The Devolution Tortoise and the Centralization Hare

John Kincaid

Robert B. and Helen S. MeynerProfessor of Government and PublicService, and Director of the MeynerCenter for the Study of State and LocalGovernment, Lafayette College.

The DevolutionTortoise and theCentralization Hare

There has been much talk in recent years of devolving powers andfunctions from the federal government to the states. Some observ-ers even proclaim a “devolution revolution” (Nathan 1996), the

result of which will be a more efficient and effective federal governmentand more robust and responsive states. The generally recognized objec-tives of devolution include (1) more efficient provision and production ofpublic services; (2) better alignment of the costs and benefits of govern-ment for a diverse citizenry; (3) better fits between public goods and theirspatial characteristics; (4) increased competition, experimentation, andinnovation in the public sector; (5) greater responsiveness to citizenpreferences; and (6) more transparent accountability in policymaking.

These are ambitious objectives, although, to date, no consensus ondirection is apparent, no plan of execution is in place, and examples ofdevolution are scarce. Indeed, there are only two commonly citedexamples of devolution: congressional repeal of the national 55-mphspeed limit and welfare reform. The prospects for significant devolutionduring the foreseeable future are not bright, largely because federalofficials are reluctant to relinquish powers they have acquired in thetwentieth century to advance national policy objectives. Consequently,devolution is plodding along at a turtle’s pace while centralization is stillracing ahead at a rabbit’s pace.

Devolution and Dual Sovereignty

Devolution is also a curious notion because, strictly speaking, therecan be no devolution in the American federal system. Unlike Britishparliamentary supremacy, whereby Parliament can unilaterally devolvepowers to regional and local authorities that possess no sovereignty orsupremacy of their own, the U.S. Congress possesses only limited,delegated powers, and the U.S. Constitution establishes a system of dual

sovereignty characterized by the U.S. Supreme Courtin 1869 as “an indestructible Union of indestructibleStates” (Texas v. White 1869, p. 725). The Congresspossesses limited, enumerated powers delegated to itby the sovereign peoples of the several states; all otherpowers are reserved to the states or to the people asstipulated by the Tenth Amendment. The U.S. Con-stitution, therefore, does not contemplate devolu-tion of powers from the federal capitol to the statecapitols.

Furthermore, while the Constitution implicitlyallows the states to exercise some powers that havebeen delegated to the federal government but notexercised by the Congress (for example, regulation ofeconomic activities not regulated by the Congressunder the interstate commerce clause), the Constitu-tion does not authorize the Congress to exercise re-served powers of the states that are not exercised by

Devolution is a curious notionbecause, strictly speaking, there

can be no devolution in theAmerican federal system. Theterm is used loosely in politicsand often used interchangeably

with “decentralization.”

some or all of the states. In other words, the stateshave historically served as backstops for the federalgovernment, reserving authority to act in the face offederal inaction so long as state action does not violatethe U.S. Constitution or run afoul of what the U.S.Supreme Court has called the “dormant commerceclause” by enacting laws that interfere with interstateor foreign commerce, even though the Congress hasnot occupied a field or acted to preempt or prohibitstate action.

Dual sovereignty is further entrenched in theconstitutional amendment processes (Article V),which guarantee that no formal transfers of constitu-tional powers can be effected between the federalgovernment and the states without the explicit, con-current consent of the Congress and the state legisla-tures (plus the people, in the never-used constitutionalconvention method of amendment). Changes in theallocation of constitutional powers require the joint

consent of both the “indestructible Union” and the“indestructible States.” If this were not the case, theUnited States would be a decentralized unitary polityrather than a noncentralized federal polity.

Defining Terms

Nevertheless, the constitutional niceties of feder-alism have not, until recently, been evident in publicdiscourse. As President Franklin D. Roosevelt put it in1937, “I am not in love with any particular methods,but I am in love with particular objectives” (quoted inMason 1952, p. 15). Since the New Deal, debate hasbeen dominated by considerations of political andpolicy objectives rather than constitutional methods.Since the late 1960s, moreover, such debate has oc-curred within the context of growing public disaffec-tion from the federal government and increasingpublic ignorance of the federal system. Given, forexample, that fewer than half of American adultsknow that their state has its own constitution (U.S.Advisory Commission on Intergovernmental Rela-tions [ACIR] 1988, p. 6), most citizens have littleconception of dual sovereignty and constitutional fed-eralism. Consequently, while “devolution” strikes adissonant note for constitutional purists, “devolu-tion,” along with related buzz words—decentraliza-tion, deregulation, delegation, deconcentration, anddecongestion—seems to strike a harmonious note forcitizens said to be “fed up” today with the politiciansin Washington, whose objectives and methods nolonger inspire much public confidence.

Defining “devolution” is not easy, however, be-cause the term is used loosely in politics and oftenused interchangeably with “decentralization.” For bi-ologists, devolution means de-evolution, namely, evo-lutionary degeneration toward greater simplicity and,often, disappearance. This is not the commonly under-stood political definition of devolution, although thebiological definition might resonate with some advo-cates of devolution who envision a steadily shrinkingfederal government, as well as with some opponentsof devolution who fear an enfeebling degeneration offederal power.

Devolution can be defined as a transfer of specificpowers or functions from a superior government to asubordinate government. The transfer is of constitu-tional magnitude even if not effected through a writ-ten constitution; it is ordinarily intended to be perma-nent; it surrenders all the powers associated with thedevolved functions (namely, political, legislative, ad-

May/June 1998 New England Economic Review14

ministrative, and fiscal); and it leaves the functionalfield vacant for occupancy by subordinate govern-ments. Devolution occurs in the context of verticalintergovernmental relations. Hence, devolution canoccur within the British parliamentary system and canalso be said to occur within the fifty American stateswhere, constitutionally, local governments are crea-tures of their state, but it cannot occur between thefederal government and the states without constitu-tional change.

Delegation can be defined as one governmentauthorizing another government to carry out func-tions on its behalf, much like a representative. Theimplementing government remains accountable to thedelegating government and may or may not enjoyfreedom to design methods of implementation. Dele-gation can occur in the context of vertical or horizontalrelationships between governments. Delegation iscommon in the American federal system and is, inreality, what is often labeled “devolution.” The 1996“welfare reform” act (The Personal Responsibility andWork Opportunity Reconciliation Act), for example,fits a delegation model better than a devolution modelbecause, even though delegation gives states muchgreater discretion to design, implement, and financecertain welfare programs, the states remain account-able to the federal government and reliant on federalfunds, and they are subject to penalties from federalauthorities for failures to achieve federal performanceobjectives. The federal government can also prohibitcertain modes of implementation, examples beingPresident Bill Clinton’s 1997 decisions to prohibit theoutsourcing of certain welfare functions to privatefirms and to extend federal labor-law mandates towelfare recipients in state workfare programs.

Deregulation (in the public sector) can be definedas the loosening or removal of regulations enacted orpromulgated by one government to control or directthe behavior of another government or set of govern-ments. The states, in principle, have plenary regula-tory authority over their local governments, exceptwhere prohibited by the U.S. Constitution. The federalgovernment has a limited range of direct regulatoryauthority over the states under the supremacy clauseof the U.S. Constitution (Article VI), although thisclause has been broadened by twentieth-century inter-pretations of the Constitution, such as the FourteenthAmendment (1868). The federal government has aneven broader range of indirect regulatory authorityover the states through grants-in-aid enacted pursuantto the Congress’s spending and general welfare pow-ers (Article I, Section 8). The rise of “regulatory”

(ACIR 1984) or “coercive” federalism (Kincaid 1993b)since the mid 1960s has become a matter of greatconcern to most state and local government officials;consequently, calls for intergovernmental deregula-tion are often intermixed and confused with calls fordevolution.

Decentralization has been subject to a host ofdefinitions, plus distinctions between legislative, ad-ministrative, political, fiscal, and spatial decentraliza-tion. Decentralization can be defined as the transfer ofresponsibilities of various kinds to subordinate ad-ministrative units or subordinate units of government,much the way most states decentralize functionsthrough counties. This was, for example, Alexis deTocqueville’s preferred model of federalism. Ratherthan “an incomplete national government” like thatcreated by the U.S. Constitution, a strong federalunion, according to de Tocqueville, needs a completenational government in which political power is cen-

What is being advocated, underthe rubrics of devolution and

decentralization? Restorations ofpowers to the states and theirlocal governments as well as

deaccessions of unwantedfunctions, which could produce arebalancing of power between thefederal government and the states.

tralized while administrative authority is decentral-ized through subordinate units of government likecounties (Tocqueville 1969). Given that the U.S. statesare not county-like administrative arms of the federalgovernment and given that the federal government is,in effect, an incomplete national government, theconcept of decentralization, like devolution, does notfit the American federal polity, even though the netresults of devolution, delegation, and deregulationmay look like decentralization (Mayer 1976).

The difference is subtle but important. Decentral-ization, like devolution, requires a complete nationalgovernment able to decentralize and recentralizepower, as metropolitan councils discovered duringMargaret Thatcher’s tenure as prime minister. Decen-

May/June 1998 New England Economic Review 15

tralization and devolution are fully applicable, forexample, to the former Soviet Union and its satellitecountries wherein power was highly centralizedthrough the Communist party. The American federalunion, however, is really noncentralized (Elazar 1984)because it is a system of dual sovereignty constitutedof multiple centers of power independently account-able to the people.

A prime example is taxation. No centralized taxpower exists in the United States. The power to tax isnoncentralized, and some tax powers are concurrent(that is, exercised by both the federal government andthe states, such as income taxation). The federal Con-stitution imposes only a few limits on state tax powers(for example, states cannot tax imports or exportswithout the consent of the Congress) while, in turn,delegating only a few tax powers to the federalgovernment (however broad and deep the de factoimpacts of the federal income tax). Indeed, the U.S.Constitution had to be amended in 1913 to give theCongress the power to tax income (Amendment 16).The states are free to levy any taxes not prohibited bythe federal Constitution, while the federal governmentis permitted to levy only those taxes delegated to itthrough the U.S. Constitution.

If devolution and decentralization do not, strictlyspeaking, fit the American situation, then what isoccurring, or being advocated, under the rubrics ofdevolution and decentralization? The answer wouldappear to be a process involving restoration, deacces-sion, and rebalancing: that is, restorations of powers tothe states and their local governments as well asdeaccessions of unwanted functions, which, together,could produce a rebalancing of power between thefederal government and the states. Few of the powersand functions nominated as candidates for devolutionwere even exercised or performed by the federalgovernment a generation or two ago. These powersand functions were exercised by the states or, in somecases, could have been, and perhaps should havebeen, exercised by the states under their general policepower. Hence, when the U.S. Advisory Commissionon Intergovernmental Relations advocated a shift ofmost federal highway functions and revenue sourcesto the states, the Commission used the term “turn-back” (ACIR 1987). Similarly, the “devolution” lan-guage used most often by elected officials is that ofrestoring or reviving state powers or states’ rights andrebalancing the federal system (for example, Councilof State Governments 1996).

Although we may be stuck with the term “devo-lution” because the media and some academics have

embraced it, many different things are occurring in thefederal system, some of which point, for the first timeduring this century, to the possibility of genuinerestorations of some state powers and genuine limitson certain further expansions of federal power while,at the same time, centralization or “counter-devolu-tion” continues apace.

Forces for Restoring State Powers

The most immediate force propelling discussionsof shifting powers back to the states was the 1994midterm elections, which brought a Republican ma-jority into both houses of Congress. The 1994 electionsended more than sixty years of nearly continuousDemocratic control of the Congress (with Democraticcontrol of the U.S. House of Representatives havingbeen the longest period of one-party rule in U.S.history) and also unseated an incumbent HouseSpeaker for the only time in this century.

The most immediate forcepropelling discussions of shiftingpowers back to the states was the

1994 midterm elections, whichbrought a Republican majorityinto both houses of Congress.

The Republicans’ “Contract with America” con-tained several provisions aimed at curbing federalpower and restoring state powers, beginning withmandate reform, which was enacted in 1995 withbipartisan support as the Unfunded Mandates ReformAct (UMRA). The state-friendly provisions of the“Contract with America” reflected long-standing Re-publican concerns dating back to President Dwight D.Eisenhower, who had likened the centralization ofpower and rise of the “military industrial complex” inthe United States during the New Deal and WorldWar II to the “extreme and dictatorial concentration ofpower” then occurring in communist Eastern Europe(Coleman and Goldberg 1990, p. 20).

President Eisenhower was not, however, able totilt power back toward the states. On the contrary,Eisenhower’s nomination of Earl Warren to be Chief

May/June 1998 New England Economic Review16

Justice of the U.S. Supreme Court, his support for theNational Defense Highway Act of 1956 and for theNational Defense Education Act of 1957, his dispatchof federal troops to Little Rock, Arkansas, in 1957 toenforce desegregation pursuant to the U.S. SupremeCourt’s ruling in Brown v. Board of Education (1954),and other policy actions, all enhanced federal power.Thus, federal policymaking during the supposedlyconservative 1950s reinforced Alpheus T. Mason’sobservation that: “For two generations American po-litical and economic life has been moving swiftlytoward ‘bigness,’ toward monolithic organization”(Mason 1952, p. 1).

President Richard M. Nixon’s New Federalism,which echoed Eisenhower’s federalism sentiments,was, despite General Revenue Sharing (1972 to1986), no more successful in stemming the growth offederal power. President Ronald Reagan began hisAdministration with strong, emphatic support for aNew Federalism aimed at shifting substantial powerback to the states. Reagan’s success was modest atbest, and the federal government continued to ex-pand its power through mandates, preemptions,and conditions attached to federal aid (Conlan 1988;Kincaid 1993b).

This brief history suggests two hypotheses. First,many of the Republicans elected to the Congress in1994 arrived with strong commitments to long-frus-trated desires to limit federal power, which for some,though not all, also means restoring state powers. Thiswas reinforced by the new Republicans elected fromthe increasingly Republican South and from theMountain West—both growing regions historicallysuspicious of, and often hostile to, the federal govern-ment. Second, it is the Congress more often than theWhite House that alters the balance of power in thefederal system. If these conclusions are correct, thensome significant restorations of state powers are likelyto occur if the Republicans maintain control of theCongress, if the Republicans capture the Presidency in2000, and if the currently federalism-friendly majorityon the U.S. Supreme Court is maintained or increasedafter 2000.

These prospects are further strengthened by sup-port for rebalancing federal-state power among moremembers of the Democratic party than was true in thepast. In a 1997 statement, for example, the DemocraticLeadership Council, with which President Clintonwas affiliated, said:

The New Democrat movement has consistently re-jected the old-fashioned liberal prejudice against stategovernments and state elected officials. . . . Now more

than ever, state elected officials represent the future ofour party and our country. State capitals are the battle-grounds where the big challenges of American domesticpolicy on the eve of the 21st century are being met (1997,p. 1).

The 1996 Democratic party platform even claimed that“Republicans talked about shifting power back to thestates and communities—Democrats are doing it.”

This is a far cry from Governor George Wallacestanding in the doorway of the University of Alabama in1963 defying federal authority, but the states have sinceexperienced a remarkable rehabilitation, which hasplaced them in a much more favorable light. Well intothe 1970s, most Americans expressed more trust andconfidence in the federal government than in the states.Since then, however, public trust and confidence—to theextent the public has any trust and confidence in anygovernments—has shifted gradually and substantiallyfrom the federal government to the states. For example,a 1995 poll conducted by Princeton Survey Research

Since the 1970s, public trust andconfidence—to the extent the

public has any trust andconfidence in any governments—

has shifted gradually andsubstantially from the federal

government to the states.

Associates found that 61 percent of Americans trustedtheir state government to “do a better job of runningthings” than the federal government. Nearly everysubgroup except black Americans and Jews expressedsimilar support for the states. Adults under age thirtyfavored the states by 72 to 21 percent, Democrats, by48 to 35 percent, and self-described liberals, by 49 to 36percent (Donahue 1997, p. 13). These findings areconsistent with trends in most polls asking similarquestions for the past decade.

This opinion change is due partly to the “mod-ernization” of state and local governments and polit-ical systems that has occurred since the 1950s, withmuch assistance and pressure from the federal gov-ernment. The structural changes highlighted by mostobservers—such as constitutional revision, legislativereapportionment, professionalized legislatures, four-

May/June 1998 New England Economic Review 17

year tenure for 48 governors, and strengthened judi-ciaries (ACIR 1985; Reeves 1990)—have been impor-tant, but the most significant change has been thealtered relationship between citizens and their stateand local governments. State and local political sys-tems are more electorally inclusive and representativeof diversity than in the past, and they are no lessinclusive than the federal arena. Previously excludedgroups have more opportunities to win office and gainemployment in state and local governments than inthe federal government. On average, state and localgovernments are also more accessible to citizen re-form, not simply because they are smaller arenas butalso because state constitutions and amendments re-quire popular ratification in 49 states; because stateconstitutions compel recourse to the people and to

Adding force to efforts torebalance federal-state power has

been migration to the suburbs andto the Sunbelt, along with the

perceived fiscal crisis of thefederal government.

supermajority consent on many important matters;and because a number of state constitutions givecitizens initiative, referendum, and/or recall powers.The federal government is constitutionally imperviousto such direct citizen participation. In addition, stateand local governments are better equipped today thanin the past to provide more and better services forcitizens, and with the appearance of less corruption.Furthermore, state and local tax systems are morediversified than in the past and are often less regres-sive than the federal tax system, where Social Securityand Medicare taxes—the largest tax bites for mostAmericans—are regressive. In addition, most stateand local governments have been more fiscally disci-plined than the federal government.

The case for devolution has been bolstered as wellby a recovering economy. As a result, most states havehad year-end budget surpluses during the 1990s; statebudget surpluses amounted to $24 billion at the outsetof 1998; state and local government spending hasincreased by 14.3 percent in real terms since 1990; 31states reduced taxes in 1996; and state and local

government employment has increased by about 1.8million persons since 1990 while federal governmentemployment has declined by more than 345,000.

Adding force to rebalancing efforts has been mi-gration to the suburbs and to the Sunbelt. The UnitedStates has been, and continues to be, a nation of smalland medium-size communities. Never in the historyof the United States have more than 30 percent ofAmericans lived in urban places having 100,000 ormore residents. The high points of big-city life inAmerica occurred between 1920 and 1970. By the late1980s, the United States had become a suburbannation. In addition, the center of population continuesto migrate southwestward as most Sunbelt statesexperience tremendous growth.

A political sine qua non of suburban life is localself-government, jealously guarded and combinedwith stout resistance to metropolitan consolidation orother arrangements perceived as threatening to localautonomy, as well as a proclivity for service-efficientcouncil-manager type governance free of “politics”(see, for example, Baker 1975). Hence, the suburbanmentality, whatever the shortfalls and shortcomingsof its realization, is receptive to “devolution,” and thefederal government and most state governments havebeen deferential to this suburban mentality. This def-erence is reflected, for instance, in the U.S. SupremeCourt’s unwillingness to extend court-ordered schoolbusing beyond central-city boundaries, in the Con-gress’s and the states’ overwhelmingly preferentialfunding for highways rather than mass transit, in theunwillingness of both the federal government and thestates to enforce fair housing laws with any vigor, inthe reduced federal funding since the Reagan era formetropolitan regional structures (for example, coun-cils of governments), and in the unwillingness of statelegislatures to mandate jurisdictional consolidations.

Migration to the Sunbelt is having at least twodevolutionary effects. For one, given that most Sunbeltstates have, historically, been states’ rights states,migration into those states has strengthened the states’rights voice in Washington, DC. At the same time,migration has helped to “modernize” those states, tomake many of them more progressive and innovativeand to soften the hard surfaces, such as racism,historically characteristic of politics in many of thosestates. Migration has helped to pull the Sunbelt statestoward the center of the national political spectrumwhile simultaneously pulling the national politicalcenter slightly rightward.

Stemming in part from this suburban and Sunbeltmigration has been the emergence of Republican con-

May/June 1998 New England Economic Review18

trol of a majority of the governorships (32 as of March1998). On average, Republican governors have beenmore supportive of restoring state powers than Dem-ocratic governors, and Republican governors, for ex-ample, played a major role in shaping “welfare re-form” in 1996. The governors, however, includingRepublican governors, do not champion state prerog-atives across the board. Most governors continue toadvocate federal preemptions of state powers in manyareas affecting interstate commerce, and most gover-nors supported the preemption-laden North Ameri-can Free Trade Agreement (NAFTA) in 1993 and theUruguay Round of the General Agreement on Tariffsand Trade (GATT) in 1994.

Both Democratic and Republican state and localofficials have given strong support to deregulatorymeasures and other efforts to reduce federal “micro-management” of state and local affairs. Here, electedstate and local officials have been nearly unanimous,as reflected in their concerted lobbying for the Un-funded Mandates Reform Act (UMRA). The state-localcoalition constructed to support UMRA was unusualbecause state and local officials rarely speak with onevoice in Washington, DC.

Certain historical imperatives, especially the per-ceived fiscal crisis of the federal government, haveadded force to rationales for shifting power towardthe states. This was the reality facing President Clintonwhen he declared the end of the era of “big govern-ment.” Regardless of one’s views about federal budgetdeficits and the size of the federal government, theweight of public opinion perceives these to be seriousproblems, even though the public does not endorsedraconian remedies. The federal government lacksthe fiscal resources, and perhaps the political will aswell, to sustain, let alone augment, the kinds ofexpansive and state-intrusive policies associatedwith the New Deal of the 1930s and the GreatSociety of the 1960s. By a combination of default,deaccession, and disinvestment, therefore, certainpowers and responsibilities must necessarily flowinto the states, and the Congress will likely dumpcertain fiscally onerous and politically volatile func-tions onto states and localities.

The growth of entitlement spending, which nowdominates the federal budget, adds more pressure onCongress to off-load “discretionary” programs thatcarry real budget costs. Although the 1996 “welfarereform” ended certain long-standing entitlements,their costs pale in comparison to the costs of SocialSecurity, Medicare, and Medicaid. These entitlementsare likely to remain entrenched federal responsibilities

not only because they are advocated by influentialreformers (Rivlin 1992; Peterson 1995) but also, andmore important, because they are predominantly mid-dle-class entitlements. Even Medicaid has become amiddle-class entitlement because of the high costs oflong-term health and nursing home care for the el-derly. Entitlement pressures on the federal budgetwill, moreover, begin to skyrocket in 2010 when babyboomers commence retirement.

Public disaffection from the federal governmenthas also contributed to support for restoring statepowers. The federal government has experienced anastonishing erosion of legitimacy in recent decades,coupled with disillusionment and skepticism aboutthe efficacy of many federal programs—an attitudereinforced by the generally critical media. Althoughmany federal programs are more successful than isgenerally believed by voters, reality is what citizensperceive it to be, and this perceived reality under-

Support for restoring state powersand rebalancing the federal system

arises not only from factorsparticular to the United States

but also from a worldwide trendtoward decentralization, whichsome link to a broader trend of

“flattening hierarchies” across allsectors of postmodern life.

wrote President Clinton’s pledge to “reinvent” gov-ernment and, among other things, “to end welfare aswe know it.” Whatever its past successes, “welfare aswe knew it” became widely viewed as a counterpro-ductive failure.

Undoubtedly, some advocates of restored statepowers desire little more than to shrink, even cripple,the federal government, while still others wish toabolish or reduce certain federal policies and pro-grams by using the subterfuge of federalism to turnthem over to the states where, they hope, the policiesand programs will be dismantled or cut back. Suchopportunistic forum shopping has long been commonin the federal system, and one can expect the debate

May/June 1998 New England Economic Review 19

over restorations of state powers to continue to beclouded by opportunism.

But such motives are overshadowed by anotherlegitimizing force: namely, the power of an idea.Demands for decentralization, devolution, deregula-tion, federalism, power sharing, and the like havebecome prominent worldwide. These demands oftencome from ethnic, tribal, linguistic, religious, andpolitical groups within multinational states, such asthe Baltic states, Belgium, Canada, India, Malaysia,Mexico, Nigeria, Pakistan, the Philippines, Rwanda,Spain, Ukraine, the former Soviet Union, the Sudan,Uganda, the United Kingdom, the former Yugoslavia,and others. Elsewhere, though, as in Denmark, France,Japan, Norway, and Sweden, pressure for decentrali-zation or “deconcentration” has stemmed from esca-lating national welfare costs, rising unemployment,and a growing belief that regional or local govern-ments can provide services more efficiently and effec-tively (see, for example, Baldersheim and Stava 1993).

An interesting and somewhat surprising twist onthe enthusiasm for decentralization has been its sup-port from the left, such as the Socialist party in France,which promoted extensive devolution (Bernier 1992),and the British Labour party’s promotion of successfuldevolution referenda in Scotland and Wales in 1997.At the same time, the rise of the more rightwardPublic Choice school of thought has provided impor-tant intellectual underpinnings for devolution (see, forexample, Buchanan 1995).

Decentralization and even federalism havegained substantial legitimacy from the perceived fail-ures of the centralized state (Wunsch and Olowu1995), failures of planned economies, and weakeningof the historic nation-state (Elazar 1995). Pressures fordecentralization and federalism are usually linkednow to political democratization, economic deregula-tion, denationalization and dereservation of state en-terprises and lands, and decriminalization of previ-ously forbidden market and voluntary sectoractivities. Hence, while federalism in the United Statesis often viewed as a conservative idea because of thehistorical backdrop of reactionary states’ rights, feder-alism (or at least decentralization) in many othercountries is frequently viewed as a liberal or progres-sive concept because of the historical backdrop ofcentralized tyranny (Kincaid 1995). As Carlos Fuenteswrote along these lines in 1990: “The Federalist Papersshould be distributed in the millions.”

In summary, support for restoring state powersand rebalancing the federal system arises not onlyfrom factors particular to the United States but also

from a worldwide trend toward decentralization,which some observers link to a broader trend of“flattening hierarchies” across all political, social, eco-nomic, and cultural sectors of postmodern life.

Federal Actions Restoring State Powers

Despite predictions of a “devolution revolution”(Nathan 1996) and prognostications of doom that“wholesale devolution invites the balkanization . . . ofAmerica” (Donohue 1997, p. 14), there are few exam-ples of “devolution,” although a discernible crawltoward some rebalancing of the federal system isunder way in the Congress, the Presidency, and theSupreme Court.

Congressional Federalism

Early steps toward rebalancing were GeneralRevenue Sharing (GRS), enacted as the State and LocalFiscal Assistance Act of 1972, and the enactment oftwo of six block grants proposed by President Nixon.GRS was created when it was still credible to arguethat the federal government was positioned to gener-ate surplus revenue that could be disbursed to fiscallyhandicapped state and local governments. GRS fundswere delivered to state and local governments byformula and with few regulatory strings. The basicidea was that state and local governments, with citizeninput, were best positioned to set priorities for spend-ing GRS monies in ways most useful and appropriatefor their jurisdictions. The Congress expected GRSfunds to be expended primarily for capital invest-ments; however, many states and localities used GRSto support current operations and/or to reduce taxes.Such “misuses” of GRS confirmed the belief of skep-tics that state and local governments cannot be trustedto behave responsibly without congressional directionand that restoring powers to state and local govern-ments will not enhance policy efficiency, effectiveness,or equity. GRS was terminated for the states in 1980during President Carter’s Administration and thenallowed to expire for local governments in 1986 dur-ing President Reagan’s Administration.

Another step toward providing states withgreater flexibility in using federal funds occurredwhen 77 categorical grants were consolidated intonine new or amended block grants in the OmnibusBudget Reconciliation Act of 1981. Although PresidentReagan had sought to consolidate about 85 categori-cals into only seven block grants, the Congress refused

May/June 1998 New England Economic Review20

to go that far. It also declined to remove as manygrant-reporting and oversight rules as had been pro-posed by President Reagan. Nevertheless, the numberof block grants leaped from four to twelve, andadditional block grants were enacted up through the1996 Temporary Assistance for Needy Families(TANF) block grant and the 1997 State Children’sHealth Insurance Plan block grant.

Block grants, however, are not sterling examplesof devolution for three reasons: (1) the Congress tendsto re-categorize block grants over time, not only forpolitical reasons but also for accountability reasons; (2)block grants are more vulnerable to funding reduc-tions than many categorical grants-in-aid, and (3) onlyabout 15 percent of federal aid flows to state and localgovernments through block grants. The Congress hasa distinct preference for more tightly controlled cate-gorical grants, and 618 such grants were funded infiscal year 1995 compared to only 15 block grants(ACIR 1995).

A discernible crawl toward somerebalancing of the federal system

is under way in the Congress,the Presidency, and the

Supreme Court.

A de facto fiscal dumping of certain functions canbe said to have occurred from 1978 to 1990 whenfederal aid to state and local governments declinedfrom 17.0 percent of federal outlays to 10.8 percent,and federal aid as a proportion of state and localoutlays declined from 26.5 percent to 18.9 percent.Although federal aid has since increased to about 23percent of state and local outlays, functional dumpinghas continued because nearly two-thirds (63 percent)of federal aid is now dedicated for payments toindividuals (mostly social welfare), whereas in 1978more than two-thirds (68 percent) of federal aid wasdedicated for state and local government policy func-tions and operations.

Most of the discussion of devolution, however,has been stimulated by more recent enactments. TheIntermodal Surface Transportation Efficiency Act of1991 (ISTEA), President Bush’s 1992 Executive Order12803 on infrastructure privatization, and President

Clinton’s similar 1994 Executive Order 12893, are citedby some observers as harbingers of devolution. ISTEAcreated a multimodal surface transportation blockgrant, gave states more discretion over certain aspectsof surface transportation while also requiring state-wide transportation plans, and placed more emphasisthan previously on “performance” standards for statesand localities. ISTEA also permitted funds transfersbetween transportation programs when requested bystate governments or metropolitan planning organiza-tions (MPOs), authorized the same state-local fundingmatch for most programs so as not to disadvantagecertain transportation choices, and provided the coun-try’s 339 MPOs with greater authority over transpor-tation planning and federal funds allocations. Al-though the outcomes of ISTEA appear to be generallysatisfactory, initial evidence also suggests that “regu-latory and workload burdens have grown signifi-cantly” (Gage and McDowell 1995, p. 148), particu-larly for MPOs.

Advocates for highway devolution argue, amongother things, that:

First, the responsibility for building, owning and operat-ing [transportation] systems is primarily regional or local,not national. Now that the interstate highway system hasbeen completed, the federal role in highways can bedramatically reduced. Second, there are major disadvan-tages with the centralized federal trust-fund approach tofunding transportation infrastructure. Third, it is statesand cities—not the federal government—that have beenmost innovative in seeking new and better ways to investin infrastructure and improve its performance, by mak-ing use of public-private partnerships (Poole 1997, p. 20).

Federal highway aid, moreover, remains a primetarget for congressional “crossover sanctions” (ACIR1984, p. 43). These are reductions of aid disbursementsto states that fail to comply with aid conditions thatare arguably not germane to the aid’s central objec-tives and original intentions. One of the first crossoversanctions was the Highway Beautification Act of 1965,which was prompted by Lady Bird Johnson’s dislikeof highway billboards. A recent prominent example isthe 21-year-old drinking age requirement that wasattached to federal highway aid in 1984 and upheld bythe U.S. Supreme Court in 1987 (South Dakota v. Dole).However, the Congress did repeal a crossover sanc-tion in 1996, namely, the 55-mph speed limit enactedunder the Emergency Highway Energy Act of 1976,thus giving states freedom to set or not set highwayspeed limits, but it also enacted a new conditionrequiring states to prohibit persons under age 21 fromdriving with a blood alcohol level of 0.02 percent or

May/June 1998 New England Economic Review 21

higher. Any state failing to enact a law to enforce thisprohibition within three years of 1996 will lose 5percent of its federal highway aid in the first year and20 percent each year thereafter.

Amendments enacted in 1996 to the Safe DrinkingWater Act of 1974 substantially changed the regula-tory regime within this field by giving states morefunding to comply with environmental standards andalso more flexibility to exercise authority over drink-ing water standards and their enforcement. The actalso responded favorably to growing complaints fromlocal governments, especially small localities, aboutout-of-reach costs of bringing water systems intocompliance with regulations and time limits promul-gated by the U.S. Environmental Protection Agency(EPA). Small jurisdictions can now choose to complywith alternative treatment and monitoring standards.Environmental remediation is likely to be a continuingintergovernmental issue because the EPA has esti-mated that local governments will bear about 88percent of the cost of public sector compliance. It isalso “projected that by 2000, local governments willhave to spend an extra $12.8 billion per year, or 65percent more than they did in 1988, just to maintaincurrent levels of environmental protection” (Habicht1992, p. 27).

Other recent measures viewed as restoring somestate and local powers include food-safety legislationthat withdrew an earlier standard of “zero cancerrisk” for pesticide residue on processed food andprovided instead for a uniform health-oriented stan-dard for chemical residues on agricultural produceand processed food. The Selected Housing ProgramExtensions Act of 1996 delegated more powers to localpublic housing authorities to evict residents whoabuse drugs or alcohol and to set aside housingexclusively for senior citizens and persons with dis-abilities. The commuter policies included in the 1990Clean Air Act were made voluntary; these were pre-vious mandates that required states with carbon mon-oxide and ozone non-attainment areas to reduce em-ployee commuting trips through car pooling and othermeasures.

The case of welfare reform. The premier specimen of“devolution,” however, is “welfare reform,” whichone observer terms “the most vivid example of au-thority cascading to lower levels of government”(Donahue 1997, p. 7).

The devolutionary centerpiece of “welfare re-form” is the Temporary Assistance for Needy Families(TANF) block grant. This reform eliminated the open-ended Aid to Families with Dependent Children

(AFDC) entitlement, along with JOBS and EmergencyAssistance. TANF will provide $16.38 billion per yearto states during fiscal years 1997 to 2003 to operatetime-limited cash assistance for needy families cou-pled with work requirements for most recipients.States can use their federal money in any way “rea-sonably calculated to accomplish the purposes ofTANF” (Section 404(a)). States have broad discretionto determine eligibility, methods of assistance, andbenefit levels; to operate new food stamp and employ-ment and training programs; and to decentralize wel-fare functions to local governments. Many state offi-cials, especially governors, helped to shape TANF andsupported its enactment. The legislation also madesubstantial though not always state-friendly changesin child care, the Child Support Enforcement program,food stamps, SSI for children, and benefits for legalimmigrants. It also set funding for the Social ServicesBlock Grant at $2.38 billion in fiscal years 1996 to 2002and at $2.8 billion in fiscal year 2003 and thereafter. Anumber of technical changes were made in 1997, mostof which were favorable to the states.

Three other elements of TANF are interesting fortheir implications of future restorations of state pow-ers. One is Section 417, which reads: “No officer oremployee of the Federal Government may regulate theconduct of the States under this part or enforce anyprovision of this part, except to the extent expresslyprovided in this part.” This is a signal to the executivebranch not to interpret the law liberally so as to extendits regulatory reach beyond the boundaries expresslyset out by the Congress.

Second, TANF overturned Shapiro v. Thompson(1969) by allowing states to treat newly arrived wel-fare claimants differently than resident recipients byapplying, for up to 12 months, the welfare rules of themigrant’s former state of residence. This provision isintended to reduce rent-seeking migration that mightdrive states into “a race to the bottom,” but the largersignificance is the willingness of the Congress tooverturn a 27-year-old state-restrictive and rights-protective ruling of the U.S. Supreme Court.

Third, the Brown Amendment added to the Per-sonal Responsibility and Work Opportunity Act of1996 enhances the authority of state legislatures overbudgetary and nonbudgetary aspects of devolution.The Amendment affirms the authority of the legisla-tures to appropriate block grant monies for Tempo-rary Assistance for Needy Families and for child-caredevelopment, rather than leaving these matters in thehands of the governors who, in the past, had servedessentially as administrative agents for the federal

May/June 1998 New England Economic Review22

government. The Amendment, therefore, also in-creases legislative leverage over the governor, placesstate legislatures on a more equal policymaking foot-ing with governors and the Congress, and exposeswelfare policymaking to greater voter input throughlegislative elections and popular initiatives.

While the above provisions constitute the major“devolutionary” elements of “welfare reform,” otherprovisions suggest a continuing and enhanced federalregulatory role. For example:

• States must meet maintenance-of-effort rules re-quiring them to spend on TANF-related activities80 percent of what they spent with state and localfunds on AFDC and related programs in fiscalyear 1994.

• States cannot use federal funds to aid familiesthat have received TANF-related assistance for acumulative total of five years, although states canset time limits under five years, and they canexempt up to 20 percent of their caseload in anyone year from the five-year limit.

• States can lose up to 25 percent of their TANFallotment for failure to meet work-participationrequirements, which increase annually so that by2002, 50 percent of all recipients from all welfarefamilies should be working at least 30 hours perweek, and 90 percent of all recipients from two-parent welfare families should be working atleast 35 hours per week. (Although most statesare rhetorically emphasizing “work first,” mosthave used a waiver in the law that allows them toopt out of the workfare mandate. For example,Massachusetts “has excused 82 percent of the72,500 families on the rolls from the 60-dayworkfare deadline” [Whitman 1998, p. 26]. Manystates are also diverting families from multiyearcash assistance to part-time work and to short-term service provision, such as job-search pro-grams and education.)

• States can be penalized for failure to participatein the Income and Eligibility Verification Systemand for failures to submit required reports tofederal officials.

• States must comply with paternity establishmentand child-support enforcement rules set by thefederal government, and states can be penalizedup to 5 percent of their TANF grant for failure todeduct a minimum of 25 percent from a family’scash assistance when the recipient does not co-operate with child-support rules without goodcause.

• States must enact laws to suspend the driver’s,

professional, occupational, and recreational li-censes of individuals whose child-support pay-ments are in arrears.

• Pursuant to welfare reform as well as immigra-tion reform, Social Security numbers must berecorded on a variety of official documents, in-cluding driver’s licenses and birth certificates.

• States must maintain assistance when parentscannot locate child care for a child under age six.

• States must use at least 4 percent of their com-bined grant funds under TANF and the newlyconsolidated Child Care and Development BlockGrant to increase the availability and improve thequality of child care.

• States must implement Electronic Benefit Trans-fer programs by October 1, 2002, unless waivedby the U.S. Department of Agriculture.

• Beginning in fiscal year 1998, a mandatory for-mula grant was added to the Maternal and ChildHealth Block Grant to provide $50 million annu-ally to states to operate abstinence-only sex-education programs.

• No more than 15 percent of a state’s TANF grantcan be used for administration.

If “welfare reform” is a vividillustration of “devolution,” thenthe states may find themselves inthe proverbial predicament of the

dog who finally catches thespeeding car. The states won newdiscretion and flexibility, but they

also inherited onerous andprescriptive requirements.

If “welfare reform” is a vivid illustration of “dev-olution,” then the states may find themselves in theproverbial predicament of the dog who finally catchesthe speeding car. The states won new discretion andflexibility to design, fund, and operate welfare pro-grams, but they also inherited onerous and prescriptiverequirements, especially in work participation and childsupport (although the 43 states that had alreadyreceived waivers to reform welfare may not have tocomply with every aspect of the new act for five years).

May/June 1998 New England Economic Review 23

In effect, the federal government is requiringthe states to accomplish what the federal govern-ment could not accomplish during its sixty-someyears of co-funding, overseeing, and tinkering withwelfare as we knew it. When states fail to meet thework-participation requirements, or to meet themadequately with upwardly mobile jobs that do notdisplace currently working low-income persons, theywill likely be criticized as being unfit for devolution.

Several other aspects of “welfare reform” areproblematic too. The anticipated reduction of $54billion in federal welfare spending during the sixyears following reform and the caps on federal fund-ing for Temporary Assistance for Needy Families andthe Social Services Block Grant could prove costly forstates during recessions. Local governments will bethe most fiscally vulnerable, because persons who fallthrough the safety net will land on their doorsteps. Inaddition, the restriction requiring state use of federalfunds for abstinence-only sex education is an extraor-dinary dismissal of state prerogatives, although it isconsistent with the “family values” advocated bymany members of Congress today. It is also consistent,for example, with the Defense of Marriage Act of 1996,which, while relieving the states of a constitutionalobligation to recognize same-sex marriages solem-nized out of state, also establishes a national definitionof marriage as a heterosexual union, for purposes offederal benefits. The power to define marriage, includ-ing marriage for federal-benefit purposes, had be-longed exclusively to the states since 1776.

In 1997, President Clinton and congressional Re-publicans concluded a budget agreement that includesa new program to provide $24 billion to the states overfive years to cover uninsured children for health care.The law gives states implementation choices, such asbringing uninsured children into Medicaid, into astate’s largest health maintenance organization, into astate’s health insurance program for public employ-ees, or into programs comparable to health insurancefor federal employees. It is too early to assess thisprogram’s devolution success, although one newsreporter has already called it

the devolution dilemma. The good news is that thefive-year budget deal the President signed last Tuesdaygives states a free hand and $24 billion in new money toextend medical coverage to millions of uninsured chil-dren. The bad news: State officials may be unprepared toassume control of the largest expansion in health careservices in 30 years (Jeter 1997, p. 29).

One can point to other hints of devolution-likeelements in congressional policymaking, but, in the final

analysis, there have been few substantive turnbacks offunctions from the federal government to the states.

Presidential Federalism

President Clinton came into office two years be-fore the congressional Republicans, also promising torestore state powers: “I intend to do my very best,” hetold the governors, “to be faithful to the lessons that Ilearned as a governor—that most of what you doought to be done by you and not by us” (Broder 1993).Yet, conclusions similar to those for the Congresspertain to Clinton’s initiatives, beginning with his“reinventing government” campaign headed by VicePresident Al Gore under the name National Perfor-mance Review (NPR). Although NPR addressed awide range of management issues primarily withinthe federal government, the NPR report, issued inSeptember 1993, contained the following intergovern-mental recommendations.

1. Create flexibility and encourage innovation by design-ing a bottom-up solution to the problem of grantproliferation and its accompanying red tape. Also,support the pending proposal for Federal-State Flexi-bility Grants that has been developed by the NationalGovernors’ Association and the National Conferenceof State Legislatures. Establish a Cabinet-level Enter-prise Board to oversee NEW initiatives in communityimprovement.

2. Issue an Executive Order addressing the problems ofunfunded federal mandates and regulatory relief andauthorize Cabinet Secretaries and agency heads toobtain selective relief from regulation or mandates inprograms they oversee.

3. Modify OMB Circular A-87, “Cost Principles for Stateand Local Governments,” to provide a fixed fee-for-service option in lieu of costly reimbursement proce-dures covering actual administrative costs of grantdisbursements.

4. Simplify OMB’s requirements to prepare multiplegrant compliance certifications by allowing state andlocal governments to submit a single certification to asingle point of contact in the federal government.

5. Modify OMB Circular A-102, “Grants and CooperativeAgreements to State and Local Governments,” toincrease the dollar threshold for small purchases bylocal governments from $25,000 to $100,000.

6. Reinvent the Advisory Commission on Intergovern-mental Affairs [sic] (ACIR) and charge it with theresponsibility for continuous improvement in federal,state and local partnership and intergovernmentalservice delivery. Direct the ACIR to identify opportu-nities to improve intergovernmental service deliveryand develop a set of benchmarks (NPR 1993, p. 167).

May/June 1998 New England Economic Review24

Many other recommendations affecting intergovern-mental relations were embedded in other sections ofthe NPR report.

NPR’s recommendations were modest. The firstrecommendation, similar to the block grant proposalsof President Reagan’s New Federalism, called forconsolidating 55 categorical grants worth $12.5 billioninto broader “flexibility grants.” This recommenda-tion has not been enacted by the Congress, but Presi-dent Clinton did quickly endorse “bottom-up” consol-idation plans submitted by Indiana and West Virginiato coordinate 199 federal grants affecting children andfamilies, thus setting a precedent for other states tofollow suit. When the new Republican Congressfloated proposals for larger block grants that wouldinclude major entitlement programs, President Clin-ton opposed most of the proposals, arguing that theseprograms need a significant, continuing federal role.The Republicans’ major block grant proposals coveredcash welfare benefits (consolidating seven programs),child welfare and abuse (consolidating 38 programs),child care (45 program consolidations), food and nu-trition (10 consolidations), housing (27 consolida-tions), health (22 consolidations), Medicaid, employ-ment and training (154 consolidations), social services(33 consolidations), and law enforcement (12 consoli-dations). The first result was a mouse: a small blockgrant, approved in 1995, which authorized $503 mil-lion for local law-enforcement.

Federal categorical grants continued to proliferateduring the Clinton Presidency, reaching more than 640by 1996. Categoricals procreate not only because theCongress is reluctant to channel funds to states throughmore discretionary block grants, but also because smallgrants provide a means to satisfy various interest-grouppressures without large federal expenditures.

Pursuant to NPR’s first recommendation, the Ad-ministration established a Community EnterpriseBoard in 1993, chaired by the Vice President. Amongother things, the Board coordinates the Administra-tion’s implementation of the Local Empowerment andFlexibility Act of 1994 through 31 empowermentzones, two supplemental empowerment zones, and 95enterprise communities, four of which are enhancedenterprise zones. This act is one of the economicdevelopment benefits the Administration was able toobtain for local governments, and it does providesignificant flexibility for local governments (Lieb-schutz 1995). “Designated zones and communitiesreceive tax benefits and flexible grants and are entitledto apply for waivers of certain Federal regulations; theunderlying principle of the program is that communi-

ties know best how to solve their own problems butmay lack the necessary resources” (Economic Report ofthe President 1998, pp. 34–35).

Considerable progress has been made on thesecond through fifth recommendations (Galston andTibbetts 1994), in part because implementation ofthese primarily administrative recommendations re-quires little if any congressional consent. In the pro-cess, the Administration has also given the regionaloffices of many federal agencies more decision-makingauthority. This has especially been the case for the U.S.Department of Housing and Urban Development,which is so important to the President’s relations withthe country’s mayors and other local officials.

Regarding NPR’s last recommendation, the Ad-ministration initially endeavored to reinvigorateACIR, but the Commission was de-funded in 1996.The President declined to defend it, primarily becauseACIR’s staff had produced a preliminary report (re-quired by the Unfunded Mandates Reform Act) that

Categorical grants proliferate notonly because the Congress isreluctant to channel funds to

states through more discretionaryblock grants, but also because

small grants provide a means tosatisfy various interest-group

pressures without large federalexpenditures.

recommended reductions and terminations of a num-ber of federal mandates. The report sparked a politicalfirestorm; more than 300 interest groups, many repre-senting constituencies important to President Clinton,lined up against it; and the Commission killed thereport in a partisan vote (McDowell 1997).

The Clinton Administration has also sought tocreate more flexibility in federal-state relations byestablishing “performance partnerships,” wherebystates develop performance measures for implement-ing federal programs and rules; in return, they aregiven more flexibility by federal agencies to meet theirperformance objectives. The most developed perfor-

May/June 1998 New England Economic Review 25

mance partnership was forged with Oregon, whichconducted a six-year statewide initiative to identifybenchmarks against which to measure its policyprogress. The benchmarks also apply to performance-based contracts between state and federal agenciesunder conditions that eliminate or relax the require-ments attached to multiple categorical grants (Goshko1995).

The EPA has sought to build a “National Envi-ronmental Performance Partnership System” to givestates more voice in program decision-making and, inthe case of states meeting performance objectives,reduce EPA’s regulatory oversight. The EPA has alsoasked the Congress to enact performance partnershipgrants that would enable states to consolidate multipleEPA grants and use them more flexibly (U.S. GeneralAccounting Office 1996). The General Accounting Of-fice noted in early 1996 that:

the historically poor EPA-state relationship has im-proved, but it continues to be strained, and programimplementation suffers as a result. . . . EPA has takenpositive, though tentative, steps toward improving itsrelationship with the states, in particular trying to pro-vide the states with the flexibility to achieve cost efficien-cies and to address the states’ priorities. However, one ofthe root causes of the agency’s past problems—a pre-scriptive, media-based legislative framework—remainsfirmly in place (U.S. General Accounting Office 1996, pp.3–4).

Under the Government Performance and ResultsAct of 1993, the Administration also launched pilotprojects to develop performance measures in selectedunits in all Cabinet departments and a number ofindependent agencies. Each unit began to develop astrategic plan, define its mission, set goals, and estab-lish performance standards for measuring results.These pilot projects were then extended to a numberof federal grant programs, particularly Goals 2000 foreducation.

The second phase of reinventing government wasinitiated in President Clinton’s fiscal year 1996 budgetproposals, which contained sweeping intergovern-mental reforms involving greater scope for programdevolution to the states, program terminations, grantconsolidations, privatization, and more performancepartnerships. The Administration proposed, for exam-ple, to consolidate many U.S. Department of Trans-portation grants into three block-like programs: adiscretionary grant, a more unified allocation of trans-portation funds to state and local governments, andthe establishment of state infrastructure banks. TheAdministration also proposed to consolidate sixty

programs managed by the U.S. Department of Hous-ing and Urban Development into three more flexibleand performance-based programs for community eco-nomic opportunity, affordable housing, and modern-ization of public housing. Altogether, the Administra-tion proposed to consolidate 271 grant programs.These more sweeping proposals of the second wave of“reinventing government” appear to have been moti-vated in part by a belief that the Republican Congresswould be more receptive to such consolidations thanprevious Democratic Congresses and in part as acounterplay to Republican block grant proposals.

The most significant block grant signed by Presi-dent Clinton, of course, was Temporary Assistance forNeedy Families. After his own welfare-reform initia-tive failed to gain steam, President Clinton pledged tothe governors to expedite administrative waivers forwelfare experimentation and to make changes in food

Aside from the “devolutionary”elements of the TANF block

grant, very little in thePresident’s rhetoric and that ofmembers of Congress suggests

that devolution was afundamental principle or primary

objective of welfare reform.

stamp rules. From the states’ perspective, however,the waiver process moved too slowly. The averagewait for approval was 210 days (Milbank and McGin-ley 1996). Initially, President Clinton vetoed a Repub-lican welfare-reform measure in December 1995 thatincluded sizable reductions in Medicaid spending, aswell as a follow-up measure in January 1996. Aftergaining some concessions from the Congress, espe-cially the removal of Medicaid from the proposal,President Clinton signed the Personal Responsibilityand Work Opportunity Reconciliation Act of 1996.

Aside from the “devolutionary” elements ofTANF, very little in the President’s rhetoric and that ofmembers of the Congress suggests that devolutionwas a fundamental principle or primary objective ofwelfare reform. Instead, reform seemed far more mo-tivated by election-year political advantage. The

May/June 1998 New England Economic Review26

stigma attached to welfare and the weakened politicalclout of interest-group advocates for welfare madeAFDC a much easier candidate for conversion to ablock grant than any of the other block grants pro-posed by the Republicans. Indeed, opposing the otherblock grant proposals contributed to President Clin-ton’s increasingly successful reelection strategy ofportraying the Republicans as radical opponents ofbenign government and as insensitive to the needs ofchildren, women, senior citizens, minorities, and thepoor. As a Democrat, moreover, President Clintonbelieves in most of the programs targeted for reformby the Republicans; consequently, his Democraticprinciples took precedence over his federalism princi-ples. While being tough on welfare, Clinton could notappear to be a party to what columnist Carl Rowansaw as the consequences of the Republican revolution:“ ‘States rights’ is about to reign again in America, andmillions of Americans are going to be hurt by it”(quoted in Ehrenhalt 1995).

President Clinton has also supported regulatoryand mandate relief for state and local governments.Initially he responded administratively to state andlocal concerns. On September 30, 1993, he issuedExecutive Order 12866, “Regulatory Planning andReview.” A major purpose of this executive order is toease the federal regulatory burden on state, local, andtribal governments by planning the issuance of regu-lations more carefully; coordinating and streamliningregulations; consulting closely and regularly withstate, local, and tribal officials; and limiting the pro-mulgation of regulations only to those “required bylaw, or [that] are made necessary by compelling publicneed, such as material failures of private markets toprotect or improve the health and safety of the public,the environment, or the well-being of the Americanpeople.” Although this clause leaves open a wide fieldof action for federal regulators, the basic intent of theexecutive order is to limit the impacts of federalmandates on states and localities to what is reallyrequired by law or to what is reasonably necessary.

On October 26, 1993, President Clinton issuedanother executive order (E.O. 12875) entitled “Enhanc-ing the Intergovernmental Partnership.” This orderwas a preemptive response to “National UnfundedMandates Day,” a protest held by state and localofficials on October 27. The executive order requiresfederal agencies to expedite and streamline their waiv-er-application processes for state, local, and tribalgovernments. “Furthermore, E.O. 12875 declares thatno agency shall promulgate any regulation not re-quired by statute or impose a mandate on a state,

local, or tribal government, unless: (1) funds necessaryto pay the direct costs incurred by the state, local, ortribal government for complying with the mandate areprovided by the federal government, and (2) theagency provides, before the formal promulgation ofregulations, a description of consultations with state,local, and tribal representatives, the nature of theirconcerns, and the agency’s position supporting theneed to issue the regulation containing the mandate”(Galston and Tibbetts 1994, p. 28). Subsequently, Pres-ident Clinton supported and signed the UnfundedMandates Reform Act of 1995.

Nevertheless, despite President Clinton’s sympa-thy for state and local mandate concerns, he hassigned many bills containing mandates, such as theHandgun Violence Prevention Act of 1993 (the BradyBill), the Family and Medical Leave Act of 1993, theFull Faith and Credit for Child Support Order Act of1994, the Multiethnic Placement Act of 1994, the SafeDrinking Water Act of 1996, the Health InsurancePortability and Accountability Act of 1996, and thefederal Megan’s Law of 1997. In examples in otherfields, the Student Loan Reform Act of 1993 requiresstates where institutions of higher education have astudent-loan default rate of more than 20 percent toassume partial responsibility for the cost of the de-faults. The 1993 Reform Amendments to the Hatch Actprohibit any elected state official from making ortransmitting, to any officer or employee of a federalagency, any verbal or written recommendation orstatement about an employee or applicant, except forrecommendations based solely on job performance.The Comprehensive Child Health Immunization Actof 1993 requires state Medicaid programs to coversome recommended childhood vaccines and to reim-burse providers for administering vaccines. The Na-tional Child Protection Act of 1994 requires authorizedstate criminal justice agencies to report child-abusecrime information to the national criminal back-ground-check system.

The special case of crime. President Clinton has alsocontributed to the extraordinary federalization ofcriminal law that has occurred since the late 1960s.Thomas Jefferson, in drafting the Kentucky Resolu-tions of 1798 in response to the Federalists’ Alien andSedition Acts, noted “that the Constitution of theUnited States . . . delegated to Congress a power topunish treason, counterfeiting the securities and cur-rent coin of the United States, piracies and feloniescommitted on the high seas, and offenses against thelaws of nations, and no other crimes whatever.” Inshort, the U.S. Constitution specifies only four crimi-

May/June 1998 New England Economic Review 27

nal offenses punishable by the federal government.Today, the federal government can punish individualsfor more than 3,000 criminal offenses and invoke thedeath penalty for more than 50 offenses, includingfederal criminal offenses committed in the 13 states thatprohibit capital punishment. This federalization of crim-inal law is a matter of growing alarm to civil libertarians.

Although even the Administration’s 1992 policyblueprint Mandate for Change identified crime as apolicy area in which “no federal role is justified,”President Clinton strongly advocated The ViolentCrime Control and Law Enforcement Act of 1994. Thisact provides $39.3 billion to state and local govern-ments over a six-year period for prisons, 100,000 “copson the beat,” drug treatment in prisons, youth-crimeprevention, and other programs. The law requires, asa condition of full funding, that states, among other

The U.S. Constitution specifiesonly four criminal offensespunishable by the federal

government. Today, the federalgovernment can punish

individuals for more than 3,000criminal offenses.

things, enact a truth-in-sentencing statute, assure thatviolent offenders serve a substantial portion of theirsentences, recognize victims’ rights, and register vio-lent sexual offenders upon their release from prisonand being placed on parole or on supervised release.A current address for the offender must be provided toa designated law-enforcement agency. The statute alsorequires clerks of state courts to report to the InternalRevenue Service information on persons who postcash bonds in excess of $10,000.

President Clinton pressed for this legislation, inpart, because being “tough on crime” has becomepolitically popular for Presidents and members ofCongress from both political parties. Republicans areno less eager than Democrats to intrude upon tradi-tional state and local prerogatives when it comes to“fighting crime.” President Clinton also advocatedthis legislation because it was a backdoor mechanismfor channeling more federal money to local govern-ments, especially big cities—a prime Democratic con-

stituency base. Congressional Republicans attackedthis strategy and succeeded in reducing the final flowof money to local governments.

In summary, flexibility for state and local admin-istration of federal programs, not devolution of federalprograms, has been the hallmark of the Clinton Ad-ministration. Although President Clinton has sup-ported and implemented devolution-like measuresthat have provided greater administrative flexibilityand regulatory relief for state and local governments,the President, like the Congress, has not sent “author-ity cascading to lower levels of government” nor hashe acted to constrain significantly his ability to sup-port or implement virtually any politically desirablemandate. Even Clinton’s first use of the line-item vetoin 1997 was anti-devolutionary because one of thethree items struck from the budget was a waiver thathad been granted by the U.S. Department of Healthand Human Services to New York State to tax health-care providers to help cover Medicaid costs. One ofthe ironies, then, of the Clinton Presidency is thatwhen the Congress was in Democratic hands, he couldnot push his intergovernmental reform agenda farenough because of old-guard opposition. Now thatthe Congress is in Republican hands, he cannot go farenough to support the Republicans’ more ambitiousdevolution proposals.

Judicial Federalism

The United States Supreme Court will play amajor role in any restoration of state powers andrebalancing of the federal system, because the acces-sion of power by the federal government during thetwentieth century required judicial deference to broadinterpretations of the Congress’s constitutionally enu-merated powers as well as activist judicial interpreta-tions of the U.S. Constitution, especially the Four-teenth Amendment and the Bill of Rights.

This is one reason why issues of fidelity to con-stitutional federalism and of constitutional amend-ments to rebalance federal-state power have surfacedin the Congress and in the states. A proposed Enu-merated Powers Act, for example, would require the“Congress to specify the source of authority under theUnited States Constitution for the enactment of laws.”A proposed Tenth Amendment Enforcement Act isintended to have a similar effect. Neither of these billsis likely to be enacted, although the U.S. House ofRepresentatives did adopt a rule in early 1997 requir-ing committee reports on bills to cite each bill’sconstitutional authority. State legislators and gover-

May/June 1998 New England Economic Review28

nors are also debating four statutory and constitu-tional proposals developed at their 1995 FederalismSummit:

• A federalism act to enhance political safeguards offederalism and give states a more effective voice incongressional deliberations.

• A mechanism to provide the people of the states,through their legislatures, the power to require Con-gress to reconsider laws, specific provisions of laws, orregulations that interfere with state authority.

• A mechanism that would allow the states to proposespecific amendments to the U.S. Constitution subject toratification by the United States Congress.

• Statutory remedies and/or constitutional reforms toaddress the problems of conditions attached to [feder-al] spending grants, regulations, and mandates (Coun-cil of State Governments 1996, p. 75).

Portions of these statutory proposals are likely tobe enacted by the Congress, but the proposed consti-tutional amendments are unlikely to pass muster inthe Congress and the state legislatures. Indeed, inAugust 1997, the National Conference of State Legis-latures failed to achieve a three-fourths vote to en-dorse the above constitutional proposals. Many legis-lators believe that the Congress and the U.S. SupremeCourt are moving in a state-friendly direction and thatconstitutional change is not now necessary.

Consequently, in late 1997, state legislators andgovernors adopted an 11-point plan for statutoryfederalism reform: (1) declare and justify the constitu-tionality of legislation enacted by the Congress; (2)limit and clarify federal preemption of state law; (3)prohibit federal conscription and coercion of stategovernments; (4) use points of order on the floors ofthe U.S. House and U.S. Senate to help protect statesagainst excessive mandates and preemptions; (5) con-solidate many more categorical grants-in-aid intoblock grants; (6) protect state laws and procedures inexpending federal funds; (7) prohibit conditions offederal aid not germane to aid purposes; (8) clarify theintent of the Unfunded Mandates Reform Act; (9)require congressional and executive federalism-im-pact statements for proposed bills and regulations;(10) streamline federal regulatory procedures; and (11)simplify federal financial reporting requirements.

Given the reluctance of state leaders to redressstate powers through constitutional change, questionsof constitutional federalism and the parameters ofcongressional power will be decided principally bythe U.S. Supreme Court and secondarily by the Con-gress. In the 1990s, the Court has turned away from its1985 ruling in Garcia v. San Antonio Metropolitan Tran-

sit Authority, in which the Court sought to abandon itsrole as umpire of the federal system by holding thatstates could not seek Tenth Amendment redress fromthe Court for federal encroachments upon state pow-ers. Instead, the states, like interest groups, wouldhave to rely on the political safeguards of federalismto defend their powers in the national political arena.

Although the Court has not overturned Garcia, ithas increasingly ignored it by seeking to limit federalauthority and to protect or restore state authority insix basic ways (compare, Tolley and Wallin 1995).

State autonomy. First, Justice Sandra DayO’Connor has advanced a “state autonomy” defenseof federalism (Merritt 1988) based on the TenthAmendment and the U.S. Constitution’s republicanguarantee clause (Article IV, Section 4). In this view,the federal government cannot deprive citizens of astate of their essential democratic right to make fun-damental decisions about their state polity. Specifi-cally, Justice O’Connor advanced this argument inGregory v. Ashcroft (1991), which upheld a provision of

The U.S. Supreme Court will playa major role in any restoration ofstate powers and rebalancing ofthe federal system, just as the

accession of federal governmentpower in the twentieth century

required activist judicialinterpretations of the

U.S. Constitution.

the Missouri Constitution requiring state judges toretire at age 70 despite the federal Age Discriminationin Employment Act. What is especially interesting inthis age of individual rights is Justice O’Connor’sreconceptualization of the Tenth Amendment as notprotecting traditional states’ rights but, rather, asprotecting the dual citizenship rights of state resi-dents. As Justice O’Connor has argued repeatedly:

the Constitution divides authority between federal andstate governments for the protection of individuals. Statesovereignty is not just an end in itself: “Rather federalismsecures to citizens the liberties that derive from the diffusionof sovereign power. . .” (Gregory v. Ashcroft 1991).

May/June 1998 New England Economic Review 29

Nevertheless, this doctrine is still narrow, lacks major-ity support, and is weakened by Justice O’Connor’sconcession that the Congress could have extended thereach of the Age Discrimination in Employment Act tocover state judges if it had “expressly” declared itsintent to restrict state governments.

Plain statements. A second judicial strategy is torequire “express” or “plain statements” in legislationof the Congress’s intent to preempt state authority (forexample, Gregory v. Ashcroft 1991); to abrogate states’Eleventh Amendment immunity (Atascadero State Hos-pital v. Scanlon 1985); to permit civil-rights suitsagainst states under 42 U.S.C. Section 1983 (Will v.Michigan Department of State Police 1989); and to attachconditions to grants-in-aid (Suter v. Artist 1992). Theserules are also limited, however, because this strategyallows the Congress to expand its power simply by“expressly” stating its intent to do so.

Federal conscription. A third strategy is to prohibitthe Congress from “conscripting” or “commandeer-ing” state officials to carry out federal laws. Thisdoctrine was best articulated in New York v. UnitedStates (1992), wherein the Court declared unconstitu-tional the “take title” provision of the federal Low-Level Radioactive Waste Disposal Act. This statuterequires states to provide for the proper disposal oflow-level radioactive wastes within their own bordersor in other states through interstate compacts. Failureto do so by 1996 would have required state govern-ments to take title to such wastes and be liable forharms caused by them. The Court held that theCongress violated the Tenth Amendment by compel-ling state governments to enact such laws and regu-lations.

Justice O’Connor’s state autonomy argument alsofigured in this decision, but with a new twist. The“take title” provision had been enacted as part of a setof compromises negotiated between the governorsand the Congress. Hence, the National Governors’Association vetoed a proposal for the State and LocalLegal Center to file a state-supportive amicus brief inNew York. The Council of State Governments, there-fore, filed, for the first time, an amicus brief before theCourt. The majority opinion held that even though the“take title” provision was an example of cooperativefederalism, the governors lacked authority under theTenth Amendment to surrender state sovereignty tothe Congress and thereby sell out the citizenshiprights of state taxpayers.

This anti-conscription doctrine was reaffirmed inPrintz v. United States (1997) wherein the Court struckdown the provision in the Brady Handgun Control

Act that required local law-enforcement officers toconduct background checks of handgun buyers. Jus-tice Antonin Scalia delivered an impassioned opinionupholding “dual sovereignty” in the federal systemand protecting the sovereignty of the states againstcongressional encroachments through liberal interpre-tations of the “necessary and proper” clause of Article1, Section 8 of the federal Constitution.

Limits on the commerce power. A new strategy notseen since 1936 emerged in United States v. Lopez(1995), in which the Court struck down the Gun-FreeSchool Zones Act of 1990 as an unconstitutional exer-cise of the Congress’s interstate commerce power.During oral argument, the solicitor general was askedto identify a human activity that could not be broughtunder congressional regulation through generous in-terpretation of the commerce clause. The solicitorgeneral could not identify a single activity. Reversingsixty years of precedent, the majority opined: “Touphold the Government’s contentions here, we wouldhave to pile inference upon inference in a manner thatwould bid fair to convert congressional authorityunder the Commerce Clause to a general police powerof the sort retained by the States.” Although it isunlikely that the Court will roll back congressionaleconomic regulation in significant ways, Lopez sig-naled the Congress that the Court is prepared in alimited way to prohibit regulation that unduly re-stricts state powers in areas not substantially related tointerstate commerce. However, the Congress, uponurging from President Clinton, resurrected the Gun-Free Schools Zones Act through the Omnibus Appro-priations Act of 1996, wherein the Congress sought toestablish a firm legislative history for a substantialconnection between interstate commerce and gunsbrought onto local school grounds. Whether suchcongressional attempts to establish regulatory author-ity in such areas as guns in local schools will, like“plain statements” of preemption, be upheld by theCourt is uncertain, although so long as the five-member majority in Lopez remains on the Court, theSupreme Court is unlikely to overturn or ignore Lopez.

Laboratories of democracy. A fifth strategy is a“laboratories of democracy” view of state powers,derived from Justice Louis D. Brandeis’s famous opin-ion that “a single courageous state may, if its citizenschoose, serve as a laboratory, and try social andeconomic experiments without risk to the rest of thecountry” (New York Ice Co. v. Liebman 1932, p. 311).This view was expressed in a quite different contextthan today when many states were enacting innova-tive and progressive rights and social welfare laws

May/June 1998 New England Economic Review30

while the Congress and especially the Supreme Courtwere often hostile to such initiatives.

This strategy was reflected in Vacco v. Quill (1997)and Washington v. Glucksberg (1997), wherein theCourt declined to recognize physician-assisted suicideas a fundamental right under the Fourteenth Amend-ment to the U.S. Constitution, thus upholding stateprohibitions of physician-assisted suicide. The Courtdid not deny that such a right might exist; it heldinstead that hitherto unrecognized Fourteenth Amend-ment rights must be deeply rooted in the nation’shistory, legal traditions, and moral practices, not in“the policy preferences of the members of this Court.”Hence, the Court reserved to the democratic processesof the fifty states the task of deciding, over the longterm, whether physician-assisted suicide is to be rec-ognized as a fundamental right. There is “no reason tothink the democratic process will not strike the properbalance,” wrote the majority.

A corollary to this strategy is the Court’s endorse-ment of “the new judicial federalism” (Kincaid 1988)in Michigan v. Long (1983), which immunizes from

The U.S. Supreme Court hasreturned to the field to umpire the

constitutional game of federal-state power-balancing with

rulings surprisingly favorable tothe states; however, there has beenno wholesale restoration of state

powers or denigration offederal powers.

federal judicial review state judicial and legislativeexpansions of individual rights protections based on“independent and adequate” state constitutionalgrounds that go beyond rights protections recognizedby the U.S. Supreme Court under the U.S. Constitu-tion. For example, the federal Court has ruled that theright of household privacy and protection againstwarrantless searches and seizures does not cover trashplaced outside one’s home for public collection. TheNew Jersey Supreme Court, however, has ruled thatstate and local police in the Garden State do need,pursuant to the state constitution, a warrant to search

someone’s curbside trash. State high courts have is-sued about 800 such rights-expansive rulings since themid 1970s. Similarly, even though the U.S. SupremeCourt struck down the federal Religious FreedomRestoration Act of 1993 (RFRA) in June 1997, the NewYork State Assembly passed a state RFRA in earlyAugust 1997 in direct reaction to the Court’s rights-restrictive ruling.

Final arbiter of the Constitution. A sixth, relatedstrategy was reflected in City of Boerne v. Flores (1997),which struck down the Religious Freedom RestorationAct. “The power to interpret the Constitution in a caseor controversy remains in the judiciary,” opined themajority, and the Congress has no authority to expandthe scope of the Fourteenth Amendment beyond the“proportionality and congruence” of the problem be-ing addressed by legislation. Many state and localofficials contested RFRA because it required them todemonstrate a “compelling interest” justification forrestrictions on religious freedom. Justice AnthonyKennedy wrote that RFRA was a “considerable intru-sion into the states’ traditional prerogatives and gen-eral authority to regulate for the health and welfare oftheir citizens.” More generally, since the Court’s rul-ing on abortion in Roe v. Wade (1973), the Court hasoften curbed, reduced, or declined to expand individ-ual rights, especially in criminal proceedings, thatexcessively intrude or impose upon state and localgovernments.

The Court has acted in other ways to protect thestates by showing greater deference to state courtproceedings under the abstention doctrine, by re-stricting the authority of federal judges to entertainhabeas corpus claims from state prisoners, by uphold-ing various types of state economic regulation, andby loosening the constraints of some federal courtorders.

In summary, the U.S. Supreme Court has re-turned to the field to umpire the constitutional gameof federal-state power-balancing with rulings surpris-ingly favorable to the states; however, there has beenno wholesale restoration of state powers or denigra-tion of federal powers, nor is the Court in a politicalposition today to make the kind of swift “switch intime that saved nine” that it made in 1937. Instead, theCourt has breathed new life into constitutional feder-alism through emerging doctrines pregnant with im-plications for future umpiring more supportive ofstate sovereignty. The cases conveying these doctrines,however, are of limited scope, uncertain precedent,and mixed motives and are supported only by 5-to-4and 6-to-3 majorities. If only one justice in the majority

May/June 1998 New England Economic Review 31

of most of these federalism cases—especially WilliamRehnquist, Sandra Day O’Connor, or ClarenceThomas—were to be replaced by a justice with theminority views expressed in these cases, the umpiremight retire from the field again or, at best, oversee thegame from the bleachers.

Much the same can be said of the panoply ofcongressional and presidential actions taken thus farin the 1990s. No substantial shift of power to the stateshas been executed, and the most oft-cited example ofdevolution, namely, “welfare reform,” was driven asmuch or more by the political advantages of endingwelfare as we knew it as by principled arguments forrestoring federalism as we knew it. These baby stepstoward devolution are potentially significant but cur-rently limited because the debate over federalism ispart of a larger debate over the very nature and futureof American society. The United States is perhaps onthe threshold of a paradigm shift comparable to thoseof the Civil War, Progressive, and New Deal–GreatSociety eras. Consequently, emerging forces favoringrebalancing necessarily collide with obstacles to resto-ration of state powers.

Forces Against Restoring State Powers

A significant obstacle to “devolution” is publicopinion. Even though there is a generalized publicpreference for local and state performance of publicfunctions, just as there is a generalized preference fora balanced federal budget, each particular “devolu-tion” proposal, like each particular item slated for abudget cut, encounters opposition from particularbeneficiaries anxious to maintain the status quo whilebeing willing to urge devolution of someone else’sprograms. Even in welfare reform, for example, de-spite negative public attitudes toward welfare, half ofthe respondents to a survey by the Kaiser/HarvardProgram (1995) supported granting more flexibility tostates to experiment with welfare programs, but alsoagreed that the “federal government has to set guide-lines when it gives money to states . . . in order toassure that the states will treat everyone fairly and dothe right thing for poor people.”

The devolution of any function or shift of anypower to the states poses risks and uncertainties forthe interests that benefit from the maintenance offederal control. Opposition is likely to stiffen, more-over, when proposals to shift powers to the states arelinked to calls for policy reform, budget cuts, deficitreduction, and downsizing (Weaver 1996).

Devolution is also frustrated because, eventhough the Republican majority in the Congress hastended to express the most support for devolution,Republicans are often more committed to principles ofindividualism than of federalism. As U.S. HouseSpeaker Newt Gingrich has expressed it:

The last sixty years has seen so much centralizationin Washington that at this point the best we can do is startby shifting power back to the state capitals. Power in fiftydifferent cities is better than power centralized in one city.

Yet our ultimate goal is to move power even beyondthe state capitals. Many governors and county commis-sioners are deeply suspicious of state governments andwould prefer bloc [sic] grants and programs that comestraight to them. . . .

However, much as I sympathize with both state andlocal governments, what we really want to do is todevolve power all the way out of government and back toworking American families. . . . Republicans envision adecentralized America in which responsibility is returnedto the individual (Gingrich 1995, pp. 104–105).

Principles of individualism and federalism frequentlycollide with each other because the promotion of

Many policy functions thatenhance federal power and reducestate powers, such as civil rights,environmental protection, middle-

class entitlements, and crimecontrol, enjoy broad and deep

public support.

individualism often requires constraints on state andlocal governments as well as the federal government.As a result, for example, on grounds of individualliberty, many Republicans as well as some Democratssupport federal preemption of state authority to re-regulate sectors of the economy deregulated by theCongress.

Politically, the devolution agenda that was setforth by the Republicans encountered numerous ob-stacles that blocked enactment of nearly the entireagenda. Proposals to devolve various health and so-cial welfare programs, for example, jeopardize theentitlement status of programs for both states andindividuals. Many state officials have sought to main-

May/June 1998 New England Economic Review32

tain state entitlements to federal support in order toguarantee adequate and predictable federal aid whilemany congressional Republicans have sought to re-duce expenditures. In turn, numerous advocacy andclientele groups have sought to maintain individualentitlements in order to prevent states from denying orreducing benefits to currently eligible citizens. Entitle-ment advocates have also expressed concern, for exam-ple, “that removing individual entitlement in the Med-icaid program would lead states to concentrate resourceson popular and powerful clienteles, such as the frailelderly in nursing homes” (Weaver 1996, p. 49).

Conflicts over funding formulas have also frus-trated devolution. “Formula fights” are among themost vigorous in Congress, and they pit states andregions against each other. In essence, no state orregion wants to be a net loser in devolution. States,however, have generally been united in seeking fed-eral guarantees of increased funding during reces-sions, while many congressional Republicans haveopposed such guarantees. Some devolution proposalshave also pitted states against their local governments,as local officials seek to defend their prerogatives andshield their jurisdictions from negative fallouts. Dev-olution proposals have sparked vigorous debates aswell, over maintenance-of-effort requirements, main-tenance of federal standards, mandates, and account-ability. In the end, many Republicans as well asDemocrats are unwilling to let go of policy objectivesthat might be defeated by devolution to states havingdifferent policy objectives.

Additionally, many policy functions that enhancefederal power and reduce state powers, such as civilrights, environmental protection, middle-class entitle-ments, and crime control, enjoy broad and deep publicsupport. Furthermore, issues are often framed in waysthat make it politically impossible to refuse federalaction. How many members of Congress, for instance,could publicly oppose a federal Megan’s Law? How-ever strongly President Reagan wished to shrink fed-eral power, he could hardly refuse to sign the billsupported by Mothers Against Drunk Drivers thatmade it a condition of federal highway aid that statesincrease the minimum drinking age to 21 in order toreduce teenage highway fatalities and clean up “bloodborders.” This political dynamic is currently reflectedin President Clinton’s proposal, thus far approved bythe U.S. Senate, to establish—as a condition of federalhighway aid—a national blood-alcohol standard of0.08 percent for drunk driving, in contrast to the 0.10standard in force in 33 states. Proponents of theproposed standard regard it as good public policy;

many opponents argue that it intrudes upon thehistoric constitutional prerogatives of the states.Hence, the most viable candidates for devolution arepowers and functions that do not enjoy strong publicsupport as well as functions of a more administrativenature that are invisible to the general public.

Similarly, a number of federal policy functionsare highly beneficial for members of Congress andserve to benefit the constituency interests of the fed-eral government generally. Thus, for example, “incontrast to the bipartisan embrace of welfare-reformblock grants, there has been deafening bipartisandisinterest in major devolution of infrastructure activ-ities, such as highways, to the states, despite theexistence of a strong performance rationale and incen-tives for state leadership of these activities” (Posnerand Wrightson 1996, p. 107).

In addition, there continue to be strong incentivesto use three key tools of coercive federalism: man-dates, conditions of aid, and preemption (Kincaid

Conditions of aid remain one ofthe most open flanks for federal

encroachments upon state powersand for federal accessions ofpowers wholly outside theCongress’s constitutionally

enumerated powers. The singlegreatest devourer of state-local

powers, however, is preemption.

1993). Unfunded and underfunded mandates are es-pecially tempting because they allow the Congressand the White House to claim credit for “feel good”policies and to respond to interest groups while “de-volving” the costs of those policies to state and localgovernments. Although the Unfunded Mandates Re-form Act appears to be having some limiting effects onthe enactment of new unfunded mandates (NationalGovernors’ Association 1996), the act has many ex-emptions, its point of order can be overridden by asimple majority vote in either house of Congress, andit is no barrier to politically popular or compellingmandates such as the bipartisan support for increasingthe federal minimum wage in 1996, which will cost

May/June 1998 New England Economic Review 33

state and local taxpayers about $1.3 billion during thenext five years, according to the Congressional BudgetOffice. Furthermore, Republicans have proved to beno less eager than Democrats to mandate their policypreferences nationwide now that they are in power onCapitol Hill.

Conditions of aid remain one of the most openflanks for federal encroachments upon state powersand for federal accessions of powers wholly outsidethe Congress’s constitutionally enumerated powers.The U.S. Supreme Court has upheld such conditionson the ground that federal aid is voluntarily acceptedby the states (South Dakota v. Dole 1987), even though,as a practical political and fiscal matter, states cannotopt out of the large grant-in-aid programs, such asMedicaid and highways. In the case of highways,moreover, if a state were to decline federal highwayaid in order to avoid compliance with objectionableconditions, the state’s residents would, presumably,still be required to pay the federal motor fuels taxwhile deriving no benefit from it. Consequently, the

A major structural change in thefederal system has been the shift

in federal policymaking fromplaces to persons and a

reconceptualization of the federalunion as one constituted by

individuals, not sovereign states.

large federal-aid programs like highways and Medic-aid become vehicles for expanding federal power.

The single greatest devourer of state-local pow-ers, however, is preemption, namely, federal displace-ment of state law under the supremacy clause of theU.S. Constitution (Article VI). More than 53 percent ofall explicit preemption statutes enacted by the Con-gress since 1789 have been enacted only since 1969(ACIR 1992). The new Republican majority in Con-gress, like previous Democratic majorities, has ahealthy appetite for preemption (see, for example,Shenk 1997). Indeed, Republican efforts to preemptstate product liability, food and drug labeling, Internettaxation, and medical malpractice laws, among others,have made a number of congressional Democratsardent states’ rights advocates.

Pressure in the federal system for preemption isenormous, primarily for economic reasons. For one,many businesses engaged in interstate commercewould rather be regulated by one 500-pound gorilla inWashington than by 50 monkeys on steroids. Second,deregulation of the economy has increased preemp-tion so as to prohibit state regulators from rushing intoregulatory vacuums created by federal withdrawal.Third, rising concern about international economiccompetition has spurred preemptions of state andlocal barriers to business competitiveness. For exam-ple, as Secretary of the Treasury Nicholas Brady saidon introducing President George Bush’s proposals topreempt certain state powers over interstate banking,something is seriously amiss when a bank in Califor-nia can open a branch in Birmingham, England, butnot in Birmingham, Alabama (ACIR 1992, p. 38).Fourth, foreign-trade agreements eliminating not onlytariffs but also non-tariff barriers to free trade posesubstantial, long-term preemption threats to a broadrange of state and local tax, regulatory, and policypowers (Weiler 1994).

All of these mechanisms for expanding federalpower also reflect a major structural change in thefederal system that first occurred during the 1960s,namely, a shift in federal policymaking from places topersons and a reconceptualization of the federal unionas one constituted by individuals, not sovereign states.This shift reflects a long-standing debate in Americanhistory that reached a flashpoint during the 1980swhen President Reagan declared that “the FederalGovernment did not create the States; the States cre-ated the Federal Government.” President Reagan’sopponents countered along the lines of William H.Seward in 1850: “The States are not parties to theConstitution as States; it is the Constitution of thepeople of the United States.” As Justice Harry A.Blackmun later put it: “Ours . . . is a federal republic,conceived on the principle of a supreme federal powerand constituted first and foremost of citizens, notsovereign states” (Coleman v. Thompson 1991). Thisindividualist view of the union is one reason whyJustice O’Connor has sought to transform the TenthAmendment from a guarantor of states’ rights to aguarantor of individual rights.

Politically, the reorientation of federal policymak-ing from places to persons has been driven primarilyby the U.S. Supreme Court’s reapportionment deci-sions requiring “one person, one vote” electoral dis-tricts for the U.S. House and for state legislatures(Reynolds v. Sims 1964; Wesberry v. Sanders 1964).Reapportionment—coupled with other developments

May/June 1998 New England Economic Review34

during the 1960s, especially the rise of the nationalmedia and of party primaries—was fully imple-mented by 1972. The principal consequence for statepowers was that reapportionment disconnected mem-bers of the U.S. House and, as a result, members of theSenate, too, from their historic electoral ties to stateand local party organizations and thereby to electedstate and local officials. The electoral incentives formembers of Congress shifted from dependence onstate and local party and government officials todependence on national interest-group support anddirect appeals to voters. Hence, the greatest federalencroachments upon state powers occurred not dur-ing the New Deal, when the federal governmentvastly expanded its power over the economy, butduring and after the Great Society, when members ofCongress had growing incentives to legislate directlyfor the interests of persons regardless of the effects ofsuch legislation on places, namely, state and localgovernments. Virtually every empirical indicator ofexpanding federal power (for example, numbers ofmandates, conditions of aid, preemptions, and federalcourt orders) shows unprecedented increases since1969—increases that continued unabated throughDemocratic and Republican administrations and shift-ing balances of partisan power in the Congress (Kin-caid 1993a & b).

A bellwether has been the Fair Labor StandardsAct of 1938 (FLSA). The act was upheld in UnitedStates v. Darby (1941), in which the U.S. SupremeCourt dismissed the Tenth Amendment as merely “atruism.” However, the Congress specifically exemptedstate and local government employees from the FLSA,and Darby concerned only the FLSA’s applicability toprivate employers. In 1968, though, the U.S. SupremeCourt permitted the U.S. Department of Labor toapply the FLSA to a limited range of state and localinstitutions held to be enmeshed in interstate com-merce (Maryland v. Wirtz 1968). In an unexpecteddecision eight years later, however, the Court struckdown the application of the FLSA to state and localgovernments as a violation of the Tenth Amendment’sprotection of traditional state governmental functionsand sovereignty (National League of Cities v. Usery1976). A government that cannot determine the wagesof its own employees cannot be said to be sovereign.

Nine years later, the Court reversed course againby overturning National League of Cities in Garcia v.San Antonio Metropolitan Transit Authority (1985). Theprincipal doctrinal argument was that the Court isobligated to protect individual rights, including theemployment rights of state and local government

employees, but not states’ rights under the TenthAmendment. Although the states obtained some con-gressional relief, though not exemption, from Garcia,the “political safeguards” of state powers in the fed-eral system that had long protected state sovereigntyfrom the FLSA had clearly deteriorated by the 1970s,such that the Court’s National League of Cities rulingwas widely viewed as an anomaly that proved, in-deed, to be short-lived (Kincaid 1993a). It was nothappenstance, therefore, that seven of the twelveUMRA-eligible mandates identified by the Congres-sional Budget Office in 1996 and early 1997 con-cerned wage impacts on state and local governmentemployees.

Still another obstacle to devolution is the set offears often expressed about the possible consequencesof devolution, especially destructive interstate compe-

It is difficult to predict what, ifany, interjurisdictional

competition might be sparked bydevolution. It is also difficult topredict outcomes, because statesand localities are quite differentpolities and communities today

than they were in the 1930s andeven the 1960s. So is the

federal government.

tition, limited state capacities to assume responsibilityfor devolved functions, accountability for policy andexpenditure outcomes of devolved programs, andgreater disparities of service provision and qualityamong jurisdictions.

The Question of Interstate Competition

Devolution talk has already generated doomsdayfears of destructive interstate competition (for exam-ple, Donahue 1997). Such fears are exaggerated, forseveral reasons.

For one, little “devolution” has occurred thus far,and what devolution can be said to have occurred is

May/June 1998 New England Economic Review 35

too new to permit proper assessment; hence, dooms-day predictions are premature. It is difficult to predictwhat, if any, interjurisdictional competition might besparked by devolution and whether such competitionwould be constructive or destructive in different func-tional areas and under different degrees of devolution.It is also difficult to predict outcomes, because devo-lution would not entail a return to pre-federalizedcircumstances. States and localities are quite differentpolities and communities today than they were in the1930s and even the 1960s. So is the federal govern-ment, and it is difficult to argue today that thefederal government is necessarily the superior re-pository of policy wisdom and political virtue.Furthermore, with modern communications and theexistence of numerous public and private institu-tions dedicated to “watchdog” tasks and to diffus-ing policy ideas, any interstate competition stimu-lated by devolution has a greater chance of yieldingpositive outcomes—such as experimentation in state“laboratories of democracy,” diffusion of innova-tions, and efficiency enhancement—than at any timein our federal history.

Second, even “welfare reform,” the most oft-citedexample of “devolution,” provides no conclusive evi-dence, despite much research, that the states havebeen, are currently, or will become engaged in a raceto the bottom over welfare benefits (see, for example,Gresenz 1997; Schram, Nitz, and Krueger 1998). Thusfar, reports Donna E. Shalala, Secretary of the U.S.Department of Health and Human Services, “Therehas been no race to the bottom in state welfarespending” (quoted in Vobejda 1998). Too many otherfactors influence state welfare decision-making, andthe relative positions of most high-benefit and mostlow-benefit states have not changed over the years,even though the real-dollar value of AFDC has de-clined in all states. The prospects for such competitionare also constrained, first, by the authority of high-benefit states to give migrants during their first yearin-state the lower benefits they received in their pre-vious state of residence, and second, by the five-yearlimit on TANF benefits, although because states canlower this limit, there may be some competitive pres-sure to do so. Furthermore, Social Security for seniorcitizens, Medicare, Medicaid (for the most part), and arange of other welfare-type programs were not in-cluded in the “welfare reform” measures. The lion’sshare of redistributive spending still lies with thefederal government, thus alleviating disparities thatmight otherwise stem from interstate competition. It ispossible that some states might emphasize dead-end

jobs so as not to attract out-of-state TANF recipientsseeking better employment, but such a strategy wouldbe counterproductive to a state’s self-interest. If TANFbeneficiaries migrate to states offering better employ-ment opportunities, such migration would be positive.

Similarly, no systematic evidence has been seen ofa race to the bottom in environmental standards(Revesz 1997), although certain Rustbelt states appearto skimp on environmental protection standards thatincrease industrial production costs. A race to thebottom in environmental protection is also unlikelyeven with further devolution, because environmentalinterest groups have clout in most states and becauseenvironmental protection is increasingly an economicasset for many states. More generally, the shift toward

Devolution could stimulate somehealthy intergovernmental

competition between the statesand the federal government, thuspossibly enriching policymaking

with the best ideas of two worlds.

a service economy places a greater value on quality-of-life assets, such as environmental protection, thusdampening competitive pressures that might other-wise push standards downward and fueling compet-itive pressures that might push standards upward.Consequently, devolution in the context of a serviceeconomy rather than an urban-industrial economy isless likely to trigger environmentally destructive in-terstate competition.

Third, devolution policies can be designed tominimize destructive interstate competition, to theextent it might occur, as reflected in some of TANF’sprovisions. Mandates, conditions of aid, performancestandards, schedules of fines for falling short of per-formance standards or for creating negative external-ities, and other mechanisms could be fashioned forthis purpose.

Fourth, zero-sum competition for business facilitiesthrough state and local subsidies, tax abatements, and soon is not likely to be accelerated much, if at all, even withsubstantial devolution. This form of competition is al-ready common and is being driven more by concernsabout international economic competition and by state

May/June 1998 New England Economic Review36

needs to generate more revenue through economic de-velopment so as to limit or avoid tax increases.

Fifth, fears of such zero-sum competition usuallyoverlook the negative consequences that could emergefrom suppression of such competition, particularly bythe federal government. States would find new waysto compete and, under conditions of substantial dev-olution, such competition could take place primarilyin fields of devolved domestic responsibility, such aswelfare. If states cannot compete for business throughsubsidies (however wasteful or ill-advised such sub-sidies), they may feel compelled to compete throughgeneral tax and expenditure reductions and throughreductions aimed at unpopular programs and vulner-able populations. Suppression of such competitionwould also increase state and local pressure on thefederal government to redistribute resources to meetthe various competitive needs and interests of thestates. The unmediated competition of the interstatemarketplace would be replaced by mediated compe-tition within the Congress and executive agencies. Theproposition that tax resources can be allocated moreefficiently through political competition among thestates within the Congress and the federal executivebranch rather than through fiscal competition withinthe interstate marketplace is not credible (Kenyon andKincaid 1991).

Sixth, given the huge number and range of federalmandates, conditions of aid, preemptions, individualrights protections, court orders, and the like, state andlocal governments have less room within which tocompete and fewer tools with which to competeagainst each other. Economic deregulation coupledwith federal preemptions of state-local regulatoryauthority is rapidly depriving states of viable tools forregulatory competition. As such, regulatory competi-tion might become more localized because citizens arelikely to cling tightly to local regulatory powers, suchas zoning, that protect property values and lifestylechoices. Thus, even with further devolution of func-tional responsibilities, massive federal constraints oninterstate competition will continue.

Seventh, devolution could also stimulate somehealthy intergovernmental competition between thestates and the federal government, thus possibly en-riching policymaking with the best ideas of two worlds.

The Question of State Capability

The question of state capability can be addressedboth empirically and normatively. One can determine

empirically the fiscal and administrative capacities ofstates to assume responsibilities for devolved func-tions under maintenance-of-effort assumptions. Butthe short answer to the question will be: Some statescan and some cannot, and some will do a better jobthan others. Substantial devolution may result ingreater diversity and heterogeneity across the states.

The normative question of state capability iswhether states should be held to maintenance-of-effortrequirements and other federal rules governing policyobjectives. True devolution in a federal democracywould leave such decisions to the citizens of eachstate. Whether a state has the capability to assumeresponsibility for any devolved function, therefore,depends substantially on whether the state’s citizenswish to give their state that capability. Here, thenormative question intersects with the empirical ques-tion, and causes concern among opponents or skepticsof devolution.

The public’s general disaffection from govern-ment is being vented within states and localities,which are more accessible to direct citizen activism.Much of this activism might be said to reflect therevenge of the “silent majority”—the public voice thatPresident Nixon sought to activate against what heregarded as a strident liberal minority. The silentmajority struck with force in 1978 through voterapproval of Proposition 13 in California, which set offa wave of citizen efforts to restrain state taxes andexpenditures nationwide. Tax and expenditure limitshave found their way into more and more stateconstitutions along with other restraints on stateand local governments, including widespread pub-lic support for term limits. These limits do notnecessarily mean that citizens would be unwilling toallow their states to assume responsibilities for devolvedfunctions, but the limits will constrain state capacities todo so if they prove workable.

Another significant feature of the revenge of thesilent majority is its almost single-minded focus onmajoritarian rights—sometimes, though not always, atthe expense of minority rights. Tax and expenditurelimits reflect efforts to protect taxpaying majoritiesagainst “big government.” The victims’ rights move-ment was one of the first non-fiscal reflections of thedesire of the silent majority to protect its rights inthe face of then-increasing rights protections forcriminals. State and federal Megan’s Laws are thelatest manifestation of this desire to protect themajoritarian community against what it regards aspreviously overprotected predators. Similarly,growing opposition to affirmative action, to equal

May/June 1998 New England Economic Review 37

protection for non-heterosexuals, and the like areefforts to protect majority rights. This revenge of thesilent majority does not have much of a civic dimen-sion. Instead, the attitude seems to be: What’s in itfor me?

The silent majority is likely to be receptive todevolution because it would shift powers and func-tions from a largely out-of-reach government to morewithin-reach governments where devolved powersand functions could be constrained by majoritarianrule and directed more toward the interests of themajorities within states and localities. In this respect,devolution could “empower” citizens and enhancemajoritarian democracy federally rather than mono-

Devolution would shift powersand functions from a largely out-

of-reach government to morewithin-reach governments, where

they could be directed moretoward the interests of the

majorities. Whether such anoutcome would be positive or

negative depends on one’spoint of view.

lithically. Whether such an outcome would be positiveor negative depends on one’s point of view.

Being largely ignored in the devolution debate iswhat one scholar has called the challenge “to findways of restoring the sense of accountability andbelonging offered by smaller, more human-scale insti-tutions, institutions that can serve as schools of citi-zenship while retaining the benefits of national gov-ernment. This is precisely the promise of federalism”(McClay 1996, p. 24).

Of course, more technical questions of state capa-bility arise as well. Devolution will require, for ex-ample, continued intergovernmental cooperation,especially as details of separating and sharing respon-sibilities within specific programs, such as TANF, areworked out under statutory rules. Significant federalregulations are likely to accompany most devolutions,and the states are likely to remain vulnerable to new

policies and rules enacted by the Congress or promul-gated by federal agencies with little or no advancewarning. States, therefore, will need to maintain ananticipatory and defensive stance. State legislatureswill also face control and oversight issues, such asassignments of devolution implementation issues toappropriate legislative committees and, more impor-tant, decisions about how much authority to embed instate law and how much authority to delegate to theexecutive branch to implement and alter programsflexibly, pursuant to federal law (Olson 1996). Statelegislators are especially concerned that the Congresswill bypass the state legislatures and state constitu-tions by devolving control over spending and pro-grammatic implementation to governors. As notedearlier, state legislators succeeded in convincing theCongress to include the Brown Amendment in the1996 “welfare-reform” law (Section 901), which re-quires that new block grant funds be appropriated bythe state legislature, but it is by no means certain thatthe Brown Amendment will become a basic principleof devolution.

The logic and fiscal imperatives of devolution,coupled with deregulation and privatization, willrequire states to work more closely with local gov-ernments, nonprofit institutions, civic organizations,businesses, and other states as well (Feustel 1997).Devolution may also have side effects on state andlocal governments, such as increased lobbying instate capitols, county courthouses, and city halls, asthese governments assume more policy and fiscalresponsibility for public functions. This could, inturn, increase voter pressure for lobbying and cam-paign-finance reform as well as stricter ethics laws(Feustel 1997, p. 25).

Conclusion

To date, there is no evidence of wholesaledevolution, although there is a discernible and, untilrecently, unanticipated nudging toward restoringsome state powers and rebalancing federal-staterelations in the federal system. It is difficult topredict the outcome if this nudge should become asurge, because such rebalancing will occur withinhistorical circumstances quite different from thosethat prevailed at the outset of the federal govern-ment’s twentieth-century power expansion. There isno a priori reason, therefore, to expect that substan-tial devolution, should it ever occur, would be moremalignant than benign.

May/June 1998 New England Economic Review38

References

Atascadero State Hospital v. Scanlon, 473 U.S. 234 (1985).Baker, Earl M., ed. 1975. “The Suburban Reshaping of American

Politics.” Publius: The Journal of Federalism, vol. 5 (Winter), entireissue.

Baldersheim, Harold and Peter Stava. 1993. “Reforming LocalGovernment Policymaking and Management through Organiza-tional Learning and Experimentation: The Case of Norway.”Policy Studies Journal, vol. 21, no. 1, pp. 104–114.

Bernier, Lynne Louise. 1992. “Socialist Intergovernmental PolicyDuring the Mitterand Era.” Publius: The Journal of Federalism, vol.22 (Fall), pp. 47–66.

Broder, David S. 1993. “Clinton Vows Not to Forget States.”Washington Post, February 2, p. A10.

Buchanan, James, M. 1995. “Federalism as an Ideal Political Orderand an Objective for Constitutional Reform.” Publius: The Journalof Federalism, vol. 25, no. 2 (Spring), pp. 19–27.

City of Boerne v. Flores, 117 S.Ct. 2157 (1997).Coleman, William G. and Delphis C. Goldberg. 1990. “The Eisen-

hower Years and the Creation of ACIR.” Intergovernmental Per-spective, vol. 16, no. 3 (Summer), pp. 19–23.

Coleman v. Thompson, 111 S. Ct. 2546 (1991).Conlan, Timothy. 1988. New Federalism: Intergovernmental Reform

from Nixon to Reagan. Washington, DC: The Brookings Institu-tion.

Council of State Governments. 1996. Restoring Balance to theAmerican Federal System. Lexington, KY: Council of State Gov-ernments.

Democratic Leadership Council. 1997. “DLC and the States.” StateFocus: Supplement to The New Democrat, vol. 9 (July/August).

Donahue, John D. 1997. Disunited States. New York: Basic Books.Economic Report of the President. 1998. Washington, DC: Government

Printing Office, February.Ehrenhalt, Alan. 1995. “Out in the States, It’s Not the 1930s

Anymore.” Governing, vol. 9 (December), p. 7.Elazar, Daniel J. 1984. American Federalism: A View from the States. 3rd

ed. New York: Crowell._______. 1995. “From Statism to Federalism: A Paradigm Shift.”

Publius: The Journal of Federalism, vol. 25, no. 2 (Spring), pp.5–18.

Feustel, Bruce. 1997. “What Do We Do Now? Devolution and theLegislative Institution.” State Legislatures, vol. 23, no. 6 (June), pp.22–25.

Fuentes, Carlos. 1990. “Federalism Is the Great Healer.” Los AngelesTimes, December 16, p. M1.

Gage, Robert W. and Bruce D. McDowell. 1995. “ISTEA and the Roleof MPOs in the New Transportation Environment: A MidtermAssessment.” Publius: The Journal of Federalism, vol. 25, no. 3(Summer), pp. 133–154.

Galston, William A. and Geoffrey L. Tibbetts. 1994. “ReinventingFederalism: The Clinton/Gore Program for a New PartnershipAmong the Federal, State, Local, and Tribal Governments.”Publius: The Journal of Federalism, vol. 24, no. 3 (Summer), pp.23–48.

Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528(1985).

Gingrich, Newt. 1995. To Renew America. New York: HarperCollins.Goshko, John M. 1995. “As Oregon Goes, So Could the Rest of the

Nation.” Washington Post National Weekly Edition, August 14–20,p. 31.

Gregory v. Ashcroft, 501 U.S. 452 (1991).Gresenz, Charles R. 1997. An Empirical Investigation of the Role of

AFDC Benefits in Location Choice. Santa Monica, CA: The RANDCorporation.

Habicht, F. Henry II. 1992. “Environmental Regulations: Lighteningthe Load.” Intergovernmental Perspective, vol. 18, no. 4 (Fall), pp.27–29.

Jeter, Jon. 1997. “The ‘Ultimate Test’ of Devolution.” Washington PostNational Weekly Edition, August 11, pp. 29–30.

Kaiser/Harvard Program on the Public and Health/Social Policy.1995. Survey on Welfare Reform: Basic Values and Beliefs; Support forPolicy Approaches; Knowledge About Key Programs. The Henry J.Kaiser Family Foundation (January), Table 25.

Kenyon, Daphne A. and John Kincaid, eds. 1991. Competition amongStates and Local Governments: Efficiency and Equity in AmericanFederalism. Washington, DC: The Urban Institute Press.

Kincaid, John. 1988. “State Court Protections of Individual RightsUnder State Constitutions: The New Judicial Federalism.” TheJournal of State Government, vol. 61 (September/October), pp.163–169.

_______. 1993a. “Constitutional Federalism: Labor’s Role in Displac-ing Places to Benefit Persons.” PS: Political Science & Politics, vol.26 (June), pp. 172–177.

_______. 1993b. “From Cooperation to Coercion in American Fed-eralism: Housing, Fragmentation, and Preemption, 1780–1992.”Journal of Law and Politics, vol. 9 (Winter), pp. 333–433.

_______. 1995. “Values and Value Tradeoffs in Federalism.” Publius:The Journal of Federalism, vol. 26, no. 2 (Spring), pp. 29–44.

_______. 1996. “Intergovernmental Costs and Coordination in U.S.Environmental Protection.” In Kenneth M. Holland, F. L. Morton,and Brian Galligan, eds., Federalism and the Environment: Environ-mental Policymaking in Australia, Canada, and the United States, pp.70–101. Westport, CT: Greenwood Press.

Liebschutz, Sarah F. 1995. “Empowerment Zones and EnterpriseCommunities: Reinventing Federalism for Distressed Communi-ties.” Publius: The Journal of Federalism, vol. 25, no. 3 (Summer), pp.117–132.

Maryland v. Wirtz, 392 U.S. 183 (1968).Mason, Alpheus T. 1952. “American Individualism: Fact and Fic-

tion.” The American Political Science Review, vol. XLVI, no. 1(March), pp. 1–18.

Mayer, Jeffrey L., ed. 1976. “Dialogues on Decentralization.” Publius:The Journal of Federalism, vol. 6 (Fall), entire issue.

McClay, Wilfred M. 1996. “The Soul of Man Under Federalism.”First Things, no. 64 (June/July), pp. 21–26.

McDowell, Bruce D. 1997. “Advisory Commission on Intergovern-mental Relations in 1996: The End of an Era.” Publius: The Journalof Federalism, vol. 27, no. 2 (Spring), pp. 111–127.

Merritt, Deborah Jones. 1988. “The Guarantee Clause and StateAutonomy: Federalism for a Third Century.” Columbia Law Re-view, vol. 88, pp. 1–97.

Michigan v. Long, 459 U.S. 904 (1983).Milbank, Dana and Laurie McGinley. 1996. “While Washington

Fiddles, Many States Devise Solutions to Problems with Welfareand Health Care.” Wall Street Journal, May 31, p. A12.

Nathan, Richard P. 1996. “The Devolution Revolution: An Over-view.” Rockefeller Institute Bulletin 1996, pp. 5–13. Albany, NY:Nelson A. Rockefeller Institute of Government, State Universityof New York.

National Governors’ Association. 1996. “States Feel Relief fromUnfunded Mandates.” Governors’ Bulletin, vol. 30, June 17, pp. 1–2.

National League of Cities v. Usery, 426 U.S. 833 (1976).National Performance Review. 1993. Creating a Government that

Works Better and Costs Less: Report of the National PerformanceReview. Washington, DC: U.S. Government Printing Office.

New York v. United States, 505 U.S. 144 (1992).New York Ice Co. v. Liebman, 285 U.S. 262 (1932).Olson, Gary. 1996. “Federal Focus: Devolution—Ready or Not, Here

It Comes.” The Fiscal Letter, vol. 19 (Spring), pp. 5–6.Peterson, Paul E. 1995. The Price of Federalism. Washington, DC: The

Brookings Institution.Poole, Robert. 1997. “Devolving Transportation Funding.” Spectrum,

vol. 70, no. 3 (Summer), pp. 20–23.

May/June 1998 New England Economic Review 39

Posner, Paul L. and Margaret T. Wrightson. 1996. “Block Grants: APerennial, But Unstable, Tool of Government.” Publius: The Jour-nal of Federalism, vol. 26, no. 3 (Summer), pp. 87–108.

Printz v. United States, 117 S.Ct. 2365 (1997).Reeves, Mavis Mann. 1990. “The States as Polities: Reformed,

Reinvigorated, Resourceful.” Annals of the American Academy ofPolitical and Social Science, vol. 509 (May), pp. 83–93.

Revesz, Richard L. 1997. “Federalism and Environmental Regula-tion: A Normative Critique.” In John Ferejohn and Barry R.Weingast, eds., The New Federalism: Can the States Be Trusted? pp.97–127. Stanford, CA: Hoover Institution Press.

Reynolds v. Sims, 377 U.S. 533 (1964).Rivlin, Alice. 1992. Reviving the American Dream: The Economy, the

States, and the Federal Government. Washington, DC: The BrookingsInstitution.

Roe v. Wade, 410 U.S. 113 (1973).Schram, Sanford, Lawrence Nitze, and Gary Krueger. 1998. “With-

out Cause or Effect: Reconsidering Welfare Migration as a PolicyProblem.” American Journal of Political Science, vol. 42, no. 1(January), pp. 210–229.

Shenk, Joshua Wolf. 1997. “Washington’s counter-devolutionaries.”U.S. News & World Report, vol. 123, no. 20 (November 24), p. 34.

South Dakota v. Dole, 483 U.S. 203 (1987).Suter v. Artist, 112 S. Ct. 1360 (1992).Texas v. White, 74 U.S. 700 (1869).Tocqueville, Alexis de. 1969. Democracy in America. [1835–40]. Ed.

J. P. Mayer; Trans. George Lawrence. New York: HarperPeren-nial.

Tolley, Michael C. and Bruce A. Wallin. 1995. “Coercive Federalismand the Search for Constitutional Limits.” Publius: The Journal ofFederalism, vol. 25, no. 4 (Fall), pp. 73–90.

United States v. Alfonso Lopez, Jr., 115 S. Ct. 1625 (1995).United States v. Darby, 312 U.S. 100 (1941).

U.S. Advisory Commission on Intergovernmental Relations. 1984.Regulatory Federalism: Policy, Process, Impact, and Reform. Washing-ton, DC: ACIR.

_______. 1985. The Question of State Government Capability. Washing-ton, DC: ACIR.

_______. 1987. Devolving Selected Federal-Aid Highway Programs andRevenue Bases: A Critical Appraisal. Washington, DC: ACIR.

_______. 1988. Changing Public Attitudes on Governments and Taxes.Washington, DC: ACIR.

_______. 1992. Federal Statutory Preemption of State and Local Author-ity: History, Inventory, and Issues. Washington, DC: ACIR.

_______. 1995. Characteristics of Federal Grant-in-Aid Programs forState and Local Governments. Washington, DC: ACIR.

U.S. General Accounting Office. 1996. Environmental Protection: Sta-tus of EPA’s Initiatives to Create a New Partnership with States.Washington, DC: U.S. GAO (February).

Vacco v. Quill, 117 S.Ct. 2293 (1997).Vobejda, Barbara. 1998. “Fewer Welfare Recipients, More Spending on

Them.” Washington Post National Weekly Edition, February 16, p. 35.Washington v. Glucksberg, 117 S.Ct. 2258 (1997).Weaver, R. Kent. 1996. “Deficits and Devolution in the 104th

Congress.” Publius: The Journal of Federalism, vol. 26, no. 3 (Sum-mer), pp. 45–85.

Weiler, Conrad. 1994. “Foreign-Trade Agreements: A New FederalPartner?” Publius: The Journal of Federalism, vol. 24, no. 3 (Sum-mer), pp. 113–33.

Wesberry v. Sanders, 376 U.S. 1 (1964).Whitman, David. 1998. “Despite Tough Talk, States Avoid Workfare.”

U.S. News & World Report, vol. 124, no. 1 (January 12), pp. 26–27.Will v. Michigan Department of State Police, 491 U.S. 58 (1989).Wunsch, James S. and Dele Olowu, eds. 1995. The Failure of the

Centralized State: Institutions and Self-Governance in Africa. SanFrancisco, CA: ICS Press.

May/June 1998 New England Economic Review40

DiscussionDavid R. Beam, Associate Professor and Director of theGraduate Program in Public Administration,Illinois Institute of Technology

First, let me say that I agree with much of JohnKincaid’s analysis. The paper is a careful siftingof important recent intergovernmental develop-

ments. I certainly agree with his finding that we nowhave as many centralizing influences emerging asdecentralizing influences.

However, I was asked by moderator GeorgeLatimer to consider areas of disagreement with Kin-caid’s paper. One of the first I noticed was his refer-ence to the Unfunded Mandates Reform Act (UMRA)as a leading idea in the Contract with America.Certainly this was true, and in fact it was one of thefew ideas from that document to be adopted. But assomeone involved with the issue for a long time, I amtroubled to hear it described in this fashion. Similarlegislation very nearly was adopted by the precedingCongress, with the strongest support from state andlocal organizations for any issue since general reve-nue-sharing. And Newt Gingrich made it clear in hisbook that he became familiar with this issue onlyrecently. In fact, the desire in the Congress to enactthis UMRA legislation so quickly was fueled by theneed to get the governors on board for the balancedbudget amendment, not because of any deep commit-ment to its principle.

Another point the paper made that troubled mewas about trust in government. It is clear that thecurrent level of trust in the federal government is low,but I do not know if trust in state and local govern-ments should be viewed as “high”. The newspaperbrought to my hotel room today, USA Today, de-scribed a poll that tried to address this question. I waspleased to see that local fire departments are trusted:78 percent of the population say they trust their localfire department “a lot.” But city and local govern-ments are trusted by only 14 percent of the population,while state governments are trusted by 9 percent andthe federal government by 6 percent. These figuresindicate very little trust for government at any level.

A third point that troubled me about the paperwas the suggestion that suburbanization could beviewed as an important reason for devolution. Subur-banization is a post-World War II phenomenon for themost part, and it has gone on continuously since that

time. Certainly the proportions of population and jobsthat have been suburbanized have increased. Butduring that same period, the federal government hasalso grown rapidly. To try to explain devolution onthe basis of suburban growth strikes me as inade-quate. Overall, in fact, I do not think Kincaid’s paperwas able to come to terms with adequate explanationsfor devolution, and certainly I would not point tosurburbanization as a major factor.

I am in closer agreement on many other points inthe paper. First, from my own perspective, I justcelebrated my 55th birthday, and I have seen interestin federalism reform come and go four times duringmy lifetime. My perception is that devolution is anissue that periodically appears, catches fire for a shorttime, then burns itself out and is gone. Those of uswho are here today may keep up a continuing interestin intergovernmental reform and federalism issues,but on the whole they are not at the top of the generalpolitical agenda very often.

The four instances occurred once under Eisen-hower, once under Nixon, once under Reagan, andnow; the current burst of interest began with theRepublican Congress elected in 1994. None of theearlier episodes lasted longer than six years, and theperiod in which each accomplished its chief legislativeobjectives was between one and three years. Theactive period in Nixon’s time was 1972 through 1974,although he had made devolution a central theme inhis campaign in 1969; Reagan’s active period occurredduring his first year, 1981. While the major theme inReagan’s first State of the Union address in 1982 wasthe issue of federalism reform, his trade-off and turn-back proposal was never seriously considered by theCongress, in part because it was so contrary to aproposal on the same issue that the governors wereadvancing.

Each of these campaigns fell dramatically short ofits stated objectives. Some legislative accomplish-ments resulted, but they never added up to thesweeping reforms their advocates had hoped for. Iagree with the view of Plunkitt of Tammany Hall thatreformers are only morning glories, and, given thetime frame of earlier campaigns, the clock may al-ready have run out on the current wave of reform.

Furthermore, not only has each burst of federal-ism reform been brief, but each has been followed byor has even coincided with significant elements ofcentralization. Over the last four decades, a shift hasoccurred from more gentle forms of federal interven-tion through federal grants that could rightly beviewed as cooperative measures, to a form of regula-

May/June 1998 New England Economic Review 41

tory federalism that employs far more coercive mea-sures, such as unfunded mandates. Not only havefederal involvements increased, but their nature hasbecome more objectionable to many state and localofficials.

Not only has each burst offederalism reform been brief, buteach has been followed by or haseven coincided with significant

elements of centralization.

So the lesson of history is clear: Political leadersnever really get religion as a result of their briefconversions to the faith of federalism reform. Twoideals are implicit here, and I would like to distinguishbetween them: first, the constitutional ideal of a lim-ited national government, and second, the intergov-ernmental reform ideal of an effectively operatingfederal partnership. Neither one ever holds adherentsfor long. Instead, political leaders quickly return totheir pattern of making new promises to voters andorganized interests and sponsoring and voting fornew, intergovernmentally operated programs.

Categorical grants were the earliest and the mostpopular: Kincaid notes that 618 were funded in fiscalyear 1995. One could also consider the growth inmandates in recent years. In the middle of PresidentReagan’s crusade for a new federalism, just as re-cently, the widely held belief was that we were rollingback the Great Society and the New Deal. But, by theend of the 1980s we had more programs, more regu-lations, and less funding, as I in fact had predicted atthe beginning of the decade. Categorical grants grewfrom 534 to 543. There were 27 major new mandates,while grant outlays actually fell by roughly 7 percent.

Overall, the longer-run political forces for federalintervention consistently dominated the short-run buthigh-profile forces for containment of the federalgovernment this last time around. Some important,lasting changes did emerge in several policy arenasduring the Reagan era, but alongside movement inother areas toward increased centralization.

So what about this time? Is there any reason tobelieve it will be different? Kincaid hesitates to offer aclear forecast on that issue. He suggests a number of

reasons why it could be different, though. I thoughtthe most provocative was his claim that we are now inthe midst of a worldwide trend toward decentraliza-tion across all political, social, economic, and culturalsectors, that we may be on the eve of a paradigm shiftcomparable to those of the Civil War, the New Deal, orthe Great Society eras. If that is true, I do not want tomiss it!

Kincaid’s strong statement struck me in partbecause, looking through earlier accounts, I saw Mor-ton Grodzin’s explanation of why the KestnbaumCommission of the 1950s had failed in its effort toreturn power to the states. One reason he gave was thecountervailing tendency of all power to flow to centralinstitutions, not only in the United States but all overthe world; and not only in government but also inreligious, business, and labor organizations. To be-lieve these tendencies can easily be reversed is toignore the power of great historical forces, he argued.

My own expectation regarding devolution is thatthis time is not so different, and our present infatuationwith the idea will probably fade away. If it does not,the main reason would be that the Supreme Court hasfor the first time come down clearly on the side of theadvocates of devolution, a striking development. But Icannot believe that the cases seen thus far can do morethan snip away at the edges of the intergovernmentalregulatory apparatus. And there are other acceptableways of achieving the same federal objectives. One viewis that the Congress has finally acquired the nerve topass legislation it never would have before, and it hasbeen slapped down for doing it. In a sense the casesreflect congressional assertiveness, not judicial restraint.

Another important point is that not very much ineconomic, political, or administrative theory rational-izes or justifies the devolutionary proposals we haveseen recently. Let me contrast that with many priorintergovernmental reform proposals that were wellthought out and, as a result, commanded significantbipartisan support. They had been developed over along period of time, and whether or not they weregood ideas, many people at the time thought theywere, and this was what pushed them through thelegislative process. Nixon’s revenue-sharing proposalstraced their origins back to Walter Heller and JosephPechman in the Johnson Administration. Communitydevelopment and manpower training block grantshad support from people who worked in those pro-gram areas. Even Nixon’s failed family assistance planin the days of talk of nationalizing welfare—these billswere all products of lengthy consideration in theirrelevant public policy communities.

May/June 1998 New England Economic Review42

In contrast, take welfare reform, the leading ac-tion to date of the current devolution drive. Fordecades the conventional view among policymakershas been that redistributive functions are best per-formed by central governments. Paul Peterson, in oneof the major recent books on federalism, argues thatmoving income maintenance to the state level, whichthe Republicans were trying to do, would be a mistakebecause it would give the states responsibilities forwhich they are ill-suited. His view, which I believe tobe a common one, is that redistributive functions arebetter handled at the national level and developmen-tal functions at the state and local levels. Peterson goesfurther, predicting that any policy that dramaticallyshifts the responsibility for welfare downward tostates and cities will prove unworkable and short-lived.

John Donahue, who is here with us today, hasalso written that “the devolution of antipoverty policywill be seen eventually as a mistake. If welfare policyis permanently turned over to the states, I believe thewell-being of the most marginal members of societywill be adversely affected in serious ways. When thefirst serious recession arrives, state-based welfare pol-icies will reveal their built-in bias toward undueharshness, and antipoverty programs will spiral to-ward the furthest degree of austerity that citizenconsciences will permit.”

Some features of the welfare reform act did reflectthe findings of research, but the implications of closingoff the AFDC matching grant program and turning itinto a block grant have not been explored adequately.Writing in the National Tax Journal, the late StevenGold commented that the abandonment of open-ended matching grants is inconsistent with the viewthat such grants are needed to stimulate the produc-tion of services that produce external benefits. Henoted that people seem no longer to accept the ideathat there are production externalities—for example,that poverty breeds crime and that welfare tends toreduce crime. Still another important implication ofadopting block grants for welfare programs is that theamount of financial aid provided to states will notautomatically increase in response to a recession, as itdid in the past.

The most clearly stated rationale for the presentwelfare policy direction, in my view, was articulatedby James Q. Wilson. In a recent article Wilson arguedthat we know so little about the tangle of pathologiesthat produce welfare dependency that we may as wellturn welfare over to the state and local governments.His is not a strong argument based on theory. He calls

it an argument based on humility. We do not knowwhat to do at the federal level, so we may as well turnit over to the states and see if they can do better thanwe have to date.

Let me close with an observation about oneunfortunate difference affecting the current devolu-tionary drive—the high level of distrust, and evenfear, of the federal government. Similar levels ofdistrust emerged during the Reagan era, and boththen and now the drive for devolution was merely onecomponent in a host of essentially antigovernmentmeasures. These included the desire for tax, expendi-ture, and deficit cuts, all simultaneously; also forderegulation, privatization, less bureaucracy, less reli-ance on experts, and more faith in the common man.We are in a period in which Jeffersonian values arerising paramount over our basically Hamiltonian gov-ernment and way of life. The main problem, in mymind, is that nothing in this revolutionary and oftenvery cynical spirit assures that we will in fact moveany closer to the historical ideal of federalism.

For decades the conventional viewamong policymakers has been

that redistributive functions arebest performed at the nationallevel, developmental functionsat the state and local levels.

Our founders did not seek a weak national gov-ernment, but rather a limited national governmentthat would be able to operate effectively in a restrictedsphere. We must now redefine that sphere—but I seeno signs that we are trying to do so in a meaningfulfashion. President Clinton is backing away from areasI would view as high on the list of real nationalconcerns, such as health care financing and delivery,which probably require more federal intervention andindeed have already required more federal interven-tion since the defeat of his earlier proposal. Instead, heis focusing on law enforcement and education, tradi-tionally among the most local of functions.

Nor is there any reason to believe much can bedone to renew the confidence in government of thevoters who are coming to dominate the electorate. Icertainly see this in my students. Vietnam, Watergate,

May/June 1998 New England Economic Review 43

a long period of economic stagnation, and a host ofother scandals and disappointments have left perma-nent marks on our political landscape.

Conversely, it was the effort of attacking the GreatDepression, the mobilization for and successful pros-ecution of the Second World War, and the long period ofpostwar prosperity that created an environment inwhich new federal initiatives were sought and wel-comed. The era, not so much of big government but ofpopular confidence in government, probably really isover. Certainly Washington has lost much of the moralauthority that, starting in the 1950s, accounted for itstriumph over the traditional defenders of states’ rights.

The condition we face, then, is not so much onewhere we have granted increased power to the statesthrough some process of devolution. State govern-ments do have greatly improved capabilities now overthe 1960s, when the great federal growth began. Andcertainly the states have been leaders in a number ofpolicy areas since the early 1980s. But they now lookstrong and trustworthy largely because the federalgovernment appears to so many to be scandal-ridden,ineffective, wasteful, and remote. I believe this view iswidespread. And, unlike our interest in devolution,this impression is likely to be with us for years tocome.

DiscussionDavid T. Ellwood, Lucius N. Littauer Professor of PoliticalEconomy, John F. Kennedy School of Government,Harvard University

Iquite enjoyed John Kincaid’s thoughtful and com-prehensive paper. It seems an excellent place tobegin any discussion of devolution. As requested

by our hosts, I propose to amplify a few points andthen focus on some larger questions raised by thepaper.

The first part of the paper seeks to understand theorigins of devolution this time around. Kincaid prop-erly points to the long and rich constitutional debatesurrounding issues of federalism and the many battlesfought on this ground in both the Congress and thecourts. He notes that there have been some relevantcourt fights this time around, and constitutional groundsare occasionally cited by congressional proponents ofdevolution. But he rightly discounts these factors as theprimary forces motivating the current debate.

A second alternative is that devolution is theresult of shifting political power. The RepublicanCongress coupled with wins by Republican governorshas led to control by a group with different politicaland philosophical orientations, which led to a pushtoward devolution. Or more precisely, perhaps devo-lution fit the politics of the moment. By one framing,the move to devolution was based not on a carefullydrawn conclusion that state governments would do abetter job, but rather on a particular political environ-ment.

Welfare, as Kincaid and others have noted, seemsto be the only domain where a really dramatic movetoward devolution has actually occurred. And thosewho might argue that devolution in this case was asmuch a result of shifting political sands as anythingcan make a good case. If you examine the recentlegislative history, you find that in the course of twoyears the Republicans went through three welfarebills, each of which was widely supported within theparty. The first bill stipulated two-year limits andworkfare requirements within the current structure. Itwas followed by an unbelievably restrictive “we aregoing to regulate every aspect of welfare recipients’lives as well as the states that administer them” bill, akind of conservative micro-management bill. This wasfollowed by a pure devolution bill, relinquishing alldecisions to the states. The final bill was an awkwardmixture of all three. The bill is in many ways a classiclegislative product, but it is very difficult to make thecase that this bill emerged from a carefully constructedand fully evolved view of devolution. One bill alongthe way did have some of those characteristics, but itfailed to meet other objectives and was discarded.

Yet for those who would deny that devolution inwelfare had an important basis in frustration withfederal controls and a belief that states could do abetter job, let me recount one anecdote. Immediatelyafter the election of the 1994 Congress, when thePresident was trying desperately to get a foothold onthe welfare issue, he called a special meeting at BlairHouse, the building across the street from the WhiteHouse, and invited governors, members of Congress,and the like. Those of us in the Administration wereterrified about this meeting because the President,when surrounded by governors, tended to go into an

May/June 1998 New England Economic Review44

extreme devolution mode. And indeed, the first partof the meeting went precisely as anticipated, eventhough we had tried to prepare the President byemphasizing the dangers of recessions, showing whymatching grants make a difference, and so on. TommyThompson, governor of Wisconsin, aided quite effec-tively by the governor of Michigan, made a compel-ling case saying, “Look, the states have been doing allthe interesting innovation lately. You guys have had60 years and basically have made a mess of things.Give us the ball.” You could see all the heads nodding,including the President’s, as Thompson spoke. Hisarguments were basically true: Any recent innovationin welfare has been at the state level.

Then we turned to the second topic, child supportenforcement, an equally important part of welfarereform that never gets any public attention. Some of usin the Administration wanted more national control,more standards in this area, as interstate issues areenormous in child support enforcement: Over one-third of cases are interstate cases. We were braced forthe next barrage, where this issue too would pushtoward devolution. To our astonishment, TommyThompson spoke passionately and effectively aboutwhy federal standards were absolutely essential inthis area. He talked about how we needed moremandates on states because otherwise they could notreach these deadbeat dads, and how this was a hugeproblem for America. It was a well-reasoned andthoughtful argument against devolution in this area.And in fact the final welfare bill ended up includingvirtually all of the Clinton Administration’s childsupport enforcement measures in it. It was clearly acentralization of these regulations, more than any ofus had expected.

Let me turn to the question of whether this cycleof devolution is real or not. I agree with the generalthrust of Kincaid’s paper, which seems to say that inmost policy areas it is not very real. My first point isthat the paper properly assesses the legislative intent,which was that the devolution not be entirely real. Inthe welfare reform bill, the Congress did give thestates the opportunity to pursue various programs inwhatever way they see fit, but the Congress alsowanted work requirements, time limits, and so forth,in part because that is what the public really wantedout of welfare reform. The public did not care verymuch about the state versus federal question—it hasalways hated the federal government more than thestates—but it did want the bill to be about work. Sothe Congress said, “Listen, we are going to givewelfare policy to the states, but we are going to

include work requirements and a few other goodiesfor various factions in the political parties.”

So the goal seemed to be partial devolution. Yetwhen you look at the bill, in fact it is almost a puredevolution bill, because any restrictions are in fact soloose that any clever state can figure out ways aroundthem. Let me give you a couple of examples. The billincludes work requirements, but they are written inthe following way. A state must have a certain fractionof the welfare caseload working or the caseload couldhave been reduced by an equivalent amount. Let meexplain: One choice is hard, the other is easy; one isexpensive, the other is cheap. And what does “case-load” mean anyway? A state can redefine completelythe population it serves: It does not need to have thesame group of people on welfare. What is the differ-ence between an old caseload and a new caseload? Orperhaps you can now serve, using the very fewresources available, only the working poor—in whichcase they are all working.

Much of the devolution literaturemisses what is now the central

challenge in social policy, that is,that we as a society want

both the advantages of federalrule and some of the advantages

of state rule.

This is the result of a process whereby the Con-gress drafts a bill under an old concept, in this case ofthe welfare population, while simultaneously chang-ing all the old rules. Many states are operating as ifTemporary Assistance to Needy Families (TANF) keptthe same program in place but allowed a bit moreflexibility. In fact, TANF allows states to do anythingthey want. Over the long run, I think it will be muchmore interesting to see the results of TANF than atpresent. But let us be clear. TANF replaced the AFDCprogram, which required $13 billion in expendituresat the federal level: a tiny, trivial program. Given allthe current excitement about the changing nature ofgovernment, TANF does not amount to much, even ifit were an example of pure devolution.

That said, I do want to take issue with whether ornot we should be worried about the effect TANF may

May/June 1998 New England Economic Review 45

have. TANF is ambiguous and it should be. Kincaid’spaper, along with much of the devolution literature,misses what I think is the central challenge now insocial policy, that is, that we as a society want both theadvantages of federal rule and some of the advantagesof state rule. For many programs, this choice is easy.No one has suggested sending Social Security to thestates: It is a pure rule-based, national system. Andalmost no one has suggested the federal governmenttake over child welfare protective services, becauseclearly that is a local issue and a developmentalprogram. The problem is with the services in theintermediate area.

In Tables 1 and 2, I present a simplified and muchless sophisticated version of what Bob Tannenwalddoes in his paper. Table 1 looks at the poverty rateamong children in six representative states; you cansee that it varies from 15 percent in Connecticut and 14percent in Wisconsin to 31 percent in Mississippi. Thatgives us a measure of need. Median household incomealso shows approximately a two-to-one ratio betweenthese states, but highly inversely correlated with thisneed. And when we look at what these states spenton low-income children, keeping in mind the stronginducement from the federal government for poorerstates to spend more, we find that Mississippi spent$110 per year per poor child, while states like Connect-icut spent nearly $2,000 per poor child. This disparitysuggests that states really do have very different views oftheir role as welfare providers as well as differing capac-ities to provide relief to poor children. That would seemto argue for a federal role.

On the other hand, why do we want statesinvolved in welfare provision? Why not just make it apurely federal program? For a long time, we thoughtthat welfare was about redistribution and that the roleof the federal government in this arena was to deter-mine eligibility and to write checks. Welfare reformimplied that this was precisely the wrong business forthe federal government to be in. Rather, we ought tohelp people help themselves, that is, run a welfare-to-work business. But the federal government is not verygood at that type of developmental program. As aresult, we are now in a somewhat awkward position.On the one hand, the arguments for the redistributivefiscal side of the federal role are very powerful, but thearguments for state innovation and the like are pow-erful too. A gap remains, certainly, between what nowhappens and what really needs to happen at the stateand local levels, involving not only questions aboutadaptability but also about innovation. But the earlyresults of welfare reform show that it has alreadygenerated a great deal of innovation.

The real challenge, in my view, is the mixed plansthat try to combine federal and state involvement in aparticular policy arena. If the federal governmentsimply gives the states money and says spend it anyway you want, but we would like you to spend it onpoor folks, there is no reason to believe that thedistribution of state spending will be any differentfrom the way states spent their previous dollars. Thisis shown in the highly variable data in the thirdcolumn of the table on child poverty. When thefederal government sees that some states are not

Table 1Poverty Rates and Median Household Income in 1993, and State Contributions per PoorChild for AFDC/JOBS in FY 1994Selected States

StatePoverty Rate among Children

Ages 5 to 17 (1993)Median Household

Income (1993)

Approximate Annual State Contributionsper Poor Child in FY 1994 for AFDC and

Related Employment and Training (JOBS)a

Arkansas 23.3% $24,018 $ 164California 24.1% $34,129 $1,759Connecticut 15.4% $42,105 $1,815Florida 21.8% $28,230 $ 635Mississippi 31.2% $22,952 $ 110Wisconsin 14.2% $32,201 $1,170aUsing 1993 Census estimates of poor children ages 5 to 17 multiplied by 1.45 to account for children ages 0 to 4. Includes administration and emergencyassistance. Excludes IV-A child care.Source: U.S. Bureau of the Census, Internet Table P93-00; Administration for Children and Families, Internet table.

May/June 1998 New England Economic Review46

spending this money on the poor, it is likely to try toforce states to spend it in ways the federal govern-ment prefers.

Once the federal government starts the process oftrying to direct precisely for whom money is to bespent, soon it is in the business of trying to dictate howit is spent. And the states are in the business of tryingto be clever and not spend it the way the federalgovernment wanted. If you look at the Medicaidprogram, the real innovations by states in recent yearshave been in finding new ways to match the unmatch-able: that is, in finding something you are doinganyway and matching it. The states have a strongincentive to “innovate” around whatever financialrestrictions the federal government attempts to placeon them. A tension inevitably develops and this iswhy I think block grants do not work very well in thelong run. Kincaid’s paper begins with a very clevercategorization of the different strategies for devolu-tion; the mixed strategies are the most interesting anddifficult cases.

Let me say one final thing about whether weshould be nervous about TANF. Look at Table 2. Whyshould the federal government be involved in socialpolicy at all at this stage? Let’s turn the questionaround. If the best argument for federal involvementyou can come up with is equalization of fiscal needsand some desire for uniformity across states, then youwould expect that the poor states would benefit morethan the rich states from federal involvement. ButTable 2 shows that by freezing in place the existingpayments, TANF is giving vastly more money to therich states than to the poor states. Indeed, we are

probably taking tax revenue out of Mississippi to giveto Connecticut as part of this devolution. This does notstrike me as a particularly stable or logical outcome inthe long term, although it is easy to see how it cameabout politically. Couple that unstable outcome withthe possibility of a recession, and you have goodreason to be nervous about the future impact ofTANF. When things get tough, states inevitably aregoing to have to start cutting benefits or taking othermeasures to cut costs. Before long, I think we will seethat, for example, state A will say, Look, we are goingto do things right in our welfare reform: We will havetraining, child care, and so on. And state B will say,You know, we want to do those things too: Here’s aticket to state A. I think that is a real danger that mustbe faced.

We are now seeing a period ofreal innovation, because states donot have money crunch problems.

But when the welfare caseloadcomes back up again during a

recession, the danger isvery real that the states

will start cutting benefits.

That said, we are now seeing a period of realinnovation, because states do not have moneycrunch problems. The welfare caseload is comingdown in a number of states, and they are redeploy-ing their resources and moving them elsewhere. Butwhen that caseload comes back up again during arecession, the danger is very real that states will notfind the new money necessary to support their newrecipients and instead will start cutting benefits.

Let me conclude by saying I thought Kincaidwrote a very provocative and helpful paper, makingthe point that precisely because TANF is ambigu-ous, we should be careful about calling it devolu-tion. It is a mixed strategy that is not very wellformulated. That makes it a challenge for those of uswho are looking for practical solutions to devolu-tion in the future.

Table 2TANF Grants per Poor ChildSelected States

StateMedian Household

Income (1993)

Approximate TANF Dollarsper Poor Child per Year in

FY 1997a

Arkansas $24,018 $ 360California $34,129 $1,850Connecticut $42,105 $2,150Florida $28,230 $ 775Mississippi $22,952 $ 350Wisconsin $32,201 $1,550aUsing 1993 Census estimates of poor children ages 5 to 17 multiplied by1.45 to account for children ages 0 to 4.Source: U.S. Bureau of the Census, Internet Table P93-00; Administra-tion for Children and Families, Internet table.

May/June 1998 New England Economic Review 47

DiscussionWilliam F. Fox, Professor of Economics and Director ofthe Center for Business and Economic Research,College of Business Administration, University ofTennessee at Knoxville

Like David Beam, I found this to be a reallyfascinating paper. John Kincaid did a wonderfuljob of summarizing so much information, and I

learned a great deal from reading it.Kincaid concludes in his paper that what we see

now is a nudge toward rebalancing federal and staterelationships, but certainly no wholesale devolution.And when welfare becomes the prime example we canname of devolution, it is easy to agree with hisconclusion. My office was involved with the state ofTennessee on issues of both Medicaid reform andwelfare reform. The waiver processes the states had togo through and the tight constraints that remain onstates do not provide much flexibility. My office con-tinues to work with both state agencies there, puttingtogether analytical summaries of new requirementsfrom the federal agencies. In this context, it is awfullyhard to conclude that much movement toward devo-lution is occurring.

Kincaid notes that devolution is either takingplace or being discussed by governments around theworld, and he gives three reasons for that trend. Firstare the important ethnic differences to be found withincountries. He singles out the East European countriesas prime examples of ethnically driven devolution.Second, escalating social welfare costs drive devolu-tion in other countries. And third is the growing beliefthat regional or local governments can provide ser-vices more efficiently. Kincaid uses each of thesereasons to explain why devolution is taking place inparticular countries. It seems to me that all threereasons apply in the United States, so why do we haveso little devolution in fact occurring here? Why canKincaid give as many explanations for devolution nothappening as for it happening? As a bit of a cynic, Iwas led to the conclusion that any level of governmentthat has power is awfully unwilling to yield it toanother level of government.

Outside the United States, in Russia, for example,since the reform a good deal of power has landed atthe state level rather than the national or local level.And the biggest opposition to getting devolutiondown to the local level in Russia has been the state

governments, which are unwilling to cede any of theirauthority to the local governments. For example, leg-islation pertaining to local governments just cannotget through the Federation Council, which is com-posed of the governors of each of the 89 regions.Similarly, some effort is also being made to try torecentralize power, and the states are the biggestopposition there. I think we are simply seeing thesame kind of unwillingness to yield power here in theUnited States.

Devolution is difficult to achievein the United States because the

federal government has suchproductive revenue sources that

it can generate more revenuethan it needs to deliver those

services it would be mostefficient at delivering.

It seems to me, looking at this from a fiscalperspective, that devolution is difficult to achieve inthe United States because the federal government hassuch productive revenue sources that it can generatemore revenue than it needs to deliver those servicesit would be most efficient at delivering. That extrarevenue gives the federal government control thattranscends the limits it ought to observe. My majorconcern about how devolution may progress arisesfrom the inability of state and local governments tocontrol their own revenue sources, particularly as theU.S. and the world economies become more interstateand more international. The most recent example ofthis is electronic commerce. It is increasingly difficultin such an environment for state and local govern-ments to collect revenues.

If we think about which taxes state governmentscould collect easily, clearly they would be taxes onproduction that occurs within their borders. But thatis exactly where the tax competition takes place. Sowhile states could efficiently collect taxes on produc-tion, they cannot levy much of a tax in that areabecause of the interstate competition for business.Several years ago, the state of Massachusetts contem-plated imposing a tax on the production of financial

May/June 1998 New England Economic Review48

services. It is no real surprise that financial serviceswould quickly have fled, just as banking has donealready through a variety of other forms, because ofthe type of tax structure. Another perfect example ofwhat is going on is the way states are now backing offfrom sales taxes on investment equipment purchasedby manufacturing firms. I do not mean to object: It isa good idea; it was also a good idea a long time ago.But what is different today is that the increasedcompetition between states means that productiontaxes simply will not be viable over the long term. Sowhere does this leave the states? It leaves them reallyneeding to be able to levy consumption taxes, becausethe same degree of competition will not occur withconsumption taxes.

But in order to levy consumption taxes effectively,the states require empowerment from the federalgovernment, because it is the federal government thatcontrols interstate activity. And a significant share ofthe production–consumption relationship takes placein an interstate environment. At a minimum, fordevolution to succeed, the states need federal supportthat will allow the collection of revenues when theproduction activity takes place in one location and the

For devolution to succeed, thestates need federal support that

will allow the collection ofrevenues when the production

activity takes place in one locationand the consumption activity

in another location.

consumption activity in another location. Withoutsuch federal enabling legislation, a big change willoccur in the way state governments collect revenue. Infact, the change is already under way. States continueto use the sales tax as their largest source of revenue,on average raising one-third of total revenues, but thishas required increasing the median rate from just over3 percent two decades ago to 5 percent today, and it israpidly approaching 6 percent. State sales taxes havegrown only slightly as a share of personal income, butthe rates have gone up 60 percent in order to do that.Without federal enabling legislation, the base thatstates are able to tax becomes smaller and smaller. If

we do not see the enactment of federal enablinglegislation, states will be forced to rely more onincome taxes and less on consumption taxes to raiserevenue; otherwise, they must rely more on the federalgovernment to collect the taxes and send the money tothem. But this would mean that devolution couldoccur only on the expenditure side, not on the revenueside, and I would argue that you cannot have devo-lution on just one side of the coin. So my concernabout devolution in the United States is focused on therevenue side.

Let me add a few comments about our recentwork in Tennessee. As we analyzed various welfarereform issues for the state, we realized immediatelythat the current federal legislation meant that at themargin, the state of Tennessee would have to pay 100percent of any increase in welfare costs, whereas in thepast it had paid roughly 33 percent. In the event of arecession, the state would have a serious problem,particularly given its extraordinarily inelastic tax sys-tem. In the 1990–92 period, for example, in two ofthose three years the elasticity of state revenue relativeto income was zero. The state was experiencing norevenue growth in an environment in which expendi-ture demands continued to rise; and in such a situa-tion it is inconceivable that the state would be willingto increase welfare expenditures at the margin in theway that would be required.

My advice to the state fiscal officers at the timewas to pass legislation that would allow the commis-sioner of finance to reduce money paid to each welfarecase by an amount bearing some relationship to theamount by which the state’s expenditures wouldotherwise rise. In other words, simply hold statespending to what was initially appropriated by havinglegislation already in place that permits the commis-sioner of finance to make appropriate adjustments.The legislation has not passed but I predict it will pass,come the first recession, because the state simply willnot have the resources to make those increases inspending. Fortunately for the states, the caseloadshave been falling. Tennessee’s 110,000 AFDC caseshad declined to 95,000 at the point at which welfarereform was passed; the level is now 64,000. We do notknow where the caseload is going to go from herebecause we have no information by which to makereliable estimates for recessions.

Let me summarize my remaining comments onKincaid’s paper as follows. I am surprised when I hearconcerns that one of the outcomes of devolution maybe different service levels across states. I thought thatwas the point of devolution. One of devolution’s

May/June 1998 New England Economic Review 49

options for states will be higher expenditures, butanother will be lower expenditures. I do not considerit opportunism if someone supports devolution as ameans to achieve smaller government. It may be thatperson prefers smaller government and that devolu-tion is the best way to achieve that goal.

If we take a broader look atwhat devolution offers, it is

clear that competition isexactly what we want.

I am also surprised when I hear concerns thatdevolution may lead to competition. Why is that aproblem? I thought competition was what we wantedto achieve. Admittedly, in the case of welfare, the

concern may be that competition would lead us downa path that some would find undesirable. But welfareis only a minuscule part of what we are considering,and if we take a broader look at what devolutionoffers, it is clear that competition is exactly what wewant. Indeed, when the World Bank and similarorganizations suggest devolution to developing coun-tries, stimulating competition and reaping its gains areprecisely why they are pushing for devolution.

Economists have observed for many years that alllocal governments can do effectively is allocation-typefunctions, while the federal government should dostabilization- and distribution-type functions. Econo-mists have now begun to reconsider that view and toreturn to those early papers and think through theassumptions they made, many of which simply arenot valid in the economy as it operates today. There isa much bigger role for stabilization by local govern-ments than economists’ conventional wisdom hasallowed. I wonder whether the same thing may not betrue about distribution as well.

DiscussionWilliam B. Modahl, Director of Tax Affairs,Digital Equipment Corporation

Having read John Kincaid’s wonderful paperand heard these interesting analyses, it iswith a sense of humility that I offer any

comments at all. (I would add that these are my views,and not necessarily those of my company.)

Kincaid opens his paper with the observationthat, strictly speaking, there can be no devolution inthe United States because from the beginning oursystem has been one of dual sovereignty. I wouldquestion that statement. It seems to me that thecurrent movement advocating devolution is trying torecover and restore state and local governmental func-tions that over a long period of time have beenusurped and centralized. The movement’s aim is toshift us closer to the original division of powers, whichwas a very well-thought-out scheme. Whatever onemay think of its appropriateness to current conditions,it was not the product of accident, as are the presentarrangements.

Of course, any consideration of which activitiesnow carried out by the central government mightsuitably be devolved back to state or local levels reallyposes a more fundamental question, and that is, whatshould our government be doing, and why? The factthat this question has become such a common topic ofdiscussion points up the growing unease with theconsensus built up during much of this century.Perhaps because of the two Great Wars, the GreatDepression, and the Cold War, we Americans felt weneeded a powerful central force to address these sortsof problems. And if circumstances pushed us in thatdirection, I think that intellectual currents pushed usthat way as well. The early and middle parts of thecentury were characterized by a lot of thinking aboutmarket failures. Welfare economics focused on howthe shortcomings of the market could be set right bythe actions of government, which was presumed tofunction in a benign and efficient manner. The won-derful image of James Buchanan comes to mind: It wasas if a singing contest were held between two sopra-nos and, upon hearing the first singer, the jurypromptly voted the prize to the second and adjournedwithout waiting to hear anything more.

Thinking back on my days in law school nearly 40years ago, I can recall the lack of interest in the insights

May/June 1998 New England Economic Review50

of the founding fathers. Their central insight was thatgovernment has a legal monopoly on coercion, so howdo you limit its exercise? That is, how can you limit theabuse of the state’s monopoly on coercion, so that theactivities of citizens can flourish? You need the gov-ernment’s monopoly on coercion to prevent fraud,abuse, and violence, but the monopoly itself becomesa potential threat. Rather than the founders’ vision ofgovernment, this century’s vision has been of a pow-erful, omni-competent institution that would bring ina better future through planning. Government wouldbe the center of action and innovation, reducing thecitizenry to passive recipients of government largesse,the beneficiaries of good planning. Citizens and thelower levels of government were not seen as playingan important or active role in civic life.

Rather than the founders’ visionof government, this century’svision has been of a powerful,

omni-competent institutionthat would bring in a better

future through planning.

But this concentration of power and the removalof restraints on the federal government required asubstantial remodeling of the Constitution, which wesometimes forget because it has been with us for solong. For example, the Sixteenth Amendment, whichpermitted the levying of an unapportioned incometax, provided the substantial federal revenue neededfor this new approach to government. The interstatecommerce clause and the general welfare clauses wereinterpreted by the Supreme Court in such a manner asto swallow up the original idea that the federal govern-ment was limited to enumerated powers and the stateswere to serve as the front line of government. The CivilWar amendments were interpreted so as to give thecentral government power over state action. The “tak-ings” clause of the Fifth Amendment, requiring paymentof compensation for property taken for public use, wasgreatly limited. Federal mandates, preemptions, condi-tions attached to federal aid, and the creation of newindividual rights all greatly limited state activities.

While these changes may no longer appear re-markable to us, they were in fact very dramatic

changes. And far from producing the benign condi-tions once hoped for or indeed assumed, this sys-tematic centralization of power and the removal ofconstraints on its exercise have led instead to dissat-isfaction and discontent. I think there is now a grow-ing feeling of governmental incompetence: Analysesof market failures have given way to analyses ofgovernment failure. Indeed, a whole school of eco-nomics, Public Choice, pursues this line of thought,and a number of recent Nobel Prize winners havegained recognition by their contributions in this area.

Current campaign finance revelations are point-ing up the interest group, rent-seeking aspect ofgovernment at the center. Perhaps we are forgettingthat much of the same activity goes on at the state andlocal levels, too. But in this general atmosphere ofdiscontent, two movements have emerged: the move-ment to reinvent government and the devolution move-ment. The movement to reinvent government seeks todeal with an overgrown and unaccountable bureaucracyby importing a concept of better customer service. Butthis movement in no way changes the fundamentalincentives controlling human action: It simply exhortspublic officials to adopt a better attitude towards publicservice. In the end, this movement simply perpetuatesthe idea of the citizenry as passive consumers of govern-ment rather than active and independent contributors tocivic life in de Tocqueville’s sense.

We often forget how much was carried out byprivate voluntary activity prior to the rise of theAmerican Leviathan. Beneficial associations, firefight-ing, education, and militias all depended upon volun-

Devolution offers promise in itscentral notion that government

functions can best be performed atthe lowest feasible level, as moreinnovation and creativity can be

expected when that happens.

tary associations of citizens and their activities, whichadded an important dimension to our democracy.Devolution offers promise in its central notion thatgovernment functions can best be performed at thelowest feasible level, as more innovation and creativ-ity can be expected when that happens.

May/June 1998 New England Economic Review 51

Another way of evaluating devolution is to con-sider that there are two ways of approaching anysocial problem: the first, through comprehensive plan-ning that attempts to anticipate every eventuality; thesecond, through preparation so as to be able to dealflexibly with rapidly changing conditions and circum-stances, without trying to spell out all the results orexactly how everything should be done in advance. Thefirst is a centralized, bureaucratic approach, and thesecond is more in accordance with the thinking of Hayekor von Mises, and is more characteristic of the businessworld, say, Silicon Valley. The devolution model maycorrespond most closely to the second approach. Cer-tainly the twentieth century is a history of the playingout of the impossibility of the first approach.

The biggest problem now is thatthere is not sufficient national

understanding of the fundamentalissues of government, much

less any consensus infavor of devolution.

The biggest problem now is that there is notsufficient national understanding of the fundamentalissues of government, much less any consensus infavor of devolution. While we see discontent with

certain federal programs, at the same time we seethings going in the other direction—the Presidentcampaigning to require school uniforms, for example.The American people do not say, “Well, that is a goodidea, but what business is it of the President’s?” Andthe members of Congress want to deal with truly localmatters like tort rules. The problem is that the citi-zenry is very poorly educated in these basic conceptsof civics, so how can one expect much real progresstoward devolution?

Still, from small beginnings combined with suc-cess in devolving individual programs, a significantmovement may grow. Positive signs can be seen instates such as Wisconsin and Michigan and in thecities too. Instead of just waiting for the federalgovernment to act, they are trying to take the lead incertain areas. The preliminary results of their effortsappear to be positive. Certainly the country seemsstarved for a sense of community, and to the extentthat we have tried to seek community in large federalprograms, we have not found it. One turns back toBurke’s idea of the little platoons, our relationshipswith the little communities around us on a daily basis.It is there that our sense of community, our relation-ship to our fellow citizens can be found. I have troublebelieving that such local programs may somehowbecome less humane or more indifferent to the suffer-ing of others than a central program distant from thecommunities in which we each live out our lives.Devolution will develop slowly and gradually, in myview, and I remain optimistic, even though very littlehas happened thus far and it will be a long road if wedo proceed very far in that direction.

May/June 1998 New England Economic Review52