¹ Some of the most important theorizing on the politics of decentralization has grown out of these two cases, including Shirk, Susan, The Political Logic of Economic Reform in China (Berkeley: University of California Press, 1993)Google Scholar; Montinola, Gabriella, Qian, Yingyi, and Weingast, Barry R., “Federalism, Chinese Style: The Political Basis for Economic Success in China,” World Politics 48 (October 1995)CrossRefGoogle Scholar; and Treisman, Daniel, After the Deluge: Regional Crises and Political Consolidation in Russia (Ann Arbor: University of Michigan Press, 1999)CrossRefGoogle Scholar.

² For an overview, see Ter-Minassian, Teresa, ed., Fiscal Federalism in Theory and Practice (Washington, D.C.: International Monetary Fund, 1997)Google Scholar. Out of the seventy-five developing and transitional countries with populations greater than five million, all but twelve claim to be in the process of effecting some form of transfer of political power to local units of government. See Dillinger, William, Decentralization and Its Implications for Urban Service Delivery, Urban Management Programme Paper (Washington, D.C.: World Bank, 1994)CrossRefGoogle Scholar.

³ Classic statements include Tiebout, Charles, “A Pure Theory of Local Expenditures,” Journal of Political Economy 64, no. 5 (1956)CrossRefGoogle Scholar; and Musgrave, Richard A., The Theory of Public Finance: A Study of Public Economy (New York: McGraw-Hill, 1959)Google Scholar. See also Oates, Wallace E., “An Essay in Fiscal Federalism,” Journal of Economic Literature 37 (September 1999)CrossRefGoogle Scholar.

⁴ Weingast, , “The Economic Role of Political Institutions: Federalism, Markets and Economic Development,” Journal of Law, Economics and Organization 11 (April 1995)Google Scholar; Montinola, Qian, and Wein-gast (fn. 1); Parikh, Sunita and Weingast, Barry R., “A Comparative Theory of Fiscal Federalism: India,” Virginia Law Review 83 (October 1997)CrossRefGoogle Scholar; Hehui Jin, Yingyi Qian, and Barry R. Weingast, “Regional Decentralization and Fiscal Incentives: Federalism, Chinese Style,” Working Paper (Hoover Institution, Stanford University, March 1999).

⁵ Contrast Lapidoth, Ruth, Autonomy: Flexible Solutions to Ethnic Conflicts (Washington, D.C.: United States Institute of Peace Press, 1996)Google Scholar with the skeptical view in David Lake and Donald Rothchild, “Political Decentralization and Civil War Settlements” (Manuscript, University of California, San Diego, 1999)

⁶ Among these exceptions are Treisman (fn. 1); Inman, Robert P. and Rubinfeld, Daniel L, “The Political Economy of Federalism,” in Mueller, Dennis, ed., Perspectives on Public Choice Theory (New York: Cambridge University Press, 1997)Google Scholar; and Panizza, Ugo, “On the Determinants of Fiscal Centralization: Theory and Evidence,” Journal of Public Economics 74, no. 1 (1999)CrossRefGoogle Scholar.

⁷ Shah, Anwar, The Reform of Intergovernmental Fiscal Relations in Developing and Emerging Market Economies, Policy and Research Series (Washington, D.C.: World Bank, 1994)CrossRefGoogle Scholar.

⁸ We consider a high level of discretion in intergovernmental fiscal transfers to be an indicator of centralization for reasons we outline in more detail below. However, discretion may in fact provide sub-national governments with bargaining leverage and open the door to moral hazard problems and blackmail. If the subnational government knows that the center has the discretion to provide bailouts, for example, then there are few incentives to control expenditure and borrowing. Moral hazard can in principle be limited by a no-bail-out rule, but the capacity to set such a rule is endogenous to politics. See Eichengreen, Barry and Hagen, Jurgen von, “Federalism, Fiscal Restraints and European Monetary Union,” American Economic Review 86, no. 2 (1996)Google Scholar.

⁹ Riker, , “Federalism,” in Greenstein, Fred and Polsby, Nelson, eds., Handbook of Political Science, vol. 5 (Reading, Mass.: Addison-Wesley, 1975)Google Scholar.

¹⁰ Four of the five countries in this category have federal constitutions; Colombia is the exception. For a more detailed review of political arrangements in the region, see Willis, Eliza, Garman, Christopher da C. B., and Haggard, Stephan, “The Politics of Decentralization in Latin America,” Latin American Research Review 34, no. 1 (1999)Google Scholar.

¹¹ We will also put aside for the moment analysis of unitary systems without subnational jurisdictions (to our knowledge no such cases exist) or in which the executives of the subnational jurisdictions are appointed by the center (a quite common form, particularly in authoritarian regimes). However, the political dynamics outlined here are easily extended to such systems.

¹² For a review of the related literature, see Bolton, Patrick, Roland, Gerard, and Spolaore, Enrico, “Economic Theories of the Break-up and Integration of Nations,” European Economic Review 40 (April 1996)CrossRefGoogle Scholar.

¹³ We will assume for the sake of argument that the status quo is “overcentralized” compared to the optimum, although the opposite can also be the case, as with Brazil.

¹⁴ For empirical evidence of such behavior, see Horcasitas, Juan Molinar and Weldon, Jeffrey A., “Electoral Determinants and Consequences of National Solidarity,” in Cornelius, Wayne, Craig, Ann L., and Fox, Jonathan, eds., Transforming State-Society Relations in Mexico: The National Solidarity Strategy (San Diego, Calif.: Center for U.S.-Mexican Studies, 1994)Google Scholar; Treisman (fn. 1); Moritz Kraemer, “Intergovernmental Transfers and Political Representation: Empirical Evidence from Argentina, Brazil and Mexico” (Washington, D.C.: Inter-American Development Bank, 1997).

¹⁵ Seabright, Paul, “Accountability and Decentralization in Government: An Incomplete Contracts Model,” European Economic Review 40 (January 1996)CrossRefGoogle Scholar.

¹⁶ This assumption is subject to the important caveat in fn. 8.

¹⁷ The ordinal preference ranking among different outcomes (for example, centralized, decentralized with central control, decentralized without central control) remains the same, but the distance in absolute value between the respective preference value for each system increases.

¹⁸ Uncertainty about transfers from the center might also lead opposition governors to raise their own revenues by taxation.

¹⁹ For the purpose of this argument, we are assuming the existence of a median governor. In fact, governors may have different preferences depending on the wealth of their state and access to locally generated tax revenues.

²⁰ In the most comprehensive analysis of the issue Carey and Shugart identify four factors that determine these electoral incentives: party control over nominations; the existence of vote pooling; whether voters cast only a single vote for a party, multiple votes, or a single vote for a candidate; and district magnitude. Carey, J. M. and Shugart, M. S., “Incentives to Cultivate a Personal Vote: A Rank Ordering of Electoral Formulas,” Electoral Studies 14, no. 4 (1995)CrossRefGoogle Scholar.

²¹ This last factor is emphasized by David Julian Samuels, “Careerism and Its Consequences: Federalism, Elections, and Policy-Making in Brazil” (Ph.D. diss., University of California, San Diego, 1998).

²² In some countries nomination procedures vary according to political party, but in our coding we have sought the central tendency across parties.

²³ Such an influence arguably presumes that electoral districts mirror subnational boundaries. That assumption remains valid for most of the cases examined here; there are some exceptions, however. In Colombia members of both the lower and upper houses were elected in electoral districts that were coterminous with subnational political units prior to 1991. As a result of constitutional changes in 1991, however, the upper house is now elected in a single national district. In Mexico three-fifths of the lower house is elected in districts that are not coterminous with lower administrative units. In Venezuela electoral districts mirrored subnational boundaries prior to 1993, but subsequent to a 1993 reform half of the lower house has been elected in districts that are not coterminous with subnational boundaries. However, even where the overlap between electoral and subnational districts is not perfect, it is still plausible that subnational politicians could influence legislative behavior.

²⁴ Despite this effect of the electoral system, we have kept nomination rules within parties as one of our two indicators precisely because it is an indicator of party organization and lines of accountability within parties, rather than a determinant of it. Even in open-list PR systems like Brazil, where party leaders exert relatively little influence over party members in comparison with Venezuela's closed-list system, decentralized nomination rules provide an important indicator of the relative influence wielded by subnational politicians.

²⁵ Carey and Shugart (fn. 20).

²⁶ Weingast (fn. 4).

²⁷ These data are drawn from Inter-American Development Bank, Latin America after a Decade of Reforms: Economic and Social Progress in Latin America, 1997 Report (Washington, D.C.: Inter-American Development Bank, 1997), 196–97Google Scholar.

²⁸ Because Mexican revenue data include transfers from other levels, these numbers overstate the level of decentralization. We explore this issue in more detail below.

²⁹ Provincial governments in Argentin a increased their share of total tax revenues from 13.7 percent in 1983 to 15.4 percent in 1992. This expansion came at the expense of local governments, which suffered a decline from 7 to 4.6 percent in th e same period, whil e the share controlled by the federal government remained virtually unchanged. In Colombia, by contrast, local governments gained at the expense of departmental governments between 1980 and 1991. Again, the central government experienced only a small decrease in its share of revenue. Brazil was more decentralized with respect to taxing powers even prior to the transition to democratic rule, suggesting the importance of historical legacies. Between 1974 and 1988 states increased their share of total government tax revenue from 36.9 percent to 49.4 percent; the share controlled by local governments actually decreased slightly from 3.8 o t 3.6 percent. However, Brazil is also the country that experienced the most significant change in taxing powers toward the state level after 1988 as well.

³⁰ As we explain below, this explosion of transfers was primarily the result of central bank payments to provincial banks that reflected a predictable political logic.

³¹ Again, see the important reservation at th. 8.

³² For example, the high level of earmarked funds in the Colombian case reflects only the broad requirement that the funds be spent on salaries in the social sectors. Subnational governments may also evade earmarked requirements. By contrast, earmarking in Mexico often contains quite detailed stipulations regarding particular projects and close monitoring.

³³ Comisíon Presidential para la Reforma del Estado (COPRE). For an analysis of how the impetus to decentralize stemmed from declining popularity of the main political parties in Venezuela, see Vin-cente Fretes Cibils, “Venezuela: Dialogo para el desarrollo: La decentralizacion” (Manuscript, World Bank, 1999).

³⁴ Knoop, Joachim, Descentralización fiscal en Venezuela (Santiago de Chile: Comisión Económica para América Latina y El Caribe, 1994)Google Scholar. Much of the following discussion draws upon Michael Penfold-Becerra, “Institutional Electoral Incentives and Decentralization Outcomes: Comparing Colombia and Venezuela” (Ph.D. diss., Columbia University, 1999).

³⁵ Mathew Soberg Shugart and Daniel Nielson, “Liberalization through Institutional Reform: Economic Adjustment and Constitutional Changes in Colombia” (Manuscript, University of California, San Diego, 1994); Torres, Pedro Medelhín, “Reestructuración del estado y desarrollo regional: Contrainsurgencia, democracia y disciplina social,” Revista Interamericana de Planificación 25 (July 1992)Google Scholar; William Dillinger and Steven B. Webb, “Decentralization and Fiscal Management in Colombia,” Working Paper no. 2122 (World Bank, May 1999); and Hoskins, Gary, “The State and Political Parties in Colombia,” in Posada-Carbo, Eduardo, ed., Colombia: The Politics of Reforming the State (London: Macmillan, 1997)Google Scholar.

³⁶ Burki, Shahid Javed, Perry, Guillermo, and Dillinger, William, Beyond the Center: Decentralizing the State (Washington, D.C.: World Bank, 1999)CrossRefGoogle Scholar.

³⁷ The recent collapse of the traditional parties and the victory of Hugo Chavez has changed the bargaining relationship between the national and subnational governments. The new constitution passed under Chavez's tutelage appears to recentralize control although vesting it directly in the hands of the president.

³⁸ Rezk, Ernesto, “Argentina: Fiscal Federalism and Decentralization,” in Bird, Richard M. and Vaillancourt, François, eds., Fiscal Decentralization in Developing Countries (Cambridge: Cambridge University Press, 1998)Google Scholar.

³⁹ For an analysis of the relationship between local officials and federal legislators and their career trajectories, see David Julian Samuels (fn. 21).

⁴⁰ For an analysis of the Cardoso administration's efforts in relation to subnational governments, see Sola, Lourdes, Garman, Christopher da C.B., and Marques, Moises, “Central Banking, Democratic Governance and Political Authority: The Case of Brazil in a Comparative Perspective,” Revista de Economia Politico 18, no. 2 (1998)Google Scholar.

⁴¹ For an analysis of the incentives to provincial governments to increase expenditures in the absence of a revenue-sharing law, see Sanguinetti, P., “Intergovernmental Transfers and Public Sector Expenditures: A Game Theoretic Approach,” Estudios de Economia 21, no. 2 (1994)Google Scholar.

⁴² Rezk (fn. 38), 232.

⁴³ Burki, Perry, and Dillinger (fn. 36), 42–43.

⁴⁴ Meyer, Lorenzo, “Un tema anejo siempre actual,” in Torres, Blanca, ed., Descentralización y democracia en México (Mexico City: Colegio de Mexico, 1986), 23–32Google Scholar.

⁴⁵ One source of influence by Mexican presidents has been a no-reelection rule for sitting legislators and elected officials at the subnational level. This makes them dependent upon the president for bureaucratic posts or positions within the party. See Smith, Peter H., Labyrinths of Power: Political Recruitment in Twentieth Century Mexico (Princeton: Princeton University Press, 1979)Google Scholar.

⁴⁶ Such a pattern has led one scholar of Mexico's decentralization reforms to characterize the Mexican case as “centralization through decentralization.” Rodriguez, Victoria E., Decentralization in Mexico: From Reforma Municipal to Solidaridad to Nuevo Federalismo (Boulder, Colo.: Westview Press, 1997)Google Scholar.

⁴⁷ We are referring to funds established under the Ley de Coordinatión Fiscal, which mandates the federal government to allocate revenue to states through three separate funds: the Fondo General de Participaciones (FGB), the Fondo Financeiro Complementario (FDB), and the Fondo de Fomento Municipal (FFM).The FGB, the largest, is more commonly known is, participaciones.

⁴⁸ Cornelius, Craig, and Fox (fn. 14).

⁴⁹ Burki, Perry, and Dillinger (fn. 36), 46.

⁵⁰ Cornelius, Wayne A., Eisenstadt, Todd A., and Hindley, Jane, Subnational Politics and Democratization in Mexico (La Jolla: Center for U.S.-Mexican Studies, University of California San Diego, 1994)Google Scholar.

⁵¹ Burki, Perry, and Dillinger (fn. 36), 46.

⁵² See Samuels (fn. 21).

⁵³ Jennie Litvack, Junaid Ahmad, and Richard Bird, Rethinking Decentralization in Developing Countries, Sector Studies Series (Washington, D.C.: World Bank, 1998); Prud'homme, Remy, “The Dangers of Decentralization,” World Bank Research Observer 10, no. 2 (1995)CrossRefGoogle Scholar.

⁵⁴ Dillinger, William, Perry, Guillermo, and Webb, Steven B., “Macroeconomic Management in Decentralized Democracies: The Quest for Hard Budget Constraints in Latin America,” in Burki, Shahid Javed and Perry, Guillermo E., eds., Annual World Bank Conference on Development in Latin America and the Caribbean, 1999: Decentralization and Accountability in the Public Sector (Washington, D.C.: World Bank, 2000)Google Scholar.

⁵⁵ Graham, Carol and Kane, Cheikh, “Opportunistic Government or Sustaining Reform? Electoral Trends and Public Expenditure Patterns in Peru, 1990–1995,” Latin American Research Review 33, no. 1 (1998)Google Scholar; Edward L. Gibson, “Federalism and Electoral Coalitions: Making Market Reform Politically Viable in Argentina” (Manuscript, Northwestern University, Evanston, 111., 1997).