congressional legislation. Cf. Minnesota v. Clover Leaf Creamery Co., 449 U. S. 456, 464 (1981). Where, as here, the national political process did not operate in a defective manner, the Tenth Amendment is not implicated.

B

The NGA argues that § 310 is invalid because it commandeers the state legislative and administrative process by coercing States into enacting legislation authorizing bond registration and into administering the registration scheme. They cite FERC v. Mississippi, 456 U. S. 742 (1982), which left open the possibility that the Tenth Amendment might set some limits on Congress’ power to compel States to regulate on behalf of federal interests, id., at 761-764. The extent to which the Tenth Amendment claim left open in FERC survives Garcia or poses constitutional limitations independent of those discussed in Garcia is far from clear. We need not, however, address that issue because we find the claim discussed in FERC inapplicable to § 310.

*514The federal statute at issue in FERC required state utility commissions to do the following: (1) adjudicate and enforce federal standards, (2) either consider adopting certain federal standards or cease regulating public utilities, and (3) follow certain procedures. The Court in FERC first distinguished National League of Cities v. Usery, 426 U. S. 833 (1976), noting that the statute in National League of Cities presented questions concerning “the extent to which state sovereignty shields the States from generally applicable federal regulations,” whereas the statute in FERC “attempts to use state regulatory machinery to advance federal goals.” FERC, 456 U. S., at 759. The Court in FERC then concluded that, whatever constitutional limitations might exist on the federal power to compel state regulatory activity, Congress had the power to require that state adjudicative bodies adjudicate federal issues and to require that States regulating in a pre-emptible field consider suggested federal standards and follow federally mandated procedures. Id., at 759-767.

Because, by hypothesis, § 310 effectively prohibits issuing unregistered bonds, it presents the very situation FERC distinguished from a commandeering of state regulatory machinery: the extent to which the Tenth Amendment “shields the States from generally applicable federal regulations.” 456 U. S., at 759. Section 310 regulates state activities; it does not, as did the statute in FERC, seek to control or influence the manner in which States regulate private parties. The NGA nonetheless contends that § 310 has commandeered the state legislative and administrative process because many state legislatures had to amend a substantial number of statutes in order to issue bonds in registered form and because state officials had to devote substantial effort to determine how best to implement a registered bond system. Such “commandeering” is, however, an inevitable consequence of regulating a state activity. Any federal regulation demands compliance. That a State washing to engage in cer*515tain activity must take administrative and sometimes legislative action to comply with federal standards regulating that activity is a commonplace that presents no constitutional defect. After Garcia, for example, several States and municipalities had to take administrative and legislative action to alter the employment practices or raise the funds necessary to comply with the wage and overtime provisions of the Federal Labor Standards Act.8 Indeed, even the pre-Garcia line of Tenth Amendment cases recognized that Congress could constitutionally impose federal requirements on States that States could meet only by amending their statutes. See EEOC v. Wyoming, 460 U. S. 226, 253-254, and n. 2 (1983) (Burger, C. J., dissenting) (citing state statutes from over half the States that did not comply with the federal statute upheld by the Court). Under the NGA’s theory, moreover, any State could immunize its activities from federal regulation by simply codifying the manner in which it engages in those activities. In short, the NGA’s theory of “commandeering” would not only render Garcia a nullity, but would also restrict congressional regulation of state activities even more tightly than it was restricted under the now overruled National League of Cities line of cases. We find the theory foreclosed by precedent, and uphold the constitutionality of § 310 under the Tenth Amendment.

H — 4 I — I 1 — 1

South Carolina contends that even if a statute banning state bearer bonds entirely would be constitutional, § 310 unconstitutionally violates the doctrine of intergovernmental tax immunity because it imposes a tax on the interest earned on a state bond. We agree with South Carolina that § 310 is *516inconsistent with Pollock v. Farmers’ Loan & Trust Co., 157 U. S. 429 (1895), which held that any interest earned on a state bond was immune from federal taxation.

The Secretary and the Master, however, suggest that we should uphold the constitutionality of § 310 without explicitly overruling Pollock because § 310 does not abolish the tax exemption for state bond interest entirely but rather taxes the interest on state bonds only if the bonds are not issued in the form Congress requires. In our view, however, this suggestion implicitly rests on a rather mischievous proposition of law. If, for example, Congress imposed a tax that applied exclusively to South Carolina and levied the tax directly on the South Carolina treasury, we would be obligated to adjudicate the constitutionality of that tax even if Congress allowed South Carolina to escape the tax by restructuring its state government in a way Congress found more to its liking. The United States cannot convert an unconstitutional tax into a constitutional one simply by making the tax conditional. Whether Congress could have imposed the condition by direct regulation is irrelevant; Congress cannot employ unconstitutional means to reach a constitutional end. Under Pollock, a tax on the interest income derived from any state bond was considered a direct tax on the State and thus unconstitutional. 157 U. S., at 585-586. If this constitutional rule still applies, Congress cannot threaten to tax the interest on state bonds that do not conform to congressional dictates. We thus decline to follow a suggestion that would force us to embrace implicitly a proposition of law far more controversial than the current validity of Pollock’s ban on taxing state bond interest, and proceed to address whether Pollock should be explicitly overruled.9

*517Under the intergovernmental tax immunity jurisprudence prevailing at the time, Pollock did not represent a unique immunity limited to income derived from state bonds. Rather, Pollock merely represented one application of the more general rule that neither the Federal nor the State Governments could tax income an individual directly derived from any contract with another government.10 Not only was it unconstitutional for the Federal Government to tax a bondowner on the interest he or she received on any state bond, but it was also unconstitutional to tax a state employee on the income earned from his employment contract, Collector v. Day, 11 Wall. 113 (1871), to tax a lessee on income derived from lands leased from a State, Burnet v. Coronado Oil, 285 U. S. 393 (1932), or to impose a sales tax on proceeds a vendor derived from selling a product to a state agency, Indian Motocycle Co. v. United States, 283 U. S. 570 (1931). Income derived from the same kinds of contracts with the Federal Government were likewise immune from taxation by the States. See Weston v. City Council of Charleston, 2 Pet. 449 (1829) (federal bond interest immune from state taxation); Dobbins v. Commissioners of Erie County, 16 Pet. 435 (1842) (federal employee immune from state tax on salary); Gillespie v. Oklahoma, 257 U. S. 501 (1922) (income derived from federal lease immune from state tax); Panhandle Oil Co. v. Mississippi ex rel. Knox, 277 U. S. 218 (1928) (vendor immune from sales tax on vendor’s proceeds from sale to the United States). Cases concerning the tax immunity of income derived from state contracts freely cited principles established in federal tax immunity cases, and vice versa. See, e. g., *518Coronado Oil, supra, at 398; Indian Motocycle, supra, at 575-579; Pollock, supra, at 586. See generally Indian Motocycle, supra, at 575 (immunity of States from federal tax equal to immunity of Federal Government from state tax); Metcalf & Eddy v. Mitchell, 269 U. S. 514, 521-522 (1926); Collector v. Day, supra, at 127.

This general rule 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. The Court in Pollock borrowed its reasoning directly from the decision in Weston exempting federal bond interest from state taxation:

“‘The right to tax the contract to any extent, when made, must operate upon the power to borrow before it is exercised, and have a sensible influence on the contract. The extent of this influence depends on the will of a distinct government. To any extent, however inconsiderable, it is a burthen on the operations of government. . . . The tax on government stock is thought by this court to be a tax on the contract, a tax on the [government’s] power to borrow money . . . and consequently to be repugnant to the Constitution.’” Pollock, supra, at 586, quoting Weston, supra, at 467, 468.

Thus, although a tax was collected from an independent private party, the tax was considered to be “on” the government because the tax burden might be passed on to it through the contract. This reasoning was used to define the basic scope of both federal and state tax immunities with respect to all types of government contracts.11 See, e. g., Coronado Oil, *519supra, at 400-401 (“Here the lease . . . was an instrumentality of the State .... To tax the income of the lessee arising therefrom would amount to an imposition upon the lease itself”); Panhandle Oil, supra, at 222 (“It is immaterial that the seller and not the purchaser is required to report and make payment to the State. Sale and purchase constitute a transaction by which the tax is measured and on which the burden rests”); Gillespie, supra, at 505-506 (“ ‘A tax upon the leases is a tax upon the power to make them . . .’” (quoting Indian Territory Illuminating Oil Co. v. Oklahoma, 240 U. S. 522, 530 (1916))). The commonality of the rationale underlying all these immunities for government contracts *520was highlighted by Indian Motocycle, 283 U. S. 570 (1931). In that case, the Court reviewed the then current status of intergovernmental tax immunity doctrine, observing that a tax on interest earned on a state or federal bond was unconstitutional because it would burden the exercise of the government’s power to borrow money and that a tax on the salary of a State or Federal Government employee was unconstitutional because it would burden the government’s power to obtain the employee’s services. Id., at 576-578. It then concluded that under the same principle a sales tax imposed on a vendor for a sale to a state agency was unconstitutional because it would burden the sale transaction. Id., at 579.

The rationale underlying Pollock and the general immunity for government contract income has been thoroughly repudiated by modern intergovernmental immunity case law. In Graves v. New York ex rel. O’Keefe, 306 U. S. 466 (1939), the Court announced: “The theory . . . that a tax on income is legally or economically a tax on its source, is no longer tenable.” Id., at 480. The Court explained:

“So much of the burden of a non-discriminatory general tax upon the incomes of employees of a government, state or national, as may be passed on economically to that government, through the effect of the tax on the price level of labor or materials, is but the normal incident of the organization within the same territory of two governments, each possessing the taxing power. The burden, so far as it can be said to exist or to affect the government in any indirect or incidental way, is one which the Constitution presupposes . . . .” Id., at 487.

See also James v. Dravo Contracting Co., 302 U. S. 134, 160 (1937) (the fact that a tax on a Government contractor “may increase the cost to the Government. . . would not invalidate the tax”); Helvering v. Gerhardt, 304 U. S. 405, 424 (1938). The thoroughness with which the Court abandoned the burden theory was demonstrated most emphatically when the Court upheld a state sales tax imposed on a Government *521contractor even though the financial burden of the tax was entirely passed on, through a cost-plus contract, to the Federal Government. Alabama v. King & Boozer, 314 U. S. 1 (1941). The Court stated:

“The Government, rightly we think, disclaims any contention that the Constitution, unaided by Congressional legislation, prohibits a tax exacted from the contractors merely because it is passed on economically, by the terms of the contract or otherwise, as part of the construction cost to the Government. So far as such a nondiscriminatory state tax upon the contractor enters into the cost of the materials to the Government, that is but a normal incident of the organization within the same territory of two independent taxing sovereignties. The asserted right of the one to be free of taxation by the other does not spell immunity from paying the added costs, attributable to the taxation of those who furnish supplies to the Government and who have been granted no tax immunity. So far as a different view has prevailed, we think it no longer tenable.” Id., at 8-9 (citations omitted).

King & Boozer thus completely foreclosed any claim that the nondiscriminatory imposition of costs on private entities that pass them on to States or the Federal Government unconstitutionally burdens state or federal functions. Subsequent cases have consistently reaffirmed the principle that a nondiscriminatory tax collected from private parties contracting with another government is constitutional even though part or all of the financial burden falls on the other government. See Washington v. United States, 460 U. S. 536, 540 (1983); United States v. New Mexico, 455 U. S. 720, 734 (1982); United States v. County of Fresno, 429 U. S. 452, 460-462, and n. 9 (1977); United States v. City of Detroit, 355 U. S. 466, 469 (1958).

With the rationale for conferring a tax immunity on parties dealing with another government rejected, the government *522contract immunities recognized under prior doctrine were, one by one, eliminated. Overruling Burnet v. Coronado Oil, 285 U. S. 393 (1932), and Gillespie v. Oklahoma, 257 U. S. 501 (1922), the Court upheld the constitutionality of a federal tax on net income a corporation derived from a state lease in Helvering v. Mountain Producers Corp., 303 U. S. 376 (1938). See also Oklahoma Tax Comm’n v. Texas Co., 336 U. S. 342 (1949) (upholding constitutionality of state tax on gross income derived from Indian lease). Later, the Court explicitly overruled Collector v. Day, 11 Wall. 113 (1871), and upheld the constitutionality of a nondiscriminatory state tax on the salary of a federal employee. Graves v. New York ex rel. O’Keefe, supra.12 And in the course of upholding a sales tax on a cost-plus Government contractor, the Court in King & Boozer overruled Panhandle Oil Co. v. Mississippi ex rel. Knox, 277 U. S. 218 (1928). See also James, supra (upholding state tax on gross income independent contractor received from Federal Government). The only premodern tax immunity for parties to government contracts that has so far avoided being explicitly overruled is the immunity for recipients of governmental bond interest.13 That this Court *523has yet to overrule Pollock explicitly, however, is explained not by any distinction between the income derived from government bonds and the income derived from other government contracts, but by the historical fact that Congress has always exempted state bond interest from taxation by statute, beginning with the very first federal income tax statute. Act of Oct. 3, 1913, ch. 16, § II(B), 38 Stat. 168.

In sum, then, under current intergovernmental tax immunity doctrine the States can never tax the United States directly but can tax any private parties with whom it does business, even though the financial burden falls on the United States, as long as the tax does not discriminate against the United States or those with whom it deals. See Washington, supra, at 540; County of Fresno, supra, at 460-463; City of Detroit, supra, at 473; Oklahoma Tax Comm’n, supra, at 359-364. A tax is considered to be directly on the Federal Government only “when the levy falls on the United States itself, or on an agency or instrumentality so closely connected to the Government that the two cannot realistically be viewed as separate entities.” New Mexico, supra, at 735. The rule with respect to state tax immunity is essentially the same, see, e. g., Graves, supra, at 485; Mountain Producers Corp., supra, at 386-387, except that at least some nondiscriminatory federal taxes can be collected directly from the States even though a parallel state tax could not be collected directly from the Federal Government.14 See generally n. 11, supra.

*524We thus confirm that subsequent case law has overruled the holding in Pollock that state bond interest is immune from a nondiscriminatory federal tax. We see no constitutional reason for treating persons who receive interest on government bonds differently than persons who receive income from other types of contracts with the government, and no tenable rationale for distinguishing the costs imposed on States by a tax on state bond interest from the costs imposed *525by a tax on the income from any other state contract. We stated in Graves that “as applied to the taxation of salaries of the employees of one government, the purpose of the immunity was not to confer benefits on the employees by relieving them from contributing their share of the financial support of the other government, whose benefits they enjoy, or to give an advantage to a government by enabling it to engage employees at salaries lower than those paid for like services by other employers, public or private . . . .” 306 U. S., at 483. Likewise, the owners of state bonds have no constitutional entitlement not to pay taxes on income they earn from state bonds, and States have no constitutional entitlement to issue bonds paying lower interest rates than other issuers.15

*526Indeed, this Court has in effect acknowledged that a holder of a Government bond could constitutionally be taxed on bond interest in Memphis Bank & Trust Co. v. Garner, 459 U. S. 392 (1983), which involved a state tax on federal bond interest. Although that case involved an interpretation of 31 U. S. C. § 742, we premised our statutory interpretation on the observation that “[o]ur decisions have treated § 742 as principally a restatement of the constitutional rule.” 459 U. S., at 397. We then stated: “Where, as here, the economic but not the legal incidence of the tax falls upon the Federal Government, such a tax generally does not violate the constitutional immunity if it does not discriminate against holders of federal property or those with whom the Federal Government deals.” Ibid, (emphasis added).

TEFRA § 310 thus clearly imposes no direct tax on the States. The tax is imposed on and collected from bondholders, not States, and any increased administrative costs incurred by States in implementing the registration system are not “taxes” within the meaning of the tax immunity doctrine. See generally United States v. Mississippi Tax Comm’n, 421 U. S. 599, 606 (1975) (describing tax as an enforced contribution to provide for the support of government). Nor does § 310 discriminate against States. The provisions of § 310 seek to assure that all publicly offered long-term bonds are issued in registered form, whether issued by state or local *527governments, the Federal Government, or private corporations. See supra, at 510. Accordingly, the Federal Government has directly imposed the same registration requirement on itself that it has effectively imposed on States. The incentives States have to switch to registered bonds are necessarily different than those of corporate bond issuers because only state bonds enjoy any exemption from the federal tax on bond interest, but the sanctions for issuing unregistered corporate bonds are comparably severe. See ibid. Removing the tax exemption for interest earned on state bonds would not, moreover, create a discrimination between state and corporate bonds since corporate bond interest is already subject to federal tax.

IV

Because the federal imposition of a bond registration requirement on States does not violate the Tenth Amendment and because a nondiscriminatory federal tax on the interest earned on state bonds does not violate the intergovernmental tax immunity doctrine, we uphold the constitutionality of § 310(b)(1),16 overrule the exceptions to the Special Master’s Report, and approve his recommendation to enter judgment for the defendant.

It is so ordered.

concurring in part and concurring in the judgment.

I join in the Court’s judgment, and in its opinion except for Part II. I do not join the latter because, as observed by The Chief Justice, post, at 529-530, it unnecessarily casts doubt upon FERC v. Mississippi, 456 U. S. 742 (1982), and because it misdescribes the holding in Garcia v. San Antonio Metropolitan Transit Authority, 469 U. S. 528 (1985). I do not read Garcia as adopting — in fact I read it as explicitly disclaiming — the proposition attributed to it in today’s opinion, ante, at 512-513, that the “national political process” is the States’ only constitutional protection, and that nothing except the demonstration of “some extraordinary defects” in the operation of that process can justify judicial relief. We said in Garcia: “These cases do not require us to identify or define what affirmative limits the constitutional structure might impose on federal action affecting the States under the Commerce Clause. See Coyle v. Oklahoma, 221 U. S. 559 (1911).” 469 U. S., at 556 (emphasis added). I agree only that that structure does not prohibit what the Federal Government has done here.

concurring in the judgment.

Today the Court reaches two results regarding § 310(b)(1) of TEFRA that I believe are analytically distinct. First, the Court finds that § 310(b)(1) does not violate the Tenth Amendment by compelling States to issue bonds in registered form. Second, the majority concludes that the statute *529also does not contravene the doctrine of intergovernmental tax immunity; in doing so, the majority overrules our decision in Pollock v. Farmers’ Loan & Trust Co., 157 U. S. 429 (1895). While I agree that the principles of intergovernmental tax immunity are not threatened in this case, in my view, the Court unnecessarily casts doubt on the protective scope of the Tenth Amendment in the course of upholding §310 (b)(1).

The Special Master appointed by the Court made a number of factual determinations about the impact that the TEFRA registration requirements would have upon the States. Most notably, the Special Master found that the registration requirements have had no substantive effect on the abilities of States to raise debt capital, on the political processes by which States decide to issue debt, or on the power of the States to choose the purpose to which they will dedicate the proceeds of their tax-exempt borrowing. After an exhaustive investigation, the Special Master summarized: “TEFRA has not changed how much the States borrow, for what purposes they borrow, how they decide to borrow, or any other obviously important aspect of the borrowing process.” Report of Special Master 118.

This well-supported conclusion that § 310(b)(1) has had a de minimis impact on the States should end, rather than begin, the Court’s constitutional inquiry. Even the more expansive conception of the Tenth Amendment espoused in National League of Cities v. Usery, 426 U. S. 833 (1976), recognized that only congressional action that “operate[s] to directly displace the States’ freedom to structure integral operations in areas of traditional governmental functions,” runs afoul of the authority granted Congress. Id., at 852. The Special Master determined that no such displacement has occurred through the implementation of the TEFRA requirements; I see no need to go further, as the majority does, to discuss the possibility of defects in the national political process that spawned TEFRA, nor to hypothesize that the Tenth Amend*530ment concerns voiced in FERC v. Mississippi, 456 U. S. 742 (1982), may not have survived Garcia v. San Antonio Metropolitan Transit Authority, 469 U. S. 528 (1985). Those issues, intriguing as they may be, are of no moment in the present case and are best left unaddressed until clearly presented.

concurring.

Although the Court properly finds support for its holding in Garcia v. San Antonio Metropolitan Transit Authority, *528469 U. S. 528 (1985), the outcome of this case was equally clear well before that case was decided. See South Carolina v. Regan, 465 U. S. 367, 403-419 (1984) (Stevens, J., concurring in part and dissenting in part). It should be emphasized, however, that neither the Court’s decision today, nor what I have written in the past, expresses any opinion about the wisdom of taxing the interest on bonds issued by state or local governments.

dissenting.

The Court today overrules a precedent that it has honored for nearly 100 years and expresses a willingness to cancel the constitutional immunity that traditionally has shielded the interest paid on state and local bonds from federal taxation. Henceforth the ability of state and local governments to finance their activities will depend in part on whether Congress voluntarily abstains from tapping this permissible source of additional income tax revenue. I believe that state autonomy is an important factor to be considered in reviewing the National Government’s exercise of its enumerated powers. Garcia v. San Antonio Metropolitan Transit Authority, 469 U. S. 528, 581 (1985) (O’Connor, J., joined by Powell and Rehnquist, JJ., dissenting). I dissent from the decision to overrule Pollock v. Farmers’ Loan & Trust Co., 157 U. S. 429 (1895), and I would invalidate Congress’ attempt to regulate the sovereign States by threatening to deprive them of this tax immunity, which would increase their dependence on the National Government.

Section 310(b)(1) of the Tax Equity and Fiscal Responsibility Act of 1982 (TEFRA), 26 U. S. C. § 103(j)(1), provides that the interest paid on state and local bonds will be subject to federal income tax unless the bonds are issued in registered form. The Court readily concludes that Congress could have prohibited outright the issuance of bearer bonds without violating the Tenth Amendment. Ante, at 511-513. But regardless of whether Congress could have required registration of the bonds directly under its commerce power, I agree with the Court that Congress may not accomplish the same end by an unconstitutional means. Ante, at 515-516. *531In my view, the Tenth Amendment and principles of federalism inherent in the Constitution prohibit Congress from taxing or threatening to tax the interest paid on state and municipal bonds. It is also arguable that the States’ autonomy is protected from substantial federal incursions by virtue of the Guarantee Clause of the Constitution, Art. IV, § 4. See Merritt, The Guarantee Clause and State Autonomy: Federalism for a Third Century, 88 Colum. L. Rev. 1, 70-78 (1988) (arguing that judicial enforcement of the Guarantee Clause is proper).

The Court never expressly considers whether federal taxation of state and local bond interest violates the Constitution. Instead, the majority characterizes the federal tax exemption for state and local bond interest as an aspect of intergovernmental tax immunity, and it describes the decline of the intergovernmental tax immunity doctrine in this century. But constitutional principles do not depend upon the rise or fall of particular legal doctrines. This Court has a continuing responsibility “to oversee the Federal Government’s compliance with its duty to respect the legitimate interests of the States.” Garcia, supra, at 581 (O’Connor, J., joined by Powell and Rehnquist, JJ., dissenting). In my view, the Court shirks its responsibility because it fails to inquire into the substantial adverse effects on state and local governments that would follow from federal taxation of the interest on state and local bonds.

Long-term debt obligations are an essential source of funding for state and local governments. In 1974, state and local governments issued approximately $23 billion of new municipal bonds; in 1984, they issued $102 billion of new bonds. Report of Special Master 20. State and local governments rely heavily on borrowed funds to finance education, road construction, and utilities, among other purposes. As the Court recognizes, States will have to increase the interest rates they pay on bonds by 28-35% if the interest is subject to the federal income tax. Ante, at 511. Governmental op*532erations will be hindered severely if the cost of capital rises by one-third. If Congress may tax the interest paid on state and local bonds, it may strike at the very heart of state and local government activities.

In the pivotal cases which first set limits to intergovernmental tax immunity, this Court paid close attention to the practical effects of its decisions. The Court limited the government’s immunity only after it determined that application of a tax would not substantially affect government operations. Thus in the first case to uphold federal income taxation of revenue earned by a state contractor, this Court observed that “neither government may destroy the other nor curtail in any substantial manner the exercise of its powers.” Metcalf & Eddy v. Mitchell, 269 U. S. 514, 523-524 (1926). When this Court extended its holding to the case of a state tax on a federal contractor, it expressly noted that the tax “does not interfere in any substantial way with the performance of federal functions.” James v. Dravo Contracting Co., 302 U. S. 134, 161 (1937). In upholding the application of the federal income tax to income derived from a state lease, this Court decided that mere theoretical concerns about interference with the functions of government did not justify immunity, but that “[rjegard must be had to substance and direct effects.” Helvering v. Mountain Producers Corp., 303 U. S. 376, 386 (1938). In Helvering v. Gerhardt, 304 U. S. 405 (1938), this Court upheld the application of the federal income tax to income earned by a state employee, because there is “[no] immunity when the burden on the state is so speculative and uncertain that if allowed it would restrict the federal taxing power without affording any corresponding tangible protection to the state government.” Id., at 419-420.

The instant case differs critically from the cases quoted above because the Special Master found that, if the interest on state and local bonds is taxed, the cost of borrowing by state and local governments would rise substantially. This *533certainly would affect seriously state and local government operations. The majority is unconcerned with this difference because it is satisfied with the formal test of intergovernmental tax immunity that can be distilled from later cases. Under this test, if a tax is not imposed directly on the government, and does not discriminate against the government, then it does not violate intergovernmental tax immunity. See ante, at 523.

I do not think the Court’s bipartite test adequately accommodates the constitutional concerns raised by the prospect of applying the federal income tax to the interest paid on state and local bonds. This Court has a duty to inquire into the devastating effects that such an innovation would have on state and local governments. Although Congress has taken a relatively less burdensome step in subjecting only income from bearer bonds to federal taxation, the erosion of state sovereignty is likely to occur a step at a time. “If there is any danger, it lies in the tyranny of small decisions —in the prospect that Congress will nibble away at state sovereignty, bit by bit, until someday essentially nothing is left but a gutted shell.” L. Tribe, American Constitutional Law 381 (2d ed. 1988).

Federal taxation of state activities is inherently a threat to state sovereignty. As Chief Justice Marshall observed long ago, “the power to tax involves the power to destroy.” McCulloch v. Maryland, 4 Wheat. 316, 431 (1819). Justice Holmes later qualified this principle, observing that “[t]he power to tax is not the power to destroy while this Court sits.” Panhandle Oil Co. v. Mississippi ex rel. Knox, 277 U. S. 218, 223 (1928) (Holmes, J., joined by Brandéis and Stone, JJ., dissenting). If this Court is the States’ sole protector against the threat of crushing taxation, it must take seriously its responsibility to sit in judgment of federal tax initiatives. • I do not think that the Court has lived up to its constitutional role in this case. The Court has failed to enforce the constitutional safeguards of state autonomy and *534self-sufficiency that may be found in the Tenth Amendment and the Guarantee Clause, as well as in the principles of federalism implicit in the Constitution. I respectfully dissent.