with whom Justice White joins, concurring in part and dissenting in part.
I agree with the Court that this case presents two issues: (1) the validity of the Mayor’s executive order as applied to projects funded entirely by the city of Boston with its own revenues, and (2) the validity of the order as applied to projects funded in part with federal revenues pursuant to certain congressionally created grant programs.
— i
Respecting the second issue, I am m agreement with the Court’s conclusion that Congress, in creating the grant programs in question, specifically authorized “the type of parochial favoritism expressed in the order.” Ante, at 213. As the Court holds, Congress unquestionably has the power to authorize state or local discrimination against interstate commerce that otherwise would violate the dormant aspect of the Commerce Clause. Prudential Ins. Co. v. Benjamin, 328 U. S. 408, 418-427 (1946).1
*216HH t-H
I do not agree, however, with the Court’s holding that the executive order is immune from Commerce Clause scrutiny insofar as it applies to city activities undertaken without specific congressional authorization.
The Court rejects certain arguments advanced by the Supreme Judicial Court of Massachusetts as relevant only if the order were “regulation of,” rather than “participation in,” the market. Ante, at 210-211. The Court holds that the order is the latter rather than the former because, in the Court’s view, it “falls well within the scope,” ante, at 211, n. 7, of the Court’s decisions in Hughes v. Alexandria Scrap Corp., 426 U. S. 794 (1976), and Reeves, Inc. v. Stake, 447 U. S. 429 (1980). With due respect, this plainly is-not so.
In Alexandria Scrap, the effect of the Maryland statute was to offer a subsidy only to scrap processors located within the State. See 426 U. S., at 803, n. 13. The Court held that, a State, free from Commerce Clause scrutiny, may enter “the market as a purchaser, in effect, of a potential article of interstate commerce” and “restrict] its trade to its own citizens or businesses within the State.” Id., at 808. Alexandria Scrap thus permits a State to prefer its residents as direct recipients of certain subsidies. See Reeves, 447 U. S., at 440, n. 14 (discussing Alexandria Scrap).
In Reeves, South Dakota refused to sell cement to out-of-state consumers until the orders of all in-state customers were filled. The Court held that the Commerce Clause is not implicated when a State prefers its own residents as direct purchasers of state-produced goods. Neither Reeves *217nor Alexandria Scrap, however, went beyond ensuring that the States enjoy “ ‘the long recognized right of trader or manufacturer, engaged in an entirely private business, freely to exercise his own independent discretion as to parties with whom he will deal.”’ Reeves, 447 U. S., at 438-439, quoting United States v. Colgate & Co., 250 U. S. 300, 307 (1919).
Boston’s executive order goes much further. The city has not attempted merely to choose the “parties with whom [it] will deal.”2 Instead, it has imposed as a condition of obtaining a public construction contract the requirement that private firms hire only Boston residents for 50% of specified jobs.3 Thus, the order directly restricts the ability of private employers to hire nonresidents, and thereby curtails nonresidents’ access to jobs with private employers. I had thought it well established that, under the Commerce Clause, States and localities cannot impose restrictions granting their own residents either the exclusive right, or a priority, to private sector economic opportunities. See H. P. Hood & Sons v. Du Mond, 336 U. S. 525 (1949); Pennsylvania v. West Virginia, 262 U. S. 553 (1923); cf. Hicklin v. Orbeck, 437 U. S. 518 (1978) (decided under the Privileges and Immunities Clause).
Such restrictions are not immune from attack under the Commerce Clause solely because the city has imposed them as conditions to its contracts with private employers. In Reeves, the Court, I thought, carefully explored reasons the policy there at issue might not have been entitled to the market participant exemption, notwithstanding the policy’s essentially proprietary nature. 447 U. S., at 440-447. The *218Court also observed that the line between “market participant” and “market regulator” is not always bright: “South Dakota, as a seller of cement, unquestionably fits the ‘market participant’ label more comfortably than a State acting to subsidize local scrap processors.” Id., at 440. See id., at 440, n. 14 (“We have no occasion here to inquire whether subsidy programs unlike that involved in Alexandria Scrap warrant characterization as proprietary, rather than regulatory, activity”).
The line between regulation and market participation, for purposes of the Commerce Clause, should be drawn with reference to the constitutional values giving rise to the market participant exemption itself. As the Court recognized in Reeves, the most important of these is that historically the “Commerce Clause responds principally to state taxes and regulatory measures impeding free private trade in the national marketplace”; it was not designed “to limit the ability of the States themselves to operate freely in the free market.” Reeves, 447 U. S., at 437. The Court also observed that the distinction between participation and regulation rests on core notions of state sovereignty, coupled with the traditional right of private traders to determine the identities of their bargaining partners free from governmental interference. Id., at 438-439. The legitimacy of a claim to the market participant exemption thus should turn primarily on whether a particular state action more closely resembles an attempt to impede trade among private parties, or an attempt, analogous to the accustomed right of merchants in the private sector, to govern the State’s own economic conduct and to determine the parties with whom it will deal.
The simple unilateral refusals to deal that the Court encountered in Reeves and Alexandria Scrap were relatively pure examples of a seller’s or purchaser’s simply choosing its bargaining partners, “long recognized” as the right of traders in our free enterprise system. The executive order in this case, in notable contrast, by its terms is a direct attempt to *219govern private economic relationships. The power to dictate to another those with whom he may deal is viewed with suspicion and closely limited in the context of purely private economic relations.4 When exercised by government, such a power is the essence of regulation.
*220Attempts directly to constrict private economic choices through contractual conditions are particularly akin to regulation because, unlike simple refusals to deal but like conventional market regulation, they threaten to extend their regulatory impact well beyond the transaction in which the State has an interest. A requirement that firms wishing to deal with the State hire a certain percentage of their work force from among state residents in practice may constrict the opportunities of nonresidents to work on projects with no connection whatever with the governmental entity imposing the condition. A firm that relies to any significant degree on a permanent work force will be compelled to favor local residents for these positions. An analogous requirement that such firms purchase only from in-state suppliers the goods used in state projects also might constrict interstate trade wholly unrelated to government business. If economic considerations counsel in favor of stable relationships with suppliers, a firm wishing to deal with the State will be compelled to favor local firms across the board. The effect of such “conditions” on the ability of nonresidents to deal with affected firms would be virtually identical to the effect of a conventional market regulation requiring such practices.
In Reeves, the Court cited “considerations of state sovereignty” as another factor counseling restraint in applying the Commerce Clause to “proprietary” activity. The States have a sovereign interest in some freedom from federal interference when hiring state employees. It might be argued that because the city could have chosen to build the projects covered by the order itself and, free from dormant Com*221merce Clause restraint, could have hired local residents, the city may contract to have the work done by private firms on the condition that the firms hire local residents.5 But the Court never has suggested that the State’s special sovereign interest in determining whom it will hire, and in setting the terms and conditions of public employment, extends to dictating whom private parties with which it contracts will hire, or the terms and conditions of private employment. In my view, the State’s interest in managing its relations with its employees is fully safeguarded by its power to do the work itself if it so chooses, with such immunity from the Commerce Clause as attaches in that situation. The Court’s observation in Reeves, 447 U. S., at 438-439, tying concerns for state sovereignty to a merchant’s customary power to exercise his independent discretion as to the parties with whom he will deal, is-fully consistent with this view. But when a State attempts to arrogate unto itself the “independent discretion” of others to deal with whom they please, it exercises regulatory power that must be consistent with the requirements of the Commerce Clause. See generally Varat, State “Citizenship” and Interstate Equality, 48 U. Chi. L. Rev. 487, 560-564 (1981).
This approach fully safeguards the power of the State to limit to state residents the direct benefits of subsidy programs supported with state funds. It permits a State to prefer local businesses as providers of the goods it purchases in the marketplace, and to prefer local residents as direct purchasers or recipients of state-created bounty. But it does not permit a State to impose clear market regulations as a condition of a contract or of a subsidy, using the tremendous power of the state treasury directly to impede the free flow of private trade in interstate commerce, or, what may be worse, to discriminate against such commerce. South Dakota should not be immune from the Commerce Clause if, for *222example, it imposes as a condition on the sale of state-owned cement that purchasers employ only South Dakota residents, or resell the cement only to South Dakota customers. Cf. Reeves, 447U. S.,at444,n. 17 (“Nor has South Dakota cut off access to its own cement altogether, for the policy does not bar resale of South Dakota cement to out-of-state purchasers”). Similarly, Maryland should not be free of Commerce Clause scrutiny if it imposes as a condition of receiving a bounty, like that at issue in Alexandria Scrap, that scrap processors employ only Maryland residents, or resell the processed scrap only in-state. In my view such conditions, like the condition at issue here, directly intrude upon the historic Commerce Clause concern with “measures impeding free private trade in the national marketplace.” Reeves, 447 U. S., at 437.
I do not intend to suggest that the Court necessarily would decide these variations of Alexandria Scrap and Reeves as it has decided this case; evidently, the Court acknowledges that “restrictions that reach beyond the immediate parties with which the government transacts business” pose Commerce Clause questions more profound than did the restrictions at issue in Alexandria Scrap and Reeves. Ante, at 211, n. 7. The Court indicates that it upholds the executive order on the understanding that, with the exception of the federal grant programs, it is applied solely to construction projects funded entirely by the city. Ante, at 208-209. Because many construction contractors hire a substantially different work crew for each project they undertake, applied to such projects the Mayor's order is arguably limited, as the Court says, to a “discrete, identifiable class of economic activity in which the city is a major participant.” Ante, at 211, n. 7.6 *223This unique aspect of employment in the construction industry — and of public works construction projects — must also underlie the Court’s related justification that “[e]veryone affected by the order is, in a substantial if informal sense, ‘working for the city.’” Ibid.
I am not persuaded, however, that even the comparatively limited terms of the executive order constitute “market participation” rather than “market regulation.” The “sense” in which those affected by the Mayor’s order “work for the city” is so “informal,” in my view, as to lack substance altogether. The city does not hire them, fire them, negotiate with them or their representative about the terms of their employment, or pay their wages. In the case of the employees of subcontractors regulated by the order, the city does not even pay, or contract directly with, their employers. In short, the economic choices the city restricts in favor of its residents are the choices of private entities engaged in interstate commerce. Thus, the executive order directly impedes “free private trade in the national marketplace,” and for that reason I would not hold it immune from Commerce Clause scrutiny. I therefore reach the question whether the order imposes an impermissible burden on interstate commerce.
I — ( h — i h-(
As the Court recognizes, the order constitutes parochial favoritism” of Boston residents over nonresidents of Boston and Massachusetts for access to private sector jobs. Ante, at 213. Thus, the order is a “protectionist measure” subject to the rule of virtually per se invalidity established by many of this Court’s cases. See, e. g., Philadelphia v. New Jersey, 437 U. S. 617, 624 (1978).7
*224That the order burdens Massachusetts residents living outside Boston to the same extent as residents of other States does not save the order from this rule. First, the order derives in part from a state statute encouraging all Massachusetts communities to institute similar measures. Mass. Gen. Laws Ann., ch. 149, §26 (West 1982).8 That statute is clearly designed to benefit all Massachusetts residents at the expense of all residents of other States. In carrying out this statutory mandate, Boston, a creature of the Commonwealth, is tainted by participation in the Commonwealth’s larger and clearly discriminatory scheme.
Second, and more significant, the order would be improper under Dean Milk Co. v. Madison, 340 U. S. 349 (1951), even absent the state statute. In Dean Milk, this Court held that a Madison, Wis., city ordinance “plainly discriminate[d] against interstate commerce,” even though “Wisconsin milk from outside the Madison area [was] subjected to the same proscription as that moving in interstate commerce.” Id., at 354, and n. 4. This was so because the ordinance “erect[ed] an economic barrier protecting a major local industry against competition from without the State.” Id., at 354. The *225Court held that the ordinance was invalid because “reasonable nondiscriminatory alternatives, adequate to conserve legitimate local interests, [were] available.” Ibid.
Boston has at its disposal reasonable alternatives to accomplish its central goal — the alleviation of unemployment among Boston residents. It can create training programs for its unemployed residents or establish aggressive referral practices aimed at promoting employment for its residents at all construction projects in the city without implicating Commerce Clause concerns. It also can undertake some of the construction projects itself, and hire Boston residents to work on them, without imposing discriminatory restraints on the private market.
Moreover, as in Hicklin v. Orbeck, 437 U. S. 518 (1978), the order is ill-suited to éliminating unemployment because it applies the preference to all Boston residents, not just the underemployed or undertrained. See id., at 527-528. Finally, since Dean Milk, the Court has indicated that a discrimination against interstate commerce is unjustified unless there is a legitimate reason, apart from their out-of-state origin, to treat differently articles of commerce or individuals engaging in commerce originating outside the State. Philadelphia v. New Jersey, 437 U. S., at 626-627. No such reason has been shown in this case.
Insofar as the Massachusetts court held Boston’s executive order violative of the Commerce Clause as applied outside the context of federal grant programs, I would affirm its judgment. To this extent, therefore, I respectfully dissent.
Because the Court does not pass on the possible invalidity of the executive order under the Privileges and Immunities Clause, U. S. Const., *216Art. IV, § 2, cl. 1, it has no occasion to determine whether Congress may authorize, through affirmative legislation, what otherwise would be a violation of that Clause. This question may present considerations different from those presented by the dormant Commerce Clause. See L. Tribe, American Constitutional Law § 6-31, p. 403, n. 18 (1978). For the reasons given by the Court, ante, at 214-215, n. 12, I also decline to reach this issue.
Had the city decided to limit its own hiring to Boston residents, its decision would almost certainly have been permissible under McCarthy v. Philadelphia Civil Service Comm’n, 424 U. S. 645 (1976), as well as Reeves and Alexandria Scrap.
That the order limits the preference to 50% of the covered jobs is, of course, not relevant to the applicability of the market participant exemption. If such preferences do not implicate the dormant Commerce Clause, they are immune even if they apply to 100% of a contractor’s jobs.
Compare United States v. Colgate & Co., 250 U. S. 300, 307 (1919) (unquestioned right of trader unilaterally to refuse to deal with those retailers who do not adhere to retail price schedule), relied upon in Reeves, 447 U. S., at 439, with United States v. Parke, Davis & Co., 362 U. S. 29, 45-46 (1960) (trader violates Sherman Act by inducing wholesalers to refuse to deal with retailers who will not adhere to price schedule), United States v. Arnold, Schwinn & Co., 388 U. S. 365, 382 (1967) (“Once the manufacturer has parted with title and risk, he has parted with dominion over the product, and his effort thereafter to restrict territory or persons to whom the product may be transferred” violates the Sherman Act), and Continental T. V., Inc. v. GTE Sylvania Inc., 433 U. S. 36, 49 (1977) (overruling Schwinn in part; nonprice vertical market restrictions are not per se unlawful, but should be judged individually under the “rule of reason” to determine whether they “should be prohibited as imposing an unreasonable restraint on competition”).
Conditioning a willingness to deal with potential bargaining partners on their derivative refusals to deal with others is particularly suspect where those whom the trader attempts to isolate are its competitors. See Lorain Journal Co. v. United States, 342 U. S. 143, 154-155 (1951). Here, the citizens of Boston, through their Mayor, have sought to do just this by requiring those wishing to deal with their city government to refuse to hire nonresidents competing with citizens for jobs. This anticompetitive and suspect goal will be present whenever a unit of state or local government requires recipients of public contracts or government subsidies to deal only with that government’s constituents.
Congress, in § 8(e) of the National Labor Relations Act, has expressly prohibited labor organizations from requiring employers to agree “to cease or refrain from handling, using, selling, transporting or otherwise dealing in any of the products of any other employer, or to cease doing business with any other person,” and has declared any such agreement to be void. 29 U. S. C. § 158(e). On the other hand, permitting labor unions to refuse to deal with the primary employer is the staple of federal labor policy, and nothing prevents an employer from refusing unilaterally to deal with others for any lawful reason. To be sure, in the construction industry, at issue in the executive order, collective-bargaining agreements are expressly exempted from this proscription of “hot cargo” clauses. Ibid.; see *220Woelke & Romero Framing, Inc. v. NLRB, 456 U. S. 645 (1982). That Congress has chosen, however, for reasons peculiar to labor policy and the history and nature of collective bargaining in the construction industry, to exclude collective agreements in that industry from this restriction does not detract from my basic point: there is a world of difference between the kind of “proprietary” activity at issue here and the kind exempted from dormant Commerce Clause scrutiny in Reeves and Alexandria Scrap.
Indeed, the Court appears to rely on this argument. See ante, at 211, n. 7.
See S. Rep. No. 187, 86th Cong., 1st Sess., 27 (1959) (“The occasional nature of the employment relationship makes [the construction] industry markedly different from manufacturing and other types of enterprise. An individual employee typically works for many employers and for none of them continuously. Jobs are frequently of short duration, depending upon *223various stages of construction”). It is noteworthy, however, that in this case the parties have stipulated that the order affects some “contractors who have regular and permanent work crews.” Agreed Statement of Facts, App. to Pet. for Cert. A41.
1 reject the suggestion that the record does not establish a cognizable burden on interstate commerce. See ante, at 209-210, and n. 6. The city *224has stipulated that, as a result of this order, some construction workers who are nonresidents of Massachusetts will be unemployed; contractors . from outside Massachusetts will be discouraged from bidding on affected projects; and the costs of construction on affected projects will increase. Agreed Statement of Facts, App. to Pet. for Cert. A37.
The Mayor’s executive order itself states that one of its purposes is to satisfy the city’s “statutory obligation to give preference to its residents in hiring for public[ly] funded construction projects pursuant to [Massachusetts] G. L. c. 149, §26.” App. to Pet. for Cert. A19. The statute to which the order refers provided: “Each county, town or district in the construction of public works, or persons contracting or subcontracting for such works, shall give preference [in hiring] to veterans and citizens who are residents of such county, town or district.” Mass. Gen. Laws Ann., ch. 149, § 26 (West 1982). In its decision holding the city order unconstitutional, the Supreme Judicial Court of Massachusetts also struck down this statute. 384 Mass. 466, 476-478, 425 N. E. 2d 346, 352-353 (1981). The Commonwealth has not appealed that ruling.