Concept
Exceptions and Defenses
Negative implications of the commerce clause
Exceptions and Defenses
Discrimination against interstate commerce is nearly fatal to a state law, yet the Court has recognized settings in which a state may adopt even a blatantly discriminatory measure and see it stand. Two exceptions are principal. Congress may approve the state law, and a state acting as a market participant may favor its own citizens in receiving benefits from government programs or in dealing with government-owned businesses.1 The list is often extended to affirmative cash subsidization and to certain forms of state self-promotion that channel business activity to a state’s own commercial operations.2 The Court has also limited each of these, creating what might be called exceptions to the exceptions, so that none can become a way around the values the doctrine protects.3
The five subsections below do not all do the same work, and the difference tells you where each belongs in an answer. The market participant doctrine and congressional consent take a discriminatory law outside the dormant Commerce Clause altogether, and the rule on downstream restrictions marks the edge of the first. The rule for traditional government functions works differently: the law stays inside the doctrine, but a court finds no discrimination and applies the lenient balancing test. The Twenty-first Amendment gives states broad power over alcohol and no exemption. Liquor laws that favor local producers remain subject to the doctrine, so a state that invokes the Amendment is making an argument that usually fails.
Reach these questions after you have found discrimination and before you declare the law invalid. Ask whether the state is buying, selling, or spending its own money; whether it favors a public facility that performs a traditional government function; and whether Congress has consented. Treat a Twenty-first Amendment argument as the state’s weakest whenever the law favors local producers.
Market Participant Doctrine
The dormant Commerce Clause restrains a state that taxes or regulates private trade. It has no hold on a state that buys or sells in the market as a private business would. Hughes v. Alexandria Scrap Corp. established the rule: a state acting as a participant in the market, rather than as a regulator, may favor its own citizens in its commercial dealings. The label covers a state that buys or sells goods or services, operates its own enterprise, or subsidizes private business, and when it applies, the dormant Commerce Clause analysis does not, so the state’s conduct generally stands.4 An arrangement that would draw the strictest scrutiny under the earlier sections of this chapter therefore draws none.
The facts of Hughes show how little it took to qualify. Maryland wanted abandoned cars out of its streets and junkyards, so it paid bounties to processors that destroyed them. In 1974 the state amended the program: a processor with a plant in Maryland could collect a bounty on a hulk by submitting an indemnity agreement from its supplier, while an out-of-state processor had to produce title documentation.5 Unlicensed suppliers found the indemnity agreement easier to sign, so hulks that had flowed to out-of-state plants began to stay in Maryland, and Alexandria Scrap, a processor outside the state, claimed that the amendment burdened the interstate flow of hulks.6
The ordinary analysis would have put the state on the defensive. The Court sorts state laws into two groups. Laws that burden interstate commerce only incidentally fall only if the burden is “clearly excessive in relation to the putative local benefits,” while laws that affirmatively discriminate face more demanding scrutiny.7 Once a law is shown to discriminate “either on its face or in practical effect,” the burden falls on the state to show that it serves a legitimate local purpose that nondiscriminatory means could not serve as well.8 A rule that gave local processors an advantage over out-of-state rivals seemed headed straight for that burden.
The Court rejected the challenge by changing the question. The earlier cases on which the processor relied shared a common thread: the state had interfered with the natural functioning of the interstate market through prohibition or burdensome regulation. Maryland had prohibited nothing and regulated nothing. It entered the market itself, bidding up the price of hulks, and the hulks stayed home because market forces, including the state’s own money, made disposal in Maryland more lucrative.9 Nothing in the purposes of the Commerce Clause, the Court concluded, prevents a state from participating in the market and favoring its own citizens in the absence of congressional action.
Two separate opinions show the doctrine’s strongest defense and its sharpest criticism. Justice Stevens, concurring, reasoned that the commerce at issue existed only because Maryland had chosen to subsidize scrap processing. The state could have confined its bounty to Maryland plants from the start, and the Commerce Clause imposes no duty to subsidize out-of-state business.10 On that view the “burden” on the out-of-state processor was only the absence of a subsidy. Justice Brennan, dissenting, objected that a state restricting its purchases to items produced within its borders aims directly at pulling labor and industry inside the state, a form of discrimination the Court has long treated with particular suspicion, and he would have required the state to justify it.11 The majority’s view is the law. The dissent explains why the Court has been careful about where the exception ends, a subject taken up below.
Reeves, Inc. v. Stake carried the doctrine from a state that buys to a state that sells. South Dakota owned a cement plant, and its Cement Commission ordered that in times of shortage the plant turn away out-of-state customers until all orders from South Dakotans were filled.12 Reeves, a ready-mix concrete distributor organized under Wyoming law, had bought about 95 percent of its cement from the plant from 1958 to 1978.13 The Court upheld the policy. The distinction between states as market participants and states as market regulators, it said, “makes good sense and sound law,” and it relied on the long-recognized right of a private trader or manufacturer to choose the parties with whom to deal.14 A state that sells cement from its own plant fits the participant label more comfortably than one that pays bounties, which is why Reeves reinforces the rule in Hughes.15
The exemption rests on two reasons. The first is the purpose of the Clause. The Commerce Clause responds principally to state taxes and regulations that impede free private trade in the national market, and nothing in the constitutional plan limits the ability of the states themselves to operate freely in that market. A private manufacturer may pick its customers, and a state acting as a proprietor bears the same restrictions that bind private participants, so evenhandedness suggests that it should share their freedoms. The Court added that the competing considerations in proprietary cases are subtle, complex, and politically charged, and that adjusting them is a task better suited to Congress than to the courts. The Reeves dissenters answered that withholding a state’s cement from interstate commerce to benefit local business is the economic protectionism the Clause was meant to prevent.16
The second reason explains why the doctrine covers only the state’s own dealings. When a state sells cement only to its residents or pays bounties only to local processors, it gives up its own money or product, which is the economic equivalent of handing something of value to its citizens.17 The in-state taxpayers who fund that benefit can weigh its cost against its value, so politics inside the state checks the program. Regulation and taxation work differently, because they can shift the cost of a local benefit onto out-of-staters who have no vote in the enacting state, and the Commerce Clause was drafted to eliminate trade barriers of that kind.18
The same reasoning reaches services and a state’s own operations. A public university may charge nonresidents higher tuition than residents, because the state is dispensing its own services.19 South Dakota would equally have been exempt if its plant had hired only South Dakota residents, bought raw materials only from South Dakota companies, or given free cement only to South Dakota charities.20
The exemption stops where its rationale stops. It lets a government choose its own customers and suppliers, as a private firm may, and it does not let the government regulate private actors.21 Conditions that a state attaches to its own purchases raise a narrower question, taken up in the next subsection.
The Court’s cement decisions draw the same line. A state may confine sales from its own plant to residents, but it may not compel privately owned plants to do so, because the state does not own what those plants make. Four limits mark the points at which a state’s conduct stops being participation.
The first limit concerns how a subsidy is funded. A true subsidy paid from a state’s general revenue should not be invalidated under the dormant Commerce Clause, because the state’s own taxpayers carry the cost.22 In West Lynn Creamery, Inc. v. Healy, Massachusetts imposed a special tax on all liquid milk sold to retailers in the state, whether produced there or elsewhere, and placed the proceeds in an earmarked fund that subsidized Massachusetts dairy farmers.23 The Court invalidated the scheme. Linking the tax to the earmarked fund made it work like an ordinary tariff, because it effectively added costs to the purchase of out-of-state products, and the Court noted that it was not squarely confronting subsidies.24 The difference from Maryland is the source of the money. Maryland paid its bounties from its own funds, while Massachusetts raised its fund through a levy that reached out-of-state milk, and a state that imposes a levy is regulating, not participating. That is why a cash payment to a company that locates a factory in the state is ordinarily safe, while a mandate that a private company keep its factory in the state or prefer local workers is a regulation, and the participant exemption does not shelter it.25
A state that participates in one market may not use that position to regulate another. The minnow ban in Hughes v. Oklahoma illustrates the point. The state might have allowed only its own residents to take minnows from state waters, but it lost its participant immunity when it sought “to prevent privately owned articles of trade from being shipped and sold in interstate commerce.”26 The same limit sank Alaska’s timber-processing rule in South-Central Timber, the subject of the next subsection.
A third limit applies when a state controls access to the channels of commerce in which it holds a proprietary interest. States pay for, build, and own their roads, so courts might treat road-use rules as participant conduct, yet cases such as Bibb, Raymond, and Kassel condemn state laws that discriminate against or unduly burden interstate users of roadways. The refusal to apply the exemption in those cases appears to reflect the special importance of the infrastructure of commerce to interstate movement.27
The fourth limit is that the exemption belongs to the Commerce Clause and does not carry over to the Privileges and Immunities Clause of Article IV. The Court has declined to recognize a market participant exception under that clause, and in the Camden case it found potential merit in a challenge to an ordinance requiring that at least 40 percent of the employees of contractors and subcontractors on city construction projects be Camden residents.28 The city’s spending of its own funds does not remove the ordinance from review. The state must show a substantial reason for the discrimination: some demonstration that out-of-staters are a peculiar cause of the mischief and that treating them unequally is the only reasonable remedy.29 Discrimination in a matter the clause does not protect, such as the price of a hunting license, is not reached at all.30
The second step does most of the work, since nearly every dispute turns on whether the state spent its own resources or reached for private parties.
Downstream Restrictions
A state that sells its own property may choose its customers, but the market participant doctrine stops at the edge of the market in which the state actually deals. South-Central Timber Development, Inc. v. Wunnicke marks that edge. Alaska owned the timber it sold, yet it could not require buyers to process the logs in Alaska before shipping them out. A condition that controls what a purchaser does with the goods after the sale is a downstream restriction, and when it discriminates against interstate commerce the exemption does not protect it.32 The previous subsection listed this as the second limit on participant status, and the case supplies the reasoning behind it.
Alaska wrote into its timber sale contract a requirement that the successful bidder partially process the timber in Alaska before export, and it charged a significantly lower price for timber sold on those terms. South-Central, an Alaska corporation that exported unprocessed logs and ran no mill in the state, sued to enjoin the requirement.33 The Court first held that Congress had not authorized the requirement, a point taken up under congressional consent below. That holding left the market participant defense as the state’s principal argument.34
The defense looked strong because Alaska really was a seller. Alaska argued that its requirement was the same kind of subsidy to local interests that the Court had accepted in Alexandria Scrap. The Court answered that Maryland had been a purchaser of scrap, while Alaska participated in the timber market and imposed conditions downstream, in the separate market for processing timber. The difference shows in what the buyer remains free to do. A true processing subsidy would let a timber buyer take the benefit by processing in Alaska or forgo it and export raw logs. Under the Alaska contract the state made that choice for the buyer, because no one who bought the timber could take it out of Alaska before processing. The state paid for compliance through a lower price, so it bundled a discount with a command.
The Court stated the limit in terms an exam answer can reproduce. The doctrine permits a state to impose burdens on commerce within the market in which it participates, and it permits no more: the state may not impose conditions, whether by statute, regulation, or contract, that have a substantial regulatory effect outside that market.35 The dispute therefore turned on how to define the market. Alaska said it participated in the market for processed timber, although it took no part in the processing itself. South-Central replied that the state, a participant in the timber market, was using its leverage there to exert a regulatory effect in the processing market, where it had no role, and the Court agreed with South-Central. The market must be defined narrowly because nearly every sale contract can carry a condition, and nearly every condition can be described as part of some larger market. If the market were the whole chain from standing tree to finished lumber, a state that owned a sliver of the supply could regulate every later stage through its sale terms.
Two reasons explain why post-sale conditions deserve different treatment from the choice of whom to sell to. A state seller has a real stake, as a private trader would, in the immediate transaction, but it has no comparable stake in what a buyer does with the goods once the state no longer has an interest in them, and the Court drew an analogy to the common-law limits on restraints on alienation and to the antitrust limits on vertical restraints. The second reason concerns the reach of the control. A state that refuses to deal with a customer chooses its own trading partners, while a state that dictates what the buyer does afterward governs the buyer’s separate economic relationships, here with processors who were strangers to the sale. A sale ordinarily ends the seller’s say over the goods, but the Alaska contract bound the buyer after it had taken delivery and paid.
Alaska also read Reeves as leaving a state free to set any terms it likes when it sells. The Court called that reading unwarranted, because Reeves endorsed a state’s right to deal with whomever it chooses without sanctioning every term a state might desire. Reeves also differed on its facts: it involved no foreign commerce, it concerned a finished product made in a state plant rather than a natural resource, and South Dakota did not bar resale to out-of-state purchasers. The resale point is the most useful on an exam, because a restriction on what the buyer may do with the goods after delivery is the signature of a downstream condition.
The sale limited to residents sorts customers inside the market the state occupies, and the resale clause reaches outward from that market, which is why the first survives under Reeves and the second draws ordinary scrutiny.36
The permitted side of the line appears in White v. Massachusetts Council of Construction Employers, Inc., where the mayor of Boston ordered that city-funded construction projects be performed by a work force at least half composed of Boston residents.37 The order reached past the city’s own contracts, since it affected whom private contractors hired, yet the Court upheld it because everyone the order affected was in a substantial if informal sense working for the city.38 Boston was buying construction services with its own money, so the hiring condition operated inside the market the city had entered. Alaska’s condition took effect after the sale, once the timber had come to rest in private hands, and it governed activity in a market where Alaska bought and sold nothing. The same sorting works for a state that builds its own plant. A requirement that the contractors building the plant hire residents stays inside the construction market, whereas a requirement imposed on those who later buy the plant’s output reaches a different market.39
Labels do not move the line between participation and regulation. The plurality refused to let a state evade its reasoning by providing in the contract that title to the timber would not pass until processing was complete, because the substance of the transaction governs rather than the label attached to it.40 It also rejected the view that a condition is valid whenever the state has the economic power to impose it by contract, because on that view it would have been irrelevant in White that the employees were in effect working for the city.41 The practical reason is plain. If a contract clause could turn regulation into participation, any state could convert a regulation into a sale term by adding a sentence to the agreements it signs.
The Justices did not all agree about the line. Justice Rehnquist, dissenting, argued that Alaska was merely paying the buyer indirectly, through a reduced price, to hire Alaskans to process the timber. The state, he noted, could reach the same result by selling only to companies that maintain processing plants in Alaska, by subsidizing processors directly, or by paying for processing and then selling the processed logs, and he thought it “unduly formalistic” to forbid the one path the state had chosen.42 The plurality replied that a state’s action as a regulator cannot be upheld on the ground that it could have achieved the same end as a participant, just as it is no defense to a vertical restraint that vertical integration could reach the same end. On an exam, apply the plurality’s rule as the law and mention the dissent only to explain why the line is contested.
Because the outcome depends on where the market boundary falls, hard cases arise when the favored group straddles two markets. A state that gives scholarships to high school graduates on condition that they attend in-state colleges plainly participates in the market for public higher education. The condition also reaches students at private colleges, a market in which the state’s participation is much harder to claim, and whether the program is a downstream restraint depends on which market a court says the condition regulates.43 An exam answer should name both markets and argue which one the condition reaches. Losing participant status does not by itself condemn the condition; it sends the condition back to ordinary analysis. In South-Central Timber the Court then held that Alaska’s local-processing requirement, because of its protectionist nature and its burden on commerce, fell within the virtual per se invalidity of laws that block the flow of interstate commerce at a state’s borders. It added that the burden on foreign commerce called for even more rigorous and searching scrutiny.44
Traditional Government Functions
In United Haulers Association, Inc. v. Oneida-Herkimer Solid Waste Management Authority, the Court held that a flow control ordinance, meaning a local law that requires trash haulers to deliver their waste to one designated processing facility, does not discriminate against interstate commerce when a public entity owns and operates the facility and the law treats every private business, in-state or out-of-state, exactly the same. Because such a law favors a government performing a traditional government function, the Court reviewed it under the lenient test reserved for nondiscriminatory laws instead of the virtually per se rule that governs discrimination.46 The ordinances closely resembled the one struck down in C&A Carbone, Inc. v. Town of Clarkstown, and the Court identified a single salient difference: the favored facility belonged to a government. That one fact moved the case from the discrimination track to the balancing track, and the reasons it could do so are the subject of this subsection.
By the 1980s Oneida and Herkimer Counties in central New York faced what they could credibly call a solid waste crisis. Many local landfills operated without permits, many had to be closed and cleaned up at great public cost, and the local waste industry suffered from price fixing, pervasive overcharging, and the influence of organized crime. At the counties’ request, New York created the Oneida-Herkimer Solid Waste Management Authority, a public benefit corporation empowered to collect, process, and dispose of the counties’ solid waste. The counties agreed that the Authority would manage all of it, and private haulers remained free to pick up trash at the curb. To support the Authority, the counties required haulers to deliver waste to its facilities, and the haulers sued, arguing that the ordinances discriminated against interstate commerce under Carbone.
The haulers’ argument had force because the two ordinances looked alike. In Carbone, Clarkstown arranged for a private contractor to build a transfer station, which the contractor would operate for five years at an above-market tipping fee of $81 per ton before selling it to the town for one dollar, and the town passed a flow control ordinance to guarantee the station a steady volume of trash. The Court struck that ordinance down because it hoarded solid waste, and the demand to get rid of it, for the benefit of the preferred processing facility. The haulers read Carbone as treating favoritism toward a public facility and favoritism toward a private one alike, while the counties read it as leaving the public-facility question open. The majority accepted the counties’ reading for reasons found in Carbone itself: the local-processing cases on which Carbone relied involved discrimination in favor of private enterprise, and three Justices had argued at length in dissent that Clarkstown’s station was in substance public, a point the majority never answered. A decision that said nothing about public facilities had not decided how they are treated, so the question was now squarely presented.47
The Court’s first reason for answering differently concerns what discrimination means. Any notion of discrimination assumes a comparison of substantially similar entities, and states and municipalities are not similar to private businesses, because unlike a private business a government carries the responsibility of protecting the health, safety, and welfare of its citizens. Rigorous scrutiny of favoritism toward in-state business is appropriate because that favoritism is so often the product of simple economic protectionism, which makes scrutiny a screen for a suspect motive. The screen fits poorly when the favored party is a government, since laws favoring local government may serve many legitimate goals unrelated to protectionism.48 The counties show how that works: the ordinances let them pursue chosen policies for handling and treating waste and spread the cost of those policies across citizens and businesses according to the volume of waste each one generates.
The second reason concerns the role of the courts. If a public beneficiary drew the same suspicion as a private one, every public service that required citizens to use it would face near-fatal review, and federal judges would end up deciding which activities government may undertake and which must be left to private competition. The Court refused that task and said that the dormant Commerce Clause is not a “roving license” for federal courts to make that judgment. The voters of the two counties could have left waste management to the private sector, in which case any regulation they adopted could not discriminate against interstate commerce, but they were equally free to give the job to their own government and to adopt flow control ordinances in support of it. The choice deserved particular respect because waste disposal is typically and traditionally a local government function, and Congress itself has declared that collection and disposal of solid waste should remain primarily the function of state, regional, and local agencies.
The third reason is political and echoes the logic taught earlier for even-handed laws. The dormant Commerce Clause cases often find discrimination when a state shifts the costs of regulation onto outsiders, because a burden that falls on interests outside the state is unlikely to be eased by the political restraints that operate when interests inside the state are affected. Here the most palpable harm, more expensive trash removal, fell on the citizens and businesses of the counties, the same people who voted for the ordinances, so the Court saw no reason to hand local businesses a victory they could not win in the political process. The reasoning also suggests a contrast with Carbone. A private firm that holds an exclusive position under a flow control ordinance has a concentrated stake in defending it, whereas a public program’s competitors can be expected to line up against it.49
Finding no discrimination did not end the case, because a nondiscriminatory law remains subject to the balancing test from Pike v. Bruce Church, Inc., under which the law stands unless its burden on interstate commerce is “clearly excessive in relation to the putative local benefits.”50 After years of discovery, neither the magistrate judge nor the district court could detect any disparate impact on out-of-state as opposed to in-state businesses, and the Court found it unnecessary to decide whether the ordinances imposed any incidental burden at all, because any arguable burden did not exceed their public benefits. Two benefits carried the balance. The ordinances gave the counties a convenient way to finance an integrated package of waste services, and they increased recycling, because the counties accept recyclables and many hazardous wastes for free and routing all waste to the Authority’s facilities made recycling laws far cheaper to enforce. The financing benefit shows why the two tracks matter. Revenue generation cannot justify a discriminatory ordinance, which is why it failed to save Clarkstown’s, yet it is a cognizable benefit in the Pike balance, because the question there is whether an incidental burden is out of proportion to the benefits and not whether an interest can excuse favoring insiders.51
The balancing portion of the opinion did not command a majority. Justice Scalia joined the holding that the ordinances do not discriminate but declined to join the balancing, observing that weighing competing values is generally left to Congress. Justice Thomas concurred only in the judgment, saying that he no longer believed Carbone was correctly decided and that he would discard the negative Commerce Clause doctrine altogether. The exam structure stays the same, a no-discrimination finding followed by Pike, but these opinions show how much of the result rests on the first step.
This rule is easy to confuse with the market participant doctrine, and the two should be kept apart. The market participant doctrine lets a government pick its own customers and suppliers as a private firm would and removes that conduct from dormant Commerce Clause review, so that if it applied, not even the balancing test would be needed. A flow control ordinance does something different, because it requires private haulers to use a public service and so forbids citizens to contract with other suppliers, which places it outside the proprietary exception.52 Justice Alito’s dissent pressed that point, arguing that the counties acted as market participants in a fee-for-service business while also regulating that market in a discriminatory way. The majority asked whether the ordinances discriminate at all, answered no because the favored entity was a government and every private business was treated alike, and then applied Pike. The exemption it recognized therefore reaches only the discrimination rule and leaves the balancing test in place.
The dissent regarded the line between the two cases as thin. Justice Alito, joined by Justices Stevens and Kennedy, called the distinction one without a difference, because the government was still favoring an in-state facility at the expense of out-of-state competitors.53 In his account the Carbone station was private only in name: the contractor built it for the town, received a guaranteed minimum waste flow and an above-market fee, and was to sell it to the town for one dollar after five years, so the only real difference was that title had not yet formally passed, and the majority exalted form over substance.54 He also argued that favoring a government-owned facility serves local economic interests, because the local employees and suppliers who depend on the facility benefit from it, which makes such discrimination protectionism in any realistic sense.55 Tests that turn on whether a governmental function is traditional or integral, he added, were abandoned as unsound in Garcia v. San Antonio Metropolitan Transit Authority.56 The question will also recur, since roughly 80 percent of landfills receiving municipal solid waste in the United States are state or locally owned.57 The majority view is the law, and the dissent marks where an exam answer should acknowledge that ownership of the facility decides which of two near-identical ordinances falls.
The rule has a firm edge: a charge that sorts private parties by origin is discriminatory no matter how legitimate its aim. Oregon’s legislature imposed a surcharge on solid waste generated out of state, and the commission set it at $2.25 per ton,58 almost three times the $0.85 charged on in-state waste.59 Oregon said the surcharge merely recouped the cost of disposing of out-of-state waste. The Court answered that the purpose of, or justification for, a law has no bearing on whether it is facially discriminatory, so the surcharge discriminated even if it only recouped costs.60 Oregon then had to justify the charge, and it had offered no higher cost for out-of-state waste and no safety or health reason unique to that waste.61 The county flow control ordinance differed because it sorted no private business by origin.
The reasoning has already been extended once. In Department of Revenue of Kentucky v. Davis, Kentucky exempted from its income tax the interest on its own bonds while taxing interest on comparable bonds of other states, and a majority upheld the scheme on the strength of the waste case, reasoning that issuing debt securities to pay for public projects is a quintessentially public function with a long history. Dissenting Justices argued that a traditional-government-function standard restates the police power, which is too broad to sort laws, and that Kentucky’s bond tax discriminated openly in a competitive market. Two questions remain unsettled. A favored enterprise run as a public-private joint venture with substantial private involvement probably falls on the Carbone side of the line, since the reasoning depends on the enterprise being the government’s own. And a government that builds an enterprise outside the traditional sphere, such as a cement plant, and then orders local builders to use only its products presents a case the decisions do not resolve, even though cities commonly provide water services and so fit the traditional label.62
The first and third steps carry most of the contest, since ownership separates this case from Carbone and equal treatment of private businesses keeps the rule from becoming a cover for protecting local firms. A fact pattern that blurs ownership, such as a private operator under a long contract with guaranteed volume, invites an argument on each side, and the dissent’s critique supplies the counterargument.
Congressional Consent
The Commerce Clause is a grant of power to Congress, and the limit it places on the states is one the Court has inferred from that grant. Because the dormant Commerce Clause is the Court’s reading of congressional silence, Congress can remove the limit simply by speaking. Congress has plenary power to regulate commerce among the states, so its approval validates even a clearly discriminatory state law.64 Consent matters because without it the usual rule applies: once a law is shown to discriminate, the burden falls on the state to show a legitimate local purpose that nondiscriminatory means could not serve as well.65 As the Court put it in South-Central Timber, Congress may redefine the distribution of power over interstate commerce by permitting the states to regulate in ways that would otherwise be impermissible.66 The Constitution uses the same device in express terms elsewhere, as when it forbids the states to lay taxes on imports and exports “without the Consent of Congress.”67
Insurance is the most important area in which the Court has found congressional approval of state laws that the doctrine would otherwise condemn.68 In Western & Southern Life Insurance Co. v. State Board of Equalization, California taxed out-of-state insurers more heavily when their home states taxed California insurers more heavily. The Court began from the principle that if Congress ordains that the states may freely regulate an aspect of interstate commerce, action within the scope of the authorization is “invulnerable to Commerce Clause challenge.”69 Prudential Insurance Co. v. Benjamin shows the same result for a plain discrimination: a state taxed out-of-state insurers at a higher rate than local insurers, and the tax survived because Congress had adopted a statute leaving the regulation of insurance to the states.70 With Congress silent, that tax would meet the strictest scrutiny. With the statute in place, the state’s conduct falls within the authorization and the doctrine has nothing to say about it.
Consent can arrive after a judicial defeat or before one. In In re Rahrer the Court upheld a state law restricting the importation and sale of alcoholic beverages, although it had earlier declared an almost identical law from another state unconstitutional, because Congress had in the meantime adopted a statute expressly permitting such state regulation.71 In New York v. United States the Court upheld provisions of the Low-Level Radioactive Waste Policy Amendments Act of 1985 that authorized states with federally approved waste sites to reject waste from states that had not conformed to the Act. The statute made that discrimination against interstate commerce permissible before any lawsuit was filed.72 Rahrer shows Congress undoing the practical effect of a ruling, and New York shows Congress writing the permission into the program from the start. In both, Congress spoke directly to what the states could do.
What Congress says has to be clear. If there is any doubt whether Congress has given its permission, the state’s conduct remains subject to the normal limits of the doctrine.73 The Court put the standard this way in South-Central Timber: for a state regulation to be removed from the reach of the dormant Commerce Clause, congressional intent must be “unmistakably clear.” The standard asks for clarity and does not demand particular words, so a statute need not use the phrase “expressly authorized” to qualify. The reason is the same political-process point that explains the doctrine itself. Interests outside a state have no vote in its legislature and so often bear the brunt of its regulation, whereas a decision by Congress draws on representatives from every part of the country. Absent a clear expression of approval by Congress, any relaxation of the restrictions on state power increases the risk that unrepresented interests will be hurt by restraints on commerce.74 If vague signals could substitute, a state could assemble consent from a stray committee remark or a loosely related federal program, and the national market would be opened to state barriers by inference, without Congress ever having chosen that result.
Alaska’s claim of consent shows the standard in operation. The state required buyers of its timber to partially process the logs in Alaska before export. The Ninth Circuit upheld the requirement on the theory that Congress had implicitly authorized it through the federal policy of imposing a primary-manufacture requirement on timber taken from federal land in Alaska.75 The Supreme Court reversed. A state policy that is consistent with federal policy, or that furthers goals the Court might believe Congress had in mind, is an insufficient sign of congressional intent. Because Congress acted only with respect to federal lands, nothing supported an inference that it meant to authorize the same policy for state lands. The contrast with the insurance and radioactive-waste statutes lies in what Congress addressed. Those statutes spoke to what the states themselves may do, while Alaska pointed only to what Congress had done with its own land.
The clear-statement requirement applies to partial relaxations of the doctrine as well. In Maine v. Taylor, Maine argued that the 1981 Lacey Act Amendments directed federal courts to treat state wildlife laws more leniently, which would have eased the scrutiny of its ban on importing live baitfish.76 The Court refused. An unambiguous indication of congressional intent is required whether the claimed authorization is a flat exemption from Commerce Clause scrutiny or the less direct form of a reduction in the level of scrutiny.77 The Court found Maine’s reading of a Senate report “positively strained”: the more natural reading was that the passage refers only to the availability of federal investigative and prosecutorial resources to enforce valid state wildlife laws.78 A statute that supplies federal help in enforcing valid state laws says nothing about laws that are invalid without federal approval, so the ban had to satisfy the ordinary justification burden.
Consent has a boundary on the other side as well. When Congress approves a state law, the law no longer violates the dormant Commerce Clause, but it can still be challenged under other constitutional provisions, such as the Equal Protection Clause or the Privileges and Immunities Clause of Article IV.79 The reason is that consent works through Congress’s commerce power, and a statute enacted under that power cannot waive limits that lie outside it. Congress could not authorize the states to make treaties or coin money, and it could not waive the Article IV guarantee of privileges and immunities.80
On an exam, treat consent as a question to reach once discrimination or an excessive burden has been found. Look for a federal statute that addresses what the states may do, ask whether its language is unmistakably clear, and then check whether some other provision independently condemns the law.
Twenty-First Amendment Claims
Section 2 of the Twenty-first Amendment bans the importation or possession of intoxicating liquors into a state “in violation of the laws thereof.”82 Read by itself, that sentence sounds like a grant of unlimited state control over alcohol, and states defending liquor laws that favor local producers have argued that it switches the dormant Commerce Clause off for that one product. In Granholm v. Heald the Court rejected the argument. It held that state liquor laws enacted under the Amendment remain subject to the dormant Commerce Clause, and that a state violates the Clause when it lets in-state wineries ship directly to consumers while denying out-of-state wineries the same privilege.83 The Amendment therefore gives a state extra regulatory authority over alcohol without excusing discrimination, which is why the overview at the start of this section sets it apart from the doctrines that actually take a discriminatory law outside the Clause.
Section 2 was adopted to give each state the authority to address alcohol-related public health and safety issues in accordance with the preferences of its citizens.84 The Court has refused to read it as more than that. It treats the Amendment as one part of a single constitutional scheme, and it has held that state liquor laws violating the First Amendment, the Establishment Clause, the Equal Protection Clause, the Due Process Clause, or the Import-Export Clause are not saved by the Amendment.85 The history in Granholm shows that section 2 restored to the states the powers they had under the Wilson and Webb-Kenyon Acts, and that those powers never included authority to pass nonuniform laws discriminating against out-of-state goods. Decisions soon after ratification read the Amendment more broadly, and many states used the authority those decisions bestowed to expand their trade barriers, with rivalries and reprisals following. The Court observed that those early decisions declined to consider the Amendment’s history and that contemporaries had offered ample opinion to the contrary.86 A privilege to favor local liquor producers cannot belong to one state alone, because any state that uses it invites the others to answer in kind. The Court described the patchwork of existing laws as essentially the product of “an ongoing, low-level trade war.”
The modern cases under section 2 fall into three categories, and the third controls here.87 In the first, a liquor law that violates another constitutional provision is not saved by the Amendment. In the second, section 2 does not cut back Congress’s own commerce power over liquor. In the third, state regulation of alcohol is limited by the nondiscrimination principle of the Commerce Clause. The leading example is Bacchus Imports, Ltd. v. Dias, where Hawaii exempted certain locally produced alcoholic beverages from an excise tax, and the Court answered that the central purpose of the Amendment was not to empower states to favor local liquor industries by erecting barriers to competition. The Court has repeated that discrimination eliminates whatever immunity the Amendment would otherwise supply.88 The price-affirmation cases fit the same line. Brown-Forman Distillers Corp. v. New York State Liquor Authority acknowledged that the Amendment gave New York considerable authority to regulate the domestic sale of alcohol but held that it did not immunize a statute that regulated the sale of alcohol in other states.89 Healy then struck Connecticut’s statute on the same footing: the Commerce Clause precludes the application of a state statute to commerce that takes place wholly outside the state’s borders.90 Granholm later noted that Healy explicitly relied on the discriminatory character of the Connecticut statute.91
The Michigan and New York schemes show what discrimination looks like when the product is wine. Michigan let in-state wineries ship directly to consumers subject only to a licensing requirement, while out-of-state wineries faced a ban, so every bottle of out-of-state wine had to pass through an in-state wholesaler and retailer, two extra layers of overhead that raised its price to Michigan consumers. New York did not ban direct shipment outright, but it required an out-of-state winery to open a branch office and warehouse in the state first, an expense most wineries could not bear. The in-state presence condition runs into the long-standing rule that a state cannot make an out-of-state firm become a resident in order to compete on equal terms. Both schemes drew the line by origin, and the Court treated them as “straightforward attempts to discriminate in favor of local producers” that the Amendment does not save.92
None of this endangers the structure of liquor regulation itself. The Amendment leaves states virtually complete control over whether to permit the importation or sale of liquor and how to structure its distribution, so a state may ban alcohol altogether, run its own outlets, or funnel sales through the three tiers of producer, wholesaler, and retailer, and the Court has called the three-tier system “unquestionably legitimate.” The line between protected and unprotected policies is evenhandedness: state policies are protected when they treat liquor produced out of state the same as its domestic equivalent.93 The reason follows from what section 2 restored, which was the power to regulate liquor uniformly. A rule that sends a local vintner and a distant one through the same three tiers gives the local producer no advantage to exploit, so the Amendment’s grant of authority and the Commerce Clause’s ban on favoritism point in the same direction.
Because the Amendment did not authorize the discrimination, the Michigan and New York schemes still faced the justification burden: once a law is shown to discriminate, the burden falls on the state to show that it serves a legitimate local purpose that nondiscriminatory means could not serve as well.94 The states offered two main justifications, keeping alcohol from minors and collecting taxes, and both failed for want of proof. On minors, the states offered little evidence that minors buy wine over the internet, a staff study found that the 26 states then allowing direct shipment reported no problems with minors’ access, and the Court reasoned that minors are less likely to drink wine than beer or spirits, have easier ways to break the law, and gain little from a channel that cannot deliver alcohol immediately.95 Without concrete evidence, the Court was left with unsupported assertions, and those fall short of the clearest showing that discriminatory regulation requires.96 Age verification or the use of common carriers that verify age would address the concern with far less restriction.97
The tax-collection justification fared no better. New York could require a permit as a condition of direct shipping, as it already did for in-state wineries, with regular sales reports and tax remittance, and other states that tax direct interstate wine shipments this way reported no problems.98 Federal law adds incentives to comply, since a winery that violates state law can lose its federal license, and state attorneys general may sue wineries in federal court to enjoin violations. The remaining rationales, orderly markets, health and safety, and regulatory accountability, could be served by an evenhanded licensing requirement, and electronic background checks and mailed or emailed records had eased the burden of monitoring distant wineries. Courts have upheld discriminatory regulations only after concrete record evidence showed that nondiscriminatory alternatives would prove unworkable. In Maine v. Taylor, an 8-1 decision, the Court upheld a ban on out-of-state baitfish.99 It deferred to findings that Maine had a legitimate and substantial purpose, given the uncertain effects of baitfish parasites on the state’s wild fish,100 and that no scientifically accepted inspection procedures for baitfish parasites existed.101 Maine proved that no workable inspection method existed, while Michigan and New York offered assertions that permits, reports, and age checks would fail. The Court summarized the result in one sentence: if a state chooses to allow direct shipment of wine, it must do so on evenhanded terms.102
The decision came by five votes to four, and the two dissents read the Amendment differently. Justice Stevens, joined by Justice O’Connor, argued that the Eighteenth and Twenty-first Amendments placed commerce in alcohol in a special constitutional category, so laws that would plainly fail for an ordinary article of commerce can stand when they regulate wine, and he relied on Justice Brandeis’s statement in Young’s Market that requiring imports to compete on equal terms would rewrite the Amendment instead of construing it. Justice Thomas, joined by Chief Justice Rehnquist and Justices Stevens and O’Connor, rested instead on the Webb-Kenyon Act, which prohibits shipping liquor into a state when it is intended to be received, possessed, sold, or used in violation of that state’s law, and he argued that the Act’s language displaces any negative Commerce Clause barrier to the Michigan and New York laws.103 The majority answered that the Act’s text can be read to forbid shipment only where it runs afoul of a state’s generally applicable laws, a phrase the dissent understood to mean all state laws except those that discriminate against out-of-state liquor.104 The Webb-Kenyon argument is in substance a congressional consent argument of the kind taught in the preceding subsection, and it succeeds only if the federal statute speaks clearly enough to reach the discriminatory law.
The dissents did not change the law. A later decision involving Tennessee’s two-year residency requirement for liquor-store licenses explained that section 2 gives each state leeway in choosing alcohol-related public health and safety measures but is no license for protectionist restrictions, and it struck the requirement because it blatantly favored residents and bore little relationship to public health and safety.105 Where the predominant effect of a law is protectionism, not the protection of public health or safety, section 2 does not shield it.106 On an exam, a Twenty-first Amendment argument arrives after discrimination has been found, and it is the state’s weakest argument when the law favors local producers.