DUE 3/26.

10CatoJ59.pdf

https://heinonline.org | holsupport@wshein.com

CITATIONS:

Bluebook 22nd ed. Thomas R. Dye, The Policy Consequences of Intergovernmental Competition, 10 CATO J. 59 (1990).

ALWD 7th ed. Thomas R. Dye, The Policy Consequences of Intergovernmental Competition, 10 Cato J. 59 (1990).

APA 7th ed. Dye, T. R. (1990). The policy consequences of intergovernmental competition. Cato Journal, 10(1), 59-74.

Chicago 18th ed. Dye, Thomas R. "The Policy Consequences of Intergovernmental Competition." Cato Journal 10, no. 1 (1990): 59-74. HeinOnline.

McGill Guide 10th ed. Thomas R. Dye, "The Policy Consequences of Intergovernmental Competition" (1990) 10:1 Cato J 59.

AGLC 4th ed. Thomas R. Dye, 'The Policy Consequences of Intergovernmental Competition' (1990) 10(1) Cato Journal 59

MLA 9th ed. Dye, Thomas R. "The Policy Consequences of Intergovernmental Competition." Cato Journal, vol. 10, no. 1, Spring/Summer 1990, pp. 59-74. HeinOnline.

OSCOLA 4th ed. Thomas R. Dye, 'The Policy Consequences of Intergovernmental Competition' (1990) 10 Cato J 59

Date Downloaded: Mon Mar 23 11:42:05 2026

Source: https://heinonline.org/HOL/Page?handle=hein.journals/catoj10&id=61

Terms, Conditions & Use of PDF Document: Please note, citations are provided as a general guideline. Users should consult their preferred citation format's style manual for proper formatting. Your use of this HeinOnline PDF indicates your acceptance of William S. Hein & Co., Inc. and HeinOnline's Terms & Conditions: https://help.heinonline.org/kb/terms-conditions/ . The search text of this PDF is generated from uncorrected OCR text. To obtain permission to use this article beyond the scope of your license, please use: https://www.copyright.com.

Please note: citations are provided as a general guideline. Users should consult their preferred citation format's style manual for proper citation formatting.

HeinOnline is a product of William S. Hein & Co., Inc. | 2350 North Forest Road, Getzville, NY 14068

THE POLICY CONSEQUENCES OF INTERGOVERNMENTAL COMPETITION

Thomas R. Dye

Competition and Public Policy Most debates over federalism are only lightly camouflaged debates

over policy. Philosophers and economists (Rawls 1971, Buchanan and Tullock 1962) may assert a distinction between constitutional issues-deciding how issues should be decided-and policy issues-deciding the issues themselves. But in politics, constitu- tional decisions are never separated from policy outcomes. People do know what the policy consequences of various constitutional arrangements will be. Citizens as well as political leaders consis- tently subordinate constitutional questions to immediate policy con- cerns. Indeed, history is replete with examples of the same political leaders arguing one notion of federalism at one point in time to achieve their immediate policy goal, and then turning around and supporting a contradictory notion of federalism at a latter time when it fits a new policy goal. No American politician, from Thomas Jefferson onward, has ever so strongly supported a view of federalism that he ended up conceding a policy battle.

Debates about federalism must acknowledge policy conse- quences. Policies have been usefully classified as allocational, devel- opmental, or redistributional (Lowi 1964). Allocational policies produce and distribute public goods and services to consumer-tax- payers. These policies encompass the provision of a broad range of state and local government services-education, health, welfare, streets and highways, police and fire protection, sewers, water and utilities, garbage disposal, parks and recreation, and so on. Develop- mental policies are those that directly enhance the economic well-

The author is McKenzie Professor of Government at Florida State University.

CATO JOURNAL

being of the state or community. These policies are directed toward economic growth; they include attracting industry, building trans- portation facilities, providing utilities, renewing urban areas, train- ing the labor force for work, and so on. Redistributional policies are designed to redirect wealth to benefit particular segments of society to satisfy equity concerns. These policies include traditional welfare services, health care for the poor, unemployment compensation, and low-income housing, as well as progressive taxation. Note that these are analytical distinctions among types of policies; any specific gov- ernment activity may have allocational, developmental, and redis- tributive elements within it.

Allocational Policy Intergovernmental competition directly strengthens the allocative

functions of government. Decentralization permits governments to match services with variations in demand. Greater overall citizen satisfaction can be achieved with multiple governments offering dif- ferent packages of public services at different prices. Competition forces governments to become more efficient in their allocative activ- ities, providing better services at lower costs. Competition forces government to be more responsive to citizens' preferences than monopoly government.

The major thrust of the original Tiebout model (Tiebout 1956) centered on allocational policies. And the value of intergovernmental competition in allocational policies has never been seriously chal- lenged. Paul E. Peterson, one of the few political scientists to incor- porate the notion of competition into a coherent theory of federalism, appears to agree:

Allocation is the functional that local governments can perform more effectively than central governments, because decentraliza- tion allows for a closer match between the supply of public services and their variable demand. Citizens migrate to those communities where the allocation best matches their demand curve [Peterson 1981, p. 77].

Developmental Policy Intergovernmental competition inspires state and local govern-

ments to be concerned with the impact of their taxing and spending policies on economic growth and to become directly involved in economic development activities. In recent years, international com- petition began to inspire similar concern in Washington. Prior to 1981, discussion of federal tax policy almost always centered on redistributional issues. Scholars and politicians who raised questions

INTERGOVERNMENTAL COMPETITION

about the growth-retarding effects of federal tax policies were ignored or ridiculed. But developmental issues have long been cen- tral to debates over taxing and spending in state capitols and city halls, and it is competition that has kept developmental concern in focus.

Monopoly governments have no direct way of estimating the growth-retarding effects of their tax policies. Cross-national compari- sons are confounded by vast cultural differences, and longitudinal examinations of the effects of past tax policies are obscured by histori- cal circumstances. Competing governments are in a better position to observe the economic consequences of their policy decisions.

Of course, we must distinguish between economic growth policy and state industrial policies. Economic growth policy is the provision of truly public goods and services with minimum burdens in taxes and regulations. State industrial policy generally involves special treatments-loans and subsidies, guarantees, and tax or regulatory exemptions-that in the long run contribute to economic ineffi- ciency. But the effects of various industrial policy decisions are not easy to identify. The literature to date has failed to observe any significant aggregate growth effects for any direct state industrial development activities (ACIR 1967; Peirce, Hagstrom, and Stein- bach 1979; Rasmussen, Bendick, and Ledebor 1984; Rubin and Zom 1985; Ambrosius 1986).

There are good theoretical reasons for predicting that direct state involvement in capital formation, as well as special tax treatments, will inevitably produce inefficiencies. But competition among the states in industrial policies is self-correcting in the long run. States will gradually learn from their mistakes as they compare their prog- ress with other states. Consider how much worse a national industrial policy would be-a monopoly government allocating capital, dispensing subsidies, and granting special privileges and protec- tions. (Not that the federal government does not already do so in innumerable tax code provisions and subsidy programs, but a national industrial policy would legitimize and enlarge the scope of these subsidies, privileges, and protections.) The only corrective to a national industrial policy would be international competition, and far more damage would be done to the nation before global competi- tion would exert its correcting influence. Perhaps the results of the industrial policy experiments currently being conducted in Ameri- ca's "laboratories of democracy" will succeed in discouraging the federal government from pursuing national industrial policy.

Thus, development policies are well served by federalism. Peterson et al. (1986, p. 230) argue convincingly that the federal

CATO JOURNAL

government would be wise to leave economic development policy to state and local government:

Since state and local governments are well equipped to pursue developmental objectives, most public efforts of this type should be left to them. By delegating responsibility for most developmental programs to state and local governments, the federal government would frankly admit its incapacity to use those programs to help populations with special needs.

Even strong proponents of state industrial policies warn against federal intervention. According to Osborne (1988, pp. 283-84):

The America economy-indeed the world economy-is made up of a series of regional economies, each of which radiates out from a city or network of cities. Each regional economy is unique. Each has a different mix of industries a different labor market, a different set of educational institutions, and different capital markets. In this country, the governmental unit that most closely matches the regional economy is the state.

Redistribution Policy The most serious challenge to intergovernmental competition

arises in redistributional policy. Can multiple, competing govern- ments undertake redistributive policies without creating unbearable free-rider problems for themselves? Will states and communities be restrained from providing the welfare services they would otherwise prefer because of the threat of an inundation of poor people from less-beneficent "free-riding" jurisdictions? Will each state and com- munity wait for other states and communities to provide welfare services in the hope that their poor people will migrate to the more generous jurisdictions? Will tax burdens to support generous welfare services encourage the nonpoor-both households and business- to migrate to jurisdictions that impose lighter tax burdens because they provide frugal welfare services?

Welfare and Competition in Theory

Median Voter Model Median voter models have provided economists and political sci-

entists with simplified sets of assumptions about democratic political systems-assumptions that have expressly or implicitly formed the basis of a great deal of empirical research in state and local govern- ment. In the basic model, each individual consumer-taxpayer tries to maximize the tradeoff between various public goods and services and disposable private income. Each consumer-taxpayer balances

INTERGOVERNMENTAL COMPETITION

the "utilities" obtained from higher levels of public services against the "disutilities" of higher taxes, in order to choose an "optimal" benefit level. After each consumer-taxpayer has made a choice based on his or her own preferences, tastes, needs, and so on, the choice of the median voter in any democratic political system should deter- mine public policy. The "median voter" label is taken from the early work of Duncan Black (1958), who suggested that democratically elected governments would seek to make the median voter's indiffer- ence curve the state's indifference curve.

A simple version of the median voter model treats the political system itself as a neutral conversion mechanism, transforming the needs, demands, and preferences of the median voter into public policy. The model is similar to the early systems model in political science literature (Easton 1965, Dye 1966). The preferences of voter- taxpayers were usually specified in state policy studies in terms of median or average characteristics of the population-for example, median family income, median school level completed, or percent- age of the population aged, young, black, urban, and so on. This required an inferential leap that characteristics of the populations determined median voter preferences and, hence, community prefer- ences. When variations in public policies among the states were shown to correlate with variations in population characteristics, this simple median voter model gained empirical support. The traditional literature on the "determinants" of state and local government taxing and spending policies relied implicitly on this simple median voter model (Sachs and Harris 1964, Fisher 1964, Bahl 1965, Dye 1966, Hofferbert 1966).

Redistribution and the Median Voter The median voter has an equity preference for redistribution. Few

of us want to see poverty, hunger, homelessness, ill-health, or depri- vation in our society, even if we do not expect to suffer these maladies ourselves. Moreover, some of us may wish to establish a floor for the human condition as an insurance policy against our own potential misfortunes. We may disagree on what minimal standards should be incorporated into this floor, but few would deny the existence of a general public interest in eliminating the worst conditions of poverty, or the greatest insecurities and afflictions of old age or disability, or the worse effects of discrimination.

The simple model of a median voter with a preference for some distributional government activities on behalf of poor, aged, young, and black populations can be developed further by assuming that these groups will contribute independently to the demand for redis-

CATO JOURNAL

tribution. Thus, increases in the poor populations should result in increased welfare spending through the direct political demands of the poor, as well as the charitable instincts of the median nonrecipi- ent taxpayer. It turns out, however, that empirical tests reveal that increasing numbers and proportions of poor (and black) populations have the opposite effect. This suggests that the redistributional pref- erences of the median nonrecipient taxpayer vary inversely with the proportion of poor and black in the population (Dye 1966, Orr 1976).

Redistribution and Policy Responsiveness The argument that intergovernmental competition disadvantages

the poor rests on the assumption that competition itself forces state and local governments to reduce welfare benefits below those pre- ferred by the median voter in their jurisdictions because of potential free-rider costs. States and communities are in economic competition with each other, and people, both rich and poor, can move freely from one jurisdiction to another. Governments are restrained from implementing the median voter's equity preferences for fear that so doing will attract poor people from other competing jurisdictions and repel affluent, productive, workers, firms, and investors.

Note that it is not sufficient that the proponents of this argument merely show that overall welfare benefits in the United States would be higher under a centralized government than they are under a competitive, federalism system. The true test of the effect of inter- governmental competition is whether it constrains the voter- preference-welfare-policy relationship, not whether welfare benefits would be raised by centralizing policymaking in Washington. Propo- nents of centralization must show that lower welfare benefits under decentralization reflect a poorer match between the true equity pref- erences of the median voter and public policy, than the match that would be obtained under centralized government.

My argument is that intergovernmental competition encourages policy responsiveness. Welfare policies, like all policies, are more responsive to citizen demands when undertaken by multiple compet- itive governments rather than monopoly government. The effect of competition is neither to lower nor to raise welfare spending but to bring it into line with citizens' demands. Competitive governments seek to match their welfare policies with both the compassion and the prudence of their citizens.

Competition and Redistribution The argument that intergovernmental competition disadvantages

the poor rests on three assumptions about human behavior-assump- tions that are empirically verifiable albeit with some difficulty.

INTERGOVERNMENTAL COMPETITION

First, the argument assumes that welfare benefits influence the decisions of the poor about where to live-that generous welfare benefits attract in-migrating poor and discourage resident poor from out-migrating in search of a better life.

Second, the argument assumes that affluent individuals, investors, and firms are repelled by the higher costs imposed by larger numbers of welfare recipients attracted by high benefits. By attracting "wel- fare seekers," states discourage productive labor and capital from coming to their states and may even encourage them to leave in search of lighter welfare burdens.

Third, the argument assumes that state and local policymakers act to avoid the free-rider problem-providing less redistribution than preferred by their median voter. This means that policymakers are aware of the benefit levels provided by sister states, and they use

these levels as guidelines in their own decisionmaking. Fearing an influx of poor, they deliberately restrict welfare benefits to levels

provided by their sister states, or perhaps even deliberately set bene- fit levels below those of their sister states to encourage the poor to

leave their own state.

Interpreting the Evidence Evidence relevant to these questions is important in discussions

of American federalism-in determining the appropriate level of

government for the provision of welfare services.

Early Research The early research literature contradicted the assumption that the

poor moved to maximize welfare payments. In a comprehensive review of this literature in 1974, sociologist Larry H. Long concluded that "no study has presented empirical evidence for the hypothesis that welfare payments themselves have attracted large numbers of

persons to states and cities with high benefit levels" (Long 1979, p. 46). Long's own research reached the same conclusion. Poor peo- ple are less mobile than affluent people. When poor people move, they do so for family reasons or job opportunities. The migrating destinations of the poor are no different from the affluent: They tend to migrate to locations that promise employment opportunities. Knowledge of welfare benefit levels and eligibility rules in distant states or cities is very limited and is rarely a part of the decision of the poor to move.

Shapiro v. Thompson The U.S. Supreme Court decision in Shapiro v. Thompson in 1969

altered the calculus of the poor in considering migration. The Court

CATO JOURNAL

ruled that States could not constitutionally impose residency require- ments on persons seeking welfare assistance. After 1969 welfare benefits became available to migrating poor in their new location as soon as their applications could be processed. Thus, it could be argued that earlier findings of the- irrelevance of welfare benefits in the locational decision of the poor were no longer valid, that welfare benefits would have a more substantial effect on the migration of the poor after 1969.

Migration to Low-Benefit States However, migration patterns in the United States over the past two

decades confirm that people have been moving to perceived job opportunities and ignoring welfare benefits. Net migration has been away from the high welfare benefit states of the Northeast and Mid- west and toward the low-benefit states of the South and West. Net migration in the United States is inversely related to welfare benefit levels in the states. The simple correlation between the proportion of new residents in the states and average monthly AFDC payments per family is -. 38.' This relationship was reported in the early litera- ture on labor migration (Galloway 1967).

It is possible, of course, that the poor could migrate opposite the net migratory flow. Net migration figures fail to tell us specifically about the migrating patterns of the poor. Several studies have attempted to disaggregate net migration figures, usually by race. These studies report that there are distinct differences between black and white migration patterns among the states. While the large white migration flow is toward states with low welfare benefits, black migration flows, are toward states with high welfare benefits (Som- mers and Suits 1973; Cebula, Kohn, Vedder 1973). These studies suggest that the poor do respond to welfare policy incentives by migrating to states offering higher welfare benefits. (For a summary, see Cebula 1979). These findings were offered some years ago as a justification for centralizing welfare policy (Orr 1976).

Higher Benefits and Immobility A strong theoretical argument can be made that high benefits create

immobilities among the poor. Generous benefits may weigh against the uncertainties of moving in search of job opportunities, thus caus- ing a state to retain a large poverty population. While theoretically

'New residents are those who moved to the state in the previous five years, from 1980 Census of Population. Average monthly AFDC payments are these reported for 1980 by Social Security Administration's Social Security Bulletin and reprinted in U.S. Bureau of the Census's Statistical Abstract of the United States 1982.

INTERGOVERNMENTAL COMPETITION

plausible, this thesis is difficult to study empirically. To learn why the poor do not move requires the causal study of a nonevent, always a difficult task.

Welfare Burdens and Migrating Wealth

Perhaps the weakest link in the argument that competition disad- vantages the poor is the assumption that high welfare costs discour- age economic growth. While theoretically plausible, it turns out that there is little convincing evidence that welfare costs raise tax levels to the point that affluent individuals, investors, or firms are encour- aged to leave states or discouraged from coming to them. Welfare expenditures are only 13 percent of total state and local government expenditures. (Education is 35 percent of total state and local govern- ment spending.) And welfare spending burdens are not related to economic growth in the states. We might observe, for example, in Table 1 that the welfare burdens (total state-local welfare spending as a percentage of personal income) are not significantly related to economic growth (growth in per capita personal income), either over longer or shorter time periods. In contrast, highway (and infrastruc- ture) spending is positively associated with growth in per capita personal income.

Of course, it is possible that the mere perception that welfare costs discourage economic growth may force policymakers in competitive states to reduce welfare benefits. We really need to see whether policymakers in the separate states eye each other in determining welfare benefit levels.

TABLE 1

STATE-LOCAL SPENDING BURDENS AND ECONOMIC GROWTH

Growth in Per Capita Personal Income Spending Burdens (Spending as a % of Income) at Beginning of Time Period 1960-84 1960-70 1970-80 1980-84

Welfare -. 04 -. 02 .12 .21 Education .31 .11 .38* .06 Highways .31* .07 .54* -. 13

NOTE: Figures are simple correlation coefficients for 50 states. An asterisk indicates a statistically significant relationship.

CATO JOURNAL

Variation in Benefits

Large variations among the states in welfare benefit levels suggests that decisionmakers do not give great weight to the benefit levels paid by other states in their own decisionmaking. If state comparisons had any significant effects on decisionmakers, we would expect to find greater convergence over time in levels of welfare benefits. No state would wish to "get out of line" in welfare benefits, fearing an influx of poor to generous states. Competing states would gradually move toward an equilibrium in benefit levels. But, in fact, differences among the states in welfare benefits are quite large and there is no trend toward convergence.

Average monthly AFDC payments in 1986, the principle state- administered welfare program, varied from highs of $532 in Califor- nia and $507 in Minnesota to lows of $116 in Mississippi and $114 in Alabama, a difference of nearly 400 percent. Moreover, these great differences have persisted over time. The coefficient of variation for average monthly AFDC payments among the states has remained fairly constant since 1940. (See Table 2; see also Peterson and Rom 1989, Table 1; Dye 1990, Table 2-4.) Continuing wide variation in benefit levels strongly suggests that welfare policy in each of the states is made independent of other states.

Peterson's Argument

The most influential report purporting to show that American fed- eralism disadvantages the poor is Peterson and Rom's recent analysis of the relationship between welfare benefits and the size of poverty

TABLE 2

CONTINUING POLICY VARIATION AMONG THE STATES, 1952-84

Per Capita State and Local Government Expenditures

Total Total Education Highways Welfare Health Taxes

1984 .45 .32 .55 .40 .32 .41 1980 .41 .32 .44 .36 .28 .21 1975 .28 .40 .42 .39 .38 .21 1970 .24 .20 .39 .43 .32 .21 1960 .20 .23 .34 .38 .31 .21 1952 .26 .44 .37 .49 .37 .23

NOTE: Figures are coefficients of variation: standard deviation divided by the mean. N = 50, except for 1952, when N = 48.

INTERGOVERNMENTAL COMPETITION

populations in the states (Peterson and Rom 1989). They present evidence that an increase in welfare benefits results in an increase in a state's poverty rate. Specifically, they found that a change in the combined maximum cash and food stamp guarantee was a significant, positive, independent determinant of a change in the percentage of a state's population living below the poverty line.

But does the fact that higher welfare benefits result in more poverty in a state really prove that the poor were attracted from other states? Certainly a strong case can be made that increased welfare benefits can increase the size of the within-state poverty populations. A con- vincing argument can be made that generous benefits and relaxed eligibility rules encourage the formation of female-headed house- holds, discourage recipients from obtaining entry-level jobs and the skills that these jobs could teach, and thereby increase poverty levels (Murray 1984). Increasing welfare benefits could increase the pov- erty population without any migration at all. Peterson and Rom do not present evidence that welfare benefits attract poor people to generous states.

Interestingly, Peterson and Rom report wide variations in welfare benefits among the states, yet continue to assert that the behavior of sister states motivates decisionmakers in state capitals. For evidence, they offer the fact that welfare benefits decline with increases in a state's poverty population. They dismiss a very plausible explanation of this relationship-that state policymakers tend to view welfare spending as a fixed proportion of state spending and, therefore, decrease benefits when the number of potential recipients increases in order to hold total costs relatively constant-because the savings from reduced benefits does not equal the costs imposed by a larger number of recipients. Instead, Peterson and Rom offer the less plausi- ble explanation that "politicians, interest groups, and the news media may ask the question where poor people are coming from and what is bringing them" (p. 723). Presumably these actors then decide to reduce benefits to discourage more in-migration of poor people.

Summary

In short, the evidence to support the critical assumptions of the argument that competitive federalism disadvantages the poor is weak. Early studies suggested that the poor migrate forjob opportuni- ties and family reasons with little knowledge of welfare rules and payments in various jurisdictions. More recent studies suggest that the poor migrate opposite net flows and toward high benefit states. But even if we concede this assumption, there is no convincing evidence to support the notion that migrating poor and their welfare

CATO JOURNAL

burdens repel affluent individuals or firms. Generous welfare bene- fits may have a modest effect on the poverty rate, but that effect is more plausibly attributable to the direct and indirect poverty-induc- ing effects of transfer payments. Finally, continuing wide disparities among the states in welfare benefit levels, notably AFDC cash bene- fits, indicate that the convergence we would expect if policymakers took their cues from other states has not occurred.

Equity Preferences and Welfare Benefits It is certainly true that the principles of efficiency and equity are

frequently in conflict in political and economic affairs (Okun 1975). American federalism may involve some tradeoffs between these val- ues. The political and economic benefits of intergovernmental com- petition may come at the costs of some equity considerations. If this is true, it should not necessarily doom the idea; society makes many such tradeoffs. Indeed, it is frequently argued that our society has traded off efficiency in so many of its redistributional programs that it threatens to impoverish both rich and poor. But let us examine the equity effects of federalism more closely. Perhaps it does not require a tradeoff of equity after all. Perhaps the poor will benefit from intergovernmental competition.

If the welfare policies of the national or state governments depended exclusively on the voting power of the poor, there would be very few welfare services in the United States. Welfare policies in the United States are primarily an expression of the equity prefer- ences of middle-class Americans, not a response to the demands of the poor themselves. Indeed, when we examine the determinants of the welfare policies of the states, we find that the states with the greatest numbers and percentages of poor people provide the least in the way of welfare services. The states with the most generous welfare programs, in both absolute terms and in proportion to their incomes, are those with the larger personal incomes. The most impor- tant determinant of public welfare benefit levels is the income level of the population. Our equity preferences rise with increases in personal income. It is better to be poor in a rich state than in a poor one.

If multiple competing governments are seeking to maximize the true preferences of consumer-taxpayers, they must do so in the provi- sion of welfare services as well as any other public goods. Is it reasonable to assume that migrating families and firms would wish to locate in areas of squalor, poverty, homelessness, and human degradation simply to avoid the costs of welfare services? The cross-

INTERGOVERNMENTAL COMPETITION

sectional evidence-that wealthy states provide more welfare serv- ices than poorer states-suggests that welfare services are preferred by many middle-class residents and wealthy firms.

The view that competitive federalism would result in the under- production of welfare services rests largely on the belief that the poor will migrate to the most generous jurisdictions. But the poor are the least mobile segments of society. Whatever their impact on state welfare services, it is certainly less than the impact of mobile upper- middle-class Americans and wealthy firms. Under competitive feder- alism, the most mobile sectors of the population and the economy exercise the greatest influence over public policy. And there is con- vincing evidence that these affluent segments of the population are more likely to support welfare services than society generally.

In short, intergovernmental competition tends to ensure policy responsiveness, Admittedly, it is not designed to maximize equity, that is, to encourage redistributional policies. But the preferences of consumer-taxpayers are not necessarily narrow and selfish. Competi- tive federalism ensures responsiveness to redistributional prefer- ences, as well as other policy preferences.

Matching public policy to citizen preferences is the essence of responsive government. Competitive elections and political parties were designed to achieve this goal. But decentralized government is also a way to match citizen preferences with public policies. Given variation in policy preferences across the nation and given indepen- dent governments with the authority to offer a wide range of policies at different costs, a truly federal government can achieve a far better match between preferences and policies than a centralized govern- ment imposing uniform policies throughout the nation.

References Advisory Commission on Intergovernmental Relations (ACIR). State-Local

Taxation and Industrial Location. Washington, D.C.: ACIR, 1967. Ambrosius, Margery M. "Effects of State Economic Development Policies on

the Health of State Economies." Paper presented at the Midwest Political Science Association, Chicago, 1986.

Bahl, Roy W. "Determinants of Changes in State and Local Government Expenditures." National TaxJournal 18 (March 1965): 50-57

Black, Duncan. Theory of Committees and Elections. Cambridge: Cam- bridge University Press, 1958.

Buchanan, James M., and Tullock, Gordon. The Calculus of Consent. Ann Arbor: University of Michigan Press, 1962.

Cebula, Richard J. "A Survey of the Literature on the Migration Impact of State and Local Government Policies." Public Choice 1 (1979): 69-83.

CATO JOURNAL

Cebula, Richard J.; Kohn, M.; and Vedder, Richard K. "Some Determinants of Interstate Migration of Blacks." Western Economic journal 11 (December 1973); 500-5.

Downs, Anthony. An Economy Theory of Democracy. New York: Harper and Row, 1957.

Dye, Thomas R. Politics, Economics, and the Public. Chicago: Rand McNally, 1966.

Dye, Thomas R. American Federalism: Competition among Governments. Lexington, Mass.: Lexington Books, 1990.

Easton, David. A Framework for Political Analysis. Englewood Cliffs, N.J.: Prentice Hall, 1965.

Fisher, Glenn W. "Interstate Variation in State and Local Government Expenditures." National Tax journal 17 (March 1964): 57-74.

Galloway, Lowell E. "The Economics of Labor Mobility." Western Economic Journal 5 (June 1967): 211-23.

Hofferbert; Richard. "The Relation Between Public Policy and Some Struc- tural and Environmental Variables in the American States." American Political Science Review 60 (March 1966): 73-82.

Long, Larry H. "Poverty States and Receipt of Welfare among Migrants and Nonmigrants in Large Cities." American Sociological Review 30 (February 1979): 46-56.

Lowi, Theodore J. "American Business, Public Policy, and Political Theory." World Politics 16 (July 1964): 677-715.

Murray, Charles. Losing Ground: American Social Policy, 1950-1980. New York: Basic Books, 1984.

Okun, Arthur M. Equality and Efficiency: The Big Tradeoff. Washington, D.C.: Brookings Institution, 1975.

Orr, Larry L. "Income Transfers as a Public Good." American Economic Review 66 (June 1976): 359-71.

Osborne, David. Laboratories of Democracy. Boston: Harvard Business School, 1988

Peirce, Neal R.; Hagstrom, Jerry; and Steinbach, Carol. Economic Develop- ment: The Challenge of the 1980's. Washington, D.C.: Council of State Planning Agencies, 1979.

Peterson, Paul E. City Limits. Chicago: University of Chicago Press, 1981. Peterson, Paul E.; Rabe, Barry G.; and Wong, Kenneth K. When Federalism

Works. Washington, D.C.: Brookings Institution, 1986. Peterson, Paul E., and Rom, Mark. "American Federalism, Welfare Policy,

and Residential Choices." American Political Science Review 83 (Septem- ber 1989): 711-28.

Rasmussen, David; Bendick, Bart; and Ledebor, Larry. "A Methodology for Selecting Economic Development Incentives." Growth and Change 15 (January 1984): 18-25.

Rawls, John. A Theory of Justice. Cambridge: Harvard University Press, 1971.

Rubin, Barry M., and Zom, C. Kurt. "Sensible State and Local Economic Development." Public Administration Review 45 (March/April 1985): 333-39.

INTERGOVERNMENTAL COMPETITION

Sachs, Seymour, and Harris, Robert. "The Determinants of State and Local Government Expenditures and Intergovernmental Flow of Funds." National Tax Journal 17 (March 1964): 75-85.

Sommers, Paul M., and Suits, Daniel B. "Analysis of Net Interstate Migra- tion." Southern EconomicJournal 40 (October 1973): 193-201.

Tiebout, Charles M. "A Pure Theory of Local Expenditures." Journal of Political Economy 64 (October 1956): 416-24.