The Dilemma of Fiscal Federalism: Grants and Fiscal Performance around the World Author(s): Jonathan Rodden Reviewed

work(s): Source: American Journal of Political Science, Vol. 46, No. 3 (Jul., 2002), pp. 670-687 Published by: Midwest Political Science Association Stable URL: http://www.jstor.org/stable/3088407 . Accessed: 27/03/2012 05:13
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Midwest Political Science Association is collaborating with JSTOR to digitize, preserve and extend access to American Journal of Political Science.

http://www.jstor.org

The

Dilemma

of

Fiscal

Federalism:

Grants

and

Fiscal

Performance

around

the

World

Joitathan

Rodden

Massachusetts

Institute

of Technology

This article uses cross-national data to examine the effects of fiscal and political institutions on the fiscal performance of subnational govern? ments. Long-term balanced budgets among subnational governments are found when either (1) the center imposes borrowing restrictions or (2) subnational governments have both wide-ranging taxing and borrow? ing autonomy. Large and persistent aggregate deficits occur when subnational governments are simultaneously dependent on intergovernmental transfers and free to borrow? found most frequently constituent units in federaamong tions. Time-series cross-section analysis reveals that as countries increase their reliance on intergovernmental transfers over time, subnation? al and overall fiscal performance decline, especially when subnational governments have easy access to credit. These findings illuminate a key dilemma of fiscal federalism and a more precise notion of its dangers: When constitutionally or politically constrained central governments take on heavy cofinancing obligations, they often cannot credibly commit to ignore the fiscal problems of lowerlevel governments. a combination

of state and lo? growth in the autonomy and responsibilities cal governments is one of the most noteworthy trends in goverArapid nance around the world in recent decades. This trend, along with the growing autonomy of supra-national bodies like the European Union, has encouraged analysts to reexamine some basic issues facing multi-tiered systems of government. As experiences with federalism unfold, an abstract literature emphasizing welfare economies has its efficiency advantages to a more balanced literature that draws political given way economy to questions of institutional design. Much of this new literature can be dangerous, especially in developing points out that decentralization countries. Above all, skeptics point out the difficulties of macroeconomic attention systems (e.g., 1995) especially when they feature formally federal constitutions that empower states with veto authority over central government de? cisions (Treisman 1999; Wibbels 2000). This article addresses one of the most formidable challenges facing fiscal indiscipline systems of government: among subnational A strikingly similar pattern has emerged in developed and governments. countries alike: free-spending have subnational developing governments deficits and called upon central governments to built up unsustainable multi-tiered assume their liabilities. provide special bailout transfers or otherwise These episodes have been extremely costly in countries like Brazil, where stability by into systemic financial crises. An impressive array of case snowballing that decentralization studies has recently demonstrated may be dangerous indeed if it allows subnational to expand their expenditures governments the costs to others (Rodden, while externalizing Eskeland, and Litvack 2002; Von Hagen et al. 2000). subnational fiscal crises have undermined macroeconomic management, Prud'homme adjustment, and reform in decentralized

Jonathan Rodden is Assistant Professor of Political Science, Massachusetts Institute of Technology> E53-433, 77 Massachusetts Avenue, Cambridge, MA 02139 (jrodden@mit. edu). While accepting full responsibility for any mistakes, the author gratefully acknowledges helpful comments from Carles Boix, Keith Dougherty, Gunnar Eskeland, Geoffrey Garrett, Robert Inman, John Kincaid, Ronald Kneebone, Al Montero, Jorn Rattso, Susan Rose-Ackerman, Frances Rosenbluth, David Samuels, Mariano Tommasi, Erik Wibbels, anonymous reviewers, and seminar participants at Austral University, Columbia, MIT, Ohio State, the University of Minnesota, the University of Sao Paulo, and the University of Washington. American Journal of Political Science, VoL 46, No. 3, July 2002, Pp, 670-687 ?2002 by the Midwest Political Science Association ISSN 0092-5853

670

THE DILEMMA OF FISCAL FEDERALISM

67I subnational have access to credit, higher governments levels of dependence on intergovernmental grants will be associated with larger subnational deficits. Second, it that this commitment hypothesizes the relationship between transfers be most pronounced eral systems?especially problem, and hence and deficits, should state governments in fed?

case studies have generated a single-country and plausible hypoth? deal of useful information good crosseses, this article breaks new ground by conducting national quantitative analysis. Virtually all cross-national While of public sector deficits and debt have empirical At first glance this subnational governments. ignored may not seem problematic; during the period from 1986 studies deficit was only .42 per? to 1996 the average subnational cent of GDP for a sample of sixty-three countries. How? include ever, in eleven formally federal systems?which subna? world's largest economies?average tional deficits exceeded 1 percent of GDP and accounted several ofthe deficits.1 In for nearly 20 percent of total government some countries, like Argentina and Brazil, the aggregate subnational deficit routinely surpassed that ofthe central 2.5 percent of GDP. In rapidly countries like Mexico, Spain, and South decentralizing deficits are increasing at an alarming Africa, subnational government and exceeded rate. Moreover, recent studies have shown that increasing deficits lead to higher central government subnational and debt (Fornisari, Webb, and Zou 1998), expenditures along with higher rates of inflation (Treisman 2000). This article is a first attempt to answer a question of accounts for cross-country growing importance?what and diachronic outcomes? fiscal in aggregate subnational do some subnational ap? governments Why pear to behave as fiscal conservatives, while others run up deficits? It weaves together dangerous and unsustainable variation argument from the threads of public eco? and political science and tests it using a large data from OECD, transition, of observations set consisting and developing countries from around the world. An im? an institutional nomics portant goal is to move beyond some of the simple generalizations in the new literature stressing the dangers of and add some insti? fiscal and political decentralization tutional detail. mindful of the situational this article factors identifies often ema basic in case studies, While phasized

disproportionately chamber. The third section plains the econometric

among when the states are directly and in the upper legislative represented introduces approach. The fourth on cross-section time-series section aver-

the data and ex?

presents the results of regressions ages, the fifth section examines

cross-secand ex-

section summarizes tional data, the penultimate tends the results, and the final section concludes.

Fiscal

Federalism

and

Commitment problem

The intergovernmental All multi-tiered subnational mon

commitment

face the possibility that governments will try to over-fish the com? governments revenue pool by shifting their costs onto others.

The problem can be captured by a simple strategic inter? in action between a central and lower-level government, decides whether or which the lower level government not to play cost-shifting strategies without knowing the gov? payoffs of the center. For instance, the lower-level a must decide whether or not to undertake ernment costly new project that will lead to dangerous debt lev? els, or when faced with a permanent negative revenue shock, it must decide whether to undertake politically measures or fund current expendipainful adjustment tures with borrowing. If it fiinds the project or refiises to adjust, it increases the likelihood that it or one of its successors will be forced to eventually request a special or ask that the central government debt-reduction grant The decision directly take over some of its obligations. about whether to fund the project or adjust to the shock response of the central depends upon the anticipated in the next stage of the game, when the cen? government ter decides whether or not to provide a bailout. are beneficial to the recipient Since bailouts but costly to taxpayers as a whole, the central government will wish to announce firmly ex ante that it is resolute? that it never prefers bailouts. For a number of reasons this commitment when defaults may not be credible ex post, however, loom or schools are about to close. If the

dilemma that can cause sub? institutional underlying national officials to view public revenue as a common is heavily involved pool. When the central government it incurs moral, in financing subnational governments, that make it difficult and practical obligations political, to commit to "say no" to entities that overspend, gener? ate unsustainable deficits, and demand bailouts. The sec? ond section explains this basic commitment problem and then structures examines the fiscal that exacerbate incentive and political that if it. First, it hypothesizes

Source: IMF, Government Finance Statistics Yearbook (various years), International Financial Statistics (various years), and author's calculations.

has access to the requisite funds, lo? central government cal governments may believe that the central govern? in this instance it will prefer the ment is irresolute?that bailout to a painful local default or reduction in service

672 or past Even in the absence of externalities provision. "no bailout" bailout episodes, the central government's commitment might be undermined by its own incen? The remainder of this ar? tives, powers, and obligations. of institutional ticle attempts to identify the confluence the center's commitment and factors that undermine thus encourage to over-borsubnational governments row. That is, it examines some key fiscal and political and their factors that allow subnational governments voters to believe that their fiscal burdens may eventually be borne by others. sustained. between An empirical literature

JONATHAN RODDEN

has established

a link

and the growth of govern? transfer-dependence ment (e.g., Winer 1980; Stein 1998; Rodden 2001; Rattso 2000). I go further and assert that transfer-dependence to local revenue mobilization) also alters (as opposed of subnational deficits by with their voters and

beliefs about the sustainability allowing local politicians?along creditors?to believe be unable adverse

will that the central government to ignore their fiscal woes. When a ultimately local government faces an highly transfer-dependent unexpected flexibility cut services, fiscal to raise additional it may not have the revenue, forcing it either to or rely on arrears to employees shock, escalates into a fiscal

Intergovernmental H1: Vertical Fiscal Imbalance

Transfers has a negative effect on sub?

run deficits, If the situation and contractors. the subnational or service

crisis in which pay workers

national fiscal performance. Intergovernmental grants He at the heart of the commitment If subnational problem. were financed purely by local taxes, charges governments and borrowing, voters and creditors would very likely view the obligations of local governments as "sovereign" like those of central governments?bailout expectations would be irrational. As a matter of both normative and descriptive fact, however, intergovernmental systems always involve the vertical flow of funds between Theoretical and empirical studies in pub? governments. theory lic economies "own-source" through of the "fiscal illusion" literature is that key proposition when the link between taxes and benefits is distorted or broken, voters are less likely to sanction by overspending politicians. Intergovernmental grants create the appearance that local public spending is funded by nonresidents.2 Grant programs often supply concentrated local that are funded by a common (national) pool of resources (see Weingast, Shepsle, and Johnsen 1988). Lo? cal voters, local politicians, and regional representatives the central legislature all receive fiscal or political from grant programs without internalizing their them to demand more than own-source on the so-called expenditures taxation. The vast effect" suggest local revenues that individuals view grants and different lenses. A

is unable to government its debt, it can claim with some

that it is not responsible for the situation. justification If successful in this strategy, eventually pressure from voters and creditors which government, to resist may be very difficult for the central government banks, local parents, political pressure from bondholders, or public sector unions. Knowing this, transfer-depen? dent governments face weak incentives to be fiscally re? sponsible ex ante. Even if such subnational governments could take simple but politically costly steps to avoid an fiscal crisis, it may be more rewarding to po? impending sition themselves for bailouts. In fact, credit rating agen? cies are very explicit in assuming that in countries with (transfers as a high levels of "vertical fiscal imbalance" of total subnational revenue), the central govern? percent ment implicitly backs the debt of the subnational gov? ernments.3 In such systems, the central government's own creditworthiness might be called into question if it national will likely be directed at the central quite likely can resolve the crisis. It

benefits

fails to enforce

within benefits funded

a loan contract against a defaulting sub? government. Approached by creditors and facto enforce the prospect of failing in its obligation ing the central government property rights, might see a bail? out as the simplest solution. It is likely that rationality depends on the structure discretionality, tiveness?and of bailout expectations

full cost, causing

"flypaper in intergovernmental grants rarely lead to tax reductions, and increases in transfers stimulate much higher expenditures than do similar increases shows that increases in locally-generated and Thaler 1995). The common revenues theme (for an overview, see Hines

by grants literature empirical

system?e.g., and redistribumatching, earmarking, the flexibility built into the local tax struc?

of the transfer

ture in each country. However, for the purpose of crossnational analysis, Hl makes the simple but very plausible hypothesis that the perceived probability of future bail? outs?and hence subnational deficits?increases with overall transfer-dependence. 3Thus at high levels of vertical fiscal imbalance, subnational credit ratings may reflect the creditworthiness of the central government or entire public sector rather than that of the individual govern? ment. Witness the uniform triple A ratings of the German Lander (in spite of widely divergent fiscal health).

in this literature

is the notion

that intergovernmental liefs about the levels

and begrants alter perceptions that can be of local expenditure

2This literature is too large to review here. For an overview of concepts and measurements of fiscal illusion and a literature review, see Oates (1991). For a theoretical application to intergovernmen? tal grants in particular, see Oates (1979).

THE DILEMMA OF FISCAL FEDERALISM

673 Fiscal Imbalance,

Figure _Borrowing

Hypothesized

Between Vertical Relationships and Fiscal Restraint_ Autonomy,

High

"Sovereignty"

Bailout Expectations

Fiscal restraint Subnational Borrowing Autonomy imposed by voters and creditors

Fiscal indiscipline

Subordination Fiscal restraint imposed by central government Low Low Vertical Fiscal Imbalance High

Borrowing

Restrictions

or whether

H2: Central governments will place restrictions on subna? tional borrowing autonomy when vertical fiscal imbalance is high. Aware of its vulnerability to manipulation, the first line of defense is to make a government's commitment no-bailout (Inman 2002). If this is undermined its co-financing commitment by obliga? tions in a system with high vertical fiscal imbalance, the central credible central government will likely turn to a second line of de? fense. Like a vulnerable parent who takes away a child's credit card, the central government may head off the moral hazard problem by formally restricting local gov? ernments' spending and access to credit. A wide range of strategies have been used around outright prohibitions on the use of debt, numerical debt ceilings, restrictions debt, and balanced budget requirements (Ter-Minasian and Craig 1997). In fact, empirical evidence seems to are a direct response to suggest that these restrictions the commitment and von Hagen problem?Eichengreen (1996) examine tions are indeed H2 and demonstrate that fiscal restric? with high most often found in countries the world, including on borrowing, limits on foreign

fiscal behavior they restrict subnational in practice.4 If they are effective, one should modify Hl and expect the interactive relationship between transferand fiscal perfor? dependence, borrowing autonomy, mance suggested by H3. If vertical fiscal imbalance is in? deed associated fiscal indiscipline, the hold when subnational govern? relationship only ments have relatively unrestricted access to borrowing. That is, subnational fiscal indiscipline should be most should pronounced in cases where vertical fiscal imbalance and borrowing autonomy are both high. This is represented corner of Figure 1, which deby the upper right-hand vertical fiscal imbalance on the horizontal axis and picts borrowing autonomy on the vertical axis. At low levels of vertical fiscal imbalance and high levels of borrowing au? tonomy (the upper left hand corner), voters and creditors view subnational obligations as "sovereign" and face incentives to keep local governments on a tight leash. Creditors and voters, them, punish profligacy with higher interest rates, that the costs ultimately fail on knowing with subnational

levels of vertical fiscal imbalance.

at the polls. Thus subnational punish politicians are not tempted to play cost-shifting stratepoliticians is low (both gies. When formal borrowing autonomy lower quadrants in Figure 1), deficits are kept under con? trol by the heavy hand of the central government.

will only affect subnational fiscal performance at high levels of borrowing autonomy. Previous studies have not asked whether hierarchical imbalance borrowing restrictions are mere parchment barriers

H3: Vertical fiscal

4Studies of the US states have addressed voter-imposed local re? strictions, but not hierarchical restrictions imposed by central governments.

674 But if H3 is correct, it merely raises an addition? should any cases fall into the upper al question?why right-hand cell? Why would a vulnerable central govern? ment with heavy co-financing obligations ever allow sub? national governments to borrow? H6: The effect of federalism pline without is conditional

IONATHAN RODDEN

on subnational fiscal

fiscal

disci? Even

on vertical

imbalance.

an effect on borrowing autonomy, one might exof federalism to pect the unique territorial representation that the center is irincrease the perceived probability resolute. Policy making in federations includes an ele-

Political Territorial H4: Political federalism

Federalism

and

Representation undermines the central

ment of bargaining among territorial units that often obviates any notion that decisions are made by a national median voter (Cremer and Palfrey 1999). The complex regional bargaining and log-rolling that often charactermight allow disunrelated regional size projects for bailouts. The asymmetry of jurisdiction in federations the commitment might also exacerbate problem if the failure of a large state might create nega? tressed votes on tive externalities for the rest of the federation?the "too (Wildasin 1997). At the same big to fail phenomenon" time, a small overrepresented jurisdiction might be "too small to fail" if it is in an especially favorable position to that would be relatively inexpensive for the other constituent units to provide (Von trade votes for bailouts recent Hagen et al. 2000). Based on such considerations, studies by political scientists posit a direct link between federal political institutions and fiscal indiscipline (Triesman 2000; Wibbels 2000). In short, political federalism might weaken lines of defense. H4 suggests that it undermines both the ize the legislative process states to trade in federations

government's ability to restrict subnational borrowing. "For an economist, nearly all public sectors are more or less fed? eral in the sense of having different levels of government that provide public services, irrespective of formal constitution" scientists, (Oates 1999, 1121). For political however, federalism is much more than mere fiscal dethat the autonomy of the cen? tral government is effectively limited, either by constituIn most federal constraints. tional rules or informal centralization?it implies systems, the constituent units had at least some influence in the formulation ofthe original constitutional contract. As a result, federal institutions often restrict the authority of the central government with explicit constitutional of the subunits, which are often enforced by courts. Perhaps the most central feature of independent federalism is the fact that in at least some policy political protections is unable to change the areas, the central government status quo without the agreement of a majority, policy even unanimity of the and sometimes supermajority, constituent units. Often this is the case because in the upper chamber the units of Con? are directly represented gress or Parliament. portant cies is the extensive difference

center's ability to restrict subnational borrowing. That is, states and provinces in federations will be higher in Fig? ure 1 than municipalities in unitary systems. But federal? ism might have an independent effect on the center's ability to commit in the first place (H5). That is, the presence of federal institutions might be associated with poor subnational fiscal performance no matter where a coun? falls in Figure 1. try Alternatively, H6 suggests an interactive relationship. Hl argues that at low levels of vertical fiscal imbalance, the center can credibly commit to remain uninvolved in the fiscal affairs of subnational and voters governments, and creditors hold local politicians responsible for their own fiscal management. If federalism places credible re? strictions on the center, this might actually bolster its commitment when the constituent units are self-financing, but undermine central government suggests fiscal discipline it when they are dependent on the for funds. Returning to Figure 1, H6 that federalism should undermine subnational

As a result of federal constitutional between

bargains, one im? and federal democra? unitary deviation of the latter from the "one

devote" principle. While most democracies person-one viate from this principle to some extent through the leg? of small, usually rural districts, islative overrepresentation small states in most federations have been able to secure representation in the upper house vastly disproportionate of the legislature, and sometimes the lower house as well (Stepan 1999; Samuels and Snyder 2001). Virtually all of of political federalism the distinguishing characteristics ability to reguimply limits on the central government's late the fiscal activities ofthe states or provinces. Not only of the provinces generally is the expenditure autonomy in but their representation protected by the constitution, the upper chamber often gives them veto power over any proposals that would limit their funding or autonomy. H5: Political discipline. federalism undermines subnational fiscal

only on the right-hand side. 2 summarizes all of these possibilities, Figure using bold lines to represent direct relationships and dashed Hl hypothesizes a lines for interactive relationships.5 5The author wishes to thank an anonymous reviewer for suggesting this presentation.

THE DILEMMA OF FISCAL FEDERALISM

675 Main Variables The first task is to come of subnational fiscal up with a comparable measure to use as a dependent discipline

Figure

Summary

of Hypotheses

Vertical Fiscal Imbalance H2 Subnational Rscai Baiance

Borrowing Autonomy H4 Federalism

variable. Recall that the argument does not predict actual bailouts, but rather a higher tolerance for deficits and debt stemming from rational bailout expectations. Sub? national debt data are unavailable, but the IMF's Govern? ment Finance Statistics (GFS) collects yearly data on sub? budget balance. Of course short-term budget deficits may reflect inter-temporal tax- or expenditurenational smoothing or counter-cyclical budgetary policy. One way to minimize the impact of economic cycles is by using over a sufficiently long time period. Another is averages to include controls for exogenous macroeconomic fluctuations. comparison, the surplus data might be divided either by expenditure, revenue, or GDR While appropriate for time-se? ries analysis within countries, GDP is a less desirable denominator for cross-national differences large cross-national For the analysis of crossdegree of decentralization. it makes sense to use surplus as a share country averages, of subnational or expenditure. Since revenues are partially determined by the central government, the most appropriate cross-national measure of subna? tional fiscal discipline expenditures. To operationalize is the deficit/surplus as a share of revenue because of comparison in public sector size and To facilitate Both strategies are employed below. cross-national and time-series

-? -?

Direct Relationship Interactive Relationship

between vertical fiscal imbalance simple relationship H5 asserts a simple and subnational fiscal performance. and subnational defi? between federalism relationship H3 sug? hypotheses: that the effects of vertical fiscal imbalance and bor? gests are conditional on one another, and rowing autonomy H6 suggests that the effects of vertical fiscal imbalance and federalism are conditional on one another. H2 and H4 acknowledge borrowing autonomy. the potential Finally, of endogeneity cits. H3 and H6 are the interactive

Data

the most important independent variable, it is necessary to distinguish between intergov? ernmental grants and "own-source" subnational revenue. Previous studies that attempt to quantify this distinction these propositions, first and then using time-series do so by using the GFS (e.g., Watts 1996; Fukasaku and de Mello 1998), which codes revenues from tax-sharing (taxes that are levied and collected by the arrangements central government and automatically transferred to the states) as "own-source" revenues. While these data might be useful for tracking changes in grants over time, they badly overestimate local revenue autonomy for a number of countries in which subnational governments have very little taxing authority. For this reason, I have created a more useful (for the task at hand) measure of vertical fis? cal imbalance that codes shared rev? (grants/revenue) enues a variety of additional by consulting between this (see Appendix 1). The correlation measure and that used elsewhere is only .46. The disadsources vantage because of this measure some of the sources is that it does not vary over time did not include sufficient as grants

The rest of this article examines

averages using cross-section cross section analysis. The data set is composed of yearly observations for forty-three cases drawn from a crosssection

of OECD, developing, and transition countries for the period between 1986 and 1996. Each observation sec? represents an aggregate state or local government federal countries

tor.6 Some

provide two separate data and local.7 Given the arguments above points?state and the important differences between states and local in federations, it is necessary to include governments both states and local governments for the same country separately, introducing appropriate controls and testing for separate effects. The sample contains all state or local sectors for which complete data could be government obtained. 6For a list of cases and data sources, see Appendix A. 7The exceptions are Argentina and India, for which only state-level data were available.

variation. However, as long as the empirical controls for cross-section effects, the GFS "grants" set-up variable may be useful for the analysis of time-series variation.

time-series

676 autonomy is measured by building on a index created by the Inter-American legal-institutional Bank for a sample of Latin American Development Borrowing countries. IDB formula I have used a slightly modified version to measure borrowing autonomy of the for a

IONATHAN RODDEN

for providing primary education often responsible and retirement benefits, it is useful to control for the portion of the population that is either too old or too young to work?the mon so-called "dependency ratio." Another com? control, population demographic density, is in? cluded as well (both are from WDI). Other aspects of a country's institutions might also affect the central government's ability to commit not to bailouts. Above all, it might be easier to commit provide if the center itself faces a hard budget constraint in the form of an independent central bank (Dillinger and Webb 1999). Bailout expectations are more rational if the central government can "resolve" a subnational fiscal cri? sis by printing (1992) more money. Thus I include Cukierman's index of central bank au? legal-institutional since Persson and Tabellini (1998) tonomy. Additionally, find important differences in fiscal behavior between I include a democracies, presidential and parliamentary variable from the World Bank's Database of Political In? stitutions (DPI) that takes on 0 for pure presidential sys? tems, 1 for systems with assembly-elected presidents, and 2 for pure parliamentary systems. Furthermore, it may be useful to control for partisan fragmentation in the cen? tral government. One might hypothesize that the central is in a better position to "say no" to bailout government requests if the president presides over a unified legisla? systems, or if the Prime Minister in a parliamentary system need not hold together a diverse coalition. Thus I include the index of political cohesion developed by Roubini and Sachs (1989).11 The fiscal woes of subnational governments might also be related to those of higher-level government, so I include the cen? tral government's surplus/expenditure ernments and an additional variable that measures ratio for all gov? the ture in presidential

Taken togovernments.8 larger sample of subnational transfers these new data on intergovernmental gether, imand borrowing autonomy represent a significant data sets dealing provement over previous cross-national the cases for which with fiscal decentralization. Among the fiscal data are available, the following countries are coded as federal because ofthe Austus of the states or provinces: Swit? tria, Brazil, Canada, Germany, India, Mexico, Spain, zerland, and the United States.9 special constitutional Argentina, Australia, sta?

Control

Variables

It is possible that central governments in federations make less credible commitments to "say no" to states not be? cause of legislative politics, but simply because states and provinces are larger and more difficult to ignore than muTo evaluate this claim, I nicipalities or local governments. calculate the average number of persons per jurisdiction in each subnational sector.10 This variable ranges from around 1500 for the French municipalities to over 25 million for the Indian states. It is also plausible that political is merely a byproduct federalism of large country size. Thus I include for area (logged square kilomeIt may be more difficult for ters), and logged population. subnational to balance their budgets when governments they are responsible for a wider range of expenditure ac? tivities. Thus a control is included for the overall level of decentralization?subnational total expenditures as a share of from the (calculated public sector expenditures It is also important to control for economic and de? GFS). fiscal that may affect subnational mographic conditions the log of real GDP per performance. international dollars from WDI), expecting a capita (PPP, are positive relationship. Since subnational governments Thus I include controls

state or province's ratio for local surplus/expenditure in federal systems (and takes on zero for governments other observations).

Cross-Section 8The index is explained in Appendix 2. It is similar to the IDB (1997,173-176) formula, but instead of calculating a weighted av? erage of state and local governments in federal systems, I calculate separate values for state and local governments and include restric? tions placed on municipal governments by state-provincial gov? ernments. In addition, I do not count borrowing restraints im? posed by state and local governments on themselves. In accordance with the argument, this index seeks to capture the attempts of higher-level governments to restrict local borrowing. 9This is in accordance with other recent attempts to code federal? ism. See, e.g., Watts (1996), Elazar (1995), and Treisman (2000). 10Population data are from the World Bank's World Development Indicators (henceforth WDI) and jurisdiction data are taken from the World Bank's World Development Report 1999/2000, Table A.l.

Analysis

from section two would be Ideally, the propositions tested using time-series data disaggregated to the level of individual states and localities. In order to differentiate nTaken from the DPI, this variable takes on 0 for presidential sys? tems under unified government, and 1 under divided government. For parliamentary systems, it takes on 0 for one-party government, 1 for a two-party coalition, 2 for a coalition with three or more parties, and 3 for minority government. Similar results to those presented below are obtained using a variety of other "government fragmentation'' variables from the DPI, including a more complex "veto player" index and the effective number of political parties.

THE DILEMMA OF FISCAL FEDERALISM

677 The results are reported in the first column of Table 1. First, note that the borrowing autonomy equation performs quite well. Recall that the Eichengreen/von Hagen hypothesis (H2) assumes that the central government is a rational, unconstrained unitary decision maker, and as it would choose to tightly regulate subnational bor? such, is high. H4 rerowing when vertical fiscal imbalance laxes these assumptions and proposes that federal institu? tions constrain the central government's range of choices. Countries Strong support is found for both propositions. with higher levels of vertical fiscal imbalance indeed demonstrate lower levels of subnational borrowing autonomy, do have signifi? and states and provinces in federations cantly freer access to deficit finance than local or municipal governments. The results also suggest that central gov? ernments in wealthier, more populous, and presidential allow subna? countries (as opposed to parliamentary) freer access to credit markets. tional governments on the other In the subnational surplus equation, the variables of interest do not approach statistical hand, if insignificant in any specification?even significance are dropped, and even if a simpler control variables OLS model is used. Thus no support is single-equation found for Hl or H5. Though vertical fiscal imbalance au? help explain levels of borrowing effects on subna? do not have independent tonomy, they tional fiscal performance. and federalism 2 estimates the same structural model, but it interaction H3 by including a multiplicative term raises the R2 of the term. Adding the interaction Model examines surplus equation from .68 to .77, and the variables of in? terest are individually and jointly highly significant. The is with reference to best way to interpret the interaction Figure 3, which plots the conditional effect of vertical fis? cal imbalance with a bold line and the 95 percent confi? dence interval with dotted lines. The horizontal axis displays the range of the borrowing autonomy index (from one to five). Figure 3 shows that when subnational gov? ernments face strict formal limitations on their ability to (2) borrow, vertical fiscal imbalance fect on fiscal balance. has a small positive ef? But as subnational governments

fiscal management and fiscal counter-cyclical it would also be useful to differentiate between ex? laxity, shocks. While such analysis is pected and unexpected would in single-case studies, data limitations possible between make cross-national virtually impossible. comparison data My goal is to make the most ofthe cross-national described above. This is best achieved by combining two strategies. This section examines long-term, purely crossOLS re? relationships using "between-effects" gressions on ten-year averages.12 While the disadvantages are obvious, this approach has some advantages: it allows sectional for the use of a more appropriate measure of vertical fis? cal imbalance that cannot vary over time, and it allows about the kinds of sys? for some broad generalizations tems in which Moreover, subnational deficits results are most the cross-section persistent. help provide back?

ground for the second empirical strategy?time-series cross-section analysis that (by necessity) uses a narrower definition of vertical fiscal imbalance, controls for crosseffects, and examines changes over time. The first goal is to estimate a model of average sub? national surplus and ascertain whether vertical fiscal im? section balance teractive and federalism effects. have direct or more complex in? there are good reasons to Furthermore,

is complicated by an intersuspect Thus the empiri? autonomy. vening variable?borrowing H2 and H4 by allowcal model must accommodate that the relationship ing federalism and vertical fiscal imbalance to affect bor? This calls for a system of equations rowing autonomy. is endogenous. in which borrowing Leaving autonomy for the moaside H3 and H6 (the interactive hypotheses) ment, the following structural model makes it possible test Hl, H2, H4, and H5 simultaneously: Surplus a{ + a2VFI + a3Borrow Autonomy a4Federalism + a- Controls + v = = + (1) + to

Borrow Autonomy bl + b2VFI + b3Federalism GDP per Capita + b4 Log b5Log Population + b$System + w, where federalism, GDP per capita, vertical

fiscal imbal?

ance, population, system (the presidential/parliamentary variable) and all control variables are treated as exog? enous. Using three-stage least squares, the parameters of equations 1 and 2 are estimated simultaneously.13 12Aslightly shorter time-series is available for some of the cases. The results presented below are not affected by the deletion of these cases, nor are they affected by limiting the data period to the years that are common to all cases. 13Avariety of other right-hand side variables have been included in equation 2, but only these approached statistical significance. To

access to credit, vertical fiscal imbal? gain independent ance has an increasingly negative impact on budget bal? ance. To get a sense of the substantive predictions of the model, it is useful to calculate predicted values of longrun deficits cal imbalance percentile when borrowing autonomy and vertical fis? and 80th are at low and high levels?20th all other variables are held at values?and

meet the order condition, population (which never achieves sig? nificance in any model of subnational fiscal balance) is not in? cluded in Equation 1. A variety of alternative three-stage specifications yielded very similar results.

678 Table 1 Simultaneous and Borrowing Estimates Autonomy of Average Subnational (1986-1996) Fiscal Balance

JONATHAN RODDEN

3-stage least squares, Standard Errorsin parentheses, ***p<.01,**p<.05, *p<.10

their mean values. At both low and high levels of vertical fiscal imbalance, the model predicts balanced budgets or face sub? small surpluses when subnational governments stantial borrowing restrictions (the lower cells in Figure 1 above). It predicts reasonable deficits (around 5 percent are largely self-fiwhen governments of expenditures) authority borrowing nancing and have wide-ranging (the upper left cell), and large long-run deficits (around 14 percent of expenditures) where high levels of borrow? ing autonomy and vertical fiscal imbalance combine (the upper right cell). au? Moving on to H6, model 3 holds borrowing constant and examines separate effects of vertitonomy

cal fiscal imbalance

for constituent

units in federations

and local governments. Consistent with H6, vertical fis? has a significant cal imbalance only negative effect on subnational fiscal outcomes among states and provinces. Substantively, once again holding all control variables at their mean values, this model predicts long-term deficits of only around one percent of expenditures among local and high at both low (20th percentile) governments values of vertical fiscal imbalance. (80th percentile) units in federations, the model pre? Among constituent dicts a 3 percent deficit when vertical fiscal imbalance is at the 20th percentile value and a 7 percent at its 80th percentile value. deficit when

THE DILEMMA OF FISCAL FEDERALISM

679 Fiscal imbalance on Subnational

Figure

Conditional

Effect

of Vertical

Surplus/Expenditure

Borrowing Autonomy Index

Given the results of models that the best model would

term. triple interaction the (VFI) x (Borrowing Autonomy) result in model 2 is driven primarily by federated units. However, a model separate effects for federated units and local including (not shown) demonstrates government tive coefficients for both that resemble the interaction fect of vertical fiscal imbalance significant, nega? the coefficient for

2 and 3, it seems possible them by using a it is possible that Specifically, combine

the results, vertical fiscal imbalance affect long-term fiscal balance, but in a and contingent way.15 First of all, there is no complicated fiscal imbalance does not have support for Hl?vertical and federalism a direct independent effect on subnational fiscal out? comes; but there is support for H2?at higher levels of vertical fiscal imbalance, central governments attempt to cut off subnational access to credit markets. Perhaps the most important result is in support of H3?when relafree to borrow, more transfer-dependent subna? tively tional sectors are likely to run larger long-term deficits. As for H5, other things equal, federated units do not run But significantly larger deficits than local governments. greater access to credit than local governments (H4), and the largest subnational defi? cits in the sample are found among federations with high levels of transfer-dependence of (H6). The coincidence 15The performance ofthe control variables can be summarized as follows. "Persons per jurisdiction" has the hypothesized negative sign in each model, but statistical significance is quite sensitive to the precise specification. Land area is unrelated to subnational fis? cal performance. As expected, the models show that expenditure decentralization is associated with larger deficits, but statistical significance is sensitive. There is no evidence that wealth affects subnational fiscal performance, and the coefficient for the "dependency" ratio, though negative as expected, does not achieve sig? nificance in many specifkations. Coefficients for population den? sity, central bank autonomy, executive-legislative relations, and central government political cohesion are not significantly differ? ent from zero. Surprisingly, the central government's long-term fiscal performance is not positively correlated with subnational fiscal performance, but the fiscal performance of local govern? ments in federal systems is intertwined with that of the states and provinces. federated units have much

To summarize

term in Model 2. This suggests that the ef? is contingent on borrow?

(and vice-versa) ing autonomy among both federated units and local governments, but the result should be apwith caution because of the small number of proached observations. More generally, gression bustness using tonomy point), should one should be skeptical about re? indexes. As a roanalysis using noncontinuous check, models 1 through 3 have been estimated a simpler dummy version of the borrowing au? index (with and all ofthe be noted the median value used as the cut? results were quite similar.14 But it that with respect to the borrowing au?

tonomy index, ten of the federated units are above the median, and only one (Austria) is below. Of twenty-six local governments in the sample, nine are above and seventeen are below. This underscores between the effects tinguishing the difficulty of disof federalism and bor? analysis.

rowing autonomy

in the cross-section

14Additionally, none of the main results are affected by including or excluding control variables, including a matrix of region dum? mies, or dropping individual cases. Similar results have also been obtained using equation-by-equation OLS.

680

IONATHAN RODDEN with panel corrected data used in political science?OLS standard errors, including fixed effects and a lagged de? lead to bias. pendent variable (Beck and Katz 1995)?may In order to avoid the potential bias associated with this the approach and assuage concerns about endogeneity, derived by Arellano and Bond (1991). to reThis approach relies on the use of first-differences move the fixed effects part of the error term and instrumental variable estimation, Time-Series Cross-Section Analysis lagged explanatory variable pendent Arellano variables where the instruments are the (in differences) and the de? twice. As recommended lagged by and Bond, (1991), one-step robust results are results presented GMM estimator below are from estimations that use the

autonomy, high vertical fiscal borrowing wide-ranging and large deficits is found primarily among imbalance, units in federal systems, but the contingent constituent and vertical between borrowing autonomy relationship fiscal imbalance appears to hold up among both feder? ated units and local governments.

While cross-section results establish

averages are admittedly blunt, these of long-run subna? the key determinates

tional deficits. A natural further step is to examine the ef? transfers on the evolution of fects of intergovernmental fiscal performance on the cross-section focuses and asks whether Building results presented above, this section variation on time series rather than cross-section and under what conditions over time within countries.

presented and used for inference on coefficients. The most straightforward in model?displayed Table 2 (model 4)?explores in the same depen? changes dent variable used above: the subnational deficit/expenditure ratio. The key dependent grants as a share of subnational control variable is subnational variable revenues. revenue is the change in An important

the growth of grants over time might affect deficits. That is, it examines diachronic versions of Hl, H3, H5, and H6. Hl posits that the growth of transfer-dependence, by increasing fiscal separation and encouraging bailout expectations, leads to growth in subnational deficits. H3 and H6 posit, will be conditional on respectively, that this relationship the presence of borrowing autonomy and political feder? alism. H5 posits that subnational deficits will grow more rapidly in federations. Dynamic analysis is particularly useful from a policy expenditure au? perspective; countries are decentralizing thority in many countries around the world, and in most are funded new subnational expenditures transfers rather than increased by intergovernmental new own-source local taxes and fees. Given the present concern in the literature about the supposed macroecocases, these this section provides dangers of decentralization, un? of the fiscal and political conditions an exploration der which decentralization might lead to upward pres? sure on deficits. nomic In order to make use of time-series sary to rely on the GFS distinction and "grant" revenue. This may not be a disadvantage, however, since the GFS conception of "grants" zeros in on the more discretionary grants that show up in yearly bud? comparability gets, and any problems of cross-national will be obviated by an empirical approach that focuses exclusively on time-series variation. The goal ofthe empiri? variation and focus cal set-up is to eliminate cross-section on changes. Given the relatively short (ten exclusively years for most countries) period under analysis, the customary approach to this kind of time-series cross-country data, it is necesbetween "own-source"

as a share of total

state, central, and local) revenue. This set-up (combined allows one to compare the impact of growing revenue and that of having more ofthe revenue decentralization, tilted towards grants. The model also includes two lags of the dependent variable, changes in all of the other control variables that vary over time, levels for those that do not, and a set of dummies sector.16 In order to examine while Hl, model 4 includes only model 5 examines H3 and H6 by down and estimating separate ef? for each subnational

grants/revenue, breaking this variable fects for systems with high and low levels of borrowing (above and below the median value), and autonomy within these categories, separate effects for local/munici-

and constituent units in federations. H5 pal governments is examined in each model by including the dummy vari? able that distinguishes between federated units and local In both models, it is not surprising that the governments. coefficient for "subnational revenue/total revenue" is posi? tive and significant; as subnational receive governments shares of total government revenue, their fiscal po? larger sitions improve. In model 4, although the coefficient for 16The one-step model performs quite well. A Wald test of the null that all of the coefficients except the constant are zero is soundly rejected. A Sargan test of over-identifying restrictions cannot reject the null hypothesis that the over-identifying restrictions are valid. The presence of first-order autocorrelation in the differenced re? siduals does not imply that the estimates are inconsistent, though the presence of second-order autocorrelation would imply this (Arellano and Bond 1991). An Arellano-Bond test soundly rejects the null of no first-order autocorrelation in the differenced residu? als, but it is not possible to reject the null of no second-order autocorrelation.

THE DILEMMA OF FISCAL FEDERALISM

68l in Subnational

Table

Determinates

of Changes

Surplus/Expenditure, 4_Model

Dynamic

Panel

Data Analysis 5_

_Model Dependent Variable: A Subnational Surplus/Expenditure

Robust standard errors in parentheses at at 'significant at 10%;**significant 5%; ***significant 1% Estimates for fixed unit effects not reported. Calculated using Stata 7.0, "xtabond" procedure, one step results

revenue is negative as predicted by Hl, grants/subnational it is not significantly different from zero. However, Model 5 demonstrates very clearly that the coefficient is negative and highly significant, as predicted by H3, among cases with high levels of borrowing autonomy, regardless of sta? tus as federated units or local governments. Recall from above that there are ten state-provincial sectors and nine local sectors and the coefficients with "high" levels of borrowing autonomy, suggest that a one percent increase with .32 in fiscal The sig-

positive coefficient for federated units with low of borrowing is driven exclusively autonomy by Austria. For the remaining cases?the local government levels sectors with low levels of borrowing autonomy?the ficient is positive but not significant. coef?

nificant

The results presented in Table 3 lend support neither to H5 nor H6. There is no evidence that deficits no esti? grow more quickly among federated units?in mation does the "federal" dummy approach statistical sectors Furthermore, significance. among subnational with substantial borrowing autonomy, growing transferdependence does not have a larger effect on fiscal

is associated (decrease) in transfer-dependence percent and .54 percent declines (improvements) balance (as shares of expenditure) respectively.

682

JONATHAN RODDEN 3 Simultaneous Fiscal Balance (1986-1996) of Average Total Estimates and Borrowing Autonomy possible objection to the use of subnational fiscal balance as the dependent variable is the possibility that soft local budget constraints and bailouts might affect the finances of the central government in addition to, or perhaps even instead ofthe local governments. For this reason it is use? the key results using total (combined and local) fiscal balance as the dependent central, state, variable. Of course this requires some changes to the data since states and local gov? can no longer be individual data points. Vertical fiscal imbalance and borrowing au? tonomy in federations must now be based on a weighted ernments specifications within federations (by expenditure share) average of state and local govern? ments. "Grants/Revenue" now refer to totals for all sub? In addition, the control variables governments. fiscal balance for higher-level measuring governments must be left out. Table 3 presents the results of a model that simply Model 2 from above using average aggregate as the dependent variable.18 Though deficit/expenditure reestimates the coefficient is slightly smaller than in the subnational deficit model, the result is quite similar and survives all of the robustness checks outlined above. When subnational are free governments to borrow, higher reliance on intergovernmental trans? fers is associated with larger aggregate deficits not just for the subnational sector, but for the entire public sector. Table 4 presents two models that extend the panel data analysis to total public sector deficits. Model 7 is the simple model, and model 8 includes separate effects. First of all, note that the coefficient for subnational revenue/ total revenue is negative and significant in both models, suggesting that other things equal, revenue decentraliza? with a rather large decline in overall fis? While this lends some empirical support to the fear that fiscal decentralization can harm budget bal? cal balance. local details are ance, once again, more precise institutional As in Table 2, the coefficient for the grants/ important. revenue variable has a negative coefficient in the simple model, but it does not quite reach statistical significance. Model 8 shows that as in the subnational deficit models, the negative coefficient is driven by the cases with substantial borrowing autonomy, the coefficients for which Total Public Sector Deficits are negative, substantively Thus large, and significant. when subnational governments are free to borrow, growis associated with growing total ing transfer-dependence and once again, in unitary systems. larger deficits, contrary to H6, the effect is tion is associated on the interaction variable national set and model ful to reexamine

Table

3-stage least squares, Standard Errorsin parentheses. "*p<.01,**p<.05.*p<.10

among In fact, the negative governments. for local governments.17

outcomes

federated

units

among coefficient is larger

than

There are reasons to suspect that subnational fiscal indis? cipline affects not only the state or local government sec? tor in question, but the entire public sector. In fact, one

,7All of these results are quite robust. Similar results are obtained with and without fixed effects and year dummies, and the results are not affected by the deletion of cases. Similar results are ob? tained when the dependent variable is measured relative to GDP

rather than expenditure, and with a variety of other estimation techniques. ,8A1Iofthe results in this section are quite similar if the dependent variable is calculated as a share of GDP rather than expenditures.

THE DILEMMA OF FISCAL FEDERALISM

683 + Subnational)

Table

Determinates Dynamic Panel

of Changes in Total (Central Data Analysis

Surplus/Expenditure,

_Model Dependent Variable: A Total Surplus/Expenditure

7_Model

8_

Estimates for fixed unit effects not reported. Calculated using Stata 7.0, "xtabond" procedure, one step results

Summary

of Results

Fiscal decentralization and political federalism may in? but their deed complicate macroeconomic management, effects are contingent on other institutional factors. The empirical analysis shows that it is useful to look beyond the rather frustrating and simple binary distinction be? tween federal and unitary systems that has characterized recent blocks literature. hierarchical fiscal systems and Intergovernmental are among the important building in a more nuanced approach to the "varieties of rules

with larger or faster-growing deficits (Hl). Rather, it is clear that higher-level governments can assuage the intergovernmental moral hazard prob? lem by cutting off the access of subnational govern? ments to credit. The cross-section models show that in? deed, at higher levels of vertical fiscal imbalance, central governments attempt to restrict subnational borrowing models predict (H2). The cross-section deficits among subnational governments relatively small that either (a)

are associated

federalism." First of all, the results lend no support to the simple that higher levels of transfer-dependence proposition

face relatively strict formal borrowing limitations, or (b) are relatively fiscally independent, while the largest longterm deficits (subnational and total) are found when subnational are simultaneously transfergovernments dependent and free to borrow transfer-dependence (H3). Similarly, growing over time is associated with larger

684 deficits borrow. The role of federalism cated. Federated deficits is somewhat units display neither than local governments (H5). However, growing do have significantly higher levels of borrowing au? they much so that it is difficult to differentonomy (H4)?so and tiate between the effect of borrowing autonomy federalism. cross-section Though the degrees of freedom are low, the analysis does suggest that the conditional more complilarger nor fasteronly when subnational are free to found

JONATHAN RODDEN local governments in small, hosome interestingly,

governments

primarily

among

mogeneous unitary systems, though troubled large federations like Brazil and India have recently introduced sweeping new legislation aimed at enhancing central control over subnational spending and borrowing. Another path to fiscal discipline is found in the up? per left-hand cell of Figure 1. Here, the central govern? ment limits its co-fmancing obligations, allows local gov?

ernments

relationship between borrowing autonomy and transferdependence holds up among both federated units and lo? cal governments. Moreover, the panel data results show very clearly that when free to borrow, growing transferdependence has a negative effect on subnational fiscal performance ments. both among federated units and local govern? But H6 posits that the negative effect of transfer-

to borrow, and leaves the enforcement of hard voters and credibudget constraints up to self-interested tors. Indeed there is considerable evidence that this variworks well among governments ety of fiscal discipline the upper left-hand corner like the U.S. states occupying and Swiss Cantons Bayoumi, Goldstein, (see, e.g., Lowry, Alt, and Ferree 1999; and Woglom 1995; and Feld and One is tempted to conclude that the 1999).

will be more pronounced dependence among federated units. Here the results of the long-term averages and dynamic analysis are discordant, but understandably so. The deficits are found among largest long-term subnational federated units with relatively high levels of vertical fis? cal imbalance, though the marginal effect of increasing is larger among local governments. transfer-dependence

Kirchgaessner clearest goal for reform is to move toward this cell, increasing the tax base and revenue-raising capacity of sub? national governments and reducing borrowing restric? sending a clear signal to voters and creditors that the center is resolute in its no-bailout pledge. Indeed the seems attractive goal of increased local self-sufficiency tions, from many perspectives. cult, both as a normative But this can be extremely diffi? and practical matter, especially with weak or corrupt local government and a

Implications

and

Conclusions

in poor countries institutions and high levels of inequality. Herein lays the dilemma of fiscal federalism

From a policy perspective, these empirical results are of wide-ranging hard to ignore. The combination subna? tional and growing transfer-de? borrowing autonomy is increasingly common, especially as countries pendence decentralize expenditures by ramping up intergovern? mental transfers rather than building up the local tax base. In most cases, increases in transfers do not keep responsipace with increases in mandated subnational this has been the route to fiscal de? bilities. Unfortunately of the developing world. This be particularly severe in large formal federa? danger may tions, where the center faces practical and constitutional centralization when trying to limit the spending challenges units. rowing activities ofthe constituent and bor? in much

more precise understanding of its dangers; for a variety of political and perhaps even moral reasons, the center often gets heavily involved in the affairs of the subna? tional governments?so involved that it cannot credibly commit to ignore their problems. At the same time, the beholden center can be politically too weak, fragmented, or even to certain subnational governments to censure them or change the basic fiscal and political institutions that create bad incentives. This is most often the case in federations resentation, to formal federations. with strong, disproportionate territorial repbut by no means is the phenomenon limited

This article suggests several paths for future research. Future studies might do more to distinguish between the incentive effects of different types of intergovernmental grants, and collect better data about the tax autonomy of subnational governments.19 Perhaps using the typology in Figure 1 as a guide, more refined work can use disaggregated state- and local-level data to examine the incen? tive effects of different types of intergovernmental trans? fers and local taxes within countries. While I present a perspective, the precise role of in? transfers in shaping the perceptions and tergovernmental incentives of voters and politicians remains a notoriously 19Fora good start, see OECD (1999). useful cross-national

The results point out not only the path to persistent subnational deficits, but also a couple of distinct paths to long-term fiscal discipline. In the lower half of Figure 1, central credit. governments An important attempt to cut off local access to finding is that these prohibitions subnational deficits are negliand short-term fluctuations in

seem to work?long-term gible in such systems,

grants have no effect on deficits. However, the data also show that this method of fiscal discipline is rarely in It is units in large federations. place among constituent

THE DILEMMA OF FISCAL FEDERALISM

685 order to underespecially vertical fiscal imbalance?in stand more clearly the political economy of institutional evolution and reform. An important goal for future stud? is a richer understanding of the way in fiscal and political institutions co-evolve, and the conditions under which reform is possible. ies of federalism which

(Oates 1991)?one open question in public economics data. that is not likely to be resolved with cross-country Finally, although institutions clearly affect outcomes, an understanding into prescriptions key independent Appendix A of these effects does not translate easily for reform. It is necessary to make the variables in this study endogenous?

Years

and Sources

GFS: GovernmentFinance Statistics Yearbook IMF: Teresa Ter-Minassian, ed., Fiscal Federalism in Theoryand Practice, 1997, International MonetaryFund. IDB:Inter-American Development Bank, LatinAmerica after a Decade of Reforms, 1997 Report.

686

JONATHAN RODDEN B Construction of Borrowing Index

Appendix

Autonomy

This index is constructed based on the method developed by the Inter-American Development Bank (see IADB 1997: 188). It is built according to the following criteria: (1) (2) Ability to Borrow: If the subnational government cannot borrow, 2 points. Authorization: This number ranges from zero to one. If all borrowing by the subnational government requires central government approval (or state government approval for local governments in federal systems), 1 point. If no subnational borrowing requires approval, zero points. If the authorization constraint only applies to certain kinds of debt, or if the approval requirement is not always enforced, a score between one and zero is given according to the level of constraint. Borrowing Constraints: If there are numerical constraints on borrowing, such as maximum debt service/revenue the coverage of the constraints. Limits on the Use of Debt: If debt may not be used for current expenditures, .5 points. ratios, up to .5 points, according to

(3)

(4)

The value of the first part of the index (criteria 1 through 4) is equal to 2 minus the sum of the points from criteria (1) through (4). For example, if subnational governments in a country cannot borrow, the total for this part will be 2 - 2 = 0. Additional criteria are: (5) Subnational Government Banks: If subnational governments own banks, 1 point. If these banks have substantial importance, an additional .5 points. If subnational governments have special relationships with banks, but do not actually own them (as in the German Lander), .5 points. Public Enterprises: If subnational governments own important public enterprises, and these have liberal borrowing practices, .5 points.

(6)

To obtain the final index for each country, the scores from criteria (5) and (6) are added to the first part of the index. One is added so that the final index varies between 1 and 5.

References Arellano, Manuel, and Stephen Bond. 1991. "Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application in Employment Equations." The Review of Eco? nomic Studies 58: 277-297. Bayoumi, Tamim, Morris Goldstein, and Geoffrey Woglom. 1995. "Do Credit Markets Discipline Sovereign Borrowers? Evidence from the U.S. States." Journal of Money, Credit, and Banking27: 1046-1059. Beck, Nathaniel, and Jonathan Katz. 1995. "What to do (and not to do) with Time Series Cross Section Data." American Political Science Review 89: 634-647. Bird, Richard. 1986. Federal Finance in Comparative Perspective. Toronto: Canadian Tax Foundation. Blaas, Hans, and Petr Dostal. 1989. "The Netherlands: Chang? ing Administrative Structures." In Territory and Administration in Europe, ed. Robert Bennett. London and New York: Printer. Bury, Piotr, and Carl-Johan Skovsgaard. 1988. "Local Govern? ment Finance." In Decentralization and Local Government: a Danish-Polish Comparative Study, ed. Jerzy Regulski, Susanne Georgi, Henrik Toft Jensen, Barrie Needham. New Brunswick, NJ.: Transaction Press. Cielecka, Anna, and John Gibson. 1995. "Local Government in Poland." In Local Government in Eastern Europe: Establishing Democracy at the Grassroots, ed. Andrew Coulson. Aldershot: Edward Elgar.

Courchene, Thomas. 1994. Social Canada in the Millennium: Reform Imperatives and Restructuring Principles. Toronto: CD. Howe Institute. Cremer, Jacques, and Thomas Palfrey. 1999. "Political Confederation." American Political Science Review 93: 69-93. Cukierman, Alex. 1992. Central Bank Strategy, Credibility and Independence: Theory and Evidence. Cambridge, Mass. and London: MIT Press. Dillinger, William, and Steven Webb. 1999. "Fiscal Management in Federal Democracies: Argentina and Brazil." Policy Research Working Paper Number 2121. Washington, D.C: World Bank. Eichengreen, Barry, and Jiirgen von Hagen. 1996. "Fiscal Re? strictions and Monetary Union: Rationales, Repercussions, Reforms." Empirica 23: 3-23. Elazar, Daniel J. 1995. "From Statism to Federalism: A Paradigm Shift." Publius 25: 5-18. Feld, Lars, and Gebhard Kirchgaessner. 1999. "Public Debt and Budgetary Procedures: Top Down or Bottom Up? Some Evi? dence from Swiss Municipalities." In Fiscal Institutions and Fiscal Performancey ed. James Poterba and Juergen von Hagen. Chicago: University of Chicago Press. Fornisari, Francesca, Steven Webb, and Heng-fu Zou. 1998. "Decentralized Spending and Central Government Deficits: International Evidence." Unpublished manuscript. Wash? ington, D.C: World Bank. Fukasaku, Kiichiro, and Luiz de Mello. 1998. "Fiscal Decentrali? zation and Macroeconomic Stability: The Experience of

THE DILEMMA OF FISCAL FEDERALISM Large Developing and Transition Economies." In Democ? racy, Decentralization and Deficits in Latin America, ed. Kiichiro Fukasaku and Ricardo Hausmann. Washington, D.C.: Inter-American Development Bank and OECD. Gilbert, Guy, and Alain Guengant. 1989. "France: Shifts in Lo? cal Authority Finance." In Territory and Administration in Europe, ed. Robert Bennett. London and New York: Printer. Harloff, Eileen Martin. 1988. The Structure of Local Government in Europe. The Hague: International Union of Local Authorities. Hecht, Arye. 1988. "The Financing of Local Authorities." In Local Government in Israel, ed. Daniel Elazar, Chaim Kalchheim. New York: University Press of America. Helmsing, A.H.J. "Rural Local Government Finance. Past Trends and Future Options." In Limits to Decentralization in Zimbabwe: Essays on the Decentralization of Government and Planning in the 1980s, ed. A.H.J. Helmsing, N.D. Matizwa-Mangiza, D.R. Gasper, CM. Brand, K.H. Wekwete. The Hague: Institute of Social Studies Press. Hines, James, and Richard Thaler. 1995. "The Flypaper Effect." Journal of Economic Perspectives 9: 217-226. Inman, Robert. Forthcoming 2002. "Local Fiscal Discipline in U.S. Federalism." In Decentralization and the Challenge of Hard Budget Constraints, ed. Jonathan Rodden, Jennie Litvack, and Gunnar Eskeland. Cambridge: MIT Press. Inter-American Development Bank. 1997. "Fiscal Decisionmaking in Decentralized Democracies." In Latin America After a Decade of Reforms, Economic and Social Progress in Latin America Report. Washington, D.C.: Johns Hopkins University Press. Inter-American Development Bank. 1999. "Decentralization and Fiscal Discipline in Sub-National Governments: The Bailout Problem." Unpublished manuscript. Washington, D.C.: Inter-American Development Bank. Kitchen, Harry, and Melville McMillan. 1985. "Local Govern? ment and Canadian Federalism." In Intergovernmental Rela? tions, ed. Richard Simeon. Ottawa: Minister of Supply and Services Canada. Lowry, Robert, James Alt, and Karen Ferree. 1999. "Fiscal Policy Outcomes and Electoral Accountability in the American States." American Political Science Review 92: 759-77'4. Newton, Michael, and Peter Donaghy. 1997. Institutions of Modern Spain: A Political and Economic Guide. Cambridge: Cambridge University Press. Nurminen, Eero. 1989. "Finland: Present and Futures in Local Government." In Territory and Administration in Europe, ed. Robert Bennett. London and New York: Printer. Oates, Wallace. 1979. "Lump-Sum Intergovernmental Grants Have Price Effects." In Fiscal Federalism and Grants-in-Aid, ed. P. Mieszkowski and W. Oakland. Washington, D.C.: The Urban Institute. Oates, Wallace. 1991. "On the Nature and Measurement of Fis? cal Illusion: A Survey." In Studies in Fiscal Federalism, ed. Wallace Oates. Brookfield: Edward Elgar. Oates, Wallace. 1999. "An Essay on Fiscal Federalism." Journal of Economic Literature 37: 1120-1149. OECD. 1999. "Taxing Powers of State and Local Government." OECD Tax Policy Studies 1. Padilla, Perfecto. 1993. "Increasing the Financial Capacity of Local Governments." In Strengthening Local Government Administration and Accelerating Local Development, ed. Per? fecto Padilla. Manilla: University ofthe Philippines Press. Persson, Torsten, and Guido Tabellini. 1998. "The Size and

687 Scope of Government: Comparative Politics with Rational Politicians." Alfred Marshall Lecture, EEA Congress, Berlin. Prud'homme, Remy. 1995. "On the Dangers of Decentraliza? tion." The World Bank Research Observer 10: 2. Rattso, Jorn. 2000. "Spending Growth with Vertical Fiscal Im? balance: Decentralized Government Spending in Norway: 1880-1990." Typescript. Norwegian University of Science and Technology. Rodden, Jonathan, Gunnar Eskeland, and Jennie Litvack, eds. Forthcoming 2002. Decentralization and the Challenge of Hard Budget Constraints. Cambridge, MA: MIT Press. Rodden, Jonathan. 2001. "Reviving Leviathan: Fiscal Federal? ism and the Growth of Government." Unpublished manu? script. Cambridge, MA: MIT. Roubini, Nouriel, and Jeffrey Sachs. 1989. "Political and Eco? nomic Determinants of Budget Deficits in the Industrial Democracies." European Economic Review 33: 903-938. Samuels, David, and Richard Snyder. 2001. "The Value of a Vote: Malapportionment in Comparative Perspective." British Journal of Political Science 31: 651-671. Segodi, R. 1993. "Financing District Development in Botswana." In Subnational Planning and Eastern Africa: Approaches, Finances, and Education, ed. A.H.J. Helmsing and K.H. Wekwete. Aldershot: Grower. Stein, Ernesto. 1998. "Fiscal Decentralization and Government Size in Latin America." In Democracy, Decentralization and Deficits in Latin America, ed. Kiichiro Fukasaku and Ricardo Hausmann. Washington, D.C: Inter-American Develop? ment Bank and OECD. Stepan, Alfred. 1999. "Federalism and Democracy: Beyond the U.S. Model." Journal of Democracy 10: 19-33. Ter-Minassian, Teresa, and Jon Craig. 1997. "Control of Subna? tional Borrowing." In Fiscal Federalism in Theory and Practice>ed. Teresa Ter-Minassian. Washington, D.C: IMF. Treisman, Daniel. 1999. "Political Decentralization and Eco? nomic Reform: A Game-Theoretic Analysis." American Jour? nal of Political Science 43: 488-517. Treisman, Daniel. 2000. "Decentralization and Inflation: Com? mitment, Collective Action, or Continuity?" American Po? litical Science Review 94: 837-858. Von Hagen, Jiirgen, and Barry Eichengreen. 1996. "Federalism, Fiscal Restraints, and European Monetary Union." American Economic Review 86: 134-138. Von Hagen, Jiirgen, Massimo Bordignon, Matz Dahlberg, Bhahan Grewal, Per Petterson, and Helmut Seitz. 2000. "Subnational Government Bailouts in OECD Countries: Bank Four Case Studies." Inter-American Development R-399. Working Paper Watts, Ronald. 1996. Comparing Federal Systems in the 1990s. Kingston: Queen's University Press. Weingast, Barry, Kenneth Shepsle, and Christopher Johnsen. 1988. "The Political Economy of Benefits and Costs: A Neoclassical Approach to Distributive Politics." Journal of Politi? cal Economy 89: 642-664. Wibbels, Erik. 2000. "Federalism and the Politics of Macroeco? nomic Policy and Performance." American Journal of Politi? cal Science 44:687'-702. Wildasin, David. 1997. "Externalities and Bailouts: Hard and Soft Budget Constraints in Intergovernmental Fiscal Relations." Unpublished manuscript. Vanderbilt University. Winer, Stanley. 1980. "Optimal Fiscal Illusion and the Size of Government." Public Choice 35: 607-622.