Federalism.pdf

+(,121/,1( Citation: Sunita Parikh; Barry R. Weingast, Comparative Theory of Federalism: India, A , 83 Va. L. Rev. 1593 (1997) Content downloaded/printed from HeinOnline Thu Dec 6 05:50:19 2018 -- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and Conditions of the license agreement available at https://heinonline.org/HOL/License -- The search text of this PDF is generated from uncorrected OCR text. -- To obtain permission to use this article beyond the scope of your HeinOnline license, please use: Copyright Information Use QR Code reader to send PDF to your smartphone or tablet device

A COMPARATIVE THEORY OF FEDERALISM: INDIA Sunita Parikh*and Barry R. Weingast** I. INTRODUCTION

Why do federal states differ so widely in their economic and political performance? Some federal states, such as the United States, are among the richest and least corrupt in the world. Others, such as India and Mexico, are very poor and are mired in slow or negative growth. Mexico has also been plagued by considerable corruption. China, another de facto federal system, is very poor but is one of the fastest growing economies in the world. To address this question, we develop a comparative theory of federal performance and apply it to India. The performance of federal systems differs in part because federalism is not the only relevant variable influencing a country's economic and political success. Nonetheless, we argue that differential performance does reflect systematic differences among federal systems. Federalism is not a single type of system, but a family of disparate systems. Although all such systems have a hierarchy of governments, differences in federal architecture help to account for differential federal performance. To understand how differences in federal structure affect federal performance, we begin with the traditional arguments favoring federalism. These fall into two categories: economic benefits and non-economic benefits. The classic economic arguments about the benefits of federalism are threefold. First, Friedrich Hayek argues that the central government can never possess enough information to tailor policies to specific circumstances.' Because lower governments have better information about projects, policies, and citizen preferences, they will make better decisions about policies with a local impact Hayek's argument implies that, except for truly national public goods such as defense, a national, one-size-fits-all policy is not optimal.

"Assistant Professor of Political Science, Washington University, St. Louis. ""Senior Fellow, Hoover Institution, and Ward C. Krebs Family Professor and Chair, Department of Political Science, Stanford University. The authors wish to thank Lisa

Mclntosh-Sundstrom for editorial assistance. IFriedrich A. Hayek, The Economic Conditions of Interstate Federalism, in Individualism and Economic Order 255,268-69 (1948). 2See id. at 268.

1593

1594

VirginiaLaw Review

[Vol. 83:1593

Second, following Charles Tiebout,3 economists have built an impressive body of theory emphasizing two related benefits of federalism. 4 The first argues that the induced competition among jurisdictions forces political officials to attend to the economic and political consequences of their decisions.5 Second, to the extent that citizen preferences differ, competition among jurisdictions leads to an optimal mix of policies across jurisdictions.6 In this way, citizens and firms are matched with desirable policies of particular jurisdictions. Third, the literature on fiscal federalism focuses on the optimal assignment of policies and taxes across levels of government One principle emerging from this literature is that the provision of public goods should be assigned to the lowest jurisdiction compatible with producing that good.8 These theories have met with considerable debate and criticism. Truman Bewley shows that Tiebout's conclusions may not always hold.9 Susan RoseAckerman raises a problem ignored by most economists: local government corruption."0 James Buchanan raises serious questions about the economists' typical assumption of benevolent government." Finally, Robert Inman and Daniel Rubinfeld build a new theory on the political economy of federalism, showing how economic efficiency and democratic rights and virtues exist in tension in a confederate republic."2 Political scientists, in contrast, emphasize the political benefits that flow from federalism. Many argue, for example, that federalism helps avoid conflict in societies that are polarized geographically and provides for the protection of minorities in divided societies.'" In this view, federalism's decentralization of power helps to prevent different ethnic or religious groups from fighting over

- Charles M. Tiebout, A-Pure Theory of Local Expenditures, 64 J. Pol. Econ. 416, 416-24

(1956). 'See, e.g., Daniel L. Rubinfeld, The Economics of the Local Public Sector, in II Handbook of Public Economics 571 (Alan J. Auerbach & Martin Feldstein eds., 1987). 5 See Tiebout, supra note 3, at 419-20. 6See id. 7Wallace E. Oates, Fiscal Federalism (1972), provides the classic statement. 8 See, e.g., id. at 31-53. 'Truman F. Bewley, A Critique of Tiebout's Theory of Local Public Expenditures, 49 Econometrica 713 (1981). Several models show how intergovernmental competition limits but may not eliminate inefficiency. See, e.g., Paul N. Courant & Daniel L. Rubinfeld, On the Welfare Effects of Tax Limitation, 16 J. Pub. Econ. 289 (1981); Dennis Epple & Alan Zelenitz, The Implications of Competition Among Jurisdictions: Does Tiebout Need Politics?, 89 J. Pol. Econ. 1197 (1981). See Susan Rose-Ackerman, Corruption: A Study in Political Economy (1978). "See, e.g., Geoffrey Brennan & James M. Buchanan, The Power to Tax (1980); James M. Buchanan, Federalism As an Ideal Political Order and an Objective for Constitutional Reform, Publius, Spring 1995, at 2. 11See Robert P. Inman & Daniel L. Rubinfeld, The Political Economy of Federalism, in Perspectives on Public Choice 73 (Dennis C. Mueller ed., 1997). 13See, e.g., Donald L. Horowitz, Ethnic Groups in Conflict 646-48 (1985).

1997]

Response

1595

the policies about which they profoundly disagree. In these societies, majority rule in a centralized system could be disastrous as groups fight for power to promote their conflicting goals. This approach suggests that federalism and other forms of decentralization can sometimes prevent such fights. Both the economists' and political scientists' approaches to federalism suffer from a significant problem, however: They ignore the critical issue of how the rules of federalism are maintained. Consider the problem of assignment of political authority. This issue does not merely concern technical details of economic efficiency, as economists suggest. It is about the allocation of politicalpower to political units of jurisdiction. Jurisdictions and interests fight for power to be assigned where they have some influence; and once allocated, jurisdictions and interests fight to preserve and enhance their power. Most federal systems diverge considerably from the economists' prescription for the optimal assignment of powers over public goods and taxes. In contrast, the first 150 years of the United States remarkably paralleled the economists' prescription. Consider the Constitution's Commerce Clause. 14 The power of the federal government to regulate commerce is second nature today; so much so that, for the past sixty years, the Commerce Clause has imposed few restrictions on federal power. It is therefore remarkable that the federal government did not exercise this power during its first hundred years, until passage of the Interstate Commerce Act"5 in 1887 regulating the railroads. It is hard to imagine the federal government's activity so constrained in the modern United States-or anywhere in today's developing world. These observations raise the question, why do public officials so rarely abide by federalism's rules? For federalism to survive, political officials must have incentives to abide by federalism's rules. 6 In other words, the rules and constraints of federalism must be self-enforcing: Political officials must find it in their interests to abide by a series of rules and to respect a series of citizen rights. For example, officials in the national government must refrain from invading the policy do-

4 U.S. Const. art. I, § 8, cl. 3. 11Ch. 104, 24 Stat. 379 (1887). 16What follows draws on an emerging literature on federalism. See, e.g., Gabriella Montinola, Yingyi Qian & Barry R. Weingast, Federalism, Chinese Style: The Political Basis for Economic Success in China, 48 World Pol. 50 (1995); Barry R. Weingast, The Economic Role of Political Institutions: Market-Preserving Federalism and Economic Development, 11 J.L. Econ. & Org. 1 (1995); Jenna Bednar, Federalisms: Unstable by Design (Apr. 15, 1997) (unpublished manuscript, on file with the Virginia Law Review Association); Jenna Bednar, William N. Eskridge, Jr. & John Ferejohn, A Political Theory of Federalism (Apr. 1996) (unpublished manuscript, on file with the Virginia Law Review Association); Rui J. de Figueiredo, Jr. & Barry R. Weingast, Self-Enforcing Federalism: Solving the Two Fundamental Dilemmas (Apr. 1997) (unpublished manuscript, on file with the Virginia Law Review Association).

1596

Virginia Law Review

[Vol. 83:1593

mains of the lower jurisdictions." Correspondingly, officials in lower governments must refrain from encroaching on the common market. The issue of self-enforcing federalism reflects a more general problem of how political officials respect the rules of government and the rights of citizens. Implementing critical features of an ideal government and society will always involve this problem. Democracy, for example, requires that political officials respect citizen rights and obey a range of rules, such as holding periodic elections according to agreed-upon procedures and allowing citizens' votes to determine electoral outcomes. Also central to democracy is a respect for civil and personal rights, such as freedoms of speech and assembly, a requirement that punishment be commensurate with the crime, and a requirement that citizens must be charged in order to be jailed. Political officials in most countries today adhere only incompletely, if at all, to these rights. Finally, for markets to thrive political officials must maintain a range of private rights, notably, a complex series of private property rights, the enforcement of contract and commercial law, and a stable tax and macroeconomic regime. The rarity with which political officials adhere to a comprehensive system of the rules and rights-such as those required to sustain democracy, federalism, or private markets-implies that when they do, it is a remarkable phenomenon requiring an explanation. The answer cannot be the country's constitution alone, for this begs the deeper question of why anyone obeys a constitution. Because political officials in most countries do, at times, ignore fundamental aspects of their constitutions, a special set of circumstances must have to hold for political officials to obey a constitution. Scholars have paid too little attention to the problem of how the rules and rights associated with constitutions, democracy, markets, and federalism are enforced. A group of scholars, whom Gibbons and Rutten call the "equilibrium institutionalists,"18 have recently begun to study this problem. These works suggest two related observations. First, a series of political institutions is typically necessary to solve particular social dilemmas and thus to ensure social 17 See William H. Riker, Federalism: Origin, Operation, Significance (1964); Weingast, supra note 16, at 4; Bednar, Eskridge & Ferejohn, supra note 16, at 5. "ISee Robert Gibbons & Andrew Rutten, Hierarchical Dilemmas: Social Contracts with Self-Interested Rulers 6 (Oct. 8, 1996) (unpublished manuscript, on file with the Virginia Law Review Association). Other works in this tradition include Bednar, supra note 16; Randall L. Calvert, Rational Actors, Equilibrium, and Social Institutions, in Explaining Social Institutions 57 (Jack Knight & Itai Sened eds., 1995); Avner Greif, Paul Milgrom & Barry R. Weingast, Coordination, Commitment, and Enforcement: The Case of the Merchant Guild, 102 J. Pol. Econ. 745 (1994); Paul R. Milgrom, Douglass C. North & Barry R. Weingast, The Role of Institutions in the Revival of Trade: The Law Merchant, Private Judges, and the Champagne Fairs, 2 Econ. & Pol. 1 (1990); Barry R. Weingast, American Democratic Stability and the Civil War: Institutions, Commitment, and Political Behavior (Dec. 1996) (unpublished manuscript, on file with the Virginia Law Review Association).

1997]

Response

1597

cooperation. One such dilemma is that repeated play is often insufficient to enforce cooperation: for example, play is too infrequent, citizens do not have sufficient information, or they fail to coordinate their actions. Institutions can help resolve these dilemmas and sustain cooperation. Second, to succeed, these institutions must be self-enforcing because they are themselves objects of choice. If citizens do not have mutual incentives to maintain them, these institutions will ultimately collapse. Using the notion of self-enforcing, limited government, we show below that the form of federalism studied by economists requires a set of implicit political assumptions called "market-preserving federalism."1 9 Part II discusses the five axioms of market-preserving federalism. In brief, these are: 1) a hierarchy of governments and division of authority exists; 2) subgovernments have primary political authority over the regulatory and police powers concerning the economy; 3) the national government has authority to police the lower governments (in particular, to assure the common market); 4) governments face a hard budget constraint; and 5) some form of institution provides a credible commitment to the entire structure (that is, federalism must provide for self-enforcing government). An important benefit of the axiomatic approach to market-preserving federalism is that it provides the basis for a comparative theory of federalism. The approach allows us to predict the economic and political performance of federal systems with different characteristics. Federalisms without a common market, for example, will exhibit far less inter-jurisdictional competition and thus far less experimentation and adaptation of policies to the economy. These governments will also exhibit more corruption. A central government that violates the fifth condition is not really federal at all. In such states, the national government typically attempts to control or interfere with the subgovernments, precluding the main benefits of market-preserving federalism. In Mexico, for example, because states receive the lion's share of their revenue from the national government, they adhere to the national government's policies."' A

19 See Ronald I. McKinnon, The Logic of Market-Preserving Federalism, 83 Va. L. Rev. 1573 (1997) [hereinafter McKinnon, Commentary]; Ronald I. McKinnon, Market-Preserving Fiscal Federalism in the American Monetary Union, Spectrum, Summer 1995, at 36 [hereinafter McKinnon, Market-Preserving Fiscal Federalism]; Montinola, Qian, & Weingast, 29 supra note 16; Weingast, supra note 16. Montinola, Qian & Weingast, supra note 16, at 55. 2 Anwar Shah, an economist with the World Bank, reports that "Mexican states in 1990 derived 60% of their revenues from the federal governments as revenue sharing or transfers." Email from Anwar Shah, The World Bank, to Barry Weingast (Sept. 30, 1997) (on file with the Virginia Law Review Association). In violation of the spirit of marketpreserving federalism, Mexico's "central government exercises overwhelming control over state and local governments." John Bailey, Fiscal Centralism and Pragmatic Accommodation in Nuevo Le6n, in Opposition Government in Mexico 173, 173 (Victoria E. Rodrfguez & Peter M. Ward eds, 1995).

1598

Virginia Law Review

[Vol. 83:1593

similar conclusion holds in federal systems designed to protect minority rights: The absence of the fifth condition threatens federalism's purpose. In Part III, we apply our approach to India. In brief, we show that India fails to conform to the ideal of market-preserving federalism by failing to satisfy several of its axioms. In violation of the second and fifth provisions, the national government has the authority to impose presidential rule, by which it may take over the government of a particular state, clearly compromising state independence. Similarly, India's central planning system lodges too much control over the economy with the central government, precluding the benefits of inter-jurisdictional competition. In sum, India's federalism compromises all the benefits that should accrue from state sovereignty under marketpreserving federalism. This approach to studying Indian federalism yields different conclusions from those reached in Does Federalism Preserve Markets?22 by Jonathan Rodden and Susan Rose-Ackerman ("Rodden/Rose-Ackerman"). Rodden/Rose-Ackerman study India as a means for exploring the implications of market-preserving federalism. Although they admit that India does not conform to marketpreserving federalism, they seem to argue that because India is federal and because India has failed to foster economic growth, market-preserving federalism fails to foster markets. In the absence of a comparative theory of federalism, they appear to have trouble distinguishing among different federal systems. Our comparative theory of federalism shows why federal systems systematically differ and, in particular, why some have poorly performing economies. This approach suggests that India's systematic departure from market-preserving federalism helps explain its poor economic performance. Our comparative approach to federalism demonstrates that there is no logical connection between federalism, per se, and governmental promotion or preservation of markets. We identify one type of federalism that promotes markets, labeling it market-preserving federalism. But as various of the conditions characterizing market-preserving federalism are removed, a federal system's incentives to promote markets are weakened or eliminated. The theory makes clear that assessing a particular federal system's ability to promote markets requires assessing which conditions it satisfies. II. A

COMPARATIVE THEORY OF FEDERALISM

All federal systems involve decentralized political authority, though not all forms of decentralization constitute federal systems. To understand federalism, we must identify its principal characteristics. The first condition is a defining characteristic of any federal system: 22

Jonathan Rodden & Susan Rose-Ackerman, Does Federalism Preserve Markets?, 83 Va. L. Rev. 1521 (1997).

1997]

Response

1599

Fl: A hierarchyof governments with a delineated scope of authority (for example, between the national and subnational governments) exists so that each government is autonomous within its own sphere of authority.2Y Beyond this condition, federal systems differ enormously in their political and economic performance. To build a theory of how their political and economic characteristics differ, we begin with a special type of federalism called market-preserving federalism.4 Formalized decentralization alone is insufficient to preserve markets; rather, a system must possess further conditions concerning the allocation of authorities and responsibilities among different levels of government. These conditions also prove useful for predicting the differential performance of particular types of federal systems. F2: The subnational governments have primary authority over the economy within their jurisdictions. F3: The national government has the authority to police the common market [from encroachments by the states] and to ensure the mobility of goods and factors across subgovernment jurisdictions. F4: Revenue sharing among governments is limited and borrowing by governments is constrained so that all governments face hard budget constraints. F5: The allocation of authority and responsibility has an institutionalized degree of durability so that it cannot be altered by the national government either unilaterally or under the pressures from subnational governments.-' These conditions represent an ideal type of institutional arrangement of market-preserving federalism. From the perspective of preserving market incentives, the authority of the national government over markets is limited to policing subgovernmental shirking (here represented as F3: subgovernment encroachment on the common market) and providing national public goods, such as defense and a stable macroeconomic regime. The institutional arrangements of federalism recognize a critical difference between the national government and the subnational governments: There is only one of the former, but there are many of the latter. Competition among jurisdictions induces limits on the discretionary authority of the subnational governments. A necessary condition for this competition to be beneficial is the absence of trade barriers, so that the entire nation becomes a common market as required by F3. Without F3, each subnational government would 1-Montinola, Qian & Weingast, supra note 16, at 55. 24See McKinnon, Market-Preserving Fiscal Federalism, supra note 19, at 37; Montinola, Qian & Weingast, supra note 16, at 55; Weingast, supra note 16, at 4. 1 Montinola, Qian & Weingast, supra note 16, at 4. See also Weingast, supra note 16, at 4 (discussing the same conditions).

1600

Virginia Law Review

[Vol. 83:1593

become something of a de facto national government in its jurisdiction, shortcircuiting federalism's limits on lower governments. Condition F2 enhances the effects of Condition F3: If decentralization remained at the discretion of the national government, the latter could intervene in the economy by using its discretion (in the absence of F2) first to compromise the system of federalism and then to intervene. Condition F4 applies to both the national and subnational governments and has two parts: fiscal transfers between levels of governments and government borrowing.6 Although Condition F4 does not preclude revenue sharing among levels of government, the hard budget constraint limits the ways in which revenue can be shared or equalized. In particular, it prevents lower governments that perform increasingly poorly from getting increasingly larger subsidies from higher levels of government. The hard budget constraint restricts open-ended access to capital markets, especially borrowing from the central bank. For lower governments this is necessary to tie local revenue to local economic prosperity: A local government's financial problems remain its own. This condition provides important incentives for local officials, as their government's fiscal health is directly related to local economic prosperity. If, in contrast, local governments were readily bailed out of their financial problems, they would not need to worry about the consequences of their choices. The hard budget constraint on the national government is necessary to prevent monetary discretion and inflation as attempts to get around the constraints on its authority mandated under F2. Condition F5 provides for credible commitment to the federal system. This condition requires that, beyond simple decentralization, the federal structure must not be under the discretionary control of the national government. Condition F5 concerns the enforcement problem and is critically important. Due to different histories and unique social, political, and economic situations, each country is likely to resolve this condition in a unique way. For many large countries with diverse economies, market-preserving federalism's balance of power between the national and subnational governments is superior to either a centralized unitary government or complete decentralization with each region as an independent state. In the latter two cases, the national government's authority is not limited through internal institutional arrangements; hence, the danger exists for the discretionary authority to encroach on markets.

26See

McKinnon, Market-Preserving Fiscal Federalism, supra note 19, at 37-38; see also

David E. Wildasin, Externalities and Bailouts: Hard and Soft Budget Constraints in Intergovernmental Fiscal Relations (May 1997) (unpublished manuscript, on file with the Virginia Law Review Association) (discussing soft and hard budget constraints in relation to the size of localities).

1997]

Response

1601

Before turning to the economic implications of market-preserving federalism, we pause for a few observations about the relationship between the market-preserving federalism model and various systems of federalism throughout the world. Our approach shows that whether a nation describes its political system as federal is irrelevant. What matters for federal performance is the combination of conditions that hold. Many de jure federalisms are nothing like market-preserving federalism. For example, in Mexico, Conditions F2, F4, and F5 fail. In this federal system, the lion's share of state revenue comes from the national government.8 This raises several problems. First, it breaks the link between local economic prosperity and fiscal health. Second and perhaps more importantly, along with the revenue come restrictions, rules, and regulations from the center. Money from the central government carries the implicit threat, made explicit in Mexico,2 9 of withdrawing funds if the lower government chooses to disobey. Local governments in these systems have neither the incentive nor the ability to differentiate themselves from their neighbors. More broadly, the failure of F2 and F5 implies that the political discretion and authority retained by the central government greatly compromise its market-preserving qualities. In sum, though many forms of political decentralization exist, marketpreserving federalism characterizes only a narrow subset.? A. The Economic Effects of Market-PreservingFederalism A critical feature of market-preserving federalism is that it limits the exercise of arbitrary authority by all levels of government. Federalism limits the central government directly by placing particular realms of public policy beyond that government's reach. For lower governments, constraints are imposed in two ways. First, under Condition F3, the central government polices state abuses of the hierarchy, such as encroachments on the common market

7See Oliver E. Williamson, The Institutions and Governance of Economic Development and Reform, in The Mechanisms of Governance 322,333-35 (1996). z Bailey, supra note 21, at 173. 29 For example, when local governments elect opposition parties to power, the national government cuts off many of their discretionary funds. See Victoria E. Rodrfguez & Peter M. Ward, Conclusion: Regents From the Opposition, in Opposition Government in Mexico, supra note 21, at 223, 227. ("But both PRD and PAN municipal governments have suffered insofar as discretionary lines of funding for more elaborate special projects and programs (dams, highways, housing projects, and so forth) are concerned."). - Notably, unlike de jure federalisms such as Mexico, 18th century England was characterized by market-preserving federalism, although the English did not call their system federal. See Weingast, supra note 16, at 15-17.

1602

Virginia Law Review

[Vol. 83:1593

F3. Second, the induced competition among lower jurisdictions places selfenforcing limits on these governments' ability to act arbitrarily?' These limits have a number of salutary effects. First, no government has a monopoly of regulatory authority over the entire economy. The national government's absence of regulatory authority prevents it from creating monopolies and other forms of inefficient economic intervention that plague developing countries. Competition limits the ability of subnational governments to create monopolies and other policies that cripple markets, because doing so would place firms in its jurisdiction at a considerable disadvantage. When a particular jurisdiction imposes an onerous restriction on its firms, the firms face a competitive disadvantage relative to competing firms from less restrictive jurisdictions. A further beneficial effect of market-preserving federalism is that competition among jurisdictions extends to factors of production, such as capital and labor. This induces jurisdictions to provide a hospitable environment for factors, typically through the provision of local public goods such as secure rights of factor owners, provision of infrastructure, utilities, and access to markets. Those jurisdictions which fail to provide these goods find that factors generally move to other jurisdictions. Although old firms may not move old plant and equipment to new jurisdictions, they will locate new investments there. In less competitive jurisdictions, local economic activity and tax revenue therefore decline. For example, the economic rise of the American South since the 1960s reflects in part firms moving to jurisdictions with fewer regulatory and labor restrictions. 32 Third, the hard budget constraint (F4) implies that local governments can go bankrupt. This provides lower governments with incentives for proper fiscal management. Local enterprises, politicians, and citizens hardly want their government to spend more money than is prudent, as bankruptcy would greatly hinder the ability of local governments to finance necessary public goods, such as those needed to attract foreign capital and lower business costs. Finally, market-preserving federalism provides a secure political foundation for markets.33 By resting the regulatory authority over markets with lower governments, market-preserving federalism induces them to foster local economic prosperity. In addition, by limiting national authority over the economy, it also prevents the national government from causing the massive

3' This

Section summarizes an extensive literature in economics, including the classic

work of Tiebout and Oates. See Oates, supra note 7; Tiebout, supra note 3. For a review of this literature see Rubinfeld, supra note 4. 32See, e.g., Ronald I. McKinnon, Market-Preserving Fiscal Federalism in the American Monetary Union, in Macroeconomic Dimensions of Public Finance: Essays in Honour of Vito Tanzi 73, 89 (Mario I. Blejer & Teresa Ter-Minassian eds., 1997). 1 See Weingast, supra note 16.

1997]

Response

1603

political distortion of markets typical of developing countries. Once marketpreserving federalism is established, markets are harder for governments to plunder: National political forces are deflected from such tendencies, and, at the local level, it is hard for any one interest group to capture the lion's share of the local government. That, in turn, implies that at least some governments are always likely to retain their pro-market focus. Market-preserving federalism therefore diminishes the prevalence of rent-seeking and patronage systems. The latter can survive only in areas with political protection from market forces, a phenomenon that market-preserving federalism is designed to eliminate. Market-preserving federalism therefore holds considerable promise for developing countries. The limits enforced by this type of federalism prevent much harmful and often crippling intervention by governments in the economy. B. PredictionsFollowing the Inception of Market-PreservingFederalism Following the inception of a system of federalism, we should observe a diversity of policy choices and experiments by local governments. People in different jurisdictions are likely to have markedly different interests, expectations, and capabilities. In making their decisions, jurisdictions may also consider markedly different theories and ideologies. We should therefore observe that, after the inception of market-preserving federalism, lower governments choose a range of policies to promote their goals. As the results of the new policies and experiments become known, citizens and policymakers around the country will update their expectations about the effects of various policies. Decentralization under market-preserving federalism therefore results in an important degree of feedback that would not be present under a unitary system that imposes a single national experiment over all regions. The competitive process among jurisdictions induces those that initially chose poor strategies to adopt variants of the strategies that succeed elsewhere. To the extent that some jurisdictions are better at promoting markets, generating wealth, and caring for the needs of their citizens, their policies are likely to be imitated by others that have been less successful. Still, we do not expect the appearance of uniformity for several reasons. First, citizens and firms will sort themselves into jurisdictions. For example, to the extent that different industries require different types of public goods, they may locate in different areas, which in turn provide different types of services. Second, differing resources and access to markets among jurisdictions imply that a variety of economic and political strategies will survive.- Fi-

mSee, e.g., Paul Krugman, Geography and Trade (1991).

1604

Virginia Law Review

[Vol. 83:1593

nally, citizens are likely to vary in their tastes for public goods, as well as their ability to pay for them. C. PredictionsWhen Market-PreservingFederalismRemains Incomplete The set of axioms used to define market-preserving federalism also forms the basis for a comparative theory of federalism. The economic and political performance of federal systems varies systematically with the combinations of the conditions they satisfy. In this Section, we go beyond market-preserving federalism to discuss implications of three other types of federal systems, characterized here by the particular combination of axioms they satisfy. The first type of federalism satisfies all but the common market axiom (F3), therefore allowing lower jurisdictions to erect trade barriers. The common market is highly unlikely to be sustained without explicit protection from the central authorities."' This implies that some areas, particularly those not likely to perform well under competition with other jurisdictions, are likely to erect trade barriers against firms and products from other areas. A federalism of this sort (one which is only incompletely market-preserving) will produce seemingly contradictory results. Some areas will be observed to promote markets while others will closely control their economy, especially to prevent influence from outside the jurisdiction. The absence of a common market also implies far less pressure against political corruption, so corruption is likely to be higher in those jurisdictions that raise high trade barriers. We also expect this variance in performance across lower jurisdictions to be more pronounced just after a system of federalism is imposed than in a more mature system. Thus, for an economy like China's, with only limited experience with markets, economic performance across provinces in the late 1980s differed enormously. 6 As it has become clear that those provinces fostering markets have gotten rich, the variance among provinces has diminished as others attempted to imitate those that have succeeded.37 Nonethe-

31For example, the common market in the early 19th century United States could not have been sustained without the ever-vigilant policing of the Supreme Court. See, e.g., Gibbons v. Ogden, 22 U.S. (9 Wheat.) 1 (1824) (striking down New York State's grant of an exclusive right to operate steamboats). Policing the market against encroachments by state governments proved a major use of its constitutional powers. These cases reveal the remarkable diversity and cleverness of the states in their efforts to erect such barriers. See Barry R. Weingast, The Political Foundations of Antebellum American Growth 20 (Sept. 1, 1993) (unpublished manuscript, on file with the Virginia Law Review Association). Similarly, such barriers are a major reason underlying the movement for economic and political union in Europe. See Geoffrey Garrett, International Cooperation and Institutional Choice: The European Community's Internal Market, 46 Int'l Org. 533 (1992). 1 See Montinola, Qian & Weingast, supra note 16, at 73-79. 31See id. at 77-78.

1997]

Response

1605

less, many interior provinces remain "dukedom" economies insulated from the world, without free markets, and highly corrupt. " Limited exposure to markets naturally generates suspicion of them and of their potential for generating dependence on outsiders. The feedback provided by multiple jurisdictions that conduct independent experiments is therefore important to address these concerns. As market growth occurs in some areas, the incentives for other jurisdictions to pursue protectionism will diminish. The experience with markets will reveal new information about how they allow local governments to provide for the needs of citizens. Even in the presence of strong trade barriers, fiscal pressures will push insulated areas to substitute market mechanisms for those activities in which the market has proven elsewhere to be a superior provider of particular goods and services. A second type of federal system satisfies all the axioms except Condition F4: It lacks centralized control over the monetary system. For example, to the extent that the authority over credit is decentralized, so that it remains at least in part at the discretion of lower governments, several problems are likely to emerge. The most obvious is inflation, as each government overgrazes the commons, causing too much growth in the money supply. The second problem is a consequence of the first. Decentralized access to credit under these circumstances also softens the hard budget constraint, as governments that increase their exposure can always borrow more in the short run. This induces moral hazard problems; for example, a jurisdiction may borrow too much to finance too many investments, many of which would not be financed were it not for access to credit in this manner. Decentralized access to credit also allows lower jurisdictions to bail out ailing enterprises, compromising economic incentives imposed by market discipline. A final type of federalism consists of a center that can impose its will on the lower governments. Various types of federal systems allow this behavior: if the central government holds regulatory authority over the economy (violating F2); if it provides lower governments with revenue (violating F4); or if the federal system remains at the central government's discretion, allowing it the unilateral authority to alter or remove the authority of the lower governments (violating F5). Because each of these federal systems compromises lower jurisdiction autonomy, all of them behave more like centralized nations than nations characterized by market-preserving federalism. These nations typically prevent lower governments from deviating from national policies. With respect to the economy, lower governments in these federal systems constitute administrative units of the central government, not autonomous or sovereign governments. If the national government does not want

8Id. at 65-66.

1606

Virginia Law Review

[Vol. 83:1593

a market economy, it has the power to prevent lower governments from fostering markets. D. Does FederalismPreserveMarkets? We end this Part with the question asked by Rodden/Rose-Ackerman in the title to their article: Does federalism preserve markets? Although their title is a clever twist on the phrase "market-preserving federalism," it reflects a fundamental alteration in the nature of the issue. The theory sketched in this Part suggests that federal systems differ too considerably to have a uniform effect on markets. Indeed, the discussion above implies that many federal systems are very poor at preserving markets. The answer to Rodden/Rose-Ackerman's question must therefore be "no." Our approach suggests further that the analysis cannot stop here. The point of the theory of market-preserving federalism is to provide an approach that explains which types of federalism protect markets. Federal systems that differ from this ideal in predictable ways will have economies that differ from an ideal market's society in predictable ways. Thus, federal systems that violate a common market assumption will lose all benefits from competition among jurisdictions. Lower governments will erect trade barriers and substantial interventions in their economies. Forms of federalism which violate Conditions F2 and F5 will have too powerful a center, also compromising federalism's market-preserving qualities. III.

FEDERALISM IN INDIA

In their critique, Rodden/Rose-Ackerman use India as a case study for exploring the conditions of market-preserving federalism. They acknowledge that in India the central government is too powerful to meet marketpreserving federalism criteria, but they point out that the state-level governments have had important responsibilities since independence and that recent economic reforms now encourage competition among the states. 9 Rodden/Rose-Ackerman attempt to make two important points with the Indian case: First, they use it to cast doubt on the viability of market-preserving federalism,' ° and second, they draw on Indian examples to argue that corruption will not necessarily decrease as power is decentralized.' Rodden/Rose-Ackerman are correct in observing that India has an authentic federal system and that some choices are being devolved to the states under economic liberalization. But rather than providing an experiment for how a

19 Rodden & Rose-Ackerman, supra note 22, at 1524-25. 40 See id. at 1537-43, 1546-66. 41Id. at 1537-40.

1997]

Response

1607

developing country might function under market-preserving federalism, India instead demonstrates precisely why it is important to differentiate market-preserving federalism from other types of federalism. The comparative theory of federalism sketched in Part II allows us to suggest how different center-state relations affect economic performance. In this Part, we consider how India measures up to the criteria for market-preserving federalism. India meets Condition F1, that of minimal federalism. Its federal structure is not in dispute; it is regularly cited as an exemplar of the type.42 The third criterion (Condition F3), that of a common market, is also essentially met. But India fails to meet the second, fourth, and fifth criteria. The central government retains enormous control over the economy, setting most economic laws and regulations; state governments have little discretion in revenue raising and spending. 43 States do not face a hard budget constraint. And finally, the center has important unilateral powers with respect to state powers.' The framers of the Indian Constitution created a strong center that could, many Indians believed, direct economic growth, protect minority rights, and defuse the political and economic tensions that might arise in such a heterogeneous nation.45 The Indian revenue collection system places states at the mercy of the central government. The central government raises the bulk of its revenue through taxes and distributes that revenue to the states according to the recommendations of several central agencies.4' States may raise revenue in only a few areas on their own, chiefly sales, liquor excise, and various fees.47 States therefore have been dependent on the various commissions and ministries. Two primary central bodies hold authority over revenue disbursement: the Planning Commission and the Finance Commission. The Planning Commission was set up to implement the Five Year Plans, and it has been the cornerstone of India's economic development strategy from independence to the recent economic reforms. Although the commission's powers have ebbed and flowed, reaching their high-water mark in the 1960s, it still controls sub-

2 Arend Lijphart, Democracy in Plural Societies 181 (1977); Riker, supra note 17, at 120-22. 43See B.S. Grewal, Fiscal Federalism in India 15-18 (1974). See Asok Chanda, Federalism in India: A Study of Union-State Relations 68-72 (1965) (discussing the historical background of the center's authority over the states). 45See, e.g., Granville Austin, The Indian Constitution: Cornerstone of a Nation 189-91 (1966). 46 See Grewal, supra note 43, at 15-32. 47See K.S. Krishnaswamy, I.S. Gulati & A. Vaidyanathan, Economic Aspects of Federalism in India, in Federalism in India: Origins and Development 180, 188 (Nirmal Mukerji & Balveer Arora eds., 1992) (suggesting that, "Though States were given the exclusive right to agricultural taxation, political conditions made its exploitation extremely difficult, if not virtually impossible.").

1608

Virginia Law Review

[Vol. 83:1593

stantial portions of the total revenue transfers from the center to the states.4 These expenditures include those stipulated by central development plans, state-level plans, and centrally sponsored schemes. This last program has become increasingly important both as a patronage tool and as a way for the center to dominate investment and distribution decisions. While the Planning Commission controls Plan expenditures, the Finance Commission has authority over all non-Plan disbursements. The commission's main job is to address the gap between committed expenditures and the state's expected revenues.f It does not rank or evaluate different types of expenditures, however, and its main goal is to avoid deficits in non-Plan areas." The lack of coordination between the two commissions creates a soft budget constraint for the states. Because the Finance Commission tends to provide funds to make up shortfalls, states have an incentive to commit to expenditures regardless of funds. This behavior by states emerged soon after independence. In the 1960s, for example, an astute observer remarked that "states indulge in competitive importuning, putting up scheme after scheme to attract funds, and then happily run up big deficits by failing to collect their own share .... ."" In the 1950s and 1960s the two commissions controlled almost all revenue transfer decisions. 53 But in the last two decades, other central government agencies have acquired considerable influence over economic policies. Since all foreign projects must be routed through ministries or central financial institutions, 4 the recent introduction of economic reforms does not necessarily mean greater decentralization to the states. Moreover, the bulk of industrial policy, especially through regulation and licensing, is controlled by centrally appointed boards and agencies. This system, known in India as License Raj, means that the center retains control over the distribution of permits and licenses for new areas of economic development through the relevant central ministry.55 Although recent central governments have pledged to loosen the grip of License Raj,56 the center still reserves the right to decide which changes will be made. Apart from the distribution of revenue and returns from investments, states can potentially increase their total revenue by borrowing. But the states'

See id. at 193-95. See id. at 203-04. oChanda, supra note 44, at 188-225. I See Krishnaswamy, Gulati & Vaidyanthan, supra note 47, at 189-90. 52 W.H. Morris-Jones, The Government and Politics of India 144 (1964). 53See Chanda, supra note 44, at 188-225,260-94; Grewal, supra note 43, at 25-32. 14Jagdish N. Bhagwati, India in Transition: Freeing the Economy 49-50 (1993). 51See id. at 50-51 41

16

Id. at 86-87.

1997]

Response

1609

ability to borrow is limited. The Constitution mandates that, as long as a state owes outstanding loans to the center, it cannot raise loans without the center's consent. 7 Although this rule is not always followed, states do borrow much more from the center than from the market."8 This has two results: First, the states have racked up enormous debt to the center, with an increasing share of their resource transfers to the center taken up by debt service charges.59 Second, this makes states further reliant on center decisions. Why would provincial leaders agree to such an asymmetric division of economic control in a heterogeneous, federal country? A number of explanations emerge from the debates that took place during the drafting of the new Indian Constitution. It should be emphasized that the Constituent Assembly distinguished between economic and political decentralization. While political decentralization was seen as inevitable, economic decentralization was never seriously contemplated given the differences in language, ethnicity, and previous forms of government (e.g., between the British colonial provinces and the princely states). Nehru and the socialist wing sought to implement central economic planning because they believed it would more quickly create a strong industrial base. Leaders more receptive to capitalist theories went along because they were afraid that open markets, especially those that were permeable by international investments, would wipe out nascent local industries that had already been damaged by centuries of colonial domination. 6' They accepted central government control as the lesser of the possible evils that could befall Indian capitalism. In addition, many feared that without central control, the disparities between rich and poor states would increase. Both types of states were well represented in the Constituent Assembly, and their divergent interests were apparent from the outset. At one point the governments of the rich states of West Bengal and Bombay suggested that they receive tax shares that reflected their contributions.? Their proposal would have given these states, which represent a mere seventeen percent of the total population, sixty-two percent of all income tax revenues.? Although these suggestions were never adopted, they revealed the potential problems that could arise if states were given greater economic powers. Central dominance was thus considered to be a way to avoid destructive interstate disputes and increase equality across

India Const. pt. XII, ch. II, art. 293. See Grewal, supra note 43, at 18-19. S9 Krishnaswamy, Gulati & Vaidyanathan, supra note 47, at 196. 6 Austin, supra note 45, at 45. 61See id. at 60-61. 6 Id. at 222-23. 3 Id.

1610

Virginia Law Review

[Vol. 83:1593

disparate regions. These deliberations did result in a system in which the national government makes all the important economic decisions. But it is far from clear that centralization led to greater equality. The data reveal important differences in the resources transferred to rich and poor states; these differences are the greatest in those areas in which the central government has the greatest discretion. Between 1956 and 1981, in discretionary transfers made by central ministries, per capita transfers to low income states were only seventy-seven percent of those to high income states; middle income states did somewhat better at eighty-six percent. 4 Nondiscretionary expenditure decisions were slightly more equitable but revealed the same type of disparity: Statutory transfers were twenty percent higher for middle income than low income states, and plan transfers were ten percent higher. ' The distribution of revenue surpluses reveals similar disparities. Because of the division of responsibilities between the Financial and Planning Commissions, a few states, primarily those with higher per capita income, have regularly been left with substantial revenue surpluses.66 For example, in the 1980s, poor Bihar's non-Plan surplus was only ten percent of that of rich Punjab. These surpluses are important because they can they be plowed back into further development schemes, creating even wider disparities in the future. Finally, in violation of Condition F5, the Indian Constitution enhances the central government's economic control over the states still further. 6 Specifically, the central government has unilateral control over federal provisions in several important ways. For example, state boundaries can be redrawn by a simple majority in Parliament.4 Even more importantly, the Indian Constitution provides that the President of India may dissolve state governments in three circumstances: threats to national security by external aggressors, a breakdown of a state's constitution, and financial crisis. 69 This power was envisaged for use in infrequent and extreme cases, and it was never invoked between 1950 and 1960.70 Since 1962, however, its usage has increased, and it has become highly politicized. Although the financial crisis provision is rarely invoked (the most common is the second case, breakdown of the Constitution)," greater economic autonomy for states could change this pattern. 64

Krishnaswamy, Gulati & Vaidyanathan, supra note 47, at 195-96. Id. at 195. 6 67 Id. For a discussion of the Indian federalist system in comparison with those of other countries, see Durga Das Basu, Introduction to the Constitution of India 49-63 (9th ed. 1982). 61 India Const. pt. I, art. 3. 69See India Const. pt. XVIII, arts. 352, 356, 360. 10Austin, supra note 45, at 216. 71Basu, supra note 67, at 302-16 (discussing emergency powers generally). 64

1997]

Response

1611

If economic strategies in a state ran counter to the interests of the parties in power at the center, the financial crisis provision could become as politicized as the breakdown provision. Despite Rodden/Rose-Ackerman's contention that India provides a good case with which to assess market-preserving federalism, India's failure to meet the approach's criteria makes it an inappropriate test for the theory. The Indian case far better illustrates what occurs in the absence of marketpreserving federalism. The ways in which India has failed to develop are consistent with the predictions of our theory about federal performance when the various conditions of market-preserving federalism are not present. To summarize, India's federalism departs considerably from market-preserving federalism, failing to meet criteria F2, F4, and F5. Violating F2 grants the central government sufficient power over the economy to prevent competition among the states, which results in an absence of state policy experimentation and innovation. It also prohibits the natural matching of policies to local conditions, which in turn inhibits specialization and exchange. The absence of F4 means that states have no financial incentive to be concerned about the effects of their policies. Finally, India's violation of F5 provides the national government with additional leverage over the states. In short, India's federalism retains the hierarchy of federalism but eliminates the main mechanisms that sustain strong markets. States are not free to set their own economic policies. Nor can they capture the gains from policies that foster economic growth. Although Roddeh/Rose-Ackerman overstate the extent to which India can be considered a fair current or future example of market-preserving federalism, they are correct to point out that a certain amount of economic decentralization and state innovation is taking place there. 2 From this evidence they conclude that market-preserving federalism cannot succeed in India. Although we believe that it remains too early to tell, considerable evidence indicates otherwise. For example, Gujarat and Karnataka are two states that have taken greatest advantage of economic liberalization. As market-preserving federalism predicts, the provision of local public goods, such as infrastructure and utilities, has increased in these states, 3 and it can be argued that their advantages over other states, such as an educated labor force and a history of indigenous capital, have helped them attract foreign investment. Rodden/Rose-Ackerman emphasize the importance of political corruption in distorting prospects for economic growth. They argue that, contrary to the predictions of market-protecting federalism, corruption is likely to increase in 72

Rodden & Rose-Ackerman, supra note 22, at 1524-25. "See Charan D. Wadhva, Economic Reforms in India and the Market Economy 127-29 (1994); Energy: Need for a Comprehensive Policy, DATA india, Sept. 24, 1995, at 764, 765; Lukewarm Response to Telecom Tenders, DATA india, July 9,1995, at 556,557.

1612

Virginia Law Review

[Vol. 83:1593

developing democracies if economic decentralization occurs, and they use evidence on civil service transfers to support their claim. 4 Unfortunately for their argument, the evidence runs counter to their position: Despite high levels of civil service transfers in Gujarat, economic growth is increasing there." In order to attribute properly the effect of corruption on economic growth, it is necessary to have a theory of corruption and its effects. Although some forms of corruption, such as pervasive and widespread rent-seeking, clearly inhibit investment and growth, others become accepted as another cost of doing business. In successful Indian states such as Gujarat and Karnataka, corruption exists, but it is routinized and predictable, and it is not sufficiently high to deter capital that desires to locate in India. In conclusion, although the rhetoric of decentralization in India has increased, it remains to be seen whether state autonomy will increase in practice. In the 1997 budget announcement, the Finance Minister recommended that central government resources be consolidated into a single fund and twenty-nine percent be set aside for the states. 7 6 Although this would increase their current share, it is still far from parity. In addition, the central government is clearly moving toward market reform and away from socialist planning. Nonetheless, it is far from apparent that the central government will yield primacy to the states. IV. CONCLUSIONS

Rodden/Rose-Ackerman ask a critical question: Does federalism preserve markets? Our answer is that nothing inherent in federalism either promotes or preserves markets. Federal systems differ too widely to have a uniform effect on the economy. To address this thesis, we sketched a comparative theory of federalism and then applied it to India. In Part II, we showed how a particular kind of federalism, called market-preserving federalism, protects and husbands markets. Those federal systems that diverge from marketpreserving federalism are unlikely to foster thriving markets. Further, we ar74 75

Rodden & Rose-Ackerman, supra note 22, at 1538. Wadhva, supra note 73, at 127-28. In addition, Rodden/Rose-Ackerman do not distin-

guish between preexisting corruption and the increased corruption they hypothesize would occur with economic decontrol. Corruption is widespread in Indian politics and admini-

stration, Paul R. Brass, The Politics of India Since Independence 53 (1990), and there may well be a correlation between the central government's decision to initiate investment and a state's level of corruption. But the mere existence of high levels of civil service transfers cannot be used to conclude that corruption is increasing or is presenting a threat to economic development and social welfare. The politicization of civil service transfers is not in question; what remains to be determined is the extent to which these postings systematically inhibit economic and social welfare across diverse states. 76Finance Minister P. Chidambaram, 1997-1998 Union Budget Speech before the Lok Sabha (March 1997), available in India on Internet, Budget '97 (visited Sept. 18, 1997)

.

1997]

Response

1613

gued that the theory's predictions conform to reality. Those federalisms roughly characterized by market-preserving federalism's conditions, such as the United States or contemporary China, have thriving markets. Those that fundamentally diverge from market-preserving federalism tend to be developing states mired in poor growth, such as Argentina, Mexico, and India. In each of the latter systems, the central government has too much power over the economy and the lower governments to allow these systems to exhibit market-preserving federalism. Sustaining federalism requires that political officials obey a series of rules concerning the allocation of political power. This issue raises a larger question. Many of the principal values associated with good government and society-including democracy, the rule of law, and thriving markets-require that political officials respect a range of citizen rights. Because there are no external authorities to enforce this respect, these rights must be self-enforcing in the sense that political officials have sufficient incentives to abide by them. That political officials in most countries today fail to do so suggests that this is a profound problem, which economists and political scientists studying federalism have neglected. Given this dilemma, we know too little about how federalism is sustained. Our focus on market-preserving federalism suggests some of the political prerequisites necessary for federalism to sustain markets. To do so, it must meet more than the minimal condition of federalism, a hierarchy of governments. Beyond a hierarchy of government, the economic benefits of federalism require that: the national government's powers over the economy be limited (F2); there must be a common market (F3); all governments must face a hard budget constraint (F4); and institutions must credibly commit political officials at all levels to these restrictions (F5). These conditions provide the political foundations underlying the economists' approach to federalism, though economists rarely specify or discuss them. The theory makes explicit the allocation of political powers among the different levels of government. In particular, federalism requires that the national government must, somehow, be restricted in its powers to providing truly national public goods, considerations of equity, and policing the federal system. Market-preserving federalism's hard budget constraint and its induced competition among jurisdictions limit the exercise of powers by lower jurisdictions. Federal systems can fail to be market-preserving in a variety of ways. In the absence of Condition F3, lower governments have the power to erect trade barriers, thereby insulating their economies from competition. This power obviates many of the benefits of federalism by allowing corruption, interest groups influence, and rent-seeking. When Condition F2 fails, the national government has too much power over the economy. If this occurs in combination with national government control over lower government reve-

1614

Virginia Law Review

[Vol. 83:1593

nue, lower governments tend to become administrative arms of the national government. Federal systems of this sort remain only nominally federal, sacrificing many of the economic benefits of lower government independence and failing to exhibit the economic benefits of market-preserving federalism. We applied our perspective to India, in part responding to Rodden/RoseAckerman's claims about Indian federalism. India's federalism has never conformed to the conditions of market-preserving federalism, nor was it designed to do so. Part of the purpose of federalism in India concerns the problem of maintaining harmony among different ethnic and religious groups. Turning to the economy, India's central government simply has too much77 power over the states to be characterized as market-preserving federalism. The dominant role of the central government in economic planning, regulation, taxation, and redistribution implies that Condition F2 fails. The ability of the national government to redraw state boundaries and to take over state governments by declaring presidential rule implies that Condition F5 fails. The central government's powers thus allow it to overwhelm the states, and in doing so, to allow the interventionist national policies that cripple the Indian economy. Nonetheless, we agree with Rodden/Rose-Ackerman that India is unlikely to adopt anything comparable to market-preserving federalism.m The requisite restrictions on government would harm a large range of interests that benefit from the current system and would oppose restrictions that limit these benefits. What would happen if India moved toward greater state autonomy, as required by market-preserving federalism? Rodden/Rose-Ackerman argue that the result would be far more corruption. We responded by suggesting that this prediction is only partially correct. Even if corruption initially increased in most states, this is not likely to be universal. This is so because there are some states and areas that we argue are greatly harmed by the current set of restrictions, such as Maharashtra, Gujarat, and Karnataka. These states would be likely to use new freedoms to foster markets and to attempt to become rich in just the way that Guangdong has done under China's decentralization. Further, just as in China, the success of a few states in India is likely to change politics in other states. After observing what can happen elsewhere, citizens in other states will pressure their political officials, asking why their state has remained mired in corruption while others have been getting rich. An essential feature of the current pattern of corruption is that the central government prevents state officials from pursuing independent policies. More

nJoachim Ahrens, The Political Institutions of Economic Development: Experiences,

Failure and Prospects in East and South Asia, in 88 Diskussionsbeitrige aus dem Volkswirtschaftlichen Seminar der Universit~it Gbttingen 25-26 (June 1996). 78 Rodden & Rose-Ackerman, supra note 22, at 1525.

1997]

Response

1615

local state freedom in India might initially result in more corruption in many states, but we predict that this would not be the final political equilibrium. We therefore disagree with Rodden/Rose-Ackerman's claim that India provides an excellent opportunity to explore the validity of market-preserving federalism's predictions. Yes, India has performed poorly on many dimensions. But, along the economic dimension, this is in part because India systematically diverges from market-preserving federalism, not because the theory of market-preserving federalism fails.