abstract. This essay argues that the interpretive struggle over the meaning of American federalism has recently shifted from the Commerce Clause to two textually marginal but substantively important battlegrounds: the Necessary and Proper Clause and, to a lesser extent, the General Welfare Clause. For nearly a decade, these quieter, more structurally ambiguous federal powers—the “shadow powers,” as I term them—have steadily increased in prominence. Beginning with Gonzales v. Raich (2005) and continuing through and beyond National Federation of Independent Business v. Sebelius (2012), the Supreme Court’s federalism jurisprudence has shifted from its once-typical form of inquiry into the scope of Congress’s commerce power, refracted through the Tenth Amendment, to become an inquiry into the transsubstantive reasons for allowing Congress to regulate at all. Paradoxically, the growth of shadow powers analysis has tended to narrow the permissible scope of congressional regulatory power. The novelty of shadow powers analysis lies in the sharp line the Court appears increasingly willing to draw between solid, if controversial, Article I powers such as the commerce power, and auxiliary Article I powers such the necessary and proper power. The invocation of the shadow powers has helped the Court find room to maneuver within its federalism analysis, while also appearing to maintain its commitment to an apparently unmoving baseline of a narrow commerce power. The growth of shadow powers analysis has obscured the outlines of federalism’s map—to shroud genuine (and perhaps salutary) doctrinal changes within a fog of constitutional text, insufficiently overruled precedents, and acontextual readings of foundational cases.

author. Professor of Law and Ludwig and Hilde Wolf Teaching Scholar, University of Chicago Law School; Associate Member, Department of History, University of Chicago. I am grateful to my fellow contributors to this Feature, Jessica Bulman-Pozen, Heather Gerken, Abbe Gluck, and Cristina Rodríguez, and to William Birdthistle, Aziz Huq, Genevieve Lakier, Charles Morse, H. Jefferson Powell, Judith Resnik, David Strauss, and the participants in the 2013 University of Wisconsin Discussion Group on Constitutionalism and DePaul University College of Law Faculty Seminar for helpful discussion and comments. I thank Abigail Molitor for excellent research assistance. I am pleased to acknowledge the support of the Mayer Brown Faculty Research Fund.

Introduction

What does American federalism require? Most observers agree on a few general principles: federalism in some form is a fundamental ingredient of the U.S. Constitution; at minimum, federalism means that the powers of the federal government are not unlimited; the exercise of those powers must be grounded in text, structure, or practice; and the states should be understood as having a definite and meaningful identity, ranging from co-equal sovereign to regulatory partner. A commitment to federalism requires, in short, that Americans constantly measure their messy legal and political structure against a hazily defined and capacious idea upon which there is little agreement beyond the fact that many of the Founders regarded federalness as one of the nation’s essential attributes. Today, federalism means, at a minimum, viewing both the states and the federal government as legitimate sources of legal and political authority, but little consensus exists as to what that general principle of multiplicity should mean in practice.1

Where is American federalism to be found in the Constitution? The word is never mentioned in the document itself, in either the 1787 text or the amendments. But commentators have long recognized that the text, structure, and underlying logic of the Constitution assume and endorse a federal system of government.2 Modern constitutional law typically focuses on three main textual and doctrinal sources of federalism: (1) the enumeration principle (Article I); (2) judicial review of state law by the Supreme Court (Article III plus the Supremacy Clause of Article VI, Clause 2); and (3) the Supremacy Clause itself, especially the requirement that the Constitution, laws, and treaties of the U.S. are the supreme law of the land and bind judges in the states.3

A related but distinct potential locus of federalism in both text and doctrine is the Tenth Amendment, which states, “The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people.”4 Beginning in the 1930s, and gaining new vigor in the 1970s, the Tenth Amendment became the touchstone for the view that federalism means taking the states seriously as sovereigns.5 Indeed, for many judges and commentators, the mere invocation of the Tenth Amendment amounts to a normative statement about the value of the states in the federal structure and the concomitant limits on federal power.6 In some cases, the Tenth Amendment is treated as a constitutional guarantee of “the province of state sovereignty”7 and “local power always existing” in the states;8 in others, it is “but a truism that all is retained which has not been surrendered.”9

How do we know what federalism ought to look like today? Following the invalidation of the Child Labor Act in Hammer v. Dagenhart in 1918, and continuing for much of the twentieth century, the paradigmatic federalism question as framed by the Supreme Court was the correct balance between Congress’s power to legislate under Article I, on one hand, and the states’ large, ill-defined, and perhaps exclusive regulatory domain on the other.10 In Hammer, Justice Day, writing for the Court, set forth the robust view of the Tenth Amendment that echoed down through decades of case law:

In interpreting the Constitution, it must never be forgotten that the nation is made up of states to which are entrusted the powers of local government. . . . The power of the states to regulate their purely internal affairs by such laws as seem wise to the local authority is inherent and has never been surrendered to the general government.11

Justice Holmes’s dissenting opinion, in contrast, crystallized the opposing argument from Article I:

I should have thought that the most conspicuous decisions of this Court had made it clear that the power to regulate commerce and other constitutional powers could not be cut down or qualified by the fact that it might interfere with the carrying out of the domestic policy of any State.12

After 1937, as is well known, the Court adopted an increasingly deferential stance toward congressional regulation under the commerce power; as part of this shift, the Court overturned its decision in Hammer.13 But the Hammer Court’s framing of the debate—inherent state authority over internal affairs on one hand, plenary federal power within a defined Article I sphere on the other—continued to structure the debate for decades. Throughout the 1980s and 1990s, the federalism debate repeatedly returned to the Tenth Amendment side of the Hammer analysis. As part of the “federalism revolution” of the Court under Chief Justice William Rehnquist, several Justices routinely emphasized the role of the Tenth Amendment in limiting Congress’s power, most notably its power to regulate interstate commerce.14 Over and over again, the Court stressed the “boundaries between the spheres of federal and state authority.”15

The message seemed clear: in order to understand the meaning of federalism, one had to begin from the premise that the domains of state and federal power were fundamentally and forever distinct. Concurrent power was downplayed and sometimes derided as impractical; experiments with overlapping state and federal authority were frequently deemed hazardous to the correct constitutional structure.16 The state and the federal bailiwicks were regarded as separate spheres,17 and the Tenth Amendment functioned as a shield around the “States’ freedom to structure integral operations in areas of traditional governmental functions,”18 functions that were “essential to [the] separate and independent existence” of the states.19 The terms of the federalism debate appeared to be set: a majority of the Justices would analyze Congress’s Article I powers, especially the commerce power, through the lens of the Tenth Amendment, with a baseline commitment to protecting the special domain of state regulatory authority.20

But the terms of the federalism debate have recently changed, with important and potentially far-reaching consequences that have not been fully appreciated—even by the Court itself. The interpretive struggle over the meaning of American federalism has shifted from the Commerce Clause to two textually marginal but substantively important battlegrounds: the Necessary and Proper Clause21 and, to a lesser extent, the General Welfare Clause.22 To be sure, the higher-profile commerce power continues to attract an enormous amount of judicial attention and scholarly commentary.23 But for nearly a decade, the quieter, more structurally ambiguous federal powers listed at the head and foot of Article I have steadily increased in prominence. Today, the battles of judicial federalism are fought not across the well-trampled no-man’s-land of the commerce power or the Tenth Amendment, but in the less trafficked doctrinal redoubts of what I term the “shadow powers.” This expansion of the battlefield carries important consequences for the meaning of modern federalism.

Beginning with Gonzales v. Raich24 in 2005 and continuing through United States v. Comstock,25 National Federation of Independent Business v. Sebelius,26 and United States v. Kebodeaux,27 the Supreme Court’s “federalism revolution”28 has taken on a new form. The Court’s federalism jurisprudence has shifted from its once-typical form of inquiry into the scope of Congress’s power to regulate interstate commerce, refracted through the Tenth Amendment,29 to become an inquiry into the transsubstantive reasons for allowing Congress to regulate at all. This transformation has been especially significant when the Court views Congress as venturing into a domain not explicitly specified in the text of Article I. Analytically, the Justices in the majority in these cases seem to be motivated more by a concern about the expansion of federal regulatory power itself, and somewhat less by a “new federalist”30-style belief in a categorical distinction between the proper spheres of state and federal power.

The return to prominence of this pair of ill-defined but foundational provisions of Article I means many things: doctrinal instability, opportunities for creative litigation, opaque or oracular or overly tentative pronouncements by the Justices. But it also provides a moment to think structurally about the Constitution, and perhaps to reach some conclusions about what federalism does and does not require. The resurgence of the Necessary and Proper and General Welfare Clauses in the doctrine is not a sign of intellectual impoverishment or a mere result of crafty litigation strategies; neither is it a retreat to the weedy curtilage of the federalism field. Instead, the return of the shadow powers heralds an opportunity to take up a central question of federalism: Is it possible to conceive of the states as having significance while also recognizing the logic of Holmes’s point in Hammer? Holmes’s dissent insisted that federalism concerns were irrelevant to determining whether a particular act was within Congress’s power.31 Thus, he maintained, a court need not conduct a separate Tenth Amendment analysis in order to satisfy federalism’s demands.32 According to this view, if a particular act of Congress is within the domain of Congress’s power, then one need not keep probing to ask about an amorphous conception of state sovereignty.

Recent doctrine has thus partially revived Justice Holmes’s conviction that the best way to approach the federalism question is by inquiring into the scope of Congress’s powers. This inquiry increasingly focuses on the shadow powers of Article I: the taxing and spending powers of the General Welfare Clause, and the necessary and proper power. The revival is only partial, however, because a majority of the Justices appears to believe, unlike Justice Holmes, that the real peril is the growth and judicial legitimation of plenary federal power, or a “federal police power.”33 Concern about the growth of federal power (via the expansion of congressional power) is articulated in terms of concern about the expansion of Congress’s power under Article I, rather than in previous decades’ “new federalist” terms, which focused on the erosion of the states’ power.34 As a matter of interpretation, Article I has returned to center stage, and the Tenth Amendment is a secondary player. But unlike Justice Holmes’s Hammer dissent, which offered a broad reading of Article I in support of the federal regulation at issue, the current Court’s shadow powers analysis tends, paradoxically, to constrain federal power.

My descriptive claim is that the clauses operate as shadow powers of Article I, and that their return to the center of debate in the Court and in the broader public sphere can provide both a problem and an opportunity for understanding how the United States’ federal structure should operate. Both powers have a potentially capacious quality, unlike the other Article I powers, which are much more bounded and subject-specific (e.g., “[t]o borrow [m]oney on the credit of the United States”;35 “[t]o coin [m]oney”;36 “[t]o constitute [t]ribunals inferior to the [S]upreme Court”37). The shadow powers tend to become contested, and to become the linchpins of judicially enforced federalism, when contemporary legal and political players determine—for a variety of reasons, from overly rigid case law to political expediency—that there is no more room to move the doctrine in the domain of “real” enumerated powers, such as the commerce power.

In addition to describing the shadow powers and explaining their doctrinal evolution, I also make a normative argument. A description of shadow powers analysis might initially lead one to believe that the Court is using the shadow powers to expand, quietly, Congress’s power beyond the ostensible limits set forth in other doctrinal areas.38 But such a conclusion reads the direction of the doctrinal change exactly backward. Paradoxically, the growth of shadow powers analysis has tended to narrow the permissible scope of congressional regulatory power.

But my critique of shadow powers analysis as deployed by the Court is not based on its direction alone. The prominent role of the shadow powers in the Court’s recent decisions is both a doctrinally unprecedented and an unhelpful development that fails to set meaningful standards for how federalism should work in practice. As I will demonstrate, the novelty of shadow powers analysis lies in the sharp line the Court appears increasingly willing to draw between solid, if controversial, Article I powers such as the commerce power, and auxiliary Article I powers such as the necessary and proper power.39 In recent doctrine, the invocation of the shadow powers has helped the Court find room to maneuver within its federalism analysis, while also appearing to maintain its commitment to an apparently unmoving post-Lopez baseline of a narrow commerce power. This maneuvering might be productive if it were carried out explicitly, with some discussion by the Justices of the reasons for preferring to adjudicate federalism at its doctrinal and textual periphery rather than at its center. But the result of the growth of shadow powers analysis has in fact been to obscure the outlines of federalism’s map—to shroud genuine (and perhaps salutary) doctrinal changes within a fog of constitutional text, under-overruled precedents, and acontextual readings of foundational cases such as McCulloch v. Maryland.40

This essay proceeds in four Parts. In Part I, I explain the meaning of the phrase “shadow powers” and explore the connections among the shadow powers and the other frequently invoked categories of federal power—implied powers, incidental powers, and enumerated powers. Parts II and III discuss the two principal shadow powers: the necessary and proper power and the taxing and spending powers contained within the general welfare power. Part IV examines the consequences of the emergence of the shadow powers for the development of federalism doctrine. It also argues that the concept of “union” could serve as a productive means of reframing the debates over federalism and might provide a “rule of engagement” for courts and other interpreters dealing with questions of federalism.

I. the contours of the shadow powers

To ask what it means to have the power to “make all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers”41 and to “lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the common Defence and general Welfare of the United States”42 is to ask existential questions about the scope of federal power—and, therefore, about the meaning of the Union. One reason for viewing the necessary and proper power and the general welfare power as “shadow powers” is that they have a similar status among the Article I powers, both structurally and doctrinally. Both potentially give Congress a broad regulatory ambit beyond the typical powers over commerce, money, the army, post offices and post roads, and the like.

Since the Founding, and even during the Philadelphia Convention and the state ratifying conventions, there have been debates about (1) what each clause’s language means (what exactly does it mean for a measure to be “necessary” and “proper,” and what is the “general welfare of the United States”?), and (2) how the powers set forth in each clause relate to the other enumerated powers. The second question is fundamentally a question about how the Necessary and Proper and General Welfare Clauses fit into the broader scheme of congressional power, and therefore into the structure of federalism. The clauses’ impact can be viewed relatively narrowly, as concerning only the power of Congress, or more broadly, as implicating the entire architecture of federalism. For some Justices and commentators, this latter view necessarily implicates a mirror-image shadow power on behalf of the states that is lodged in the Tenth Amendment.43

To begin, the clauses are historically linked. Although the records of the Constitutional Convention shed little light on what, if any, debates surrounded the drafting of the Necessary and Proper Clause, the sources are more forthcoming in their discussion of the origins of the General Welfare Clause. Equally important are the pre-1787 precedents for the powers. An antecedent use of the phrase “necessary and proper” can be found in a 1765 pamphlet written by the Maryland lawyer and colonial official Daniel Dulany.44 As part of a broad critique of the Stamp Act, Dulany argued that Parliament had misunderstood the constitutional basis of the British Empire. The North American colonies were concededly “Dependent upon Great-Britain,” Dulany observed.45 But only the colonies’ own local legislatures possessed the authority to levy “internal Taxes.”46 As a solution to the mounting confrontation between the colonies and their metropolitan cousins, Dulany proposed allocating the power to lay internal taxes to the colonial legislatures, and the power to tax for broader, imperial purposes to Parliament. “May not then the Line be distinctly and justly drawn between such Acts as are necessary, or proper, for preserving or securing the Dependence of the Colonies, and such as are not necessary or proper for that very important Purpose?”47 The idea that legislation might have to satisfy a test of necessity and propriety in order to be valid was thus present in Anglo-American constitutional argument more than two decades before the Constitution was drafted. These precedents are useful not to establish a causal intellectual lineage between colonial commentators and the drafters of the Constitution, but rather to suggest that the phrases were already becoming recognized legal terms of art well before the Revolution.

Two years later, in response to the Townshend Acts of 1767, the Philadelphia lawyer John Dickinson’s Letters from a Farmer invoked the “general welfare” of the empire as a permissible justification for Parliament to tax the colonies.48 Prior to the Stamp Act of 1765, Dickinson argued, “every statute relating to these colonies from their first settlement to this time” was “calculated to preserve or promote a mutually beneficial intercourse between the several constituent parts of the empire.” To be sure, many of those acts had levied trade duties on the colonies. “[Y]et those duties were always imposed with design to restrain the commerce of one part that was injurious to another, and thus to promote the general welfare.”49 Taxation that was explicitly intended to harmonize trade within the empire, and thus to balance intra-imperial interests against one another, was permissible. The problem with the Stamp Act, however, was its overt and unprecedented goal of “raising a revenue.”50

As the examples of Dulany and Dickinson demonstrate, some conception of both powers predated the drafting of the Constitution.51 The idea that one might assess legislative power as “necessary and proper” or in service of the “general welfare” was available to British North Americans at least two decades before the members of the Constitutional Convention included the phrases in Article I. Both notions were part of the intellectual portmanteau that contemporary observers used to understand how political unions based on a significant degree of centralized legislative authority ought function.

The meaning of both clauses has been the subject of significant debate from the Founding to the present day, as commentators have tried to determine whether they were lesser, auxiliary powers that must be attached to another, underlying power or whether they are standalone powers in their own right (albeit with some limits).52 The broad phrasing of the clauses, as well as their bracketing locations at the beginning and end of Article I, suggests that they are in some sense fundamental congressional powers. The course of their doctrinal development has been as uneven and disputed as this description suggests. Commentators have argued not only about the proper subject of each clause, but also about each clause’s structural and theoretical connection to the rest of Article I, and hence to the Constitution as a whole.53

The clauses are also similar in their relationship to the rest of Article I. Commentators typically use the phrase “enumerated powers of Article I” to refer to a specific set of clauses granting authority to Congress.54 Among the powers housed in Article I, the most commonly invoked are the commerce power and the taxing and spending powers. Other frequently cited powers include the treaty power, the war power, the power to regulate immigration and citizenship, and the set of powers over domestic affairs such as bankruptcy law and the coining of money.55 Clearly, all these are enumerated powers, in that they are listed in the text of Article I, Section 8, which begins “Congress shall have power.” But the nature of the General Welfare Clause and the Necessary and Proper Clause is more ambiguous. True, both clauses describe powers of Congress, rather than limitations on power elsewhere granted.56 And the taxing and spending powers, which are the modern doctrinal nodes of the General Welfare Clause, are typically denominated as enumerated powers.57 But neither clause refers specifically to a power of government over a particular domain, such as post offices or the coining of money. Moreover, each clause describes the general contours of some permissible types of congressional regulation, but not all regulation by Congress under its enumerated powers must conform to the contours of one or the other of these two clauses. Instead, the clauses appear to offer alternative categories of regulation, in addition to the narrower subjects listed in the rest of Section 8. Most striking is the opacity of the clauses’ language: “the common Defence and general Welfare of the United States” and “all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers” are phrases that invite—and perhaps even require—interpretation.

The clauses do differ from each other in important respects, however. Significantly, the Necessary and Proper Clause refers explicitly to the powers listed in the preceding seventeen clauses of Article I, Section 8, and defines the necessary and proper power with reference to those “foregoing powers.” The necessary and proper power is therefore by its own terms connected with the other sources of congressional regulatory authority; its purpose, according to current doctrine, is to enable Congress to carry out those other powers.58 The General Welfare Clause, in contrast, stands on its own among the Article I powers, on an equal footing with the commerce, naturalization, and other powers, rather than dependent on those powers as necessary predicates to its own exercise. If the necessary and proper power is best understood as an auxiliary power to Congress’s primary powers under Article I,59 the General Welfare Clause—especially when broken down into its modern doctrinal components, the taxing and spending powers—is just such a primary power. The ability to tax and spend, even if the permissible goals of that taxing and spending are limited to the common defense and general welfare,60 is an independent Article I power in a way that the necessary and proper power has never been held to be.61

A related distinction follows from what modern doctrine regards as the auxiliary status of the necessary and proper power, as contrasted with the primary nature of the taxing and spending powers under the General Welfare Clause. This distinction turns not on the particular power’s ability to stand on its own as a basis of congressional regulation (the primary/auxiliary question), but rather on how we conceptualize the source of the power—whether it derives its authority from the text itself, or whether the text requires something more, such as common law judicial interpretation or historical practice.62

The enumerated powers of Article I are typically regarded as deriving their authority from the fact of their enumeration: they exist because they are listed in the text of Article I, which created the powers as well as the institution (Congress) that wields them.63 During the controversies of the late eighteenth and early nineteenth centuries surrounding the scope of the taxing and spending powers contained within the General Welfare Clause,64 the disagreement centered on the meaning of the phrase “general welfare,” not on the ultimate source of authority behind the provision in Article I. Certainly, commentators wrangled over the relationship between the concept of the “general welfare” and the more specific powers set forth in the other clauses.65 But they did not appear to question the theoretical justification for the Constitution’s grant to Congress of some general welfare power (whatever the scope of that power turned out to be). The fact of the enumeration was sufficient to grant Congress some set of powers to tax and spend, with some connection to a notion of the “general welfare of the United States.” The battle concerned the scope of the power, not its source.

This positivist vision of enumeration becomes more complicated, however, when applied to the language of the Necessary and Proper Clause. Given that the Necessary and Proper Clause incorporates the preceding seventeen clauses by reference (“carrying into execution the foregoing powers”), one might ask whether the necessary and proper power is an enumerated power. It might be something else altogether, categorically distinct: an “implied” or an “incidental” power,66 or a “non-enumerated” power.67 To be sure, the overarching power—to “make all laws which shall be necessary and proper for carrying into execution the foregoing powers”—is plainly enumerated in the text of Article I. But the subjects to which that power should be applied, and the ends to which the power should be put, are not enumerated.68 Unlike the Commerce Clause, which tells us that Congress possesses the power to regulate commerce among the several states, the Necessary and Proper Clause describes a power that potentially touches all the domains of congressional regulation, but that does not itself describe such a domain.69 The necessary and proper power is therefore enumerated, in the sense that it originates in words contained in Article I, but the level of the enumeration seems qualitatively different from that of other Article I powers, including those contained in the General Welfare Clause.70

II. the necessary and proper clause

A. Theory and History

To see how the shadow powers have returned to the center of the doctrinal conversation, we can look to recent case law surrounding the Necessary and Proper Clause. Before 2005, one would have been hard pressed to identify a body of doctrine on the necessary and proper power. Unlike the commerce power, which in a series of doctrinal boluses has expanded and contracted the permissible scope of federal regulation of the economic realm, the necessary and proper power has tended to ride along as a quieter, sometimes overlooked presence in the case law—the perpetual bridesmaid to the commerce power’s bride. For decades, commentators essentially threw up their hands in explaining the clause, which was seen as “haunt[ing] constitutional theory from its perch in interpretive limbo.”71 In recent years, however, the Court has increasingly brought the necessary and proper power to the center of federalism doctrine. Because the power is relatively undertheorized, it provides new avenues for doctrinal development, sometimes in surprising directions. Perhaps most surprising is the steady transmutation of the necessary and proper power from a regulatory and interpretive device that tended to expand federal power into a tool for checking that same power.

Since 1819, the lodestar for understanding the clause has been Chief Justice Marshall’s opinion in McCulloch v. Maryland upholding the constitutionality of the Second Bank of the United States as a valid exercise of Congress’s power.72 In McCulloch, Marshall sketched a broad—but not unlimited—necessary and proper power, which he described as an “incidental or implied” power.73 In so doing, Marshall planted the seed of the modern notion that the necessary and proper power is somehow not an enumerated power, even though it is plainly listed in Article I.74 Marshall framed the inquiry as whether Congress was acting within one of its enumerated Article I powers in a given case (here, in establishing the Bank); if so, then Congress had broad discretion to select the regulatory means by which it carried out that power. Hence Marshall’s famous statement: “Let the end be legitimate, let it be within the scope of the constitution, and all means which are appropriate, which are plainly adapted to that end, which are not prohibited, but consist with the letter and spirit of the constitution, are constitutional.”75

Marshall’s opinion devoted little discussion to principles of state sovereignty or the Tenth Amendment. Indeed, he rejected the argument that the Tenth Amendment’s presence in the Constitution “excludes incidental or implied powers” or “requires that every thing granted shall be expressly and minutely described.”76 Limits on Congress’s regulatory power should be seen as incorporated into the federal structure via the enumeration principle. When Congress regulated within the realm of one of its enumerated powers, therefore, its action could not be challenged based on non-Article I federalism imperatives such as those that Maryland argued had been textualized in the Tenth Amendment. Maryland might insist that the presence of the Bank and the prohibition on state taxation of the Bank compromised its sovereignty as a state. But Marshall insisted that the enumerated (“great”77) powers “to lay and collect taxes; to borrow money; to regulate commerce; to declare and conduct a war; and to raise and support armies and navies” necessarily entailed an implied or incidental power to establish a corporation (the Bank) to execute those powers.78 The nexus between the underlying enumerated powers and the implied or incidental power to carry them out effectively blocked the states from objecting that their domain had been invaded. The “government of the Union, though limited in its powers,” was “supreme within its sphere of action,” Marshall insisted.79 In other words, if an exercise of federal power fit within the bounds of Article I, it could not be challenged as an overreach or a violation of the states’ authority. The Union had a discrete sphere of action, and within that sphere, the states’ cries of trammeled sovereignty would be unavailing—perhaps even unconstitutional.

Chief Justice Marshall’s choice of terminology in McCulloch was intricate, perhaps even slippery, but it shaped the doctrinal world in which we still live today. Marshall termed what I have called the primary powers of Article I (to tax, to borrow money, to regulate commerce, to declare and conduct war, and to raise and support armies and navies) “great substantive and independent power[s]” because of their status as distinct heads of congressional power, as that power was created in Article I.80 “Greatness” was a category within the Constitution itself, not an assessment of the relative importance of a particular power as applied to governments generally.81 Marshall’s use of the phrase “great substantive and independent power” to describe “the power of making war or levying taxes or of regulating commerce” demonstrates his sense that the enumerated/non-enumerated distinction did not quite capture the difference between the primary powers of Article I (war, commerce, taxation) and the implementing power described in the Necessary and Proper Clause. The great powers were what Article I said they were, and only what Article I said they were; the necessary and proper power was an explicit statement of the basic interpretive principle that, as Marshall put it, “the powers given to the government imply the ordinary means of execution.”82

The idea of a power necessarily implying the power to execute it is connected to another important point about the foundational reading of the Necessary and Proper Clause in McCulloch. A careful reading of the opinion demonstrates that, for Marshall, the necessary and proper power was an enumerated power, notan implied power, of the federal government. It was an enumerated power because it was listed in Article I. But it was not a primary power. Enumerated powers might be primary powers, but they might also be auxiliary or implementing powers.

The case of the necessary and proper power, however, was even more complex: it was an enumerated, auxiliary power that gave Congress some additional quantum of authority besides that which it would already have in order to execute the primary powers underlying the creation of the Bank (tax, borrowing, commerce, war, armies and navies). The late-eighteenth-century mind could easily conceive of primary powers as necessarily containing the power to execute those powers,83 but the Constitution made that common law interpretive principle explicit by committing it to words in Article I. As Akhil Amar notes, “Marshall suggested that perhaps the clause was merely declaratory of what would have been the best reading of the Constitution even had the clause not existed.”84 This interpretation is borne out by James Wilson’s comments in the Pennsylvania ratifying convention. Responding to a delegate who argued that “the general clause at the end of the eighth section gives to Congress a power of legislating generally,” Wilson insisted that the clause was not “capable of giving them general legislative power.” The clause meant instead that “they shall have the power of carrying into effect the laws which they shall make under the powers vested in them by this Constitution.”85

For most of the early twentieth century, the necessary and proper power made occasional and relatively uncontroversial appearances, typically operating in tandem with the commerce power. An important line of cases upheld congressional regulation of railroad rates, milk prices, and labor standards based on a blend of commerce power and necessary and proper power analyses.86 The decisions did not separate the two doctrinal strands. Frequently, they mixed discussions of the regulation of interstate commerce itself with references to intrastate regulation that was an “appropriate means to the attainment of a legitimate end, the effective execution of the granted power to regulate interstate commerce.”87 This conceptual melding of the two species of power continued into the New Deal era, leading Justices and scholars to posit that the New Deal expansion of the commerce power was powered by an expansion of the necessary and proper power.88

Today, the auxiliary status of the necessary and proper power is a settled doctrinal fact.89 Even after the Court began limiting Congress’s commerce power in a series of cases beginning with United States v. Lopez90in 1995, the necessary and proper power appeared to be something of a doctrinal backwater. Indeed, the auxiliary nature of the power ensured that it was always overshadowed by the primary Article I power to which it was attached in any given statute or case. Parties challenging federal regulation therefore tended to focus their arguments on the primary power, reinvigorated by the Court’s holding in Lopez that only activity that was itself “economic” or “commercial” in nature could validly support congressional regulation under the Commerce Clause. Arguments based on the necessary and proper power tended to be viewed as at best routine, at worst an attempt to conjure Article I power where none actually existed. “Since the time of McCulloch v. Maryland, it has been clear that the [Necessary and Proper] Clause presents no formidable barrier to legislative activity,” one commentator noted in 1983. “The terms of the clause have always been read broadly.”91 Justice Scalia, for his part, has deemed arguments based on the clause “the last, best hope of those who defend ultra vires congressional action.”92 It is a hornbook axiom that Congress cannot regulate under the necessary and proper power alone; it must base an exercise of that power on an underlying enumerated power.93 The question whether the necessary and proper power “counts” as one of Congress’s enumerated powers continues to plague the doctrine, however.

Indeed, this lingering uncertainty about the status and scope of the necessary and proper power has become even more acute as a consequence of recent Supreme Court decisions. In a handful of cases since 2005, the Court has highlighted a new set of doctrinal consequences of Marshall’s cleverness in McCulloch. Few of these consequences, however, comport with the expansive vision of federal power as “supreme within its sphere” that Marshall limned. In beginning his justification for Congress’s chartering of the Bank with the implied powers argument—that all Article I powers include the means of exercising them—Marshall grasped for the biggest limb on the interpretive tree. Before he even reached the specific grant of power in the Necessary and Proper Clause, Marshall had essentially disposed of the issue on the basis of Congress’s implied powers. Recall his statement that “the constitution of the United States has not left the right of Congress to employ the necessary means, for the execution of the powers conferred on the Government to general reasoning. To its enumeration of powers is added” that of the Necessary and Proper Clause.94 Because Marshall appears to have based some significant part of his decision on the implied-powers principle, the scope of the enumerated necessary and proper power has remained vague and largely undefined in the doctrine, even as judges and commentators have continued to lard the clause into their opinions and articles alongside discussions of the primary enumerated powers.

Why does this latent uncertainty about the enumerated or non-enumerated nature of the Necessary and Proper Clause matter? The answer lies in the modern black-letter law that the necessary and proper power can be validly used only as an auxiliary to one of the primary enumerated powers.95 In periods of judicial deference to congressional legislation under those primary powers (for example, the period between the New Deal expansion of the commerce power in 1937 and its constriction in 1995 in United States v. Lopez96), the fuzzy status of the necessary and proper power is difficult to discern and appears to have little significance. In periods of intensified judicial scrutiny of such legislation, however, the situation is quite different.

When the Court narrows the commerce power (as it did in Lopez), the relationship between that power and the necessary and proper power matters a great deal. If our view of the necessary and proper power derives from the strong implied-powers version of McCulloch, according to which the Bank was constitutional because it was an implied means by which Congress could exercise its primary Article I powers (taxation, borrowing, commerce, war, armies/navies), then the scope of the necessary and proper power at any given moment follows directly from the scope of the primary power to which it is attached. If the Court narrows the commerce power, what happens to the argument that Congress must have an implied means of exercising that power? Almost certainly, the implied-power extension of that primary power would also be narrowed. If, on the other hand, the necessary and proper power is viewed as an enumerated power, and—importantly—even if we accept that it is auxiliary rather than primary, its basis in the text of the Constitution would likely afford it greater judicial deference than if its scope was seen as entirely dependent on the scope of the primary power.

Indeed, recent doctrinal developments bear out the hypothesis. The way the necessary and proper power has come down to us from Marshall—and specifically its ambiguous enumerated/non-enumerated status—has permitted the Court to confine it to a few specific regulatory domains, which rely on a few interconnected primary powers. Paradoxically, some of these putative primary underlying powers are themselves non-enumerated, the most notable being Congress’s power to establish a body of federal criminal law.97 Within those areas, the clause has been read fairly broadly, but the fact that it is cabined in those areas necessarily limits its overall reach. The uncertain scope and status of the necessary and proper power have permitted the Court to use the power as a means of constraining federal regulatory power, rather than—as the Antifederalists feared, and as Marshall hoped—as a textual and structural foundation for the principle that the government of the Union is supreme within its defined sphere. To see how this result came about, let us examine the case law.

B. Recent Doctrine

The Necessary and Proper Clause returned to center stage in 2005, in the somewhat surprising vehicle of Justice Scalia’s concurrence in the judgment in Gonzales v. Raich.98 In permitting the federal Controlled Substances Act to override California’s medical marijuana statutes, Justice Scalia recast much of the twentieth-century commerce power case law as turning in fact on the necessary and proper power.99 The crucial point for Scalia was that in the all-important regulatory arena of activities that substantially affect interstate commerce, the real source of Congress’s power was in fact the Necessary and Proper Clause, not the Commerce Clause.100 Whenever Congress was a step removed from an enumerated power, Justice Scalia argued, in order for its regulatory action to be valid, that action had to be based on the necessary and proper power.101 To the extent that prior cases had treated the “substantial effect” inquiry as part of Commerce Clause doctrine, they were “incomplete because the authority to enact laws necessary and proper for the regulation of interstate commerce is not limited to laws governing intrastate activities that substantially affect interstate commerce,” Justice Scalia observed.102

Indeed, on Justice Scalia’s reading, the necessary and proper power greatly expanded the reach of Congress’s regulatory power: “Where necessary to make a regulation of interstate commerce effective, Congress may regulate even those intrastate activities that do not themselves substantially affect interstate commerce.”103 But the Article I power that enabled that regulation came from the Necessary and Proper Clause, not the Commerce Clause. This distinction separated Justice Scalia’s reasoning from that of the other five Justices in the majority, all of whom took the view that Congress’s commerce power included “the power to prohibit the local cultivation and use of marijuana in compliance with California law.”104 The five Justices who joined the majority opinion105 implicitly endorsed McCulloch’s implied-powers argument, insofar as they based their view principally on the commerce power, with the necessary and proper power mentioned almost as an afterthought. Writing for the Court, Justice Stevens blended the commerce and necessary and proper discussions to such a degree that his opinion reads as though they were a single unit of analysis.106 Justice Scalia’s concurrence, in contrast, treated the necessary and proper power as a discrete power that, while auxiliary, amounted to a distinct source of congressional authority.

Justice Scalia’s Raich concurrence appeared as though it might open doctrinal space for a broader reading of the necessary and proper power, especially as the Court held to its narrow reading of the commerce power in cases following Lopez.107 As the doctrine subsequently unfolded, however, it became clear that a necessary and proper power that was closely tied to the commerce power—and perhaps too casually theorized—was susceptible to the same constrictions as the commerce power. In cases following Raich, several Justices picked up Justice Scalia’s necessary and proper analysis and applied it in other contexts to uphold congressional regulation, with the result that a distinct body of Necessary and Proper Clause doctrine and commentary appeared to be developing for the first time since McCulloch.108 But the status of the necessary and proper power remained unsettled. Moreover, it was difficult to rationalize the state of the doctrine: was it really sensible to have a broad necessary and proper power at the same time as a narrow commerce power? The difference might have been ascribed to shifting majorities on the Court, but for the crucial role Justice Scalia’s Raich concurrence had played in launching the new developments.

The newly prominent Necessary and Proper Clause again proved dispositive to the Court’s analysis five years later, in United States v. Comstock.109 In Comstock, the Court relied on the necessary and proper power in upholding a federal civil commitment statute for persons convicted of federal sex offenses.110 Writing for the Court, Justice Breyer held that the statute was a valid extension of Congress’s authority to establish federal criminal law; the fact that the federal criminal law was itself based on an implied power of Article I did not trouble the Court.111 The Court rejected the argument that “when legislating pursuant to the Necessary and Proper Clause, Congress’s authority can be no more than one step removed from a specifically enumerated power,”112 calling such an argument “irreconcilable with our precedents.”113 Justice Breyer’s opinion pointed out that even the dissenting justices acknowledged

that Congress has the implied power to criminalize any conduct that might interfere with the exercise of an enumerated power, and also the additional power to imprison people who violate those (inferentially authorized) laws, and the additional power to provide for the safe and reasonable management of those prisons, and the additional power to regulate the prisoners’ behavior even after their release.114

Justice Kennedy’s concurrence took a similar approach, stating that analysis under the Necessary and Proper Clause “depends not on the number of links in the congressional-power chain but on the strength of the chain.”115 Here, he found that the chain connecting Congress’s authority to establish the federal criminal law was sufficiently robust also to permit it to mandate federal civil commitment for some convicted sex offenders.

The result in Comstock led some observers to wonder whether the Necessary and Proper Clause of the aughts would reprise the high-profile commerce power of the period between 1937 and 1995.116 The Court appeared to anticipate this concern and took steps in its opinion to alleviate it, stating, “[n]or need we fear that our holding today confers on Congress a general police power, which the Founders denied the National Government and reposed in the States.”117

But Justice Scalia appeared not to share this comfort with the Comstock holding. Despite his crucial role in reviving the doctrinal force of the necessary and proper power in his Raich concurrence, Justice Scalia joined Justice Thomas in dissenting from the Court’s application of the power to the civil commitment statute. The dissent cited the Raich concurrence for the principle that although the Necessary and Proper Clause “‘empowers Congress to enact laws . . . that are not within its authority to enact in isolation,’ those laws must be in ‘effectuation of [Congress’s] enumerated powers.’”118 Having launched the second career of the Necessary and Proper Clause in the context of federal drug regulation (a decision from which Justice Thomas had also dissented), Justice Scalia took a different view of the power in relation to federal civil commitment of sexually dangerous persons. The dissenters’ principal objection to the statute at issue in Comstock was its attenuated relationship to any particular enumerated power.119 Absent such a connection, the dissenters appeared to view the case as though Congress had based the statute on a bare assertion of the necessary and proper power, and nothing else—that is, as though a decision to uphold the regulation would transform the power from an auxiliary into a primary tool. Indeed, the dissent expressed skepticism that a claim of congressional power grounded in the Necessary and Proper Clause might ever pass constitutional muster.120

The specter of the commerce power haunts the analysis in Comstock, with unfortunate and muddying consequences for the doctrine. At oral argument, the Justices and the advocates circled around the commerce power, sometimes distinguishing it from the necessary and proper power and sometimes using it as an analogue. Justice Kennedy pressed then-Solicitor General Elena Kagan to make the commerce power argument for the civil commitment statute; in response, Kagan noted that “the government has never argued the Commerce Clause here in the sense that it has never argued that these activities have a substantial effect on interstate commerce, and it hasn’t done so because of . . . the Morrison precedent.”121 The government appeared to have made a policy decision in Comstock not to challenge the applicability of Morrison’s narrow reading of the commerce power to the civil commitment statute, but the Justices and advocates alike seem to have recognized the awkward analytical results for the later case.122 One example was the government’s somewhat strained connection between the individual being subjected to civil commitment and the federal criminal justice system as the crucial tie between the regulation and a primary Article I power.123 But using an existing physical link between the individual being regulated and a location of federal power (here, a federal prison) as evidence that the individual should continue to be the subject of federal power is a conceptually different type of connection than the classic McCulloch-style inquiry into the nexus between the regulation (e.g., Congress’s creation of the Bank) and an enumerated power (e.g., the powers to borrow, tax, regulate commerce, make war, and raise armies and navies).

The question of the proper level of generality at which the link between the regulated activity and the relevant enumerated power ought to be evaluated has continued to bedevil the necessary and proper doctrine. The power made a brief but important appearance in the Court’s 2012 decision in National Federation of Independent Business v. Sebelius124 (NFIB), in the section of the healthcare case concerning the individual mandate provision of the Patient Protection and Affordable Care Act (commonly referred to as the ACA). The analysis in NFIB demonstrated the powerful consequences of Justice Scalia’s revival of the necessary and proper power in his Raich concurrence. Chief Justice Roberts’s opinion for the Court held that the individual mandate was not a “proper” means of carrying out Congress’s enumerated powers because it would “draw within [Congress’s] regulatory scope those who would otherwise be outside of it.”125 The result in NFIB surprised many commentators, who had assumed that the Court’s broad reading of the necessary and proper power in Comstock would lead it to uphold the individual mandate.126

The Chief Justice then devoted the remainder of his three-page discussion of the necessary and proper power to distinguishing NFIB from Raich and Comstock. Despite these efforts at distinction, however, the Court imported two of the most questionable elements from those cases. First, the Court’s main objection to the individual mandate again blurred the line between the necessary and proper power and the commerce power, as the Raich majority had done. Second, the Court adopted the Comstock Court’s approach to assessing the connection between the activity to be regulated and the enumerated power to which that activity was tied. Both these analytical strands came together in a single paragraph of the decision:

This is in no way an authority that is “narrow in scope,” Comstock, or “incidental” to the exercise of the commerce power, McCulloch. Rather, such a conception of the Necessary and Proper Clause would work a substantial expansion of federal authority. No longer would Congress be limited to regulating under the Commerce Clause those who by some preexisting activity bring themselves within the sphere of federal regulation.127

If Comstock gave us the Court adjudicating in the shadow of the Commerce Clause, NFIB brought the narrow post-Lopez/Morrison commerce power back into the doctrinal fray and made it an integral part of the necessary and proper power analysis.128 Consider Chief Justice Roberts’s statements in the quoted paragraph: upholding the individual mandate under the necessary and proper power would automatically expand the scope of the commerce power as well. There is some irony here with respect to the legacy of Justice Scalia’s expansive necessary and proper power in Raich. In Raich, Justice Scalia (but not the majority) had separated the necessary and proper power from the commerce power, insisting that any regulation under the “effect on interstate commerce” prong of McCulloch and Lopez was in fact an exercise of the necessary and proper power, not the commerce power.129 That analysis made it theoretically possible to sustain a regulation under the necessary and proper power as an auxiliary to the commerce power, even though that regulation could not be sustained under the commerce power itself. Although Justice Scalia had not gone so far as to categorize the necessary and proper power as itself an enumerated power,130 he had presented it as an analytically distinct element of Congress’s authority. The Court had extended this interpretation in Comstock. In NFIB, however, Chief Justice Roberts recombined the two strands.131 Once the Court had held the ACA invalid under the commerce power, the necessary and proper portion of the opinion followed. The contours of the necessary and proper power largely tracked those of the commerce power.

Similarly, consider the Chief Justice’s statement that Congress must be limited to regulating individuals “who by some preexisting activity bring themselves within the sphere of federal regulation.”132 In Comstock, the preexisting connection between the regulated activity and the federal sphere was the individual’s presence in federal custody. In NFIB, the individual mandate failed this test because it sought to “vest[] Congress with the extraordinary ability to create the necessary predicate to the exercise of an enumerated power.”133 In other words, Congress got the timing wrong. The implication of the Court’s argument was clear: Congress would not be permitted to bootstrap a substantial effect on interstate commerce out of the very regulation at issue in the case. But recall the skepticism with which the Court greeted the government’s claim in Comstock that the entire civil commitment scheme rested on the fact that the individuals in question were in federal custody at the time the ongoing commitment was ordered. The traditional necessary and proper framework dating from McCulloch had not looked to the nexus between particular individuals and federal power; it was not analogous to a question of personal jurisdiction. (If that had been the case, it is difficult to see how the newly created Second Bank of the United States could have been upheld—because it continued to hold assets that had once been held by the First Bank? Surely that fact alone would not have sufficed.) On the contrary, the inquiry turned on the link between the activity to be regulated and the particular primary Article I power cited as the source of the regulation.

In the brief period since the NFIB decision, the Court has continued to endorse an approach to the necessary and proper power that focuses on the regulated individual’s relationship to federal power. Timing, and a special relationship to federal power, again proved dispositive in United States v. Kebodeaux,134 in which the Court upheld the application of the Sex Offender Registration and Notification Act (SORNA) to an individual who had been convicted by a special court-martial of a sex offense. As in Comstock, the Court upheld Congress’s exercise of the necessary and proper power in the realm of federal criminal law, with the additional weight in Kebodeaux of the Military Regulation Clause.135 Again, the tie to a permissible Article I purpose came not from the connection between the activity Congress sought to regulate and a specific, primary congressional power, but from a particular individual’s prior presence within the federal system.

Yet Justice Breyer’s opinion for the Court in Kebodeaux suggested a broader vision of the Necessary and Proper Clause, hearkening to Holmes’s dissent in Hammer and the view that the scope of Article I should be determined on the basis of Article I alone, and not influenced by concerns about preserving a sphere of state sovereignty. To support this point, Justice Breyer cited a 1924 decision upholding a federal Prohibition statute for the proposition that the Necessary and Proper Clause permits Congress to “adopt any means, appearing to it most eligible and appropriate, which are adapted to the end to be accomplished and consistent with the letter and spirit of the Constitution.”136

Despite the differing holdings in NFIB and Kebodeaux, these two most recent analyses by the Court of the Necessary and Proper Clause demonstrate the clause’s increased importance for modern debates about the meaning of federalism. As part of the NFIB analysis, the Court explicitly linked the necessary and proper power and the commerce power. In so doing, the Court demonstrated the way in which Chief Justice Marshall’s implied-powers reasoning in McCulloch, which in that case had supported Congress’s creation of the Bank, could be redirected toward a constriction of federal power. If the scope of the auxiliary necessary and proper power was analytically connected to the primary power to regulate commerce, then the Court’s decision to constrain the commerce power would also narrow the necessary and proper power.

The outcome of these cases suggests a paradoxical conclusion: Chief Justice Marshall’s broad reading of Article I, according to which all the primary powers include the means of exercising them, made it possible for a later Court to restrict the reach of the necessary and proper power as a derivative of the commerce power, and in so doing to shrink the permissible domain of congressional regulation. Perhaps even more surprising is that the alternative to Chief Justice Marshall’s implied-powers argument came from Justice Scalia, whose concurrence in Raich offered a way to view the necessary and proper power as auxiliary but also as a meaningful source of federal regulatory authority beyond the scope of the primary Article I powers.137 Justice Scalia’s Raich concurrence set the stage for the inferential reasoning that supported the Court’s decision to uphold the civil commitment statute in Comstock and the notification requirement in Kebodeaux. In the latter cases, however, the concurrence’s author attempted to contain the consequences of his analysis by arguing that the exercise of the auxiliary power was too many steps removed from any identifiable primary power.138

As these cases demonstrate, working through the Necessary and Proper Clause has again become a required step in judicially enforced federalism. Contrary to its early days, however, the clause has increasingly been used to avoid the crucial issue of the proper scope of Congress’s Article I power; in some cases, it has functioned as a tool for the Court to check federal power, either by tying the clause directly to a narrowed commerce power, or else by confining the clause’s operation to situations involving a formalistic, quasi-jurisdictional nexus between specific individuals and federal instrumentalities. In short, far from the fount of federal police power that some observers have feared the Necessary and Proper Clause might become,139 the clause has become a subtle doctrinal hurdle to congressional regulation.

Indeed, the same Justices who are committed to hewing to limits on the commerce power have also tended to offer the most creative readings of the shadow powers. Consider, for example, the arc traveled by the necessary and proper power from Lopez to Raich to Comstock. The majority opinion in Lopez, which Justice Scalia joined, contained no reference to the Necessary and Proper Clause; the Court did not discuss whether the Gun-Free School Zones Act of 1990 might be permissible under that power.140 Instead, the majority confined its analysis to the commerce power and held that the statute was invalid because it was not a regulation of economic activity.141 In Raich, however, the Necessary and Proper Clause was the pivotal ingredient in both the majority opinion and Justice Scalia’s concurrence.142 After Raich, the necessary and proper power appeared to be an avenue of potentially broad congressional power that survived the Court’s avowedly stringent commerce-power reasoning in Lopez.143 The Court’s decision in Comstock upholding the civil commitment provision continued this expansive reading of the necessary and proper power, building on the sharp bifurcation between the necessary and proper power and the commerce power that Justice Scalia set forth in his Raich concurrence.144 The result in Comstock demonstrated the power of Justice Scalia’s distinction between the powers. Paradoxically, moreover, Justice Scalia’s dissent from the Comstock majority145—and his apparent skepticism at oral argument of the government’s argument that the civil commitment provision was a permissible exercise of the necessary and proper power146— served to highlight the Necessary and Proper Clause’s doctrinal transformation from a murky commerce power addendum to a distinct, visible, and therefore controversial source of congressional power.

A broad reading of a shadow power similarly accompanied, and arguably enabled, a narrow reading of a primary power in NFIB. Consider the progression of Chief Justice Roberts’s analysis of the individual mandate provision, from the commerce power to the necessary and proper power to the taxing power. Having determined that the individual mandate was not a permissible exercise of the commerce power because it failed the Lopez test of regulating economic activity,147 Chief Justice Roberts next moved to the necessary and proper power as a separate step of analysis. Here, again, the individual mandate failed to pass muster. Chief Justice Roberts seized on a phrase from McCulloch v. Maryland148and decontextualized it to create a new limitation on the necessary and proper power. The individual mandate was not analogous to the regulation of the intrastate marijuana market in Raich because “Raich . . . did not involve the exercise of any ‘great substantive and independent power,’ of the sort at issue here.”149 Why the regulation of the intrastate market in healthcare (or, more precisely, self-insurance) amounted to a “great substantive and independent power” was left unclear. Also unacknowledged was the sudden appearance of a new distinction in the case law between permissible uses of the necessary and proper power (e.g., to plug holes in existing regulatory schemes150) and uses that appeared, in some vaguely defined way, to be simply too large in scale to be permitted by such a potentially expansive provision as the Necessary and Proper Clause.151

The problem with the trajectory of shadow powers analysis is that it combines a claim of doctrinal stringency in one area (the post-Lopez commerce power) with an unacknowledged doctrinal slipperiness in another area (the necessary and proper power). Moreover, the Court’s increasingly sharp separation between the two powers suggests that regulations under the Necessary and Proper Clause are less constitutionally firm than those authorized by the Commerce Clause. But, as the long line of decisions dating from the early twentieth century demonstrates,152 the commerce power and the necessary and proper power have a long history of conceptual and doctrinal overlap. Because the necessary and proper power has traditionally been understood as an auxiliary of the commerce power, recent decisions that treat the former as a distinct basis of congressional regulation inevitably search in vain for firm precedents that contain reasoning specific to the Necessary and Proper Clause. The result is to undermine the necessary and proper power while appearing to treat it as equal in weight to the commerce power.

Both the General Welfare Clause and the Necessary and Proper Clause thus might be thought of as doctrinal workarounds,153 tools for the Court to find pressure points in the case law in order to yield new insights about the permissible scope of congressional power.154 They are shadowy, however, because of their textual and structural ambiguity, because it is not clear that they can bear the interpretive weight being placed upon them, and because their new prominence is based on unarticulated premises yet presented as dictated by precedent.

III. the general welfare clause

Like the necessary and proper power, the powers housed in the General Welfare Clause of Article I155 have assumed a newly prominent role in the Supreme Court’s recent decisions concerning the scope of congressional power. As is also the case for the Necessary and Proper Clause, the General Welfare Clause contains ambiguous language regarding the means and ends of permissible congressional regulation. The clause is the opening provision of Article I, Section 8. It states, “The Congress shall have Power To lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the common Defence and general Welfare of the United States.”156 Are the taxing and spending powers the means to achieving the ends of common defense and general welfare for the nation? Or is providing for the common defense and general welfare another means, an additional source of authority for Congress to regulate toward some other (unstated) end? Unlike the necessary and proper power, which has throughout its history been understood as auxiliary, the taxing and spending powers have traditionally been listed among the primary, enumerated powers of Congress.157 But the two clauses are similar in that both have resurfaced in recent doctrine as the Court tries different approaches to the old problem of assessing congressional regulation’s congruence with the Constitution’s commitment to federalism.158

The taxing and spending powers of the General Welfare Clause may fairly be termed shadow powers because although they appear uncontroversial, secondary, and innocuous compared to their polarizing Article I companion, the commerce power, they nevertheless contain the potential for both doctrinal expansion and contraction. Like the Necessary and Proper Clause, the General Welfare Clause appears to have a non-substantive, almost neutral orientation; neither is “about” a specific subject or establishes a particular domain for Congress. The power to tax and spend in some form is a basic tool of government generally, and the necessary and proper power is an instruction (perhaps a tautology,159 perhaps not) to the federal government that it can execute the powers confided to it.

The path of the General Welfare Clause began in early controversy, traversed a lengthy period of settlement, and has recently reached a new ground of contestation and uncertainty. Following several decades in the founding and early national periods in which the meaning of the clause—specifically, the connection between the taxing and spending powers and the “general welfare of the United States”—was hotly debated,160 the clause settled down in the early twentieth century to become arguably the second most important source of enumerated congressional authority after the commerce power.161

From the founding era through the early nineteenth century, commentators such as James Madison, Alexander Hamilton, and Thomas Jefferson argued over what the General Welfare Clause meant. Three main views emerged, holding that the clause ought to be understood as (1) a freestanding power to regulate for the common defense and general welfare (a view that was not widely held but that was nevertheless frequently invoked in debate);162 (2) a statement that Congress might tax and spend for the common defense and general welfare (Alexander Hamilton’s view);163 or (3) a statement that Congress might tax and spend only in pursuit of its other enumerated powers (James Madison’s view).164 According to all three views, the taxing and spending powers contained in the General Welfare Clause were enumerated; the difference concerned whether they were primary (Addison and Hamilton) or merely auxiliary (Madison).

In 1936, the Court adopted what it termed the Hamiltonian view of the General Welfare Clause (although the Justices went on to invalidate the tax at issue as impermissible “regulation”).165 On this view, the Court stated, “the clause confers a power separate and distinct from those later enumerated, is not restricted in meaning by the grant of them, and Congress consequently has a substantive power to tax and to appropriate.”166 Over the course of the twentieth century, therefore, the General Welfare Clause came to be awkwardly divided between the two doctrinal paths of the taxing power and the spending power.167 Beginning in the 1930s, taxing power cases typically focused on federal taxation of individuals,168 while spending power cases tended to concern Congress’s authority to use federal funds to encourage states to adopt particular programs or policies.169 These conditional spending programs and the bargains they entailed became a key tool in Congress’s regulatory arsenal and a way of working around objections to direct regulation of states qua states, although they sometimes prompted the objection that states were being coerced.170 The taxing power, meanwhile, was seen as relatively uncontroversial, albeit powerful.171

Today, however, the meaning of the General Welfare Clause is again in transition, as the doctrine works through a period of crisis of a type not seen since the 1930s, when the question of the federal government’s power to regulate the broader economy consumed both political and constitutional debates.172 A syntactically challenging clause that had, over the course of the twentieth century, achieved the status of an escape hatch for Congress became the focus of intense scrutiny in the Court’s 2012 decision in NFIB v. Sebelius.173 In upholding the ACA’s individual mandate under the taxing power component of the General Welfare Clause174 but striking down the act’s Medicaid expansion provision under the spending power,175 the Court revived an earlier, more stringent version of its spending power analysis while also expanding Congress’s taxing power. NFIB, then, was a split decision for the General Welfare Clause as a whole. But the decision plants a marker for future federalism analysis. The taxing power and the spending power, and the underlying idea that there is something called “the general welfare of the United States,” will likely be prominent sites of debates over federalism in the coming years.

The most striking element of the NFIB decision, for purposes of shadow powers analysis, is the Court’s apparent desire to use roundabout doctrinal means to test the overall quantum of federalism in the current American system. The majority opinion canvasses virtually the entire relevant landscape of Article I, trying every available doctrinal door in order to arrive at its conclusions. Two points are particularly noteworthy for this discussion: first, the change to existing spending power doctrine that the decision brought about; second, the dearth of references to the Tenth Amendment in the majority’s opinion.

In invalidating the conditional spending program attached to the ACA’s Medicaid expansion,176 the Court revived an inquiry into the potentially coercive nature of such programs that it had all but discarded in the decades since the 1930s,177 when it announced the coercion test in United States v. Butler.178 The return of coercion to the doctrine, after decades of conditional spending cases in which the Court upheld regulation despite only a nominal connection between the imposed condition and Congress’s regulatory end,179 suggests that the Court is not averse to reexamining doctrinal elements that many commentators had assumed were no longer contested.180 Especially noteworthy is the return of Butler, a pre-1937 precedent in which the Court held not only that a key piece of New Deal legislation “invades the reserved rights of the States,” but also that the “plan of regulation” contained in the act was not a permissible exercise of Congress’s taxing power.181 In NFIB the Court acknowledged that all taxes were potentially regulatory,182 but the adoption of a test derived from a period when a more formalist distinction between taxation and regulation obtained was still startling.183 Moreover, the Butler Court had developed the coercion analysis in the context of a tax provision, not a conditional spending program. Yet the NFIB majority imported the Butler coercion test from its origin in the taxing power, in an era of suspicion about taxes, into the modern conditional spending realm, despite the fact that the conditional spending power had previously been understood as offering the broadest avenue for congressional regulation under Article I.184

The lack of references to the Tenth Amendment in the Court’s opinion is also striking. With the exception of a single citation in its preliminary description of the Eleventh Circuit’s decision,185 the majority did not refer to the textual and doctrinal provision at the center of the Court’s “federalism revolution” of the 1980s and 1990s.186 Instead, the Court couched its discussion of the Medicaid conditional spending provision in terms of contract and individual freedom.187 Quoting an earlier decision holding that Congress could not “commandeer” state legislatures in which the Court had relied heavily on a robust reading of the Tenth Amendment, the Court in NFIB appeared to downplay the state sovereignty argument in favor of the individual liberty argument.188 Despite the fact that previous decisions invalidating congressional acts as unauthorized by the Commerce Clause discussed the Tenth Amendment frequently and at length,189 and despite the obvious relevance of the Court’s anticommandeering precedents to its decision about the Medicaid provision, the NFIB Court declined to discuss the Amendment.190

The lack of any substantive reference to the Tenth Amendment in the decision is additional evidence that the current form of the federalism debate has cycled back to its 1930s incarnation, rather than continuing in the line laid down by the “new federalism” cases of the 1980s and 1990s. Justice Holmes’s emphasis on Article I in his dissent in Hammer v. Dagenhart has returned to the forefront of the Court’s approach to federalism questions involving congressional regulation, but with a newly stringent application. The Court evaluated the ACA through the lenses of the commerce, necessary and proper, taxing, and spending powers, and found the act deficient in three of those four categories. Again, the narrowed commerce power—here actually present before the Court—appeared to guide the majority’s analysis of the other Article I powers. Moreover, the spending power issue directly implicated the states as states, but the Court’s emphasis on contract and individual liberty allowed it to avoid the potentially unproductive Tenth Amendment analysis of previous decades. Here again, the shadow powers of Article I entered into the doctrine when the direct lines of analysis—the Commerce Clause and the Tenth Amendment—appeared to have reached the limits of their usefulness. And yet again, the shadow powers pushed the Court toward Justice Holmes’s Article I approach in Hammer rather than the Hammer majority’s focus on the Tenth Amendment, with the counterintuitive result that the Court held the Act insufficient to meet the requirements for three of the four Article I powers at issue. In short, the NFIB decision demonstrated that it was possible to adopt Justice Holmes’s technique of analyzing the powers of Congress detached from state sovereignty concerns without also endorsing Justice Holmes’s conclusion that the Court should defer to Congress’s reasonable assessment of the scope of its powers.191 As with the return of the necessary and proper power, the Court’s interpretation of the taxing and spending powers in NFIB pushes the doctrine back toward the Article I methodology of the Hammer analysis, but with a very different result.

IV. the shadow powers and the union

What, then, is the significance of the shadow powers for modern federalism doctrine? And why should it matter whether the Court upholds a particular provision under the Necessary and Proper Clause or the General Welfare Clause as opposed to the Commerce Clause?

Viewed from a distance, the decision of the Court in NFIB that the individual mandate was a permissible exercise of the taxing power, but not the commerce or necessary and proper powers, might have little significance for the overall quantum of federal regulation. Viewed from the closer perspective of a potential litigant or an advocate, however, the choice of doctrinal pathway matters enormously for how a claim is structured. Viewed from inside Congress, moreover, the Court’s interpretive stance affects how legislation is drafted, what (if any) findings are included, and whether jurisdictional hooks such as requiring that a particular item (e.g., a gun192) have traveled in interstate commerce are incorporated. And viewed from the perspective of the Constitution’s deep background commitment to a principle of federalism, the Court’s choice to use the textually and structurally ambiguous Necessary and Proper Clause or General Welfare Clause in a given case should be seen as an indication that the Court is using Article I to sound, and perhaps change, the contours of federal regulatory power.193 The powers set forth in the clauses are shadow powers because they appear less controversial than their substantively specific Article I neighbor, the commerce power, but they nevertheless implicate foundational questions of how federalism works—and how it ought to work—in practice.

For the Founders, Federalists and Antifederalists alike, the key issue was how to build a union. They had many different ideas of what that union might look like. The broad array of possibilities available to them included a commonwealth with a single sovereign monarch but separate legislatures, like Scotland and England between 1603 and 1707; a shared, sovereign, and entirely new legislature, like Great Britain after 1707; or a general government to handle external affairs with colonial or state governments retaining authority over internal matters, like the de facto situation of British North America and the United States under the Articles of Confederation.194 But their focus was on making a union, not on maintaining the states as states. The novelty for the Founders was conceiving of the United States as a distinct entity that might have a distinct general welfare. Recall Dickinson’s citation in 1765 of the “general welfare” of the greater British Empire as a permissible justification for previous taxes levied on the colonies by Parliament.195 Dickinson invoked the general welfare of the entire empire as a legitimate purpose of trade regulation, including taxation, by Parliament. “We are but parts of a whole; and therefore there must exist a power somewhere, to preside, and preserve the connection in due order.”196 Not only the power, but also the duty of the central legislature, then, was to make such regulations as were “necessary for the common good of all.”197 Notably, both Dickinson’s “general welfare” and Dulany’s “necessary or proper” emerged in the context of taxation, the original concurrent power that virtually all participants in the imperial and founding-era debates believed must be held concurrently by legislative bodies at both the center and the periphery.

When modern judicial federalists take as their mission the preservation of state sovereignty, they tend to look to history, particularly that of the Founding Era, for support.198 But more often than not, they misread that history. To the extent one can reasonably generalize about the views of late-eighteenth-century North American commentators, their focus on federalism derived from their preoccupation with the problem of forming a viable and enduring union. Federalism was their particular answer to the question of how to build a union. It was a tool, and later a normative value, but it was not an end in and of itself.199 As the Dickinson example demonstrates, the imperative of building a union was felt strongly by the members of the founding generation.200 Moreover, the Tenth Amendment was initially promoted by Federalists who feared that the enumeration of rights contained in the Bill of Rights would be read as an exhaustive list, rather than by the Antifederalists or Jeffersonians, who are often lionized by the “new federalists.”201

Indeed, by the early nineteenth century, appeals to an almost mystical conception of “the Union” increased in frequency and fervor, eventually replacing earlier decades’ invocations of “the Republic” with a phrase and a concept that spoke directly to the complexities of federal-state relations. As Abraham Lincoln put it in his speech before the Young Men’s Lyceum of Springfield, Illinois, in 1838, American government was sustained by “the attachment of the People” and was therefore imperiled when citizens suffered the “alienation of their affections from the Government.”202 Lincoln’s solution was for the chief texts of the Union, the “Constitution and Laws,” to join the Declaration of Independence to comprise the “political religion” of the nation.203 No less a devotee of state sovereignty than John C. Calhoun saw himself as a Union man, engaged in a dialogue with the Founders and contributing to an ongoing founding moment even after he foreswore his membership in the Federalist Party.204 At the outbreak of the Civil War, President Lincoln would invoke no less authority in his first inaugural address than “universal law and . . . the Constitution” to argue that “the Union of these States is perpetual.”205

To the extent that modern federalism doctrine relies on some combination of historical sources, originalist interpretation, and historical practice, it is crucial to understand that precedents for both the Necessary and Proper and General Welfare Clauses emerged as part of arguments for centralized legislative authority over matters pertaining to the empire, and later the Union, as a whole. Another way to put this argument is to say that American federalism has historically assumed a meaningful scope of operations in which the power of the Union will operate. Still another way to put it is to say that some areas of government power are aimed at what Dickinson termed “the common good of all,”206 and what we might also call the general welfare of the United States. Within those areas, the federal government must have the power to act—not necessarily exclusive power, but some meaningful amount of power. This is the enumeration principle of Article I—derived from pre-revolutionary sources, glossed by Marshall in McCulloch, and extended by Holmes in his Hammer dissent.

From a structural perspective, moreover, one justification for the Court to apply forceful limits to federal power is the belief that the states have a discrete domain or “sphere” of regulatory autonomy.207 Even if this claim is accurate, however, it does not preclude the existence of a specific sphere of federal regulatory authority. The whole purpose of a union, in the colonial and founding eras as today, was not to have a consolidated, unitary government, but rather to bring a group of entities together under a single overarching authority.208 Within the defined sphere of its powers and duties, that union-level government must have the means of carrying out the ends confided to it. Indeed, in McCulloch, Marshall made this point explicitly: “[T]he Government of the Union, though limited in its powers, is supreme within its sphere of action. . . . The nation, on those subjects on which it can act, must necessarily bind its component parts.”209 The government of the United States was established in order to create a species of polity denominated as a union. The purpose of that union was to function as a nation with respect both to external affairs and to a somewhat defined set of internal affairs.210 While these facts do not necessarily establish that the federal government was intended as the equivalent of a national government, they do demonstrate that a key purpose of federalism has always been to provide a basis for a union and for a nation. The question for federalism, then and now, is what the best conception of federal union should be. Federalism, in other words, has always been the United States’ distinctive species of nationalism.