39942
Intergovernmental Fiscal Relations
and Local Financial Management
A Partnership of the Asian Development Bank
and
World Bank Institute
Asian Development Bank, Manila
October 11-15, 1999
Intergovernmental Relations: Issues in Public Policy
by
Robert D. Ebel and Serdar Yilmaz
World Bank Institute
� World Bank, 1999
1
I. Introduction
Policy making at subnational level is a relatively recent phenomenon.
Decentralization of policy-making power among governments has gained grounds
especially with the European Union example, the break up of communism and micro-
nationalism (Tanzi, 1995). More importantly, advancements in economic theory, and new
financing and management tools have equipped policy makers with adequate means for
decentralization of power.
Developments in financial markets and advancements in information technology
have significantly reduced the time and cost of processing information and understanding
hierarchies in the marketplace. Today, the private sector has a wider set of tools for
financing and advancing business opportunities. And, the public sector is getting its own
share of change. Public sector reforms in developing and emerging market economies
have put strong emphasis on local participation in addressing the issues once were largely
tackled with by central government.
This paper reviews the key recent advancements in economic theory1 and
changes in practice in governance. The paper is organized as follows: The next section
discusses the definition of subnational governments and subnational policy-making in
different governance systems. Section 3 presents the traditional assignment of functions
to subnational governments and presents the results of the empirical analyses. In fourth
and fifth sections, assignments of expenditure and revenue in an intergovernmental fiscal
system are discussed. And, the final section concludes the paper with suggestions for
further inquiry.
II. What is Subnational and Subnational Policy-Making?
The term subnational covers all tiers of governmental units under central
government. In federal systems, subnational governments refer to states, provinces, or
cantons and lower level units such as counties and councils. However in unitary systems,
there are two levels of government usually. Regardless of the administrative structure of a
2
country, the design of intergovernmental relations has important implications on the
performance of public sector. An effective system should provide guidance to agents of
public administration system in fulfilling expectations of constituencies. Three defining
features of an effective intergovernmental system determine the performance of public
sector.
The legal basis provides a transparent framework for intergovernmental relations.
The legal basis of the system of intergovernmental relations should provide clarity to the
role of subnational governments in public policymaking process. The important
components of an intergovernmental system are expenditure and revenue responsibilities.
The legal framework defining intergovernmental relations should assign revenues and
expenditures to different levels of government without leaving a room for ambiguity and
negotiations.
In countries where the system of intergovernmental relations is guaranteed by the
constitution, subnational policymaking power is not susceptible to shifts in the allocation
of power. On the other hand, in countries where the legal basis for policymaking power
lies in the enactment of simple laws, the changes in the political environment affect
intergovernmental relations. In countries where the majority of the legislative branch of
government enacts simple laws for defining subnational governments' revenue and
expenditure assignments, central government has discretionary power over subnational
governments' policymaking domain.
Fiscal discipline and coordination ensure compliance with the guidelines of
macroeconomic policy objectives. Given that the potential exists for fiscal misbehavior of
subnational governments complicate the national objectives of fiscal and monetary
policies, therefore an institutional framework should be set for assisting them in
compliance. An appropriate institutional framework for intergovernmental cooperation
must ensure voluntary compliance, such as the design of intergovernmental
councils/committees as "gate-keepers" for central grants and transfers. Furthermore,
3
establishing hard-budget constraint is an important tool for fiscal discipline where
subnational governments face financial consequences of their decisions.
The system of intergovernmental fiscal relations is an important component of
the subnational policy-making. Properly structured intergovernmental transfers provide
support for policy objectives of subnational governments. The system of
intergovernmental fiscal relations should be transparent and promote fiscal
harmonization, subnational government accountability, and regional equity.
Subnational Policy-Making
In all countries, power is necessarily divided to some extent across governments.
The extent of division of power has important implications for the functioning of the
public sector and hence the macroeconomic performance of countries. The concept of
fiscal decentralization is a general term that includes broader elements of political, fiscal
and institutional arrangements, and can be characterized as devolution, deconcentration
and/or delegation. The clear distinction between three forms of decentralization practices
is decision-making power of subnational authorities. Devolution is the fullest form of
decentralization where independently established subnational governments are given the
responsibility for delivery of a set of services along with the authority to impose taxes
and fees to finance services. Deconcentration refers to decentralization of central
government offices. Delegation is mandating subnational governments with certain
services under the supervision of central government.
Decentralization is a widely used term and there are several definitions of fiscal
decentralization. In some cases, it refers to spatial decentralization, in others
administrative decentralization. However, the one used by Ebel and Hotra (1997) covers
must of the prominent features of the term and is the most appropriate one for the purpose
of this paper:
The term fiscal decentralization encompasses the political, economic, and
institutional underpinnings of intergovernmental fiscal relations, and
4
ranges from examining the efficiency of public institutions and developing
sustainable infrastructure finance, to rationalizing fiscal transfer
mechanisms and supporting the social safety net.
III. Decentralization versus Centralization
If we analyze decentralization under the light of general government functions,
the question that needs to be answered is which level of government performs different
functions of government best. The argument in favor of decentralization stems from
motivation for enhanced efficiency, accountability and autonomy. According to Stigler
(1957), a representative government works best the closer it is to the people. Carrying
Stigler's argument further, Oates (1972) formulates the decentralization theorem as "each
public service should be provided by the jurisdiction having control over the minimum
geographic area that would internalize benefits and costs of such provision."
In orchestrating decentralization, Oates formulation of decentralization theorem
and Musgraves (1959) definition of functions of government provide an intellectual basis.
In his seminal work on the theory of public finance, Musgrave (1959) separates the
functions of government into three: macroeconomic stabilization, income redistribution,
and resource allocation. In the fiscal federalism literature, there seems a consensus about
assigning resource allocation function to subnational governments, however there is a
debate about assigning the other two functions.
Macroeconomic Stability
Macroeconomic stability can be achieved by a policy prescription with two
sources of cause: monetary and fiscal. It is argued that assigning the stabilization function
to subnational governments is inappropriate because currency stability requires that
5
monetary policy can be best pursued by central government and central governments are
better equipped in dealing with spillover effects of local spending, inflationary pressures
of monetization of local debt and cyclical shocks (Oates, 1972).
The major determinant of the success of monetary policies is the price stability
(Barro, 1996) and price stability is a direct result of independence of monetary authority,
central bank, from political pressure (Alesina and Summers, 1993). The independence of
central bank might be endangered under centralized systems because in most of the
central countries, the central bank operates under the influence of ministry of finance.
However, in decentralized systems, the independence of central bank might be protected
by an institutional design. The four central banks recognized as the most independent
central banks are in highly decentralized federal countries: Swiss central Bank,
Bundesbank of Germany, Central Bank of Austria, and the United States Federal Reserve
Board (Shah, 1997). Moreover, these countries had very low rates of inflation over the
last millenium.
Huther and Shah (1998) examines statutory aspects of central bank operations,
such as the terms of office for the chief executive officer, the formal policy making
power, limitations on lending to the government and other 13 criteria and created a
central bank independence index for 80 countries. Their analysis shows that there is a
statistically significant positive correlation between central bank independence index and
decentralization index. Pearson correlation coefficient of central bank independence and
decentralization of 0.327 is statistically significant at 5 percent level.
Some argue that subnational level fiscal policy-making jeopardizes
macroeconomic stability (Oates, 1972; Prud'homme, 1995; Tanzi, 1996). In a
decentralized system, fiscal policy becomes a responsibility shared by different levels of
decision-making powers. The main concern is the financing of subnational governments'
spending which might circumvent central fiscal policy objectives. The key in the crisis
avoidance solution is a hard budget constraint.
6
Under a soft constraint regime, subnational government spending increases the
pressure on central banks to engage in inflationary finance regardless of decentralization.
Many centralized countries suffered through high inflation in the past. In this regard, the
question is whether fiscal decentralization mitigates the problem or not. In countries
where subnational governments' accountability is weak and the institutions of political
control are immature, there may be a risk that subnational governments may abuse the
fiscal power. Therefore, according to Wildasin (1997) "effective fiscal decentralization
requires an institutional structure that minimizes these adverse incentives." He argues that
fiscal decentralization can make budget constraints harder rather than softer.
An effective decentralization initiative should also promote transparency,
accountability, and predictability in the system of intergovernmental fiscal relations.
When the incentive structure is set without these institutional arrangements, policymakers
would find hard to resist making an adverse selection. In order to avoid the adverse
selection problem in the form of bailouts, hard-budget constraints should be enforced.
Although devolution of power to subnational governments is a necessary
condition for decentralization, it is not a sufficient condition. In fact, devolution of power
is an inseparable indicator of fiscal decentralization together with hard-budget
constraints. The natural consequence of devolution must be establishing hard-budget
constraints. If decentralized governments' spending increase fiscal deficits at the central
level and put macroeconomic stability into jeopardy, a genuine form of decentralization
is not achieved. The question that needs to be asked in defining a genuine decentralized
system is "Does central government absorb fiscal imbalances incurred by subnational
governments?"
If hard-budget constraint is lacking, the moral hazard problem is inevitable. Hard-
budget constraint motivates responsible behavior by subnational governments and
eliminates bail-out conditions. In a recent study Wildasin (1997) defined bailout
conditions as
7
...[i]f the level of local expenditures selected by a central government under a
bailout is 'close' to the level that the locality would choose at a social optimum,
then there is not much cost imposed on the locality from the loss of control over
local spending levels relative to the gain from having the central government
finance the totality of local public good provision (p. 26).
When Musgrave (1959) identified macroeconomic stability as one of the three
functions of government, policy makers and economists alike viewed fiscal policies as
important factors for the aggregate demand fluctuations, which in return affect prices and
employment. In traditional Keynesian theory, fiscal policy is an important tool for
manipulation of short-run demand therefore it should be left to the central government.
However, Spahn (1998) criticizes Keynesian demand management at five points:
(1) It assumes regionally symmetrical shocks: there may well be regionally
asymmetrical shocks.
(2) It assumes a closed economy: fiscal shocks would be offset by an exchange-
rate change in an open economy.
(3) It assumes segmented capital markets: as long as capital markets are informed
and rational, it might be less costly for subnational governments to borrow from
capital markets. Borrowing from central government for the sake of stabilization
policy might exceed the cost of region specific risk factors.
(4) It neglects supply-side effects of effects of fiscal policy: fiscal policy has
supply effect also, especially infrastructure investment programs. Subnational
governments are better informed about their residents' needs.
(5) It neglects built-in stabilizers: fiscal stabilization is automatic through built-in
elements of tax system.
8
In the case of regional asymmetrical shocks, Tanzi (1995) raises the willingness
and ability of subnational governments in pursuing countercyclical fiscal policy. He
highlights heterogeneity of regional economies and argues that regions with high
dependency on a particular resource, such as oil, might be unwilling to pursue
countercyclical policies contradictory to rest of the regions in the country. Gramlich
(1987) calls such situation "zero sum" shocks and argues that in the presence of such
"zero sum" shocks, there is not much that central authority would do. Opening of national
economies to foreign trade and capital flows weakened the central government's ability to
take countercyclical measures. However, Gramlich (1987) demonstrates that there are
cases which subnational governments can pursue countercyclical policy and also, built-in
stabilizers limit subnational governments to exploit fiscal policy. In addition, subnational
governments can take measures such as the use of "rainy day" funds: revenues of good
times can be used over the course of business cycle.
Although macroeconomic management must be a concern of central governments
primarily, the arguments laid out above support the idea that decentralization would not
endanger stability. In fact if appropriately designed, fiscal decentralization may indeed
help macro stability.
Redistribution
In the literature of decentralization, it is argued that subnational governments
cannot effectively carry out redistributive policies. The traditional view that subnational
governments can not achieve redistribution stems from the fact that factors of production,
capital and labor, are mobile. However, the literature about labor migration suggests that
labor is relatively immobile in the short-run (Greenwood and Hunt, 1986) and some of
the capital is tied to locality. Despite the theoretical underpinnings, the varied degree of
mobility provides an opportunity to subnational governments to engage in redistributive
policies to a certain extent.
9
In most of the federal countries, the constitution explicitly assigns social welfare
functions and redistributive taxing ability to subnational governments. Even in some of
the developing and transition countries, such as Brazil, India, Russia, subnational
governments are engaged in public health care and public education (Bird, Ebel, Wallich,
1995). In addition to those functions, Sewell (1996) argues that the regulatory power of
subnational governments has profound distributional implications. Regulatory policies
that have distributional impact include land use, rent controls, user charges, and others.
Resource Allocation
With respect to resource allocation function, Musgrave (1959) argues that policies
of subnational branches of governments should be permitted to differ in order to reflect
the preferences of their residents. If greater decentralization increases number of
alternative fiscal jurisdictions, in Tiebout's analysis (1956) any attempt to increase tax
rates in one jurisdiction would result in migration of its residents to another jurisdiction.
Therefore, the argument in favor of fiscal decentralization is twofold: (1) fiscal
decentralization will increase competition among the local governments which will
ultimately limit the size of the public sector; and (2) decentralization will increase
efficiency because local governments have better information about their residents' needs
than the central government.
In Tiebout's analysis, taxpayers migrate alternative jurisdictions in order to avoid
higher taxes and interjurisdictional competition limit excessive taxing power of the
governments. Along with the lines of Tiebout, Brennan and Buchanan (1980) developed
the "Leviathan" hypothesis where they argue that fiscal decentralization serves as a
constraint on the behavior of the revenue-maximizing government. The Leviathan model
predicts that the overall size of the public sector should inversely vary with
decentralization.
Empirical studies have tested the Brennan and Buchanan model of the
"Leviathan" government and their findings are contradictory. Oates (1985) and Nelson
10
(1986) found no supporting evidence for the Leviathan hypothesis, whereas Marlow
(1988), Grossman (1989), Joulfanian and Marlow (1990) and Grossman and West (1994)
have reported a strong positive correlation between the increase in the subnational
government expenditure in total government expenditure and reduction in total public
sector size.
These studies measured fiscal decentralization as a ratio of subcentral government
expenditure in total government expenditure. The higher is this ratio, the higher is the
degree of fiscal decentralization. However, previous attempts to conduct econometric
tests ignored differences in governance systems by using a pooled data set of federal and
unitary countries. Because of the differences in the design of the intergovernmental
relations, these two different groups of countries should be analyzed separately.
In this part of the paper, by using Oates' (1985) definition of decentralization, the
arguments in favor of decentralization are revisited by using regression analysis for
countries listed in Table-1.
Table-1 List of Countries
Unitary Countries Federal Countries
Belgium Kenya Argentina Peru
Chile Luxembourg Australia Spain
Denmark Netherlands Austria Switzerland
Finland Norway Bolivia United States
France Philippines Canada
Iceland Sweden Columbia
Ireland Thailand Germany
Israel United Kingdom India
Italy Mexico
Two models are specified for the 1971-1990 period. The first model tests
Leviathan hypothesis by looking at the impact of decentralization on government share of
GDP. In this model, dependent variable (GOV) is total government share of GDP and
11
independent variables are decentralization (DEC/DEC1/DEC2), population (POP) and
per capita GDP. Three different decentralization variables are:
DEC = share of local governments expenditure in total government expenditure in
unitary countries (local governments' expenditure divided by total government
expenditure);
DEC1 = share of state governments expenditure in total government expenditure
in federal countries (state governments' expenditure divided by total government
expenditure);
DEC2 = share of local government expenditure in total state government
expenditure in federal countries (local governments' expenditure divided by total
state government expenditure).
In the second model, the impact of decentralization on economic growth is tested
by using an intensive form production function. The dependent variable is real per capita
output (GDP) and independent variables are stock of capital per labor (STO) and
decentralization variable (DEC/DEC1/DEC2). The description and sources of variables
are given in Table-2.
Table-2 Description of Variables
Variable Description Source
DEC Share of local governments expenditure in total Government Finance Statistics
government expenditure (IMF)
DEC1 Share of state governments expenditure in total Government Finance Statistics
government expenditure (IMF)
DEC2 Share of local government expenditure in total state Government Finance Statistics
government expenditure (IMF)
POP Population World Tables (The World Bank)
GOV Government Share of GDP % (1985 International Prices) Penn World Tables 5.6
GDP Real GDP per capita (1985 International Prices) Penn World Tables 5.6
STO Capital Stock per worker (1985 International Prices) Penn World Tables 5.6
Since the true specifications of decentralization models are not known a priori,
omitted factors in the regression analysis become part of the unobservable error terms.
Furthermore, in the analysis of panel data the disturbance term in the model consists of
12
two components: overall disturbance and the individual specific error that reflects
individual differences and varies across observations. Therefore, the information
contained in the omitted variables cause error terms to be contemporaneously correlated.
Using country specific dummy variables alleviates the burden of omission of explanatory
variables but not the effects of temporal shocks. Country dummy variables capture
country specific effects, accounting for country specific characteristics of
decentralization, such as form of decentralization, hard or soft budget constraints on local
governments. Temporal shocks are taken into account by using time-period dummy
variables in the analysis. However, introduction of time and country dummies at the same
time raises identification problems, therefore one set of country and time dummy
variables are excluded arbitrarily without affecting the results and for the sake of
efficiency an overall constant term is included to the analysis (Balestra, 1996).
Results
Table-3 presents the results for both of the models. As seen in the table, there is a
negative relationship between fiscal decentralization and governments size in both groups
of countries. The point estimate is significant at 10 % level for unitary countries and at 1
% level for federal countries. The two different definitions of fiscal decentralization
variable for federal countries have the same sign (negative) and they are both significant
at 1 % level. A negative relation supports Leviathan hypothesis: the overall size of the
public sector varies inversely with fiscal decentralization.
The first model includes two control variables. Real per capita income controls
for the influence of Wagner's law that argues that rising income is positively related to
government growth. GDP variable exerts a negative effect that is statistically significant
at 1 % level of confidence for unitary states and federal states with DEC2 variable. In the
case of DEC1 variable for federal states, GDP variable has the correct sign but it is not
significant. The negative sign of GDP variable suggests that public sector size is an
income-inferior good. Population variable POP has a negative sign for unitary states and
13
it is significant at 10 % level. Although the sign of population variable is unexpected, it is
impact on government growth is trivial. In the case of federal countries, the population
variable has two opposite signs, for DEC2 variable, it has negative sign, but it is not
significant. For DEC1 variable, it has positive sign and it is significant at 1 % level.
By estimating an intensive form production function, we tested the impact of
decentralization on output. In this analysis, the output variable is GDP per capita and
explanatory variables are stock of capital per worker and decentralization. As seen in
Table-3, the stock variable has positive sign, as expected, and statistically significant in
all three models. The main point of interest is the sign of decentralization variable. The
decentralization variable has positive sign in all of the models but it is not significant in
the first two models. In the last model, decentralization variable is positive and
statistically significant at 1 percent level.
The results of regression analysis suggest that public sector size is negatively
correlated with decentralization. An increase in subnational governments' expenditure
share in the total government expenditure share would limit the size of the government in
the economy. The negative relation between decentralization and government size is
statistically significant. However, the relationship between decentralization and economic
growth is positive but not statistically significant at the conventional levels. The only
statistically significant relation between decentralization and economic growth is in the
third model where decentralization is represented by the increase of local governments
share in states' expenditure in federal systems. The lack of legal basis and fiscal
coordination might be the reasons for insignificant relation in unitary countries.
14
Table-3 Regression Results
Unitary Federal
GOV GDP GOV GDP GOV GDP
C 23.3450* 8116.393* 12.2930* 7778.371* 22.0959* 1586.733
(1.5669) (256.818) (1.9109) (1041.01) (0.9818) (2088.703)
DEC -9.1328*** 1395.196
(5.4932) (1163.61)
DEC1 -6.1820* 259.2456
(1.8939) (3139.08)
DEC2 -3.7412* 9150.633*
(0.7895) (2856.74)
POP -0.0001*** 2.5700* -1.6100
(6.7400) (2.3400) (9.9400)
GDP -0.0009* -9.1800 -0.0008*
(0.0001) (0.0002) (6.7800)
STO 10424.16* 9827.314*** 14120.17***
(3238.55) (566.2.22) (7445.02)
R2 0.914 0.981 0.972 0.979 0.972 0.956
Adjusted 0.902 0.978 0.966 0.975 0.965 0.946
R2
N 307 306 186 186 148 148
RSS 1.8605 605.4491 0.9315 781.6988 0.4295 991.9066
F-statistic 79.9619* 377.0716* 167.2822* 242.4240* 128.5568* 90.4462*
Note: Standard errors are in parenthesis.
* significant at 1% level
** significant at 5% level
*** significant at 10% level
15
IV. Expenditure Assignment
The theory of fiscal federalism does not provide a clear guidance to expenditure
assignment (Hemming and Spahn, 1997). However, an effective decentralization requires
a well-defined framework in the assignment of expenditure responsibilities. Since the
system of intergovernmental relations differs from country to country, country specific
conditions determine the framework of expenditure assignment (see the appendix for
design of expenditure assignments in different countries). However, the underlying
premise behind the expenditure assignment should be avoiding "government failure."
The basic inquiry in assigning expenditures must include identifying the services
that can be provided under competitive market conditions. If market conditions do not
prevent internalization of costs in the service provision, private provision of services,
such as telecommunications, electricity, water, etc., would alleviate the burden of
expenditure assignment. In most of the transition countries, privatization has played an
important role in defining expenditure assignments.
Under certain circumstances, competitive market conditions fall short in service
provision and government intervention is necessary. In the delivery of public goods and
the presence of externalities, markets do not function properly. Services, like education
and health care, produce positive externalities. An externality occurs when the activities
of an individual can affect others in a way that is not reflected in market transactions. In
the presence of externalities, governments intervene the marketplace and in most of the
countries provide the services. From intergovernmental system design the basic question
is which level of government meets the needs of people. Being more knowledgeable
about the local residents' preferences, subnational governments have comparative
advantage in fulfilling the expectations of constituents in these services.
In the case of public goods, the most common definition stresses two attributes:
nonexclusivity and nonrivalness. Subnational governments can provide public goods like
roads and transportation services. However, regional spillovers across subnational
16
governments are an important factor that needs to be taken into consideration in public
goods service provision. Presence of benefit spillovers is the major argument in favor of
central government intervention. However, horizontal cooperation among subnational
governments provides a solution spillover problem. Regional authorities, like
Metropolitan Transit Commissions in the U.S., would coordinate efforts among
subnational governments. Public goods, like national defense can not be provided by
subnational governments and should be left to the central governments.
Expenditure assignment is the first step in designing an intergovernmental fiscal
system. Revenue assignments without a concrete assignment of expenditure
responsibilities would weaken the decentralization process (Martinez-Vazquez, 1994).
However, there is no set of rules in defining the best system of expenditure assignment.
Instead, the proper division of expenditure responsibilities should be guided by country
specific conditions and people's preferences. Expenditure assignment is directly related
with the third function of government in Musgrave's analysis: resource allocation.
Efficiency criteria should provide guidance to the assignment of expenditure
responsibilities.
V. Revenue Assignment
The essence of decentralization is that subnational governments have the authority
and responsibility to own-finance local services at the margin. The debate about which
revenues to assign for this purpose focuses on which revenue sources are appropriate for
subnational governments given their unique economic, institutional and demographic
characteristics. There are six important points in revenue assignment (Musgrave, 1983).
High inter-jurisdictional mobility of tax base makes taxation difficult to
subnational governments. Taxes that do not reflect benefits received have the power to
distort the location of economic activity. If consumption-based taxation exceeds the value
of benefits of public services, the result is minor distortion of residential choice.
However, if production-based taxation exceeds the value of benefits, it would crowd-out
17
economic activities. Inter-jurisdictional mobility adds another degree of complexity to the
tax assignment problem. For example, if geographic area of subnational governments are
small, the result of different rates in consumption taxes would be "shopping abroad."
Steep progressive tax rates can be problematic for subnational governments.
Especially if income redistribution function is assigned to the central government,
progressive taxes at subnational level would drive-out investments and high-income and
supposedly more productive people. Uniform subnational taxes minimizes locational
distortions of economic activities.
Services provided by subnational governments can be financed through benefit
taxes. Efficient allocation of resources requires subnational governments recover their
expenses from the beneficiaries of their services. In this case, services provided by
subnational governments should be financed through user charges and other local fees
and taxes that are related to benefits. Examples of benefit related revenues include taxes
levied on motor vehicles and fuels and construction fees.
If subnational governments are assigned to provide redistributional services such
as, education and healthcare, they need to control a stream of revenues. Such revenue
items include income tax, local general business taxes and the sales tax. The proper
assignment of revenues depends on the expenditure assignment. For some of the
expenditure assignment items, it is impossible and undesirable to exclude people who are
not willing to pay or do not effort payments.
During the assignment of revenues, stabilization function of government needs to
be protected by a system of counter-cyclical rate adjustments. Built-in stabilizers are
necessary in order to avoid subnational governments exploitation of fiscal power.
Uneven distribution of tax bases among subnational governments forces the
residents of one subnational area to bear the economic burden of taxes imposed by
18
another jurisdiction. Taxation of natural resource is the best example this type of taxation
practice.
VI. Conclusion
The trend in examining the roles of government has led to a resurgence of interest
in fiscal decentralization. However, such an interest fueled controversies and debates.
This paper summarized some of the principles and practices in decentralization practice.
Decentralization is not a panacea for government failure, however it provides a greater
potential for an improved public administration system.
Decentralization practice needs to be applied with care. There are different pieces
of a decentralized system that make the system function properly. Based on the country
specific conditions, decentralization practice must be carried out with a special attention
paid to the incentive structure. Transparency, accountability and predictability are the
three most important features of an intergovernmental fiscal system. The incentive
structure must be set to promote these features.
19
Table 1: Fiscal Autonomy in Subcentral Governments
Own taxes Base and rate under local control.
Overlapping taxes Nationwide tax base, but rates under local control
Nontax revenues Fees and charges. Generally, the central government specifies where
such charges can be levied and the provisions that govern their
calculation.
Shared taxes Nationwide base and rates, but within a fixed proportion of the tax
revenue (on a tax-by-tax basis or on the basis of a "pool" of different tax
sources) being allocated to the subcentral government in question, based
on (1) the revenue accruing within each jurisdiction (also called the
derivation principle) or (2) other criteria, typically population,
expenditure needs, and/or tax capacity.
General purpose grant Subcentral government share is fixed by central government
(usually with a redistributive element), but the former is free
to determine how the grant should be spent; the amounts
received by individual authorities may depend on their
efforts.
Specific grants The absolute amount of the grant may be determined by
central government or it may be "open-ended" (that is,
depend on the expenditure levels decided by lower levels of
government), but in either case central government specifies
the expenditure programs for which the funds should be
spent.
Source: Anwar Shah, The Reform of Intergovernmental Fiscal Relations In Developing & Emerging Countries, Policy
and Research Series #23, World Bank 1994.
20
Table 2: Conceptual basis of expenditure
assignment
Responsibility
Expenditure Policy, standards, Provisional/
Category and oversight administration Comments
Defense F F Benefit and costs are national in scope
Foreign affairs F F Benefit costs are national in scope
International trade F F Benefit and costs are national in scope
Monetary policy, currency, banking F F Benefit costs are national in scope
Interstate commerce F F Benefit and costs are national in scope
Transfer payments to persons F F Redistribution
Subsidies to business and industry F F Regional development, industrial policy
Immigration F F Benefit and costs are national in scope
Unemployment insurance F F Benefit and costs are national in scope
Airlines and railways F F Benefit and costs are national in scope
Fiscal policy F, S F, S, L Coordination is possible
Regulation F F, S, L Internal common market
Natural resources F F, S, L Promotes a common market
Environment F,S,L S,L Benefits and costs may be national, regional,
or local in scope
Industry and agriculture F,S,L S,L Significant interjurisdictional spillovers
Education F,S,L S,L Transfers in kind
Health F,S,L S,L Transfers in kind
Social welfare F,S,L S,L Transfers in kind
Police S,L S,L Primarily local benefits
Water, sewage, refuse L L Primarily local benefits
Fire protection L L Primarily local benefits
Parks and recreation F,S,L F, S, L Primarily local responsibility, but national and
provincial govts. may establish own parks
Highways
Interstate F S,L Internal common market
Provincial S S,L Provincial benefits and costs
Interregional S S,L Interregional benefits and costs
Local L L Local benefits and costs
Spending power F,S F,S Fiscal transfers to advance own objectives
Note: F is federal responsibility; S is state or provincial responsibility; and L is local responsibility
Source: Anwar Shah, The Reform of Intergovernmental Fiscal Relations in Developing and Emerging Countries.
Policy and Research Series #23, World Bank 1994
21
Table 3: Assignment of local public services to municipal and regional/metropolitan governments
Allocation criteria for provision
Economic Allocation criteria for
Benefit- evaluation public vs. private production
Economies Economies cost Political Consumer of sectoral
Public Service of scale of scope spillout proximity sovereignty choices Composite Efficiency Equity Composit
e
Fire fighting L L L L L M L P G P
Police protection L L L L L M L P G G
Refuse collection L L L L L M L P P P
Neighborhood parks L L L L L M L P G G
Street maintenance L L L L L M L P P P
Traffic management L M L L L M L P P P
Local transit service L M L L L M L P P P
Local libraries L L L L L M L G G G
Primary education L L M M L M M P G P,G
Secondary education L L M M L M M P G P,G
Public transport M M M L, M M M M P, G G P,G
Water supply M M M L, M M M M P G P,G
Sewage disposal M M M M M M M P, G P,G P,G
Refuse disposal M M M M M M M P P P
Public health M M M M M M M G G G
Hospitals M M M M M M M P, G G P,G
Electric power M M M M M M M P P P
Air/water pollution M M M M M M M G G G
Special police M M M M M M M G G G
Regional parks M M M L, M M M M G G G
Regional planning M M M L, M M M M G G G
Note: L is local government; M is regional/metropolitan government, P is private sector, and G is public sector
Source: Anwar Shah, The Reform of Intergovernmental Fiscal Relations in Developing and Emerging Countries.
Policy and Research Series#23, World Bank, 1994
22
Table 4: Conceptual basis of tax assignment
Determination of
Collection
and
Type of Tax Base Rate administratio Comments
n
Customs F F F International trade taxes
Corporate Income F F F Mobile factor, stabilization tool
Resource Taxes
Resource rent (profits, income) tax F F F Highly unequally distributed tax bases
Royalties, fees, charges, severance
taxes,
production,output and property taxes S, L S, L S, L Benefit taxes/charges for state-local
services
Conservaton charges S,L S, L S, L To preserve local environment
Personal income F F, S, L F Redistributive, mobile factor;
stabilization tool
Wealth taxes (taxes on capital wealth,
wealth transfers, inheritances, and F F,S F Redistributive
bequests)
Payroll F, S F, S F, S Benefit charge, e.g. social security
coverage
Multistage sales taxes (value-added tax F F F Border tax adjustments possible under
[VAT] federal
assignments; potential stabilization
tools
Single-stage sales taxes (manufacturer,
wholesale, retail)
Option A S S, L S, L Higher compliance cost
Option B F S F Harmonized, lower compliance cost
"Sin" taxes
23
Excises on alcohol and tobacco F F F Health care shared responsibility
Betting, gambling S, L S, L S, L State and local responsibility
Lotteries S, L S, L S, L State and local responsibility
Race tracks S, L S, L S, L State and local responsibility
Taxation of "bads"
Carbon F F F To combat global/national pollution
BTU taxes F, S, L F, S, L F, S, L Pollution impact may be national,
regional or local
Motor fuels F, S, L F, S, L F, S, L Tolls on federal/provincial/local roads
Effluent charges F, S, L F, S, L F, S, L To deal with interstate, intermunicipal,
or local pollution issues
Congestion tolls F, S, L F, S, L F, S, L Tolls on federal/provincial/local roads
Parking fees L L L To control local congestion
Motor Vehicles
Registration, transfer taxes and annual S S S State responsibility
fees
Driver's licenses and fees S S S State responsibility
Business taxes S S S Benefit tax
Excises S, L S, L S, L Residence-based taxes
Property S L L Completely immobile factor, benefit tax
Land S L L Completely immobile factor, benefit tax
Frontage, betterment S, L L L Cost recovery
Poll F, S, L F, S, L F, S, L Payment for services
User charges F, S, L F, S, L F, S, L Payment for services rendered
Note: F is federal responsibility; S is state or provincial
responsibility; and L is local
Source: Anwar Shah, The Reform of Intergovernmental Fiscal Relations in Developing and Emerging
Countries.
Policy and Research Series #23, World Bank 1994
24
Table 5: Magnitude of General Government Expenditures and Portion Administered by Each Level of
Government
(Average of Latest three years available)
Social Total Expenditure Education Health Social Security and Welfare
Securit
y and
Welfar
e
Country and Ending Year Total Educatio Health Central State Local Central State Local Central State Local Central State Local govern-ment
n govern govern govern- govern- govern- govern govern govern govern govern- govern
-ment -ment ment ment ment -ment -ment -ment -ment ment -ment
(In percent of GDP) (In percent of general government) (In percent of general government)
Argentina 2 (1987)33.2 4.0 1.1 9.1 60.3 39.7 33.3 66.7 24.4 75.6 89.4 10.6
Australia (1987)39.1 5.5 5.5 9.6 52.9 40.4 6.8 8.5 91.3 0.2 43.5 55.6 0.9 92.8 6.2 1.0
Austria (1987)51.8 70.4 13.7 16.9
Belgium (1987)56.7 85.9 11.9
Bolivia (1986)11.1 85.9 10.6 3.4
Brazil (1987)34.1 65.8 24.5 9.6 4.8 34.5 60.7 2.6 89.5 7.9 65.8 31.3 2.9
Canada (1987)46.0 5.8 6.0 12.3 41.3 40.3 18.4 81.7 18.3 98.1 1.9 100.0
Chile (1987)32.3 4.9 1.9 8.8 93.8 6.2 55.5 39.2 5.3 49.0 40.2 10.8 90.0 7.8 2.2
Colombia (1984)18.0 5.5 1.3 3.2 67.4 23.9 8.7 46.8 53.2 7.1 92.9 26.1 73.9
Denmark (1986)57.6 7.1 5.2 23.1 44.9 52.9
Finland (1987)43.0 54.7 45.3 75.3 24.7 97.0 3.0 91.8 8.2
France (1985)49.3 4.6 8.3 20.9 82.2 16.5 1.0 73.8 25.2 74.4 11.2 14.4 79.0 10.9 10.1
Germany (1983)50.2 4.2 8.0 21.2 58.7 21.5 17.9 20.0 80.0 39.2 60.8 95.7 4.3
Hungary (1988)64.5 5.7 4.2 18.1 77.8 22.2 9.0 90.1 30.2 69.8 0.0 100.0
India 2 (1986)22.6 3.4 0.9 2.3 47.5 52.5 65.3 34.7 72.8 27.2 0.0 100.0
Indonesia 2 (1988)22.8 3.1 0.5 0.4 88.7 11.3
Ireland (1987)55.8 72.5 27.5 67.2 32.8 97.0 3.0 94.9 5.1
Israel (1986)62.9 5.3 2.0 10.0 90.8 9.2 94.0 6.0 91.9 8.1 75.9 24.1
Kenya (1984)29.3 5.2 2.1 1.4 94.3 5.7 74.1 25.9 92.0 8.0 97.4 2.6
Luxembourg (1987) 39.1 4.4 0.7 21.3 81.3 15.9 98.7 1.3 82.9 17.1 100.0 0.0
Malawi (1984)29.1 3.7 2.2 0.6 93.7 6.3
Mexico (1984)30.2 90.1 7.6 2.3
Netherlands (1988) 59.2 70.1 29.9
25
New Zealand (1981) 43.2 86.9 13.1
Norway (1986) 47.2 66.4 33.6
Pakistan (1979) 26.1 68.2 28.3 3.5
Paraguay (1984) 11.3 95.1 4.9
Poland (1988) 48.1 71.1 28.9 28.0 72.0 10.3 89.7 99.3 0.7
Romania (1985) 32.3 2.1 2.1 8.9 77.0 23.0
South Africa (1986) 33.3 74.8 12.5 12.7
Spain (1986) 38.2 78.8 9.9 11.3
Sweden (1987) 61.6 59.8 40.2
Switzerland (1984) 37.4 5.3 5.9 13.9 47.5 28.3 24.2 6.2 57.5 36.3 45.5 32.1 22.4 88.5 5.6 5.9
Thailand (1982) 21.2 4.1 1.1 1.2 92.3 7.7 94.8 5.2 93.5 6.5 2.6
Tunisia (1982) 34.0 5.1 2.5 4.7 94.6 5.4 100.0 100.0 100.0 0.0
United (1987) 44.8 5.1 5.1 14.3 70.9 27.2 12.7 87.3 100.0 0.0 84.0 16.0
Kingdom
United States (1987) 37.1 5.1 4.3 9.0 60.3 17.3 22.4 4.2 24.5 71.3 50.5 33.8 15.7 78.0 14.6 7.4
Yugoslavia (1987) 25.3 3.2 4.2 7.8 23.2 31.4 45.4 0.0 0.0 100.0 0.0 0.0 100 7.3 75.9 16.8
Zimbabwe (1986) 45.0 8.3 2.6 3.0 75.8 24.2 60.2 39.8 86.6 13.4 100.0 0.0
Source: Levin (1990) based on International Monetary fund, Government Finance Statistics Yearbook,Vol.13 (Washington: International Monetary Fund, 1989).
1
Excluding intergovernmental grants.
2
Data for general government do not include local government.
3
Includes supranational authorities' share of general government expenditures in Belgium (2.2 percent), Denmark (2.2 percent), France (1.4 percent), Germany (1.8 percent),
Luxembourg (2.7 percent), and the United Kingdom (1.9 percent)
26
Table 6: Tax Revenue Attributable to Each Level of
government
(In percent of general government)
Total Tax Revenue Taxes on Income Taxes on Property Domestic Taxes on
Goods and Services
Country and Year C S L C S L C S L C S L
Netherlands (1988) 97.8 0.8 100.0 100.0 99.4
Paraguay (1987) 97.6 2.4
Indonesia (1988) 97.3 2.7 100.0 55.4 44.6 95.4 4.6
Chile (1987) 96.5 3.5 100.0 100.0 95.5 4.5
Kenya (1986) 96.4 3.6 100.0 1.4 98.6 97.6 2.4
Zimbabwe (1986) 96.3 3.7 100.0 12.3 87.7 98.3 1.7
Ireland (1987) 96.0 2.3 100.0 42.9 57.1 98.7
Malawi (1984) 95.9 4.1 99.7 0.3 1.9 98.1 99.8 0.2
Israel (1986) 95.5 4.5 100.0 22.9 77.1 100.0
Thailand (1988) 95.0 5.0 100.0 58.8 41.2 92.8 7.2
South Africa (1986) 94.4 1.5 4.1 100.0 27.7 72.3 95.6 4.4
Belgium (1987) 93.5 5.0 90.0 10.0 100.0 95.9 3.3
New Zealand (1981) 93.5 6.5 100.0 17.9 82.1 98.8 1.2
Luxembourg (1987) 93.3 6.2 87.6 12.4 92.0 8.0 99.2 0.7
Hungary (1988) 91.7 8.8 73.6 26.8 55.5 45.5 99.2 0.8
France (1988) 90.0 9.2
United Kingdom (1987) 88.1 10.7 100.0 17.0 83.0 99.8
Spain (1986) 87.7 3.3 9.0 84.3 1.2 14.4 56.7 23.9 19.4 76.1 7.4 16.5
Mexico (1984) 84.6 12.7 2.7 98.2 1.6 0.2 8.7 52.4 38.9 99.4 0.3 0.3
Australia (1987) 81.1 15.3 3.6 100.0 4.9 49.2 46.0 75.6 24.4
Norway (1986) 80.9 19.1 57.1 42.9 40.6 59.4 99.8 0.2
Poland (1988) 80.9 19.1 79.6 20.4 50.8 49.2 89.2 10.8
Columbia (1984) 80.3 13.3 6.3 100.0 9.5 90.5 70.0 30.0
Bolivia (1986) 78.6 18.4 3.0 99.2 0.8 60.1 36.2 3.7
Sweden (1987) 77.7 32.6 33.2 66.8 100.0 100.0
Pakistan (1979) 76.9 18.5 4.6
Finland (1987) 76 25.3
Austria (1987) 75.8 11.5 12.7 58.0 22.6 19.4 52.1 4.3 43.6 71.3 13.1 15.6
Brazil (1987) 71.4 26.6 1.9 100.0 2.8 51.9 45.3 43.7 53.4 2.8
Denmark (1987) 71.2 28.2 54.0 43.0 57.0 43.0 99.4 0.1
Germany (1988) 69.8 21.7 7.6 39.4 40.5 20.1 5.2 54.5 40.3 69.9 29.4 0.1
India (1986) 67.2 32.8 100.0 39.0 61.0 51.9 48.1
United States (1987) 66.7 20.6 12.8 81.6 16.8 1.7 5.6 7.2 87.1 17.1 68.0 14.9
Switzerland (1987) 60.5 22.8 16.7 22.8 42.2 35.0 32.6 42.2 25.2 87.7 11.8 0.5
Canada (1988) 50.8 40.0 9.2 63.5 36.5 16.1 83.9 38.7 60.8 0.5
Source: Levin (1991).
Note: C = central government, S = state government, and L = local government; figures are the average of latest three years available.
Endnotes:
1 This is now a robust literature on what we label fiscal decentralization. More about this literature can be
found at http://www.worldbank.org/decentralization.
27
References:
Alesina, Alberto and Lawrence H. Summers. 1993. "Central Bank Independence and
Macroeconomic Performance: Some Comparative Evidence" Journal of Money, Credit
and Banking 25: 151-62.
Balestra, Pietro. 1996. "Fixed Effect Models and Fixed Coefficient Models" in Laszlo
Matyas and Patrick Sevestre (eds.) The Econometrics of Panel Data (The Netherlands:
Kluwer Academic Publishers).
Barro, Robert J. 1996. Getting It Right: Markets and Choices in a Free Society
(Cambridge: MIT Press).
Bird, Richard M., Robert D. Ebel and Christine I. Wallich (eds.). 1995. Decentralization
of the Socialist State, Intergovernmental Finance in Transition Economies (Washington,
DC: World Bank).
Brennan, G. and J. Buchanan. 1980. The Power to Tax: Analytical Foundations of a
Fiscal Constitution (Cambridge: Cambridge University Press).
Ebel, Robert D. and Karen Hotra. 1997. "World Bank Supports Fiscal Decentralization"
Transition 8 (5): 11-12.
Gramlich, Edward M. 1987. "Federalism and Federal Deficit Reduction" National Tax
Journal 40 (3): 299-313.
Greenwwod, M. J. and G. L. Hunt. 1986. "Migration and Employment Change: Empirical
Evidence on the Spatial and Temporal Dimensions of the Linkage" Journal of Regional
Science 26: 223-34.
Grossman, P. J. 1989. "Fiscal Decentralization and Government Size: An Extension"
Public Choice 62: 63-69.
Grossman, P. J. and E. G. West. 1994. "Federalism and the Growth of Government
Revisited" Public Choice 79: 19-32.
Hemming, Richard and Paul Bernd Spahn. 1997. "European Integration and The Theory
of Fiscal Federalism" in Mario I. Blejer and Teresa Ter-Minassian (eds.) Macroeconomic
Dimensions of Public Finance (London: Routledge).
Huther, Jeff and Anwar Shah. 1998. "Applying a Simple Measure of Good Governance
to the Debate on Fiscal Decentralization" Policy Research Working Paper No. 1894,
World Bank.
Joulfaian, D. and M. Marlow. 1990. "Government Size and Decentralization: Evidence
from Disaggregated Data" Southern Economic Journal 56: 1094- 1102.
28
Marlow, M. L. 1988. "Fiscal Decentralization and Government Size" Public Choice 56:
259-269.
Martinez-Vazquez, J. 1994. "The Challenge of Expenditure Assignment Reform in
Russia" Environment and Planning C: Government and Policy 12: 277-92.
Musgrave, R. A. 1959. The Theory of Public Finance (New York: McGraw-Hill).
Musgrave, R. A. 1983. "Who Should Tax, Where, and What?" in Charles E. McLure, Jr.
(ed.) Tax Assignment in Federal Countries (Canberra: The Australian National
University).
Nelson, M. A. 1986. "An Empirical Analysis of State and Local Tax Structure in the
Context of the Leviathan Model of Government" Public Choice 49: 283-294.
Oates, Wallace. 1972. Fiscal Federalism (New York: Harcourt, Brace, Jovanovich).
Oates, Wallace. 1985. "Searching for Leviathan: An Empirical Study" American
Economic Review 75: 748-757.
Prud'homme, R. 1995. "The Dangers of Decentralization" The World Bank Research
Observer 10: 201-26.
Shah, Anwar. 1997. "Fiscal Federalism and Macroeconomic Governance: For Better or
For Worse" Paper presented at Decentralization, and Macroeconomic Governance
Conference, Brasilia, 16-17 June, 1997.
Spahn, Paul B. 1998. "Intergovernmental Relations, Macroeconomic Stability, and
Economic Growth" Paper presented at the Intergovernmental Fiscal Relations and Local
Financial Management Course of the World Bank Institute, Vienna.
Sewell, David O. 1996. " 'The Dangers of Decentralization' According to Prud'homme:
Some Further Aspects" The World Bank Research Observer 11: 143-50.
Stigler, G. 1957. "The Tenable Range of Functions of Local Government" in US
Congress Joint Economic Committee. (ed.) Federal Expenditure Policy for Economic
Growth and Stability (Government Printing Office: Washington, DC).
Tanzi, Vito. 1995. "Fiscal Federalism and Decentralization: A Review of Some
Efficiency and Macroeconomic Aspects" The World Bank Research Observer 10: 295-
307.
Tanzi, Vito. 1996. "Fiscal Federalism and Decentralization: A Review of Some
Efficiency and Macroeconomic Aspects" Annual Bank Conference on Development
Economics 1995, The World Bank, Washington D.C., 295-316.
29
Tiebout, Charles M. 1956. "A Pure Theory of Public Expenditures" Journal of Political
Economy 64: 416-424.
Wildasin, David E. 1997. "Externalities and Bailouts Hard and Soft Budget Constraints in
Intergovernmental Fiscal Relations" Policy Research Working Paper No. 1843, World
Bank.
30