This article appears in the January/ February edition of 2013 for The American Interest.

For decades, Democrats and Republicans alike have invested heavily in governance schemes that erode the Constitution’s separation of powers and mar its proper functioning. The Federal judiciary has uniformly rubber-stamped these schemes. The consequence has been an unsustainable spree of borrowing, spending and overregulation at the Federal level, cyclical fiscal crises at the state level, and less accountable and less representative government at every level. 

These governance schemes are generally of two kinds: one erodes the separation of powers between Federal and state governments, while the other erodes the separation of powers within the Federal government. In the first category is “cooperative federalism”, whereby the Federal government uses monopoly powers to coerce and subvert the prerogatives of state governments. In the other is Congress’s delegation of vast rule-making authority to administrative agencies. 

These two categories of concern are often treated as being entirely distinct, but they share profound similarities. Both are methods for Congress to escape accountability by hiding its power in other institutions of government. Cooperative federalism allows Congress to hide its power within the decision-making of state governments, while its delegation of rule-making authority allows it to hide its power in the far-flung bureaucracy of the Executive Branch. 

The Federal judiciary has a crucial role to play in maintaining and policing the boundaries of America’s basic institutions of state. It is a role it abdicated when confronted with the popular nationalist programs of the New Deal. The constitutional doctrines the judiciary has invoked to let Congress blur these critical separations of power are deeply flawed as a matter of constitutional law, and they have ultimately become unsustainable as a matter of political economy. Federal courts must begin to enforce a strict separation of powers, both between the Federal and state governments and within the Federal government itself. And Congress itself must start undoing the consequences of its own self-indulgence. 

The Trojan Horse of Cooperative Federalism

There are two main species of cooperative federalism. The first is Federal assistance grants such as Medicaid, in which Congress gives money to state governments on condition that their programs comply with Federal preferences. The second is cooperative regulation, in which the Federal government allows states to implement Federal regulations themselves, also on condition that they meet certain requirements (known as “conditional preemption”). 

Both sound nice, but both violate the principle of federalism in the Constitution, under which states are declared sovereign and the powers of the Federal government are specifically enumerated and correspondingly limited. Both allow anti-competitive political cartels in Congress to strangle innovation and regulatory competition at the state level by using the federal machinery to impose an uncompetitive policy baseline on everybody.

At the Supreme Court, federalism has staged a promising comeback, though it is still little more than a rear-guard action. The Court ruled in Garcia v. San Antonio Metropolitan Transit Authority(1984) that the limits on Congress’s power to control state governments depend on the national political process itself—in other words, on Congress’s self-restraint. Backing away from this dangerous idea, two crucial cases of the Rehnquist Court established the blanket principle that the Federal government cannot command state governments to do anything. 

New York v. United States (1992) struck down part of a Federal law because it required states either to take title to low-level radioactive waste generated within their borders or to regulate its disposal according to Congress’s instruction. “In this provision”, reasoned the Court’s majority, “Congress has crossed the line distinguishing encouragement from coercion.” Congress could not force states to choose between two alternatives, neither of which Congress had the power to impose “as a free standing requirement.” Writing for the majority, Justice Sandra Day O’Connor wrote, “While Congress has substantial powers to govern the Nation directly, including in areas of intimate concern to the States, the Constitution has never been understood to confer upon Congress the ability to require the States to govern according to Congress’ instructions.” 

The rule of New York was strengthened five years later in Printz v. United States, when the Court struck down a part of the Brady Act that required states to conduct background checks on prospective gun purchasers. The Court insisted that Federal and state governments occupy separate spheres in a “structural framework of dual sovereignty” and that the states must remain “independent and autonomous within their proper sphere of authority.” In a majority opinion by Justice Antonin Scalia, the Court ruled that if a Federal law offends “the structural framework of dual sovereignty”, it is not “proper” and is “an act of usurpation.” Hence, Congress simply cannot command a state official to do anything.

If New York and Printz blasted away the foundations of Garcia’s cooperative federalism, they managed nevertheless to leave undisturbed its twin pillars: conditional Federal grants and cooperative regulation. Under New York, both are forms of “encouragement” not rising to the level of “coercion.” Where the Federal government merely “encourages state regulation rather than compelling it, state governments remain responsive to the local electorate’s preferences; state officials remain accountable to the people.” 

The distinction between the Federal commandeering of state governments, which is prohibited, and cooperative Federal programs that states are theoretically free to turn down, which is allowed, is an utterly illusory one because in either case there is a penalty. If officials disobey a legal requirement, they may be dismissed, and are subject to writs of mandamus and criminal penalties. But if they don’t accept “voluntary” Federal grants (and comply with the attached conditions), there is also a penalty: The tax revenue their citizens have already contributed to the program will be transferred to other states and they will lose all use of it. And if states don’t accept Congress’s invitation to implement Federal law “voluntarily”, the Feds take over, diminishing the state’s regulatory autonomy.

The imputation of a penalty is the difference between law and a suggestion for good conduct. It’s the difference between putting up posters to encourage schoolchildren to eat broccoli and giving them detention for not doing so. Every law presupposes the free will of those subject to it. They can choose to obey or suffer the penalty for not obeying. Hence, every law contains elements of both encouragement and compulsion. That is as true of the prohibition on murder as it is of the conditions attached to cooperative Federal-state programs. Though murder is malum in se, there is no difference in the form or operation of the command: Obey, or you will lose something you’re otherwise entitled to have. Simply put, the penalties in cooperative federalism give such programs the force of legal compulsion.

If we reverse the logic of O’Connor’s distinction between encouragement and coercion and start by asking whether a Federal law leaves elected state officials free to regulate “in accordance with the views of the local electorate”, it becomes obvious that virtually all instances of cooperative federalism boil down to the Federal commandeering of state agencies. That commandeering is no less effective or compelling just because it operates “indirectly” through the leverage of penalty.

South Dakota v. Dole (1987) upheld a Federal law that threatened states with the loss of 5 percent of Federal highway funds if they did not raise their drinking age to 21. Dole noted that the penalties attaching to such programs could not be so onerous as to “pass the point at which pressure turns into compulsion.” Dole insists that state prerogatives must be preserved, both in theory and in fact, but the ruling would have us believe that the state’s freedom to refuse the funding and its conditions constitutes freedom of choice. It does, but only in the sense that I am free to disregard the prohibition of murder. 

The fact remains that states must either obey the Federal will or accept an onerous penalty: the loss of their state’s share of funding for the program in question. Dole imagines a spectrum of penalties from mild to onerous, and some imaginary point in between that Congress cannot cross. This rule is so indeterminate that it proved impossible for courts to apply in the decades after Dole was handed down. No Federal court applying Dole ever struck down a Federal penalty as coercive, no matter how coercive it was—until the June 2012 Obamacare ruling.

The Patient Protection and Affordable Care Act, a.k.a. Obamacare, requires that states expand their Medicaid programs from arrangements to help specific categories of poor people with healthcare (pregnant women, the disabled, needy families, children) into a vast wealth-redistribution scheme for the entire nonelderly population up to 133 percent of the Federal poverty level. It threatened states with the loss of all Federal Medicaid funds if they did not comply with the new mandates. That draconian penalty was too much for the Roberts Court, which ruled that the Federal government could refuse states the subsidies that Obamacare provides for the expansion itself, but could not cut off existing Medicaid funds for states that refused to comply. 

Compared with the modest penalty in Dole, the threat of losing all Federal Medicaid funding (more than 20 percent of a typical state’s budget) was “much more than relatively mild encouragement”, wrote Roberts: “It is a gun to the head.” Wherever the point is between encouragement and compulsion, the penalty for not complying with the Medicaid expansion was well beyond it. But we are still left to wonder: Where is that all-important point? 

The answer is: Nowhere. The point doesn’t exist, or rather it exists wherever the Court chooses to place it in any given case…before it decides to move it again. The Court has not recognized that, where the conditions attached to Federal grants go beyond the narrow Federal interest in how the money is spent, and impact broader state policies and programs, such grants are coercive, not by degrees along a spectrum, but categorically. In all such cases, the Federal government taxes money away from the residents of a state and offers to give it back to the state only on condition that its government complies with Federal preferences on a whole range of quintessentially state policy matters. Even under the Court’s Obamacare ruling, states that refuse to expand their Medicaid programs will still be subsidizing the Medicaid expansion of other states. That is coercion, pure and simple. Alas, the steady erosion of state autonomy that coercion entails will continue until the Court finally confronts the fatal flaw in Dole

The essential problem in conditional Federal funds for the states arises from the lack of restraints on the purposes for which Congress may use its taxing and spending power. Under the Constitution, Congress has the power to levy taxes only in order “to provide for the common Defense and general Welfare of the United States.” But nothing in the Supreme Court’s inordinately deferential interpretation of this clause prevents Congress from using that power in order to provide for the welfare of special interest groups. The door to intergovernmental collusion and anti-competitive fiscal cartels is thus left wide open.

The Constitution originally required that all “direct taxes” (which the Supreme Court eventually interpreted to include income taxes) be apportioned among the states according to population. In order to meet the apportionment requirement, residents of poor states had to pay much higher tax rates than residents of rich states, a grossly regressive scheme that made it virtually impossible for Congress to enact any direct tax. Until early in the 20th century, Congress had to rely on indirect taxation (such as import duties and excise taxes), a severely constrained source of revenue. But the 16th Amendment gave Congress the power to levy income taxes without apportionment, putting vast sums of taxpayer money at its disposal. 

For the separation of Federal and state powers, the consequences were fatal. Congress could now offer state officials this Faustian bargain: Give up your autonomy in exchange for Federal funds, and you will be able to offer your constituents more goodies while avoiding accountability for increased taxation. As Michael Greve observes in The Upside-Down Constitution (2012), this blurring of accountability has allowed officials at the Federal and state levels to claim credit for much more spending than if they had to account for it entirely by raising taxes on their own. 

The dramatic expansion in the American public sector since World War II has occurred almost entirely at the state level: As a share of GDP, Federal revenue has remained steady while state revenue (not including Federal assistance, which has exploded in recent years) has nearly tripled and has almost pulled even with Federal revenue. Greve is almost certainly correct that this expansion has been due to the collusive intermingling of state and Federal finances. Worse, the collusion readily lends itself to the formation of state fiscal cartels in Congress—cartels designed to diminish regulatory competition and diversity among the states and impose the preferences of uncompetitive states on everybody. In 1926, for example, Congress adopted a Federal inheritance tax coupled with an offset for state inheritance taxes. This not only neutralized the competitive advantage of states with low tax rates but also strongly incentivized states to raise their rates. Among the effects of this whole arrangement is the exacerbation of inequalities between rich states and poor states (especially through the vehicle of matching funds) and the amplification of cyclical state fiscal crises. And when this fiscally fueled takeover of state governments runs into the political limits of what Congress can tax, Congress does what no state can do: It borrows to the tune of trillion-dollar deficits. 

In terms of both flawed constitutional doctrine and disastrous political economy, the problems are largely the same in the context of cooperative regulation, the other main species of cooperative federalism. This governance scheme (which legal scholars confusingly refer to as “conditional preemption”) involves the Federal government giving states “permission” to implement Federal law on condition that their programs meet Federal specifications. The penalty here is not the threat of transferring the state’s wealth to other states, but rather the threat of the Federal government implementing Federal standards itself if states cannot or will not do so. The ground for collusion is laid by a combination of the Supremacy Clause and concurrent Federal-state jurisdiction over the same range of conduct, which arose largely from the vast expansion of Federal commerce power after the New Deal.

The arrangement contemplated in the Clean Air Act is typical. The U.S. Environmental Protection Agency (EPA) says to each state, “We’ll give you permission to implement our regulations yourself, so long as you design your State Implementation Plan according to our specifications. Otherwise we will pre-empt your regulation and impose our own Federal Implementation Plan.” Because the business community is increasingly terrified of the EPA (with good reason), states usually jump at the chance to implement the regulations themselves, even at their own expense. The fiction of “voluntary” state acceptance is even more tenuous here than in the conditional grants context, because the state gets a bad deal no matter what choice it makes—hardly the paradigm of an arms’ length contract freely entered into. 

If the concerns for accountability and representative government expressed in New York and Printz have proved only marginally troubling for the Supreme Court in the face of coercive Federal grants such as the one in Dole and Obamacare, those concerns have troubled the Court even less in the arena of cooperative regulation. Its two major cooperative regulation cases, Hodel v. Virginian Surface Mining (1981) and FERC v. Mississippi (1982), appear to foreclose any constitutional challenge to such schemes on federalism grounds. Hodel upheld a Federal law that established national standards for surface coal mining and allowed each state to implement the standards through regulatory programs subject to Federal approval. Absent a federally approved state program, the Federal government would implement the standards itself in that state. The Court ruled that, because Congress could simply have pre-empted all state regulation, “We fail to see why the Surface Mining Act should become constitutionally suspect simply because Congress chose to allow the States a regulatory role.”

In fact, though the Justices failed to see it, there was a very good reason to doubt the constitutionality of the Surface Coal Mining Act. In effect, it deputized state governments into implementing Federal regulations, diminishing both accountability and representative government. The Act threatened the “States’ ability to function effectively in a federal system” and their “separate and independent existence”, prerogatives which, under Printz, are supposed to be categorically inviolable. (Incidentally, the Act also violated separation of powers within the Federal government by delegating to the states the President’s exclusive authority to see that the laws are faithfully executed.)

The Court’s decisions in this area have been unstable and contradictory for decades. There is room to argue that if Garcia is ever finally overruled on the basis of New York and Printz, thenHodel and FERC should fall with it. The “national political process” exalted in Garcia presupposes that Congress can be held accountable for the costs and consequences of its own policies. But it is precisely this accountability that cooperative regulation allows Congress to escape by deputizing state governments. That is why cooperative regulation is more than just a lesser-included power of wholesale preemption and should be viewed as a distinct kind of power. 

The institutional incentives point in the same direction. Acting through the EPA’s cooperative regulatory programs, Congress does not merely hide accountability for its own policies within state agencies. It also escapes the vital limitations that the legislative process imposes on it when it must provide for the implementation of its own programs. If the EPA didn’t have the option of subverting states into implementing its new greenhouse gas regulations, for example, it might have to hire tens of thousands of bureaucrats to process permit applications from the six million businesses that could eventually be subject to the new regulations. It is inconceivable that Congress would appropriate such vast sums to implement an imposition that, in the form of “cap and trade”, couldn’t pass Congress even with a Democratic supermajority. But if the EPA has the money to implement the scheme in just a few states, it can threaten most of them into compliance.1

Hence, both of cooperative federalism’s two major manifestations tend to create anti-competitive political cartels that diminish self-government while increasing the power of Congress. But Printz still offers hope for a historic shift in the Court’s thinking. If states must remain “independent and autonomous within their proper sphere of authority”, then it may once again be possible to trace the outer boundaries of the Federal government’s delegated powers by tracing the outer boundaries of the states’ reserved powers. 

It doesn’t really matter where the line is drawn. The key thing is to draw it somewhere, and keep the functions of state and Federal government strictly separate. The first step is for the Supreme Court to start giving real teeth to its habitual insistence (repeated most recently in the Obamacare decision) that the Federal power to regulate commerce “among the several States” presupposes a judicially enforceable distinction between what is national and what is truly local. 

Giving Self-Government Away


n 1912, one of the New Republic’s original triumvirate of writers, Walter Weyl, wrote: “In truth, the Constitution is not democratic. It was, in intention, and is, in essence, undemocratic.” The Progressives’ contempt for the Constitution went hand in hand with their exaltation of supposedly scientific knowledge as a criterion of legitimacy for legislation of all kinds. Both tendencies entailed a dangerous threat to accountable, representative government—and to the power of the Supreme Court. 

The Court’s authority to enforce the Constitution would now be assailed by an increasingly powerful constituency for expanding Federal power well beyond the limits that had been understood since the Constitution’s ratification. Once Franklin D. Roosevelt was elected President, the Court was continually on the ropes defending the Constitution against the New Deal’s encroachments, until finally, after Roosevelt’s failed “Court-Packing” proposal of 1937, it simply caved in. After Wickard v. Filburn (1942) removed all effective limits on the Federal commerce power, Congress began regulating all aspects of economic activity. 

Combined with the then-modish infatuation with science, this opened the floodgates to delegations of legislative authority to Executive Branch agencies. Given the difficulty of drawing clear boundaries between legislative, executive and judicial powers, the Court abandoned formal categories in favor of a functional approach that allowed any delegation for which Congress supplied “intelligible” guiding principles, and prohibited only “usurpation” of the powers of other branches. Yet as David Schoenbrod observes, usurpation is not the only problem: “Legislators enhance their power by delegating: they retain the ability to influence events by pressuring agencies, while they shed responsibility for the exercise of power by avoiding public votes on hard choices.”2

Congressional influence over agency actions is retained through the power of the purse, and through the Senate’s power to consent to Executive Branch appointees. But delegations of rule-making authority vitiate a cardinal purpose of having a Senate in the first place. Upon his return from the American embassy in France, Thomas Jefferson is said to have complained to George Washington about the creation of the Senate, for which he saw no need. The incident supposedly occurred over breakfast. “Why did you pour that coffee into your saucer?” asked Washington. “To cool it”, answered Jefferson. “Even so”, replied Washington, “we pour legislation into the senatorial saucer to cool it.” 

In the modern era, the Senate’s “cooling” function is captured in the power to filibuster, and the requirement of a 60-vote supermajority to override it. Delegation eliminates this constraint: It allows legislation to occur in effect even when only a minority in Congress supports it, rather than blocking legislation when a sizeable minority opposes it. This was demonstrated by the EPA’s recent greenhouse gas regulations, which never would have passed in Congress.

Walter Lippmann, the most brilliant of the New Republic’s founding triumvirate, advocated an approach to lawmaking that presupposed the superiority of scientific understanding in the cost-benefit analysis that legislatures routinely make. He assumed that scientific skepticism would win out over political prurience. He was wrong. The scientific spirit is quintessentially one of inquiry, skepticism and self-criticism. Those qualities often go begging when science is proffered as certain knowledge in the service of a political agenda. Lippmann failed to realize how vulnerable science is to manipulation as a tool of political advocacy, whether by the Right or by the Left. 

Congress has nevertheless continued to exalt supposedly scientific knowledge above popular self-government, no doubt partly because it has such a strong incentive to do so. Its broad delegations of rule-making authority have allowed it to escape accountability by placing “hard choices” in the hands of supposed experts in the Executive Branch. The Administrative Procedure Act of 1946 further blurs the separation of powers by treating rule-making as a problem of due process rather than of properly locating the legislative power. When an agency proposes new rules, it must provide people with notice and an opportunity to be heard, nothing more. Courts have strengthened these safeguards, but this has only reduced Congress’s incentive to sort out the competing interests of interest groups, which look increasingly to the courts to resolve their conflicts. And Humphrey’s Executor v. United States (1935) ruled that if delegations contain quasi-legislative or quasi-judicial functions, then the President’s power can also be curtailed, thereby paving the way for agencies that are effectively independent of any political control. 

The separation of powers within the Federal government is perhaps the most basic structural design feature of the Constitution. Article I vests the legislative power in Congress, Article II vests the executive power in the President, and Article III vests the judicial power in the Federal courts. Obviously, this arrangement presupposes clear distinctions between legislative, executive and judicial functions. 

The difficulty of drawing distinctions between the legislative and executive functions of government arises because the power to execute laws implies the power to interpret them and fill in the gaps with policy. Chevron v. NRDC (1984) embraced an extreme form of judicial deference to Executive Branch agencies’ interpretations of Federal law, supposedly out of respect for the prerogatives of the President. The Court has yet to grasp that this deference to the President actually increases the power of Congress, as Schoenbrod explains.

In Mistretta v. United States (1989), the Court declared,

Our jurisprudence has been driven by a practical understanding that in our increasingly complex society, replete with ever changing and more technical problems, Congress simply cannot do its job absent an ability to delegate power under broad general directives.

This proposition seems obvious enough—except that it contradicts the Constitution’s most basic structural design feature. Proponents of delegation point out that the Constitution does not offer much basis for a judicially enforceable separation of powers. 

And it is true: It does not. The Constitution was designed to achieve balance among the three branches of government at the Federal level. The Supreme Court has insisted that core functions remain the preserve of the respective branches, but in cases of overlap it cannot referee the conflict, which instead must be negotiated in the political arena. The desired “separation of powers” instead must result mainly from the proper functioning of the three branches, each under their own recognizance.

In Federalist No. 47, James Madison wrote, “the accumulation of all powers, legislative, executive, and judiciary, in the same hands, whether of one, a few, or many, and whether hereditary, self-appointed, or elective, may justly be pronounced the very definition of tyranny.” All three branches have contributed procedural safeguards to guard against this danger in the delegation of rule-making authority. But such safeguards cannot guard against the loss of accountable, representative government, because that loss inheres in the delegation itself.

Supposedly, any law that attempts to vest legislative power in the President or the courts is unconstitutional. Congress has gotten around that by hiding its power within the policymaking discretion of the Executive Branch, under the justification that it can enact laws “necessary and proper for carrying into Execution” its constitutional powers. Members of Congress can still influence agency action through the power of the purse; their constituents can manipulate rulemaking through citizen lawsuits or by “capturing” the agencies wholesale, but they can still disclaim responsibility for the results. The net result is that agencies are able to ram through sweeping regulations of society that would never pass Congress. Members of Congress can claim credit for the benefits, while escaping accountability for the costs. That is convenient for them, but very inconvenient for the people, whose ability to control the government through the ballot-box is significantly diluted. 

Under current constitutional law, Congress can delegate virtually all of its functions to the Executive Branch. Indeed, it mostly has. Executive agencies add 60,000 pages of new rules every year, vastly more than the volume of new laws passed by Congress. 

The Supreme Court must take a broader view of what constitutes the “core functions” of each branch. But given the largely non-justiciable overlap at the margins, the political branches must also do their part. The incoming Congress will showcase a significant contingent of conservative legislators (from the elections of 2010 and 2012) who are wary of Federal power and fiercely protective of the Constitution. Their challenge will be to understand where Congress has gone too far, and thus where it must start policing its own boundaries. Congress must reassert its role as the principal rulemaker. It could start by passing the REINS Act, which would require congressional approval of major new Executive Branch regulations.

Significant reforms of the laws on administrative procedure could give the Federal courts a renewed basis for policing the separation of powers within the Federal government. It will be especially crucial to find Federal court appointees who understand that Congress increases its power by hiding it in the Executive Branch, even when that appears to augment rather than “usurp” executive power. In the meantime, Republicans and Democrats in Congress will continue to profit from the work of a faceless, nameless bureaucracy controlled by nobody and whose handiwork is untraceable to them. 

The national crisis the United States faces today has resulted from specific, deeply flawed constitutional decisions of the Supreme Court, and the expansions of congressional power those decisions have blessed. Tinkering at the margins of such decisions in the interests of judicial restraint makes sense when there is time to wait upon the glacial evolution in the Court’s constitutional jurisprudence. But when, as now, various deeply flawed lines of precedent have combined to produce a national crisis, the Court must be willing to revisit and reverse precedents that are proving unworkable. 

Congress also has a crucial role to play. Committee hearings and reports can flesh out and help develop institutional consensus on the dangers of cooperative federalism and excessive delegations of rule-making authority. Proposals such as the REINS Act will have difficulty gaining adherents in the absence of a broad consensus on the urgent need for sweeping government reform. 

The separation of powers within the Federal government serves the same function that the Supreme Court has ascribed to federalism: As the Court said recently, “federalism secures to citizens the liberties that derive from the diffusion of sovereign power.” This argues for a renewed national commitment to accountable and representative government, and to the separation of powers on which it so vitally depends.