The obstruction doctrine holds that state laws run afoul of the spirit of the Constitution when they foreclose access to national markets, fragment subjects demanding a uniform national rule, or impede sustained federal initiatives. None of these principles are novel. Each has been applied across cases spanning railroad regulation, milk reciprocity agreements, and interstate trucking. What has been missing is a unifying framework that makes these principles legible.

The doctrinal anchor is Buck v. Kuykendall, a decision that constitutional scholars have largely passed over. Justice Brandeis identified all three features of the obstruction doctrine in a single opinion, distinguishing laws that merely burden interstate commerce from those that structurally obstruct it. That distinction is the doctrine’s organizing principle.

The doctrine differs meaningfully from the dormant commerce clause (DCC). Where the DCC asks whether a state law discriminates against or unreasonably burdens out-of-state commerce, the obstruction doctrine asks a prior structural question: Does the law undermine the constitutional prerequisites for a functioning union? That distinction matters considerably after National Pork Producers Council v. Ross narrowed the scope and applicability of the DCC.

The practical stakes of recognizing and acting on the doctrine are high. State legislatures are moving aggressively to regulate artificial intelligence and digital infrastructure in ways carrying substantial extraterritorial consequences. The obstruction doctrine does not resolve every contested question in this space, but it provides a principled, historically grounded, and judicially workable framework for identifying when states have exceeded their constitutional authority.

The Constitution prevents states from obstructing the conditions a national market requires: free movement of goods and services across state lines, room for competition and innovation, and protection of Americans in every state from regulatory regimes imposed by lawmakers in one. This prohibition is woven into the letter and spirit of the document. Yet, a failure to precisely name this limitation and identify its substance has given rise to a troubling doctrinal and popular development. An increasing number of lawmakers and public policy stakeholders assert that, in the absence of affirmative congressional activity, a state may claim extraordinary authority to defend what its residents deem to be in the best interest of the whole of the nation, even in ways that disrupt the freedom of residents of other states. Particularly, state officials have seemingly determined that the alleged high stakes of several issues have increased the extent to which states may indirectly burden interstate commerce when exercising their police powers.

The sponsor of an artificial intelligence (AI) bill in New York that, as originally drafted, imposed several transparency requirements on AI developers across the nation explained that the bill was necessary “in order to keep everyone in the city, state, country, and world safe from some pretty extreme risks.”¹ Lawmakers in California have likewise defended far-reaching state AI legislation on the basis that Congress’s failure to tackle certain hypothetical risks posed by AI means that states must step in. According to Gov. Gavin Newsom, states “cannot afford to wait for a major catastrophe to occur before taking action to protect the public [from AI risks].”²

State and local climate regulations have followed the same logic. New York City attempted to pass the costs of upgrading the city’s infrastructure for new climate conditions onto five fossil fuel producers it deemed disproportionately responsible for causing alterations to the city’s climate and landscape. In legal action, the city explicitly defined climate change as the sort of “existential threat” that demands such action.³ These sorts of laws are becoming more and more common.

The Founders anticipated states acting in this fashion and sought to preclude them from doing so. The transition from the Articles of Confederation to the Constitution was, in no small part, a rejection of the idea that states could pursue their own priorities without regard for national consequences. This policy analysis introduces the obstruction doctrine to resolve that ambiguity. The doctrine identifies three categories of state action the Constitution forecloses: laws that eliminate access to national markets, laws that fragment a subject demanding national uniformity, and laws that hinder federal prerogatives. Courts have applied each of these constraints for over a century. They have done so without a unifying label.

This policy analysis supplies one. By naming this doctrine and beginning the process of identifying supporting case law, other scholars can assist with filling out its contours and locating constitutional principles and judicial decisions that bolster its legitimacy. The “Developing the Obstruction Doctrine” section introduces the obstruction doctrine through Buck v. Kuykendall, a 1925 Supreme Court decision that best captures the doctrine’s logic and identifies its three operative prongs. The remainder of the paper explores these prongs in more detail: The Prong I section contends that the doctrine bars states from limiting interstate competition; the Prong II section argues that the doctrine denies states the authority to dictate national policy; and the Prong III section asserts that the doctrine broadly protects federal policy prerogatives from state interference. The paper concludes by highlighting areas for further study.

This is not meant to be an exhaustive analysis but rather a catalyzing paper that sparks additional inquiry. It’s also important to flag from the outset that the obstruction doctrine is explicitly and intentionally synthesizing existing theories and frameworks. As demonstrated by the slow emergence and eventual wide recognition of the major questions doctrine, it may be the case that certain constitutional principles are dormant or underutilized until they are more clearly expressed and explained. The struggle to precisely locate where the Constitution bans state intrusion into national affairs warrants this effort to tie together loose threads. Whether this doctrine brings together the right threads in the proper order is very much a matter that should be debated.

This analysis proceeds by inductive constitutional reasoning. It reads across judicial decisions applying the commerce clause and related provisions to identify principles courts have applied without naming, and synthesizes those principles into a doctrine that courts and scholars can use and refine. A fuller account grounded in constitutional text and original meaning is among the tasks this paper calls for others to take up. That account may strengthen the doctrine, complicate it, or reorder its prongs. Any of the three outcomes would advance the underlying inquiry being explored here.

What is not subject to debate, however, is the current ambiguity as to whether and how states may intervene in matters of national concern. As long as this ambiguity endures, states will persist in testing the boundaries of their authority and their ability to dictate national outcomes, especially when they believe that doing so is a matter of the nation’s survival. The US tried a system in which states can advance their self-serving interests without due regard for the implications for the rest of the country. It proved unworkable. The stakes today—in AI, energy, and the infrastructure of a competitive national economy—are too high to risk finding out how an imbalance could derail overall liberty and innovation.

Two clarifications are important to establish at the outset: First, the obstruction doctrine is a negative constraint on state authority; it is not an affirmative grant of federal authority. A finding that a state law obstructs interstate commerce says nothing about whether Congress may regulate the same subject. That question turns on the enumerated powers and nothing else. The doctrine clears state-level barriers to a national market; it does not build or compel a federal regulatory regime in their place. Where the doctrine strikes down a state law and Congress has no enumerated basis to act, the result is no rule at all. The Constitution tolerates that result.

A second clarification follows from the first. When this analysis refers to the national interest or to subjects requiring national treatment, the reference is not to a centrally determined policy agenda. It is to the structural prerequisites for a functioning union, one that avoids the pitfalls of the Articles of Confederation. Minimally, that includes free movement of goods and services across state lines, competition that rewards innovation, and protection of out-of-state residents from regulatory regimes they had no role in shaping. The doctrine guards those prerequisites. It does not guard any particular vision of what the federal government should do with the space the obstruction doctrine clears, and it is important that the federal government not abuse such authority in ways that would themselves interfere with individual liberties.

The obstruction doctrine rests first on the text of the Constitution. The commerce clause grants Congress authority to regulate commerce among the several states. That grant carries a structural corollary the Founding generation understood well and very much intended. Although not expressly spelled out in the text of the Constitution, a grant of power to Congress to regulate commerce among the states and otherwise address matters that implicate the viability of the union would be meaningless if states retained authority to obstruct that commerce themselves and thereby undermine national prosperity.

The failure of the Articles of Confederation was, in significant part, a failure of coordination. States erected barriers against one another’s goods, discriminated against out-of-state residents, and pursued local advantage at national expense. That failure prompted the drafters to instruct the Committee of Detail to grant Congress the power to regulate “in all cases for the general Interests of the Union, and also in those Cases to which the States are separately incompetent, or in which the Harmony of the United States may be interrupted by the Exercise of individual Legislation.”⁴ The commerce clause was the manifestation of that direction. It embodies a constitutional commitment to an integrated national economy, one that state regulatory power may fragment or foreclose only at constitutional peril. Courts have given that commitment various judicial expressions over time. The dormant commerce clause (DCC) is one, but as presently interpreted, it does not wholly address the issues that motivated the drafters. The obstruction doctrine, as this paper argues, is a more complete and reliable vehicle for ensuring fidelity to the vision of the union held by the drafters and ratifiers of the Constitution.

The obstruction doctrine is a constitutional rule against state laws that obstruct, rather than merely burden, the national interest. The doctrine emerges from several disparate doctrines that previously have not been formally tied together. The existing threads—market access, national uniformity, federal prerogatives—run through more than a century of commerce clause jurisprudence, in addition to decisions related to the privileges and immunities clause, the guarantee clause, and other federalism-related principles. But no single decision draws these cases and doctrines together as cleanly as a 1925 highway case that most constitutional scholars have largely passed over. This section introduces that case and explains why it serves as the doctrine’s foundation.

Constitutional doctrines sometimes operate for decades before anyone bothers to name them. The obstruction doctrine is one such principle. Courts have applied its logic, litigants have argued its substance, and legislators have occasionally felt its bite, all without a unifying label to make the doctrine legible. This section begins the work of providing one. The sources of the doctrine are many, and its threads run through a long line of commerce clause decisions. But a 1925 case, Buck v. Kuykendall, does the work of drawing those threads together with unusual clarity.⁵ Buck is worth dwelling on not merely because it struck down a state highway regulation, but because Justice Brandeis identified, in a single opinion, the three core features of what this paper calls the obstruction doctrine.

Buck concerned a State of Washington law that foreclosed common carriers from using the state’s highways for deliveries over fixed routes or between designated endpoints without approval from the state’s director of public works.⁶ Directors would grant such permission only if they deemed the service required by both public convenience and necessity. This kind of regulation was not unheard of at the time. Early highways were expensive to maintain and prone to deterioration, especially by heavy vehicles used to transport goods, people, or both.

The Supreme Court regularly upheld state laws intended to aid with infrastructure upkeep.⁷ In Hendrick v. Maryland, for instance, the Court upheld a Maryland law that mandated nonresidents obtain a certificate of registration before making use of the state’s highways.⁸ The general rationale for the constitutionality of such laws, despite being challenged under the commerce clause and equal protection clause, was that the state had authority to incidentally burden interstate commerce when clearly regulating with the intent of advancing the safety of state residents.⁹

In Buck, however, the Court struck down Washington’s law.¹⁰ The reasoning is important because it establishes an oft-neglected means of analyzing state laws when challenged for interfering with congressional authority to regulate interstate commerce. The Buck Court concluded that the law was more a “prohibition on competition” than a means to promote safety on highways or to facilitate their upkeep. A brief recitation of the facts illuminates the Court’s departure from its generally favorable approach to state highway regulations up to that point. Buck, a resident of Washington State, wanted to launch a bus line between Seattle and Portland. The line would operate on a highway that was built with federal funds. Buck secured the requisite approval from Oregon to start the line and otherwise satisfied Washington’s motor vehicle requirements. Yet, the relevant state officials did not grant him the final certificate because they contended that there were already enough means of transportation between the two cities.

Buck challenged the denial under the commerce clause and equal protection clause but lost in the lower courts.¹¹ The Supreme Court took the case. Justice Brandeis, writing for the majority, first summarized existing case law: State highways are the property of the state, and states may set fees to maintain and service those highways. The fees may vary according to proxies for the likelihood that the auto use in question poses a danger to others and results in “wear and tear.”

These powers, however, are not without limits. They must align with the Fourteenth Amendment, be designed to promote safety, and not be “obnoxious to the commerce clause,” as would be the case if the regulatory burden proved “unreasonable” relative to its benefits.¹² Washington’s law did not fit within those constraints. “The provision here in question is of a different character,” Brandeis reasoned, because its “primary purpose is not regulation with a view to safety or to conservation of the highways, but the prohibition of competition.”¹³ In short, the law obstructed interstate commerce rather than merely burdening it indirectly.

This obstruction was the product of three faulty bases for state regulation. First, Brandeis faulted the law for dictating “the persons by whom the highways may be used” rather than the manner of use. Second, the state’s inquiry—assessing the quality and sufficiency of interstate competition—is a “test which is peculiarly within the province of the federal action.” And third, limiting use of the highways more generally defeated the purpose of Congress as “expressed in the legislation giving federal aid for the construction of interstate highways.”¹⁴

That the Founders built such protections into the constitutional order makes it all the more noteworthy that consistently enforcing them has proven so difficult over time. Despite the lengths the Founders went to in developing a new system in which the central government had the requisite authority to advance the nation’s economic and general prosperity, challenges to expansive claims of state authority have commonly fallen between the cracks of disparate and contested constitutional ground, with a few exceptions.¹⁵

Some point to the DCC as a check on state authority.¹⁶ Such arguments are swiftly (and understandably) made less compelling by citation to the Court’s recent decision in National Pork Producers Council v. Ross.¹⁷ That decision revealed doubt among the justices as to whether the courts could strike down nondiscriminatory state laws on the basis of balancing a state’s moral considerations against broader economic factors.¹⁸ It is unclear the extent to which the DCC remains a viable route for striking down state laws that indirectly or directly impose constraints on nonresidents, at least so long as the ideological posture of the Court persists.

One objection deserves a direct answer. Buck is a DCC case. If National Pork Producers has narrowed the DCC’s reach, does the obstruction doctrine simply inherit that vulnerability, much as Lochner v. New York cannot be rescued from the collapse of substantive due process by attaching a different label to its reasoning?¹⁹ That does not seem to be the case: Lochner struck down a state law limiting the hours bakers could work because the Court majority held that the law was a labor regulation masquerading as health policy, and the Court asserted that the right to buy and sell labor was a fundamental economic liberty. But Lochner later fell because the Court repudiated the underlying doctrine wholesale. National Pork Producers did no such thing. It questioned the judicial manageability of balancing state moral preferences against economic burdens on interstate commerce. The market access and uniformity principles on which this doctrine primarily rests were left untouched by this ruling. The privileges and immunities clause, the guarantee clause, and the structural corollary the drafters intended to bake into the commerce clause do not rise and fall with the Supreme Court’s understanding of one strand of a much broader constitutional conversation. No DCC decision, narrowing or otherwise, has reached those grounds. The obstruction doctrine is broader than the DCC; that is the point of naming it and detailing it in this policy analysis.

The obstruction doctrine differs from the DCC in the question it asks. The DCC focuses on whether a state law discriminates against out-of-state commerce or imposes unreasonable economic burdens. The obstruction doctrine focuses on whether a state law structurally undermines the national interest the Constitution was designed to protect, including but not limited to the integrity of the national market. Discrimination and burden are symptoms; obstruction is the disease. A unified doctrine built around that distinction gives courts a more precise and durable tool than one built around balancing tests that shift with the ideological composition of the bench.

The case for a unified doctrine grows stronger when the remaining alternatives are assessed on their own terms. There is an argument to be made that such laws violate the guarantee clause by denying out-of-state residents the chance to hold accountable the officials in other states enacting de facto nationwide laws.²⁰ That line of attack—with possibly a few exceptions—has been more predominantly confined to law review articles than to judicial scrutiny. One could also flag the privileges and immunities clause where a state unnecessarily or disproportionately discriminates against out-of-state residents.²¹ Though that angle has worked in some contexts, it’s far from a reliable means of ensuring that states observe the constitutional boundaries on their authority.²² Scholars and jurists, including Justice Kavanaugh, could presumably cite several other provisions that may bolster the case against state laws that interfere with the nation’s well-being.²³ Each of these theories has its uses, but none provides a reliable, unified framework. Buck offers a way out of that difficulty. Though Justice Brandeis may not have intended to stitch together various constitutional threads into a cohesive doctrine, this opinion nevertheless serves as a useful guide for building out such a framework.

In sum, Buck signals the need to distinguish between mere impediments to interstate commerce and actual obstructions.²⁴ Brandeis set forth three such blocks: A state law obstructs interstate commerce when it eliminates access to national markets, when it forecloses uniform regulation of a subject that demands it, or when it hinders an established federal prerogative. Those three tests are the obstruction doctrine. But the doctrine does not authorize Congress to regulate any subject the doctrine reaches; rather, it identifies state laws that fail constitutional scrutiny. In short, the obstruction doctrine is not a backdoor to federal power. The sections that follow spell out the doctrine’s tests.

The constitutional prohibition on state laws that foreclose market access is well established in theory and inconsistently applied in practice. Decisions spanning more than a century—from railroad cases to milk regulations to waste disposal ordinances—have each enforced some version of this principle without assembling it into a coherent rule. This section connects those doctrinal strands.

Market access is the foundation of the constitutional order the commerce clause established.²⁵ The clause was designed to prevent states from treating their borders as economic weapons. Relatedly, the DCC serves to “establish a free market across the entire nation by restricting states and localities from impeding the free flow of goods across state lines.”²⁶ The first prong of the obstruction doctrine holds that line. A state law that blocks out-of-state competitors or otherwise redirects interstate traffic to other states—whatever its stated rationale—is an obstruction in the most direct sense of the word.

Fifty years after Buck, the Court doubled down on the idea that prohibition of market participation served as an obstruction per Buck’s understanding of interstate obstructions. In Great A&P Tea Co., Inc. v. Cottrell, the Court assessed the constitutionality of a Mississippi law that prevented out-of-state milk from being sold in the state unless a regulatory agency of the other state “accepts Grade A milk and milk products produced and processed in Mississippi on a reciprocal basis.”²⁷ A Maryland company with a milk production operation in Louisiana challenged the law as a violation of the commerce clause.

The company submitted an application to the requisite state authorities to sell its products in Mississippi. The application was denied because it did not include a reciprocal agreement between Louisiana and Mississippi. Notably, the state found no faults with the milk or the dairy’s operations. Still, a paper barrier concocted by the state blocked the free exchange of milk.

In reviewing the dispute, the Court highlighted that the “very purpose of the Commerce Clause was to create an area of free trade among the several states.”²⁸ It further clarified that the clause, “even without implementing legislation by Congress, is a limitation upon the power of the States.”²⁹

The district court tried to water down that constitutionally designated zone by framing the law as a nondiscriminatory, cost-cutting mechanism. The thinking went that “as long as Mississippi mutually exchanges standards of inspection with other states, there can be no burden on interstate trade.”³⁰

Justice Brennan, writing for the Court, saw through the idea that as long as all states agreed to a mutual means of raising barriers to exchange, such hindrances were constitutionally sound. Brennan faulted the lower court for “attach[ing] insufficient significance to the interference effected by the clause upon the national interest in freedom for the national commerce, and attached too great significance to the state interests purported to be served by the clause.”³¹ Protection of the “common market created by the Framers” demands that states not be allowed to initiate the sorts of trade wars that reciprocity mandates will inevitably spark.

On the whole, this prong of the obstruction doctrine dictates that a state “may not interfere with transportation into or through the State, beyond what is absolutely necessary for its self-protection,” as recognized way back in 1877 in and in several subsequent cases.³² In Toomer v. Witsell, for instance, the Court relied on the privileges and immunity clause to void a South Carolina law that “impose[d] an artificial rigidity on the economic pattern of the industry.”³³ More recently, in C & A Carbone v. Clarkstown, the Court struck down a local ordinance because, if others were to replicate that law, “the free movement of solid waste in the stream of commerce [would] be severely impaired,” as summarized by Justice O’Connor in a concurrence.³⁴ The Court in Morgan v. Virginia may have summarized this principle best in recognizing that the promotion and protection of national travel is a constitutional imperative.³⁵ That case dealt with a state law attempting to dictate a bespoke arrangement for the segregation of bus passengers. Even Justice Black, often a skeptic of the Court’s rulings related to protecting interstate commerce, recognized the faults with this law.³⁶ In a concurrence, he drew on Court precedent from 1877 to conclude that “uniformity in the regulations by which [a carrier] is to be governed from one end to the other of his route is a necessity in his business.”³⁷ Whether it be people, goods, or data, the upshot is that the Supreme Court has on several occasions found some means to thwart state laws that constituted an impediment to the otherwise steady flow of a national economy.

As indicated by Great A&P and related cases, market access is not a courtesy that states extend to out-of-state competitors. It is a constitutional guarantee. A state that forecloses that guarantee—whatever the justification—has crossed from regulation into obstruction.

But market access is only one dimension of the problem. A state can leave the market technically open while still doing serious constitutional damage by fragmenting the policy environment in which that market operates. That is the concern addressed in the next section.

The first prong of the obstruction doctrine polices the border between state regulation and market access. This second prong addresses a different but related problem: the state that regulates not to block a competitor but to fill a void the Constitution reserves for Congress. The difficulty is that the line between legitimate local regulation and impermissible national policymaking is not fixed. It moves. For most of the 20th century, states exercised broad and largely unchallenged authority over power generation, power grid policy, and zoning. That made sense when those decisions had only local consequences. It makes less sense today given the economies of scale of many industries, and given that certain industries—data centers, the physical infrastructure of AI and modern military logistics, are the most obvious example—sit at the intersection of state land-use law and national security. Recent drone strikes on Amazon Web Services cloud computing facilities in the Persian Gulf region are a reminder that what may look like a zoning question can be a strategic one with national security implications.³⁸ If data centers must be increasingly housed on American soil to advance national and economic security, then it cannot be the case that hundreds of counties can ban the development of that critical infrastructure. The physical boundaries of a state are an unreliable guide to whether a regulatory subject is local or national in character.

A railroad case from the mid-20th century, Southern Pacific Co. v. Arizona, established the analytical framework for spotting and stopping this kind of jurisdictional creep by states into national affairs. Southern Pacific Co. centered on the constitutionality of an Arizona law that subjected railroad companies to a financial penalty if they operated a train of more than 14 passenger cars or 70 freight cars. The state sought to recover fees from Southern Pacific for allegedly violating the law. The company admitted it ran a train of excess length but contested the law as incongruous with the commerce clause. The trial court sided with the railroad. The Supreme Court of Arizona, however, reversed for two reasons: First, it found the law was enacted to advance the health, safety, and well-being of residents. Second, it found that Congress had yet to specifically regulate the length of interstate trains, thereby leaving the door open for Arizona and other states to legislate in that domain.

The decision was appealed to the Supreme Court. The Court did not contest that Arizona was indeed acting pursuant to police powers. The Court also did not find that any specific federal law or regulation precluded the state from regulating train length. This latter finding is especially noteworthy because the Court thoroughly assessed whether related federal regulations stretched to cover this domain. What’s more, the Court noted that Congress had explicitly been invited to pass legislation related to train length and had opted not to. Still, the law was struck down under the commerce clause. The majority’s reasoning provides evidence for the existence of this prong of the obstruction doctrine.

Chief Justice Stone, writing for the majority, observed that the “residuum of power” left to states to regulate matters of local concern was subject to specific bounds.³⁹ First, regulations enacted pursuant to that residuum must pertain to issues “of local concern” that are “local in character and effect.”⁴⁰ Second, as addressed in the first prong of the obstruction doctrine, use of that power must not seriously interfere with the operation of national commerce. And third, “the consequent incentive to deal with [those local issues] nationally [must be] slight.”⁴¹ The Arizona law crossed that third line. The Court drew on a long line of cases to clarify that states cannot “regulate those phases of the national commerce which, because of the need of national uniformity, demand that their regulation, if any, be prescribed by a single authority.”⁴²

The Court acknowledged the state’s very real interest in ensuring safety over its rail network as well as the fact that the state had evidence that supported this intervention.⁴³ Nonetheless, one state’s economic preferences cannot disturb a national policy, whether established in practice or by legislation. Arizona was very much in danger of doing just that. The standard practice at the time was to run trains well beyond the length dictated by Arizona law. In fact, freight trains often stretched to more than 125 cars and even reached 160 with some regularity. This common practice meant that imposition of the law would result in upwards of $1 million in compliance costs for railroads operating in Arizona. And the negative consequences did not end there. The Court added:

Against this backdrop it is obvious that Arizona’s law was plainly incompatible with national practice. This conflict indicated the need for a nationwide policy. “If the length of trains is to be regulated at all,” explained the Court, “national uniformity in the regulation adopted, such as only Congress can prescribe, is practically indispensable to the operation of an efficient and economical national railway system.”⁴⁵

The holding in Southern Pacific did not emerge from thin air. In an earlier dissent, then-Justice Stone contended that such cases demand an extensive inquiry, including “a consideration of all the facts and circumstances, such as the nature of the regulation, its function, the character of the business involved and the actual effect on the flow of commerce,” as well as an evaluation of whether “the regulation concerns interests peculiarly local, and does not infringe the national interest in maintaining the freedom of commerce across state lines.”⁴⁶

Well before then, in Cooley v. Board of Wardens, the Court also flagged that “some [regulatory fields] imperatively demand[] a single uniform rule … and some, like the subject now in question [pilotage fees], as imperatively demanding that diversity, which alone can meet the local necessities of navigation.”⁴⁷ The Cooley Court unequivocally declared that regulation pertaining to the former—“subjects of this power [that] are in their nature national, or admit only of one uniform system, or plan of regulation”—is of “such a nature as to require exclusive legislation by Congress.”⁴⁸ More generally, Chief Justice Stone asserted that there have been many cases in which state laws “permissible because only local in their effect” were ultimately deemed to go too far “when they attempted to impose standards of equipment and operation.”⁴⁹ Justice O’Connor reached a similar conclusion, acknowledging that the Court’s willingness to defer to health and safety regulations imposed by the state often centered on the idea that “their burden usually falls on local economic interests … thus insuring that a state’s own political processes will serve as a check against unduly burdensome regulations.”⁵⁰ Yet such deference may not be warranted where, for whatever reason, those burdens are borne by out-of-state residents and may hinder competition that serves to benefit the national public.

The lesson of Southern Pacific and its antecedents is straightforward: When a regulatory subject demands uniform national treatment, the Constitution does not permit states to fill the void simply because Congress has not yet acted. The absence of federal legislation is not an invitation. It is a signal that the subject may be one the Constitution reserves for a single national authority. That is especially true where, as in the railroad context, the practical consequence of state-by-state regulation is to hand the most restrictive state effective control over the entire national system. Courts applying this prong of the obstruction doctrine must ask not whether Congress has spoken, but whether the subject is one that can tolerate the cacophony of 50 different answers.

The third prong of the obstruction doctrine is that state law may not unduly undermine existing federal initiatives such as extensive infrastructure projects. This prong adds another bound on state action by demanding analysis of the extent to which the federal government is affirmatively pursuing some larger effort implicated by the state law.

Several decisions provide the contours of this prong. In an oft-overlooked case, Kassel v. Consolidated Freightways Corp., the Court again struck down a state law that, while grounded in the state’s police powers, entered a domain that demanded a national approach.⁵¹ In relevant part, the Court held that

Notably, the Court did not specify a single congressional act or federal program when pointing to a potential conflict between the state law and “the federal interest.” Instead, the justices seem to have inferred from decades of sustained and significant investment that the federal government was especially dedicated to allowing for the free flow of commerce over the nation’s roadways. To allow states to detract from such efforts would strike at the effectiveness the Founders sought for the federal government.

Justice Brennan, joined by Justice Marshall, concurred in Kassel and accentuated the need for states to collectively and fairly contribute to the maintenance of a national market infrastructure. Citing legislative history that revealed protectionist intentions of Iowa legislatures, Brennan faulted Iowa for trying to reap the benefits of the interstate trucking network while directing its costs to others: “Iowa may not shunt off its fair share of the burden of maintaining interstate truck routes, nor may it create increased hazards on the highways of neighboring States in order to decrease the hazards on Iowa highways.”⁵³ He emphasized that this sort of unneighborly legislation—the kind referenced by Alexander Hamilton in the Federalist Papers—undercut the traditional deference states received when purportedly exercising their police powers.⁵⁴ “The decision of Iowa’s lawmakers to promote Iowa’s safety and other interests at the direct expense of the safety and other interests of neighboring States,” explained Brennan, “merits no such deference.”⁵⁵

Relatedly, in Raymond Motor Transportation, Inc. v. Rice, the federal government’s general stake in robust interstate travel served as a basis to attack a state law that imperiled such traffic.⁵⁶ Under the relevant statute, which limited the length of tractor-trailer trucks that could travel on Wisconsin roads to prohibit tandem trailers, a truck company would have to stop before entering the state, detach a trailer, hire an additional driver with a tractor to separately haul that single trailer, and then coordinate at the other end of the state to bring the trailers back together. In some instances, companies found it more cost-effective to simply route entirely around the state (the 23rd-largest in the US by area). The state offered no sound empirical evidence that this law would have any significant safety outcomes, nor did it contradict findings that longer trucks may actually pose fewer safety risks. As an apparent final justification, the state defended the law by claiming it was what the people wanted.

The Raymond Motor Court was unpersuaded. Even though the Court admitted that state legislation pertaining to safety has often received heightened protection, it could not overlook the significant ways in which the law eroded the nation’s long-standing, albeit abstract, commitment to a free-flowing transit network. The increased costs of moving goods, the delays in moving those same goods, and the elimination of certain deals that trucking companies could have reached absent this policy all proved too much for the Court to stand behind Wisconsin.

Still, in some instances, the Court has failed to adequately consider pervasive expressions of the federal government’s policy priorities. Prior to the aforementioned cases, it was occasionally true that a state’s mere reference to police powers permitted legislative actions that impeded federal initiatives. The Court has fallen into this trap in previous eras. Consider, for example, its decision in Bradley v. Public Utilities Commission of Ohio.⁵⁷ In that case, the Court permitted the state commission to block interstate trucks from driving on a specific state route because of the possibility of undue and dangerous traffic congestion.⁵⁸ While such an aim may seem a proper use of police power, the state advanced this policy based solely on a commission’s report that analyzed highway conditions along just 2.5 miles of road. Given that the federal government has serially and substantially invested in interstate vehicular traffic, decisions like Bradley likely amount to a misstep by the Court. If states are permitted to continually foreclose their own roads from use, then the federal government may find itself spending more and more on maintaining roads that nevertheless benefit that state.

That misstep, however, has been the exception rather than the rule. Instead, the tendency has been to ensure that when states legislate around the infrastructure essential to the nation’s economic prosperity, they do so only in a manner that does not unduly disrupt the use of that infrastructure or that encourages said use. For example, in County of Mobile v. Kimball, the justices drew a line between state and federal responsibilities and powers over harbors, bays, and navigable rivers. The state (Alabama) could improve those critical commercial avenues and channels—“improve” meaning, in this instance, to make more commercially useful, such as by removing obstacles. The Court explained that the state may “promote the growth of that internal commerce and insure its safety.”⁵⁹ This promotional power includes “an undoubted right to remove obstructions from their harbors and rivers, deepen their channels, and improve them generally”; such interventions, however, may “not impair their free navigation as permitted under the laws of the United States, or defeat any system for the improvement of their navigation provided by the general government.”⁶⁰ This latter limitation is an expression of prong three of the obstruction doctrine.

It would be a mistake to read these cases as being only about transportation. The roads and rails at issue were proxies for something larger: the commercial networks that the federal government has built, maintained, and in some cases deliberately left open through a combination of spending, regulation, and restraint. The internet, the power grid, and the physical infrastructure of the digital economy are the modern equivalents. They did not emerge from state action alone but reflect decades of federal decisions—some affirmative, some intentional abstentions—about how commerce should flow across the country. A state law that disrupts those networks raises the same constitutional concern that guided the Court in Kassel, Raymond Motor, and Kimball.

In sum, the federal government does not need to have passed a statute addressing every state law that undermines a national initiative. Decades of sustained investment speak for themselves. A state that free-rides on that investment while simultaneously erecting barriers to its effectiveness has no constitutional ground to stand on.

The argument for the obstruction doctrine is, at its core, an argument about what kind of federal system the Constitution created. It is not an argument against states: States retain broad authority to regulate in the interest of their residents, and nothing in the doctrine disturbs that authority where it is genuinely and appropriately exercised. Nor is it an argument for federal regulation. Striking down a state law under the doctrine does not necessarily authorize Congress to fill the space that law occupied. The doctrine is indifferent to whether Congress legislates, whether it declines to legislate, or whether the resulting market is governed by no statute at all. Its concern is whatever state law may be running afoul of the doctrine’s three tests. The doctrine is an argument against a particular kind of state behavior: the kind that treats congressional silence as permission, that treats local majorities as proxies for national consensus, and that treats the urgency of a problem as a license to ignore the constitutional allocation of power.

The Founders were familiar with that behavior. It was the defining failure of the Articles of Confederation, as detailed above. The Constitution answered that failure by ensuring national governance of national issues.

The obstruction doctrine recovers that understanding. Its three prongs are expressions of principles that courts have applied, inconsistently and without a unifying label, across more than a century of commerce clause jurisprudence. Giving those principles a name is an effort to make the doctrine easier to apply and harder to evade. States that have grown accustomed to exploiting the ambiguity in this area of the law should expect that the apparent gray zone will not last indefinitely.

The constitutional case for the doctrine is strong. The policy case is perhaps even stronger. And the practical case—given the pace at which state legislatures are moving to regulate AI, energy, and digital infrastructure—is strongest of all. Consider what the doctrine would say about the laws described at the outset of this paper: New York’s RAISE Act—which requires frontier AI developers operating anywhere in the state to publish safety protocols, submit to a new state oversight office within the Department of Financial Services, and report qualifying safety incidents to state authorities within 72 hours—poses a conflict with prong two.⁶¹ Frontier AI governance, which involves setting the rules for who can develop leading models and how and when they can be developed, is precisely the kind of subject that demands uniform national treatment.⁶² A developer operating across state lines cannot simultaneously train models that comply with New York’s mandates, California’s expectations, and whatever standards other states may yet impose.⁶³ The consequence is exactly what Southern Pacific warned against: the most restrictive state acquiring effective regulatory authority over the entire national system, with the market aligning downward to the lowest common denominator.

New York Gov. Kathy Hochul celebrated the RAISE Act as “nation-leading” and explicitly designed to fill a void left by federal inaction. But those qualities do not save the law from the constitutional scrutiny: Under the obstruction doctrine, congressional silence on a subject demanding national uniformity is not an invitation for state intervention.⁶⁴ It is a signal that the subject may be one the Constitution reserves for a single authority. California’s 2023–2024 Senate Bill 1047, though ultimately vetoed, illustrated the same problem in sharper relief.⁶⁵ The bill’s expansive jurisdictional scope covered developers with only marginal connections to California, meaning its kill-switch mandates, third-party audit requirements, and pre-release risk assessments would have applied to any company doing business in the state regardless of where development occurred. Such a predictable and substantive extraterritorial effect cannot be dismissed as merely incidental. It is prong-one obstruction: a state conditioning market access on compliance with a regulatory framework that Congress has not authorized and that no single state is constitutionally positioned to impose.

This policy analysis has traced the obstruction doctrine through more than a century of commerce clause decisions. The doctrine holds that a state law is unconstitutional when it forecloses access to national markets, when it fragments a subject that demands a uniform national rule, or when it impedes a sustained federal prerogative. Together, those three prongs supply a more reliable constitutional check on state overreach than any existing alternative.

While this paper and the author’s expertise lie in AI policy, the obstruction doctrine is not limited to that domain. For example, New York’s Climate Change Superfund Act presents yet a different area in which the doctrine would apply.⁶⁶ By imposing $75 billion in retroactive liability on fossil fuel companies for extraction and refining activities that occurred predominantly outside New York’s borders—and without any finding of wrongdoing—the law operates as both a prong-one and prong-three violation. It disrupts the national energy market by forcing producers to internalize costs that will inevitably be passed through to consumers nationwide, and it does so by unilaterally assigning liability for conduct that the federal government has regulated, and in some respects deliberately left unregulated, through decades of energy policy. As the federal government argued in its own challenge to the act, the law threatens to raise energy costs for consumers across the country and to undermine the uniform national energy market that the commerce clause was designed to protect.⁶⁷ That is obstruction—of markets, of national policy, and of federal prerogatives—by any name. Stopping such obstruction by states should in no way be presumed to provide the federal government with its own green light to obstruct the rights or freedoms of its citizens.

More important scholarly tasks lie ahead to more fully map out and apply the doctrine. The first is to determine whether the three prongs identified here are exhaustive. (The case law surveyed above suggests they may not be.) There are decisions that resist clean classification under any of the three, and that resistance may signal the existence of additional prongs rather than analytical error. Locating and naming those prongs is the next step. The second task is evidentiary. Each prong identified here rests on a handful of cases. A doctrine capable of doing the work this paper claims requires a far denser evidentiary foundation: more cases, more constitutional provisions, more engagement with the historical record.

This future work should not obscure what the doctrine already provides. Courts do not need a fully mapped framework to apply its principles; they need a coherent account of what those principles are and where they come from. This policy analysis has supplied that account. The three prongs identified here are grounded in more than a century of commerce clause jurisprudence, and they are sufficient to resolve the most pressing constitutional questions now arising from state efforts to regulate AI, energy, and digital infrastructure. The pace of state legislation in these areas means that those questions will reach courts sooner rather than later. When they do, the obstruction doctrine will be available, not as a novel theory, but as a recovery of principles the Constitution has always contained.

Frazier, Kevin T. “The Obstruction Doctrine,” Policy Analysis no. 1022, Cato Institute, Washington, DC, July 28, 2026.