United Haulers Ass'n v. Oneida-Herkimer Solid Waste Management Authority

concurring in the judgment.

I concur in the judgment. Although I joined C & A Car-bone, Inc. v. Clarkstown, 511 U. S. 388 (1994), I no longer believe it was correctly decided. The negative Commerce Clause has no basis in the Constitution and has proved unworkable in practice. See Camps Newfound/Owatonna, Inc. v. Town of Harrison, 520 U. S. 564, 610-620 (1997) (Thomas, J., dissenting); Tyler Pipe Industries, Inc. v. Washington State Dept, of Revenue, 483 U. S. 232, 259-265 (1987) (Scalia, J., concurring in part and dissenting in part); License Cases, 5 How. 504, 578-586 (1847) (Taney, C. J.). As the debate between the majority and dissent shows, application of the negative Commerce Clause turns solely on policy considerations, not on the Constitution. Because this Court has no policy role in regulating interstate commerce, I would discard the Court’s negative Commerce Clause jurisprudence.

I

Under the Commerce Clause, “Congress shall have Power ... [t]o regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes.” U. S. Const., Art. I, §8, cl. 3. The language of the Clause allows Congress not only to regulate interstate commerce but also to prevent state regulation of interstate commerce. State Bd. of Ins. v. Todd Shipyards Corp., 370 U. S. 451, 456 (1962); Gibbons v. Ogden, 9 Wheat. 1, 210 (1824). Expanding on the interstate-commerce powers explicitly conferred on Congress, this Court has interpreted the Commerce Clause as a tool for courts to strike down state laws that it believes inhibit interstate commerce. But there is no basis in the Constitution for that interpretation.

*350The Court does not contest this point, and simply begins its analysis by appealing to stare decisis:

“Although the Constitution does not in terms limit the power of States to regulate commerce, we have long interpreted the Commerce Clause as an implicit restraint on state authority, even in the absence of a conflicting federal statute. See Case of the State Freight Tax, 15 Wall. 232, 279 (1873); Cooley v. Board of Wardens of Port of Philadelphia ex rel. Soc. for Relief of Distressed Pilots, 12 How. 299, 318 (1852).” Ante, at 338.

The Court’s reliance on Cooley v. Board of Wardens of Port of Philadelphia ex rel. Soc. for Relief of Distressed Pilots, 12 How. 299 (1852), and Case of the State Freight Tax, 15 Wall. 232 (1873), is curious because the Court has abandoned the reasoning of those cases in its more recent jurisprudence. Cooley and State Freight Tax are premised upon the notion that the Commerce Clause is an exclusive grant of power to Congress over certain subject areas.1 Cooley, supra, at 319-320 (holding that “[wjhatever subjects of this [Commerce Clause] power are in their nature national, or admit only of one uniform system, or plan of regulation, may justly be said to be of such a nature as to require exclusive legislation by Congress” but holding that “the nature of th[e] subject [of state pilotage laws] is not such as to require its exclusive legislation” and therefore upholding the state laws against the negative Commerce Clause challenge); State Freight Tax, supra, at 279-280 (applying the same rationale). The Court, however, no longer limits Congress’ power by analyzing whether the subjects of state regulation “admit only of one uniform system,” Cooley, supra, at 319. Rather, *351the modern jurisprudence focuses upon the way in which States regulate those subjects to decide whether the regulation is permissible. E. g., ante, at 338-339, 345. Because the reasoning of Cooley and State Freight Tax has been rejected entirely, they provide no foundation for today’s decision.

Unfazed, the Court proceeds to analyze whether the ordinances “discriminat[e] on [their] face against interstate commerce.” Ante, at 338. Again, none of the cases the Court cites explains how the absence or presence of discrimination is relevant to deciding whether the ordinances are constitutionally permissible, and at least one case affirmatively admits that the nondiscrimination rule has no basis in the Constitution. Philadelphia v. New Jersey, 437 U. S. 617, 623 (1978) (“The bounds of these restraints appear nowhere in the words of the Commerce Clause, but have emerged gradually in the decisions of this Court giving effect to its basic purpose”). Thus cloaked in the “purpose” of the Commerce Clause, the rule against discrimination that the Court applies to decide this case exists untethered from the written Constitution. The rule instead depends upon the policy preferences of a majority of this Court.

The Court’s policy preferences are an unsuitable basis for constitutional doctrine because they shift over time, as demonstrated by the different theories the Court has offered to support the nondiscrimination principle. In the early years of the nondiscrimination rule, the Court struck down a state health law because “the enactment of a similar statute by each one of the States composing the Union would result in the destruction of commerce among the several States.” Minnesota v. Barber, 136 U. S. 313, 321 (1890); see Foster-Fountain Packing Co. v. Haydel, 278 U. S. 1, 13 (1928) (stating that a Commerce Clause violation would occur if the state statute would “directly . . . obstruct and burden interstate commerce”). More recently, the Court has struck *352down state laws sometimes based on its preference for national unity, see, e. g., American Trucking Assns., Inc. v. Michigan Pub. Serv. Common, 545 U. S. 429, 433 (2005) (justifying the nondiscrimination rule by stating that “[o]ur Constitution was framed upon the theory that the peoples of the several states must sink or swim together” (internal quotation marks omitted)), and other times on the basis of antiprotectionist sentiment, see, e. g., Oregon Waste Systems, Inc. v. Department of Environmental Quality of Ore., 511 U. S. 93, 98 (1994) (noting the interest in “avoid[ing] the tendencies toward economic Balkanization”); New Energy Co. of Ind. v. Limbach, 486 U. S. 269, 273-274 (1988) (stating that the negative Commerce Clause “prohibits economic protectionism— that is, regulatory measures designed to benefit in-state economic interests by burdening out-of-state competitors”); see also Carbone, 511 U. S., at 390 (“The central rationale for the rule against discrimination is to prohibit state or municipal laws whose object is local economic protectionism, laws that would excite those jealousies and retaliatory measures the Constitution was designed to prevent”); Toomer v. Witsell, 334 U. S. 385, 403-404 (1948) (striking down a law that “impose[d] an artificial rigidity on the economic pattern of the industry”)..

Many of the above-cited cases (and today’s majority and dissent) rest on the erroneous assumption that the Court must choose between economic protectionism and the free market. But the Constitution vests that fundamentally legislative choice in Congress. To the extent that Congress does not exercise its authority to make that choice, the Constitution does not limit the States’ power to regulate commerce. In the face of congressional silence, the States are free to set the balance between protectionism and the free market. Instead of accepting this constitutional reality, the Court’s negative Commerce Clause jurisprudence gives nine Justices of this Court the power to decide the appropriate balance.

*353II

As the foregoing demonstrates, despite more than 100 years of negative Commerce Clause doctrine, there is no principled way to decide this case under current law. Notably, the Court cannot and does not consider this case “[i]n light of the language of the Constitution and the historical context.” Alden v. Maine, 527 U. S. 706, 743 (1999). Likewise, it cannot follow “the cardinal rule to construe provisions in context.” United States v. Balsys, 524 U. S. 666, 673 (1998). And with no text to construe, the Court cannot take into account the Founders’ “deliberate choice of words” or “their natural meaning.” Wright v. United States, 302 U. S. 583, 588 (1938). Furthermore, as the debate between the Court’s opinion and the dissenting opinion reveals, no case law applies to the facts of this case.2

Explaining why the ordinances do not discriminate against interstate commerce, the Court states that “government is vested with the responsibility of protecting the health, safety, and welfare of its citizens.” Ante, at 342. According to the Court, a law favoring in-state business requires rigorous scrutiny because the law “is often the product of ‘simple economic protectionism.’” Ante, at 343. A law favoring local government, however, “may be directed toward any number of legitimate goals unrelated to protectionism.” Ibid. This distinction is razor thin: In contrast to today’s deferential approach (apparently based on the Court’s trust of local government), the Court has applied the equivalent of strict scrutiny in other cases even where it is unchallenged that the state law discriminated in favor of in-state private entities for a legitimate, nonproteetionist reason. See Barber, swpra, at 319 (striking down the State’s inspection *354law for livestock even though it did not challenge “[t]he presumption that this statute was enacted, in good faith,... to protect the health of the people of Minnesota”).

In Carbone, which involved discrimination in favor of private entities, we did not doubt the good faith of the municipality in attempting to deal with waste through a flow-control ordinance. 511 U. S., at 386-389. But we struck down the ordinance because it did not allow interstate entities to participate in waste disposal. Id., at 390-395. The majority distinguishes Carbone by deciding that favoritism of a government monopoly is less suspect than government regulation of private entities.3 I see no basis for drawing such a conclusion, which, if anything, suggests a policy-driven preference for government monopoly over privatization. Ante, at 344 (stating that “waste disposal is both typically and traditionally a local government function” (brackets and internal quotation marks omitted)). Whatever the reason, the choice is not the Court’s to make. Like all of the Court’s previous negative Commerce Clause cases, today’s decision leaves the future of state and local regulation of commerce to the whim of the Federal Judiciary.

Ill

Despite its acceptance of negative Commerce Clause jurisprudence, the Court expresses concern about “unprecedented and unbounded interference by the courts with state and local government.” Ante, at 343. It explains:

“The dormant Commerce Clause is not a roving license for federal courts to decide what activities are appropriate for state and local government to undertake, and *355what activities must be the province of private market competition.

“There is no reason to step in and hand local businesses a victory they could not obtain through the political process.” Ante, at 343, 345.

I agree that the Commerce Clause is not a “roving license” and that the Court should not deliver to businesses victories that they failed to obtain through the political process. I differ with the Court because I believe its powerful rhetoric is completely undermined by the doctrine it applies.

In this regard, the Court’s analogy to Lochner v. New York, 198 U. S. 45 (1905), suggests that the Court should reject the negative Commerce Clause, rather than tweak it. Ante, at 347. In Lochner the Court located a “right of free contract” in a constitutional provision that says nothing of the sort. 198 U. S., at 57. The Court’s negative Commerce Clause jurisprudence, created from whole cloth, is just as illegitimate as the “right” it vindicated in Lochner. Yet today’s decision does not repudiate that doctrinal error. Rather, it further propagates the error by narrowing the negative Commerce Clause for policy reasons — reasons that later majorities of this Court may find to be entirely illegitimate.

In so doing, the majority revisits familiar territory: Just three years after Lochner, the Court narrowed the right of contract for policy reasons but did not overrule Lochner. Muller v. Oregon, 208 U. S. 412, 422-423 (1908) (upholding a maximum-hours requirement for women because the difference between the “two sexes” “justifies a difference in legislation”). Like the Muller Court, today’s majority trifles with an unsound and illegitimate jurisprudence yet fails to abandon it.

Because I believe that the power to regulate interstate commerce is a power given to Congress and not the Court, I concur in the judgment of the Court.

with whom Justice Stevens and Justice Kennedy join, dissenting.

In C & A Carbone, Inc. v. Clarkstown, 511 U. S. 383 (1994), we held that “a so-called flow control ordinance, which require[d] all solid waste to be processed at a designated transfer station before leaving the municipality,” discriminated against interstate commerce and was invalid under the Commerce Clause because it “deprived] competitors, including out-of-state firms, of access to a local market.” Id., at 386. Because the provisions challenged in this case are essentially identical to the ordinance invalidated in Carbone, I respectfully dissent.

I

This Court has “interpreted the Commerce Clause to invalidate local laws that impose commercial barriers or discriminate against an article of commerce by reason of its origin or destination out of State.” Id., at 390. As the Court acknowledges, a law “ ‘ “discriminat[es]” ’ ” in this context if it mandates “‘differential treatment of in-state and out-of-state economic interests’ ” in a way “ ‘that benefits the former and burdens the latter.’ ” Ante, at 338 (quoting Oregon Waste Systems, Inc. v. Department of Environmental Quality of Ore., 511 U. S. 93, 99 (1994)). A local law that discriminates against interstate commerce is sustainable only if it serves a legitimate local purpose that could not be served as well by nondiscriminatory means. Maine v. Taylor, 477 U. S. 131 (1986).

“Solid waste, even if it has no value, is an article of commerce.” Fort Gratiot Sanitary Landfill, Inc. v. Michigan Dept, of Natural Resources, 504 U. S. 353, 359 (1992). Accordingly, laws that “discriminate against [trash] by reason of its origin or destination out of State,” Carbone, 511 U. S., at 390, are sustainable only if they serve a legitimate local purpose that could not be served as well by nondiscriminatory means.

*357In Carbone, this Court invalidated a local ordinance requiring all nonhazardous solid waste in Clarkstown, New York, to be deposited at a specific local transfer facility. The Court concluded that the ordinance discriminated against interstate commerce because it “hoard[ed] solid waste, and the demand to get rid of it, for the benefit of the preferred processing facility.” Id., at 392.

The Court explained that the flow-control ordinance did serve a purpose that a nonprotectionist regulation would not: “It ensures that the town-sponsored facility will be profitable, so that the local contractor can build it and Clarkstown can buy it back at nominal cost in five years.” Id., at 393. “In other words ... the flow control ordinance is a financing measure.” Ibid. The Court concluded, however, that “revenue generation is not a local interest that can justify discrimination against interstate commerce.” Ibid.

The Court also held that “Clarkstown has any number of nondiscriminatory alternatives for addressing the health and environmental problems alleged to justify the ordinance”— including “uniform safety regulations” that could be enacted to “ensure that competitors... do not underprice the market by cutting corners on environmental safety.” Ibid. Thus, the Court invalidated the ordinance because any legitimate local interests served by the ordinance could be accomplished through nondiscriminatory means. See id., at 392-393.

This case cannot be meaningfully distinguished from Car-bone. As the Court itself acknowledges, “[t]he only salient difference” between the cases is that the ordinance invalidated in Carbone discriminated in favor of a privately owned facility, whereas the laws at issue here discriminate in favor of “facilities owned and operated by a state-created public benefit corporation.” Ante, at 334. The Court relies on the distinction between public and private ownership to uphold the flow-control laws, even though a straightforward application of Carbone would lead to the opposite result. See ante, *358at 342-344. The public-private distinction drawn by the Court is both illusory and without precedent.

II

The fact that the flow-control laws at issue discriminate in favor of a government-owned enterprise does not meaningfully distinguish this case from Carbone. The preferred facility in Carbone was, to be sure, nominally owned by a private contractor who had built the facility on the town’s behalf, but it would be misleading to describe the facility as private. In exchange for the contractor’s promise to build the facility for the town free of charge and then to sell it to the town five years later for $1, the town guaranteed that, during the first five years of the facility’s existence, the contractor would receive “a minimum waste flow of 120,000 tons per year” and that the contractor could charge an above-market tipping fee. 511 U. S., at 387. If the facility “received less than 120,000 tons in a year, the town [would] make up the tipping fee deficit.” Ibid. To prevent residents, businesses, and trash haulers from taking their waste elsewhere in pursuit of lower tipping fees (leaving the town responsible for covering any shortfall in the contractor’s guaranteed revenue stream), the town enacted an ordinance “requiring] all nonhazardous solid waste within the town to be deposited at” the preferred facility. Ibid.

This Court observed that “[t]he object of this arrangement was to amortize the cost of the transfer station: The town would finance its new facility with the income generated by the tipping fees.” Ibid, (emphasis added). “In other words,” the Court explained, “the flow control ordinance [wa]s a financing measure,” id., at 393, for what everyone— including the Court — regarded as the town’s new transfer station.

The only real difference between the facility at issue in Carbone and its counterpart in this case is that title to the *359former had not yet formally passed to the municipality. The Court exalts form over substance in adopting a test that turns on this technical distinction, particularly since, barring any obstacle presented by state law, the transaction in Car-bone could have been restructured to provide for the passage of title at the beginning, rather than the end, of the 5-year period.

For this very reason, it is not surprising that in Carbone the Court did not dispute the dissent’s observation that the preferred facility was for all practical purposes owned by the municipality. See id., at 419 (opinion of Souter, J.) (“Clarkstown’s transfer station is essentially a municipal facility”); id., at 416 (describing the nominal “proprietor” of the transfer station as “essentially an agent of the municipal government”). To the contrary, the Court repeatedly referred to the transfer station in terms suggesting that the transfer station did in fact belong to the town. See id., at 387 (explaining that “[t]he town would finance its new facility with the income generated by the tipping fees” (emphasis added)); id., at 393 (observing that the challenged flow-control ordinance was designed to “ensur[e] that the town-sponsored facility will be profitable”); id., at 394 (concluding that, “having elected to use the open market to earn revenues for its project, the town may not employ discriminatory regulation to give that project an advantage over rival businesses from out of State” (emphasis added)).

Today the Court dismisses those statements as “at best inconclusive.” Ante, at 340, n. 3. The Court, however, fails to offer any explanation as to what other meaning could possibly attach to Carbone’s repeated references to Clarkstown’s transfer station as a municipal facility. It also ignores the fact that the ordinance itself, which was included in its entirety in an appendix to the Court’s opinion, repeatedly referred to the station as “the Town of Clarkstown solid waste facility.” 511 U. S., at 396, 398, 399. The Court likewise *360fails to acknowledge that the parties in Carbone openly acknowledged the municipal character of the transfer station. See Pet. for Cert., O. T. 1993, No. 92-1402, p. 5 (“The town’s designated trash disposal facility is operated by a private contractor, under an agreement with the town” (emphasis added)); Brief for Petitioners, O. T. 1993, No. 92-1402, p. 26 (arguing that “it is clear that the purported safety and health benefits of [the flow-control ordinance] derive simply from the continued economic viability of the town’s waste facility” (emphasis added; internal quotation marks omitted)); Brief for Respondent, O. T. 1993, No. 92-1402, p. 8 (“The Town entered into a contract with Clarkstown Recycling, Inc., which provided for that firm to build and operate the new Town facility” (emphasis added)).

I see no ambiguities in those statements, much less any reason to dismiss them as “at best inconclusive”; they reflect a clear understanding that the station was, for all purposes relevant to the dormant Commerce Clause, a municipal facility.

Ill

In any event, we have never treated discriminatory legislation with greater deference simply because the entity favored by that legislation was a government-owned enterprise. In suggesting otherwise, the Court relies unduly on Carbone’s passing observation that “‘offending local laws hoard a local resource — be it meat, shrimp, or milk — for the benefit of local businesses.’” Ante, at 341 (emphasis in original). Carbone’s use of the word “businesses,” the Court insists, somehow reveals that Carbone was not “extending” our dormant Commerce Clause jurisprudence “to cover discrimination in favor of local government.” Ante, at 341.

But no “extension]” was required. The Court has long subjected discriminatory legislation to strict scrutiny, and has never, until today, recognized an exception for discrimination in favor of a state-owned entity.

*361A

This Court long ago recognized that the Commerce Clause can be violated by a law that discriminates in favor of a state-owned monopoly. In the 1890’s, South Carolina enacted laws giving a state agency the exclusive right to operate facilities selling alcoholic beverages within that State, and these laws were challenged under the Commerce Clause in Scott v. Donald, 165 U. S. 58 (1897), and Vance v. W. A. Vandercook Co., 170 U. S. 438 (1898). The Court held that the Commerce Clause barred the State from prohibiting its residents from purchasing alcohol from out-of-state vendors, see id., at 442, but that the State could surmount this problem by allowing residents to receive out-of-state shipments for their personal use. See id., at 452. The Court’s holding was based on the same fundamental dormant Commerce Clause principle applied in Carbone.1 As the Court put it in Vance, a State “‘cannot discriminate against the bringing of [lawful] articles in and importing them from other States’ ” because such discrimination is “ ‘a hindrance to interstate commerce and an unjust preference of the products of the enacting State as against similar products of other States.’” 170 U. S., at 443 (quoting Scott, supra, at 101). Cf. Carbone, supra, at 390 (the Commerce Clause bars state and local laws that “impose commercial barriers or discriminate against an article of commerce by reason of its origin or destination out of State”).

Thus, were it not for the Twenty-first Amendment, laws creating state-owned liquor monopolies — which many States maintain today — would be deemed discriminatory under the *362dormant Commerce Clause. See Granholm v. Heald, 544 U. S. 460, 489 (2005) (explaining that the Twenty-first Amendment makes it possible for States to “assume direct control of liquor distribution through state-run outlets”); see id., at 517-518 (Thomas, J., dissenting) (noting that, although laws creating a “state monopoly” in the sale of liquor “discriminate]” against interstate commerce, they are “within the ambit of the Twenty-first Amendment” and are therefore immune from scrutiny under the dormant Commerce Clause). There is, of course, no comparable provision in the Constitution authorizing States to discriminate against out-of-state providers of waste processing and disposal services, either by means of a government-owned monopoly or otherwise.

B

Nor has this Court ever suggested that discriminatory legislation favoring a state-owned enterprise is entitled to favorable treatment. To be sure, state-owned entities are accorded special status under the market-participant doctrine. But that doctrine is not applicable here.

Under the market-participant doctrine, a State is permitted to exercise “‘independent discretion as to parties with whom [it] will deal.’” Reeves, Inc. v. Stake, 447 U. S. 429, 438-439 (1980). The doctrine thus allows States to engage in certain otherwise-discriminatory practices (e.g., selling exclusively to, or buying exclusively from, the State’s own residents), so long as the State is “acting as a market participant, rather than as a market regulator,” South-Central Timber Development, Inc. v. Wunnicke, 467 U. S. 82, 93 (1984) (emphasis added).

Respondents are doing exactly what the market-participant doctrine says they cannot: While acting as market participants by operating a fee-for-service business enterprise in an area in which there is an established interstate market, respondents are also regulating that market in a discriminatory manner and claiming that their special gov*363ernmental status somehow insulates them from a dormant Commerce Clause challenge. See ibid.

Respondents insist that the market-participant doctrine has no application here because they are not asserting a defense under the market-participant doctrine, Brief for Respondents 24-25, but that argument misses the point. Regardless of whether respondents can assert a defense under the market-participant doctrine, this Court’s cases make clear that States cannot discriminate against interstate commerce unless they are acting solely as market participants. Today, however, the Court suggests, contrary to its prior holdings, that States can discriminate in favor of in-state interests while acting both as a market participant and as a market regulator.

IV

Despite precedent condemning discrimination in favor of government-owned enterprises, the Court attempts to develop a logical justification for the rule it creates today. That justification rests on three principal assertions. First, the Court insists that it simply “does not make sense to regard laws favoring local government and laws favoring private industry with equal skepticism,” because the latter are “often the product of ‘simple economic protectionism,’ ” ante, at 343 (quoting Wyoming v. Oklahoma, 502 U. S. 437, 454 (1992)), while the former “may be directed toward any number of legitimate goals unrelated to protectionism,” ante, at 343. Second, the Court reasons that deference to legislation discriminating in favor of a municipal landfill is especially appropriate considering that “ ‘[w]aste disposal is both typically and traditionally a local government function.’” Ante, at 344 (quoting 261 F. 3d 245, 264 (CA2 2001) (Calabresi, J., concurring)). Third, the Court suggests that respondents’ flow-control laws are not discriminatory because they “treat in-state private business interests exactly the same as out-of-state ones.” Ante, at 345. I find each of these arguments unpersuasive.

*364A

I see no basis for the Court’s assumption that discrimination in favor of an in-state facility owned by the government is likely to serve “legitimate goals unrelated to protectionism.” Discrimination in favor of an in-state government facility serves “‘local economic interests,’” Carbone, 511 U. S., at 404 (O’Connor, J., concurring in judgment) (quoting Raymond Motor Transp., Inc. v. Rice, 434 U. S. 429, 444, n. 18 (1978)), inuring to the benefit of local residents who are employed at the facility, local businesses that supply the facility with goods and services, and local workers employed by such businesses. It is therefore surprising to read in the opinion of the Court that state discrimination in favor of a state-owned business is not likely to be motivated by economic protectionism.

Experience in other countries, where state ownership is more common than it is in this country, teaches that governments often discriminate in favor of state-owned businesses (by shielding them from international competition) precisely for the purpose of protecting those who derive economic benefits from those businesses, including their employees.2 Such discrimination amounts to economic protectionism in any realistic sense of the term.3

*365By the same token, discrimination in favor of an in-state, privately owned facility may serve legitimate ends, such as the promotion of public health and safety. For example, a State might enact legislation discriminating in favor of produce or livestock grown within the State, reasoning that the State’s inspectors can more easily monitor the use of pesticides, fertilizers, and feed on farms within the State’s borders. Such legislation would almost certainly be unconstitutional, notwithstanding its potential to promote public health and safety. See Philadelphia v. New Jersey, 437 U. S. 617, 627 (1978) (noting that the Court has repeatedly invalidated legislation where “a presumably legitimate goal was sought to be achieved by the illegitimate means of isolating the State from the national economy”).

The fallacy in the Court’s approach can be illustrated by comparing a law that discriminates in favor of an in-state facility, owned by a corporation whose shares are publicly held, and a law discriminating in favor of an otherwise identical facility that is owned by the State or municipality. Those who are favored and disfavored by these two laws are essentially the same with one major exception: The law favoring the corporate facility presumably benefits the corporation’s shareholders, most of whom are probably not local residents, whereas the law favoring the government-owned facility presumably benefits the people of the enacting State or municipality. I cannot understand why only the former law, and not the latter, should be regarded as a tool of economic protectionism. Nor do I think it is realistic or consistent with our precedents to condemn some discriminatory laws as protectionist while upholding other, equally discriminatory laws as lawful measures designed to serve legitimate local interests unrelated to protectionism.

For these reasons, I cannot accept the proposition that laws discriminating in favor of state-owned enterprises are *366so unlikely to be the product of economic protectionism that they should be exempt from the usual dormant Commerce Clause standards.

Proper analysis under the dormant Commerce Clause involves more than an inquiry into whether the challenged Act is in some sense “directed toward ... legitimate goals unrelated to protectionism”; equally important are the means by which those goals are realized. If the chosen means take the form of a statute that discriminates against interstate commerce — “ ‘either on its face or in practical effect’ ” — then “the burden falls on [the enacting government] to demonstrate both that the statute ‘serves a legitimate local purpose,’ and that this purpose could not be served as well by available nondiseriminatory means.” Taylor, 477 U. S., at 138 (quoting Hughes v. Oklahoma, 441 U. S. 322, 336 (1979)).

Thus, if the legislative means are themselves discriminatory, then regardless of how legitimate and nonprotectionist the underlying legislative goals may be, the legislation is subject to strict scrutiny. Similarly, the fact that a discriminatory law “may [in some sense] be directed toward any number of legitimate goals unrelated to protectionism” does not make the law nondiseriminatory. The existence of such goals is relevant, not to whether the law is discriminatory, but to whether the law can be allowed to stand even though it discriminates against interstate commerce. And even then, the existence of legitimate goals is not enough; discriminatory legislation can be upheld only where such goals cannot adequately be achieved through nondiseriminatory means. See, e. g., Philadelphia, supra, at 626-627 (“[T]he evil of protectionism can reside in legislative means as well as legislative ends,” such that “whatever [the State’s] purpose, it may not be accomplished by discriminating against articles of commerce coming from outside the State unless there is some reason, apart from their origin, to treat them differently”); Hunt v. Washington State Apple Advertising Comm’n, 432 U. S. 333, 352-353 (1977) (explaining that “we *367need not ascribe an economic protection motive to” discriminatory laws; such laws are subject to strict scrutiny even “if enacted for the declared purpose of protecting consumers from deception and fraud in the marketplace”).

Dean Milk Co. v. Madison, 340 U. S. 349 (1951), is instructive on this point. That ease involved a dormant Commerce Clause challenge to an ordinance requiring all milk sold in Madison, Wisconsin, to be processed within five miles of the city’s central square. See