Prev Top Next

III. The Economics of Trends in United States Federalism

In this section we shall leave the economic theories of the previous two sections and turn to a discussion of current trends in federal-state relations in the United States. I begin with a brief survey of recent developments and then discuss two recent explanations of these developments that stress the role of economic factors in these trends.1

A. Recent developments in the Supreme Court

Let me boldly characterize the developments in federal-state relations in the United States between the New Deal of the 1930s and the mid-1970s as follows: the power of the central government expanded greatly, while that of the states diminished. This expansion of power to the center was greatest in civil rights and voting rights but was also noticeable in broader federal regulatory authority. However, since the mid-1970s the Supreme Court has been struggling to find a new balance between federal and state authority that generally involves protecting a larger sphere of sovereignty for the states and reducing the sphere of sovereignty for the federal government.

In National League of Cities v. Usery, 426 U.S. 833 (1976), the Court for the first time in forty years overturned an Act of Congress regulating economic relationships: it held unconstitutional a Congressional statute that applied minimum wage and maximum hours limitations to state and local government employees. This decision marked the beginning of the current unsettled period in federalism jurisprudence in the United States Supreme Court.

The Court reversed the Usery decision ten years later in Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528 (1985). The reason for this reversal was that the Court discovered and applied a new model of federalism—the "federal process" model—in the period between Usery and Garcia.2 That model holds that the Constitution protects the federal-state balance of power not just through explicit provisions (such as the Commerce and Supremacy Clauses) but also through political institutions. Therefore, the Supreme Court does not need to become involved through judicial review in the thorny political issue of "enforc[ing] limits on the scope of national regulatory power."3 Applying the federal process model, one could argue that the Court erred in Usery. Debate in Congress and state representation in the Senate had adequately exposed the nature and strength of the relevant state interests; because the Senate (and House, of course) passed the statute, the ex post objections of the states to the statute should have been discounted.

This plausible argument notwithstanding, the Supreme Court then reversed field again in New York v. United States, 112 S. Ct. 2408 (1992), rejecting the federal-process model. Why? Further reflection apparently convinced the Court that the federal-process model was not, after all, convincing: federal political institutions could not serve as an adequate protection against federal impingement on state sovereignty. Critics of the federal-process model argued that the Senate does not protect state governments per se but rather protects private citizens from the less-populous states.4 Another problem with the federal-process model is that Congress responds to conflicting interests of many kinds, not just to state interests. In this political jostling, the states might receive inadequate protection of their interests vis-à-vis the federal government. Congressional representatives, for instance, may perceive a conflict between their own political interests and the institutional interests of state governments. If pressed to achieve some national goal, Congress will try to do so without raising taxes but rather, perhaps, by imposing unfunded mandates on the states. In New York the Supreme Court held that these and other flaws of the federal-process model were fatal to the judicial "hands off" view of federal-state relations enunciated in Garcia. Instead, the Court chose a view of the federal-state relationship in which the federal government may compel the states to implement federal policies only if it pays for them or gives the states realistic options of non-compliance.

Late in its 1994 term the United States Supreme Court handed down two opinions that illustrate how unsettled the issue of federalism still is. The battle was most forcefully joined in U.S. v. Lopez, 1995 WL 238424 (U.S.).5 At issue was the constitutionality of the Gun-Free School Zones Act of 1990. In that Act Congress made it a federal offense "for any individual knowingly to possess a firearm at a place that the individual knows, or has reasonable cause to believe, is a school zone."6 The defendant Lopez, a twelfth-grade student at Edison High School in San Antonio, Texas, arrived at school with a .38 caliber handgun and five bullets. The authorities, acting on an anonymous tip, arrested and charged him under Texas law with firearm possession on school premises. The following day the state dropped its charges after federal agents charged Lopez with violating the Gun-Free School Zones Act. The District Court found him guilty and sentenced him to six months' imprisonment and two years' supervised release. Lopez appealed on the ground that the Act unconstitutionally went beyond Congress' power to legislate under the Commerce Clause. The Court of Appeals for the Fifth Circuit agreed and reversed the respondent's conviction.

The U. S. Government appealed, justifying the Act as falling within Congress' Commerce-Clause power in that firearms in or near schools may lead to violent crime, and violent crime affects the national economy in three ways. First, the costs of violent crime are large and are spread across state lines, principally by insurance markets. Second, violent crime makes some areas of the country less attractive and therefore reduces travel across state lines. And, third, firearms in or near schools damage the educational environment, and that damage will result in a less productive labor force and, therefore, a less healthy national economy.7

Chief Justice Rehnquist wrote the majority opinion, in which he spoke for a closely-divided Court that had found the Act to be unconstitutional. Under the government's theory, he wrote, there was almost no limit to the federal government's power under the Commerce Clause: with respect to crime, the government could regulate anything that would cause violent crime to increase, so long as it could find some connection between the regulated activity and interstate commerce. Rehnquist's implication is that, with respect again only to crime, this connection is so easy to find that there is no effective limitation on Congress' ability to regulate local activity so as to reduce crime. By further implication, what is true for violent crime would be true for any other activity that the federal government chose to regulate. The result, Rehnquist said, was that under the government's argument there would be no effective limits on the Congress' commerce-power jurisdiction. The Chief Justice wrote that this went too far; instead, he proposed a test for limiting that jurisdiction: "the proper test [of the appropriate reach of the Commerce Clause] requires an analysis of whether the regulated activity substantially affects interstate commerce."8 [My emphasis.]

B. State objections to federal power

Other manifestations of this movement to give increased recognition to the sovereignty of the states may be seen outside the Supreme Court. In 1994 four states—Colorado, Hawaii, Illinois, and Missouri—passed "sovereignty" resolutions, whose purpose was to remind the federal government that the states have their own claims on sovereignty and are not mere agents of the national government. Several other states are contemplating the adoption of similar resolutions. Additionally, there is a movement to create a "Conference of the States," which would be an ongoing convention with powers to propose amendments to the United States Constitution. Two such amendments that the organizers of the Conference have proposed are one that would allow the states to bypass Congress and suggest amendments of their own and another that would allow the states the power to veto some federal laws by a vote of three-fourths of the states. A model resolution calling for the convening of the Conference is before all 50 state legislatures.

Among the states' complaints against the federal government is their objection to unfunded mandates.9 A United States Conference of Mayors survey recently found that ten of these mandates accounted for an average of 11.7 percent of city budgets. The City of Chicago has said that it spends $190 million per year complying with federal mandates.10 Some states have refused to follow federal mandates; some have brought actions against the federal government for relief from the costs imposed on them through these unfunded mandates. For example, California, Pennsylvania, and Illinois originally refused to pass legislation to implement the federal Motor Vehicle Registration Act of 1993, called the "motor voter" law.11 An additional example of this newly-arisen will on the part of the states to assert their sovereignty is to be found in an action filed by the State of California against the federal government—California v. United States, 94-0674K(CM)—to recover the cost of increased services that the state must provide to illegal immigrants, who have arrived, the state alleges, because the federal government is not doing an adequate job of policing the borders.

C. Reforming federal-state relations

The developments in the Supreme Court and the newly-aggressive activities of the states against the federal government clearly indicate that the state-federal relationship in the United States is in flux. Why has this happened? As I noted earlier, the traditional view is that the United States is experiencing an ideological change-of-heart towards more conservative views of government, and one way in which that change is finding expression is in the form of a desire for a relatively smaller central government and relatively larger state governments. One measure of the strength of this traditional view is that political passions can be seen to run high on these matters.12

However, my contention is that this unsettled period in federal-state relations is due in large part to a change in real economic factors. I cannot fully document this contention now, but I can offer some collateral evidence from the work of others. I shall briefly discuss two recent studies of United States federalism that lend support to the economic theory of federalism and that make recommendations for restructuring the federal system that are in line with the economic theory.

The first study examines how the states and the federal government have specialized in the provision of services in the federal system. The author, Professor Paul Peterson of Harvard, says that states and localities generally should control physical and social infrastructure—roads, education, mass transit systems, public parks, police and fire services, and sanitation systems. He calls these developmental concerns, and the policies addressing them developmental policies. By contrast, the federal government should be in charge of distributional matters, transferring resources from those who have to those who have not; he calls the policies addressing these concerns redistributive policies.13

Against this rough normative theory of governmental responsibility in a federation, Professor Peterson evaluates recent trends in the United States. He finds that since 1962 state and local spending on developmental matters has been double what it has been at the federal level. The total amount spent on development by all levels of government, as a percentage of GDP, has increased over the last 30 years, but the relative shares of the various segments of government have been constant. By contrast, the division of redistributive expenditures has changed dramatically. State and local expenditures on those in need have steadily declined, while federal government redistributive expenditures have more than doubled as a percentage of GDP, from 4.9 percent in 1962 to 10.3 percent in 1990.

Professor Peterson holds that these trends were not the result of ideological differences but rather were the result of non-political, perhaps economic, forces. He discounts ideology as a causal influence because, during the period in which these trends developed, Democrats controlled the state legislatures, the governorships, and the national legislature. Because Democrats generally prefer redistributive policies, their control of state government should have led them, all other things equal, to increase state expenditures on redistribution. That this did not occur suggests that factors other than ideology determined the trends in the division of redistributive authority between states and the federal government over the last 30 years.

Why are state and local governments particularly efficient at providing developmental infrastructure? For all the reasons elaborated in the economic theory of federalism. States and localities are sensitive to local business interests and to the interests of their inhabitants. If voters and businesses like local policies, they will stay in the locality or migrate into it. If they do not, they will go elsewhere. If a state or local government finds an attractive policy innovation, other state and local governments will copy, and the innovation will spread. If a state or locality implements a new policy that does not work, other governments will not follow. This does not necessarily mean that policies will be uniform across all members of a federation. Because citizen tastes and preferences differ from region to region, state and local governments can be sensitive to these differences and accommodate their policies to them.

Peterson sees the national government as the natural provider of redistributive policies (just as the economic theory did). Therefore, he thinks that sending welfare back to the states (as has recently been proposed by the new Republican Congressional leadership) is a serious mistake. The evidence supports Peterson's and the economic theory's prediction that member-state governments will under-provide redistributive policies. State Aid to Families with Dependent Children (AFDC, the principal U.S. welfare program) benefits are jointly funded by the national and state governments. AFDC was set up in 1937, and, during its first 33 years, federal aid was relatively generous, and federal controls were relatively strong. As a result, states raised the real value of the benefits continuously. The mean monthly benefit paid to a family (in real 1993 dollars) in the average state was $287 in 1940, $431 in 1950, $520 in 1960, and $608 in 1970. Then in 1970 the federal government allowed the states much more freedom to set AFDC benefit levels, and these levels became relatively less generous. From their peak in 1970 of $608 per month, AFDC benefits have declined. They fell in real terms (in real 1993 dollars) to $497 in 1975, $437 in 1980, $409 in 1985, $379 in 1990, and to $349 in 1993.

State-level welfare reform has generally been nothing more than the reduction of benefits. The federal government in the 1988 Family Support Act gave the states the opportunity to experiment with AFDC. Not surprisingly, the first three proposals to come to the Department of Health and Human Services all proposed new restrictions on welfare benefits. For example, Wisconsin and New Jersey asked, among other things, for permission to withhold the increase in benefits that normally comes with the birth of an additional child. California wanted an immediate 25 percent reduction in benefits for all welfare recipients and a second reduction for all families remaining on welfare after six months, and a restriction that limited the level of benefits to new in-migrants to the level in the state from which they came. In summer, 1994, Wisconsin said that 20 percent of its new welfare recipients were newcomers to the state, mostly from Illinois, a state with lower welfare benefits. Wisconsin asked the federal government for permission to restrict payments to these newcomers for their first six months to the level of benefits they would have received in their previous state of residence.14

Professor Alice Rivlin, currently Director of the Office of Management and Budget, also recently proposed a restructuring of the federal-state division of responsibilities.15 The central reform she proposes is that the "states should take primary responsibility for a productivity agenda involving education, work force skills, and public infrastructure, while the federal government should retreat from these areas [, and that] a new system of common shared taxes ... should be adopted to help put state financing on a more secure and less unequal footing."16 She gives four reasons for restructuring federalism:

(1) the increase in global interdependence, which creates a need for the United States economy to become more productive in order to thrive;

(2) the need for less top-down reform (i.e., reform dictated from the federal government) and more bottom-up reform (i.e., reform that results from a multitude of experiments at the state level);

(3) the need for more tax revenue, which the states are reluctant to provide but which the federal government can provide; and

(4) general dissatisfaction with government, in large part because the federal government seems so distant.

Rivlin proposes that the restructuring should have the states taking charge of the public investment needed to raise productivity and incomes. We have already seen that an argument in favor of devolution is that the states can serve as experimental laboratories for different policies.17 In the recent past, a frequent argument against devolution was that the states did not have the abilities or the revenues to take on new governmental duties, even if economic considerations argued for devolution. Professor Rivlin says that this argument against a shift in federal responsibilities is no longer valid. In the past the federal government was, relatively speaking, far more competent than state governments. Governors had very short terms and few powers. Legislatures met for only a few weeks every other year, and their members served part-time and had almost no staff. Rural areas dominated these state legislatures, in some cases preventing reapportionment. For instance, Tennessee had no reapportionment between 1901 and 1962; as of the early 1960s eight other states had not redistricted in 50 years. Then, partly under the compulsion of Supreme Court decisions, the states undertook serious legislative reform in the 1960s.18 Governors were typically given more power. In 1955 only 29 governors had four-year terms; by 1988 that number was 47. During that same time period the number of states that barred a second term for their governors dropped from 17 to three. The general trend in the 1960s and 1970s was toward more professional and better-qualified governors and toward state legislators of a higher quality. As the workload imposed on state legislatures increased, the part-time state legislator gave way to the full-time legislator with a small but professional staff.

States have also found new ways to obtain revenue. The traditional sources of state and local governmental funds have been property and sales taxes. Whatever the other arguments that could be made against the property tax, a sound complaint against it was that it was not very flexible, so that as long as that tax was the main source of their income, states could not increase their revenues to take on new responsibilities. But sales taxes now bring in more money to the states than do property taxes, and income taxes are an increasing source of funds. Between 1960 and 1990 property taxes dropped from 37.7 percent to 21.8 percent of state and local government revenues; individual income taxes rose from 5.7 percent to 14.8 percent of state revenues; and fees and charges grew from 16.8 percent to 28.9 percent.19 As a result of these fund-raising innovations, state and local revenues, exclusive of federal grants, have gone up from 7.6 percent of GDP in 1960 to 10.3 percent in 1990.20

There have been two prior efforts, both recent, to restructure the division of responsibilities between state and federal governments. In the early 1970s Nixon's "New Federalism" proposal envisioned a stronger federal role in income maintenance and health care but a weaker federal role in the performance of almost every other governmental duty. Under that plan, there would have been fewer categorical grants from the federal government to the states, and the federal government would have given that reduced financial aid to the states with fewer restrictions.21 But the New Federalism failed. The beneficiaries of the existing categorical programs fiercely resisted the proposal. In the end Congress enacted only two block-grant programs—the Comprehensive Employment and Training Act of 1973 and the community development block grant program.

The second recent effort to restructure the federal relationship was President Reagan's "swap proposal" made in the mid-1980s, under which the federal government would have taken over some responsibilities from the states if the states would have taken over some from the federal government. For example, the federal government would have taken over the full costs of the Medicaid program in exchange for devolution of AFDC and the food stamp programs to the states. Most governors stoutly resisted the proposal (although, in retrospect, it appears that most of the states would have been better off financially if they had accepted). In addition, the proposal called for the devolution of 125 separate categorical grants to the states in exchange for the phasing-out of some federal taxes (alcohol, tobacco, telephone, part of the federal gasoline tax, and the windfall profits tax on oil and gas sales). Revenues from those taxes would have been put in a federal trust fund for the use of the states between the time of the implementation of the proposal and the elimination of the taxes in 1991. States would thereafter have had to raise their own taxes to pay for the programs or terminate them. The states were not at all enthusiastic about the proposal, and it died.

Professor Rivlin suggests that if a restructuring leads to many programs' being sent back to the states, it is unlikely that the states will raise taxes in order to pay for those programs. States are reluctant to impose new taxes for two reasons. First, they have unequal resources. Some may be wealthy and willing to increase taxes; others may be poor and unwilling to do so. Second, interstate competition generally precludes a state from imposing new taxes unless all other states do likewise.22 Because of this reluctance on the part of states, Rivlin proposes common shared taxes between federal and state governments. No doubt she is correct in asserting that a supra-state source of revenue would be necessary in order to restructure the federal-state relationship that she contemplates. However, whatever economic sense there may be to restructuring federalism in accordance with Professor Rivlin's suggestions, its political prospects are not bright.

Prev Top Next

1 There is an extensive literature in the law and in political science on federalism. For leading examples in the political science literature, see Vincent Ostrom, The Political Theory of a Compound Republic (2d ed., 1987) and Samuel Beer, To Make a Nation: The Rediscovery of American Federalism (1993). Some leading examples in the law literature on federalism are Herbert Wechsler, "The Political Safeguards of Federalism: The Role of the States in the Composition and Selection of the National Government," 54 Colum. L. Rev. 543 (1954); Jesse Choper, Judicial Review and the National Political Process (1980); Akhil R. Amar, "Of Sovereignty and Federalism," 96 Yale L.J. 1425 (1987); Deborah J. Merritt, "The Guarantee Clause and State Autonomy: Federalism for a Third Century," 88 Colum. L. Rev. 1 (1988); Deborah J. Merritt, "Three Faces of Federalism: Finding a Formula for the Future," 47 Vand. L. Rev. 1563 (1994); and David L. Shapiro, Federalism: A Dialogue (1995). 2 One of the seminal pieces in this literature is Choper, supra n. 48. Central to Professor Choper's argument is the view that the Court should avoid politically-charged issues whenever it can, that federalism is a politically-charged issue, and that the Court can safely leave the resolution of federal-state controversies to the political process. See also D. J. Merritt, "Three Faces of Federalism: Finding a Formula for the Future," 47 Vand. L. Rev. 1563, 1566-70 (1994). 3 Id. at 1566. 4 The connection between the Senate and the states per se was strong through the early part of this century. Until the passage of the seventeenth amendment in 1913 state legislatures, not citizens, elected Senators. 5 The other case was U.S. Term Limits, Inc., v. Ray Thornton, 1995 WL 306517 (U.S.), decided on May 22, 1995. At issue was the constitutionality of an amendment to the Arkansas Constitution that restricted the ability of a candidate to appear on the ballot for Congress if that person, although otherwise eligible to serve, had already served three terms in the House of Representatives. The Supreme Court held, 5-4, that neither Congress nor the states could determine the conditions of eligibility of Congressional candidates. Those conditions are fixed by the Constitution; the Constitution allows the states (through U. S. Const., Art. I, § 4) to determine only the time, manner, and condition of elections for Congress. 6 18 U.S.C. § 922(q)(1)(A) (1988 ed., Supp. V). The Act defined a "school zone" as "in, or on the grounds of, a public, parochial or private school" or "within a distance of 1,000 feet from the grounds of a public, parochial, or private school." 7 Justice Breyer's dissent found much to admire in the federal-process model. He said that Congress had felt that the effects of handguns in schools were deleterious, that there was adequate protection for the states in the Congressional debate and representation, and that, in light of these things, the Court should show extreme deference to Congress in these matters. 8 1995 WL 238424 (U.S.) at 6 (emphasis added). In dissent, Justice Souter says that the appropriate standard that the Court should apply is whether or not Congress had a rational basis for regulating the activity in question. This is a far more deferential standard toward Congress than the one that the majority approved in Lopez. 9 Both chambers of the Congress passed bills earlier this year to limit unfunded mandates. Neither bill is retroactive, and each permits future unfunded mandates so long as they do not impose costs in excess of $50 million per state per year. For an excellent discussion of the economic issues in unfunded mandates, see Kathleen Segerson, Thomas J. Miceli, and Lih-Chyi Wen, "Intergovernmental Transfers in a Federal System: An Economic Analysis of Unfunded Mandates" (Working Paper, Department of Economics, University of Connecticut, 1995). 10 Richard Reuben, "The New Federalism," ABA J. 76 - 81 (April, 1995). 11 Pennsylvania and Illinois originally refused to pass implementing legislation. California passed such legislation, but Governor Pete Wilson, who is seeking the Republican nomination for President, vetoed the bill and sued the federal government for a declaration that the law was unconstitutional. California estimated that it would cost the state $20 million per year to implement the motor voter law. A public-interest group sued the governor of Illinois for failure to comply with the law. In Association of Community Organizations for Reform Now (ACORN) v. Edgar, 1995 WL 329726 (7th Cir. (Ill.)), decided June 5, 1995, Judge Richard A. Posner held that the federal mandate to provide voter registration as part of motor vehicle registration was not violative of the U.S. Constitution. The State of Illinois chose not to appeal the decision. However, its announced scheme of implementation of the motor voter law may itself be challenged. The State will have employees of the Department of Motor Vehicles inform applicants for a driver's license that they may register for an election for federal office and will assist them to do so but will not inform or assist them in registering for state and local elections. 12 As an example of this passion, consider Arthur Schlesinger, Jr.: "The Republican Party has apparently embarked on a crusade to destroy national standards, national projects, and national regulations and to transfer domestic governing authority from the national government to the states. A near majority of the Supreme Court even seems to want to replace the Constitution by the Articles of Confederation. There has not been so basic an assault on the national government since the Civil War." "In Defense of Government," Wall Street Journal, June 7, 1995, A14, col. 3. 13 Paul E. Peterson, "Who Should Do What?" The Brookings Review (Spring, 1995), pp. 6-11. This is an excerpt from The Price of Federalism (Brookings, 1995). 14 There are important issues of federalism involved in these restrictions. For several decades many states have attempted to pay new migrants to their states lower welfare and other governmental benefits than they pay to their settled citizens. Those thereby disadvantaged have challenged those state limitations as unconstitutional restrictions on the right of United States citizens to travel. Thus far, the United States Supreme Court has always found these restrictions to be unconstitutional. 15 Reviving the American Dream: The Economy, the States & the Federal Government (1992). 16 Id. at 8. 17 We have also seen two arguments against this. First, Madison felt the states would be more likely to be in thrall to tyrannical factions than would the national government. Second, there may be circumstances in which national uniformity is more important than is satisfying local preferences. 18 In Baker v. Carr (1962) the Court held that legislative malapportionment was violative of the fourteenth amendment to the United States Constitution and that state legislative districts should be apportioned according to the principle of "one person, one vote." Reynolds v. Sims (1964) held that the fourteenth amendment required both houses of state legislatures to be apportioned according to population. 19 Advisory Commission on Intergovernmental Relations, Significant Features of Fiscal Federalism, v. 2: Revenues and Expenditures 132 (1991). 20 Budget of the United States Government, Fiscal Year 1992, table 15.1 (1991). 21 Under the proposal 129 separate programs (or "categorical grants") would have been lumped into six block grants with very few constraints on how the state-recipients spent that money. 22 For example, Florida attempted to impose a tax on services in 1987. The tax went into effect, but businesses then boycotted Florida as a place for conventions. The impact was so great that the governor and legislature agreed to repeal the tax within six months of its implementation.