Policy, Research, and External Affairs]
WORKING PAPERS
L Public Economics
Country Economics DepartmentThe World Bank
July 1991WPS 726
Perspectiveson the Design
of IntergovernmentalFiscal Relations
Anwar Shah
Practical guidelines on fine-tuning the structure of transfersbetween federal, state, and local governments - without reas-signing spending and taxing responsibilities.
The Policy, Research, and External Affairs Complex distributes PRE Working Papers todisseninate the findings of work in progress andto encourage the exchange of ideas among Bank staff and all others interested in development issues. lhesc papers carry the names ofthe authors, reflect only their views, and should be used and cited accordingly. The findings, interpretations, and conclusions are theauthors' own.r Tey should not be attributed to the World Bank, its Board of Directors, its management, or any of its member countries.
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Plc,Research, and External Affairs
Pubilc Economics
WPS 726
This paper - a product of the Public Economics Division, Country Economics Departrnent - is part ofa larger effort in PRE to reform public sector management in developing countries. It is one of a series ofdiscussion papers prepared for the Intergovernmental Fiscal Relations Project of the Division. Copies areavailable free from the World Bank, 1818 H StreetNW, Washington DC 20433. Please contact Ann Bhalla,room N1O-059, extension 37699 (107 pages).
The literature on flscal federalism provides much certain taxes, especially sales and excise taxes.useful guidance in the design of The solution is often to fine-tune existingintergovernmental fiscal relations. But few assignments rather than redesign the system. Onedeveloping countries have paid serious attention alternative often ignored is for the higher-levelto this guidance in designing their transfers. government to determine the tax base and for the
lower-level government to levy supplementaryShah provides a framework for assessing (piggyback) rates on the uniforr. -x base.
intergovernmental fiscal arrangements anddevelops some blueprints for helping developing Most countries, says Shah, ignore the basicnations chart a course for reforn. Except for rule of intergovermmental transfers, that grantcentrally planned economies in transition to programs be designed to meet grant objectives.market economies, most of these arrangementsdo not require fully restructuring the economy so Almost invariably, developing countriesmuch as fine-tuning the existing structure of have excessive specific-purpose programs -transfers (without reassigning spending and often the result of pork-barrel politics- fortaxing responsibilities). many of which program objectives are vague or
unspecified or are decided after funds areShah observes that assignment problems go released. This increases flexibility and discre-
from one extreme to the other. In Yugoslavia, for tionar spending at the cost of transparency,example, decentralization went too far and objemivity, and accountability. Some have acircumvented the federal government's role of pen erse economic effect - for example,stabilization and redistribution. A conscious covering lower-level deficits or salaries, thuseffort is needed to restore that federal role. But discouraging tax efforts at the lower level.in most countries the national government's role Reviewing these programs must be high onis too pervasive and intrusive - reaching agendas for public sector reform.beyond the important roles of national defenseand security to such purely local functions as Federal-local and state-local transfers inpothole repair and rat control. most developing countries need restructuring.
National governments are not equipped toUsually these problems arise not because monitor local use of national funds. Moreover,
constitutional assignment conflicts with theory local jurisdictions are better suited than nationalbut because de facto assignment conflicts with to administering to local needs and should bede jure responsibilities. Often, major reform is encouraged to raise local taxes to finance them.possible with administrative orders, so constitu-tional amendments are not needed. Also, local goveirnments are not generally
allowed to borrow in credit markets. Autono-Problems often arise, for example, from mous bodies should be set up to supervise and
overlapping and uncoordinated administration of help local borrowing for capital projects.
The PRE Working Paper Series disseminates the findings of work under way in the Bank's Policy, Research, and ExtemalAffairsComplex. An objective ofthe scries is to get thesef mdings out quickly, even iifprcsentations are less than fully polished.The findings, interpretations, and conclusions in these papers do not necessarily represent official Bank policy.
Produced by the PRE Dissemination Centcr
Perspectives on the Design of Intergovernmentl ia RIdaobw
by Anwar Shah, Pubflc Economias Division, World Bank
Table of Contents
Executive Summary
1.0 Introduction
1.1 Expenditure Assignment: Theory and Practice
1.11 Conceptual Basis of Expenditure Assignment1.12 Expenditure Assignment Issues in Selected Countries
1.2 Tax Assignment: Theory and Practice
3.1 Theory of Tax Assignment3.2 Tax Assignment Issues in Selected Countries
1.3 Tax Base and Revenue Sharing Mechanisms
2.0 Intergovernmental Transfers: Theory and Practice
2.1 Intergovernmental Transfers: A Review of The Theoretical Underpinnings
2.11 Intergovernmental Transfers: A Conceptual Perspective2.111 Grant Types
2.1111 Non-Matching Transfers2.1112 Selective Matching Transfers
2.12 The Economic Rationale For Intergovernmena Transfers2.13 Criteria For the Design of Intergovernmental Fiscal Arrangements2.14 The Design of Fiscal Equalization Grants: Issues and Options
* The author is grateful to Antulio Bomfim for excellent research assistance and to Bela Balassa, DieterBiehl, Richard Bird, Raja Chelliah, Wallace Oates and Peter Knight for comments.
2.2 Intergovernmental Transfers in Practice: An Overview
2.21 A Critical Look at Revenue Sharing Practices in Developing/Transition Economies2.22 Federal Transfers to States and Municipalities2.23 Special Issues in State-Municipal Transfers2.24 The Relevance of Developed Country Experiences
3.0 Vertical and Horizontal Fiscal Imbalances: Some Evidence
4.0 Fiscal Federalism Practices: An Impressionistic Overview
5.0 Blueprints and Pathways For Fine Tuning, Restructuring and Reform
Appendix A: Municipal-Local Fiscal Needs and State-Municipal UnconditionalTransfers in Australia, Canada and the USA
Appendix B: The Practice of Federalism in Selected Developed Countries
References
A LIST OF TABLES AND FIGURES
TABLES
1.1 Conceptual Basis of Expenditure Assignment
1.2 Expenditure Assignment in Selected Countries
1.3 Expenditures By Function By Level of Govermnent
1.4 Conceptual Basis of Tax Assigmnent
1.5 Tax Assignment in Selected Countries
2.1 The Conceptual Impact of Conditional Grants
2.2 Federal/Central Revenue Sharing Mechanisms in Selected Countries
2.3 Federal/Central Transfers to Lower Levels of Government
2.4 State Transfers to Local Governments in Selected Countries
3.1 Vertical Imbalances in Selected Countries
3.2 Coefficients of Vertical Imbalance For Selected Countries
4.1 An Impressionistic Evaluation of Selected Fiscal Systems
A. 1 Canada: Basis of Provincial Unconditional Assistance To Local Governments
EIGLURES
1. Conditional Non-Matching Grant
2. Unconditional Non-Matching Grant
3. Open Ended Matching Grant
4. Closed Ended Matching Grant
PERSPECTIVES ON THE DESIGN OF INTERGOVERNMENTAL FISCAL RELATIONS
by Anwar Shah, World Bank
Executive Summary
The way taxing and spending authorities are delineated and the manner in which
intergovernmental transfers are structured in a country have come to be recognized as of fundamental
importance in the efficient and equitable provision of public services. In view of this recognition, this
paper reviews fiscal federalism principles and practices in both the developing and the developed world
with a view to distilling principal lessons of practical policy nature emerging from this literature and
reflects upon their relevance for a reform of intergovernmental fiscal arrangements in
developing/transition economies. In doing so, the paper provides a broad framework in assessing
intergovernmental fiscal arrangements in individual countries and also develops some blueprints in helping
developing nations chart their own courses of action for their reforms. With the major exception of
centrally planned economies in transition to market economies, one does not expect that these
arrangements will be fully restructured in any developing economy, yet the paper concludes that major
economic gains may be possible in some countries by simply fine tuning the existing structure of
intergovernmental transfers without reassigning expenditure and taxing responsibilities. A discussion of
the major conclusions of this paper follows:
The Assignment Issues
The assignment problems in developing/transition countries cover the whole spectrum. In some
countries like Yugoslavia, there is a growing recognition that the past decentralization effort may have
gone too far and circumvented the federal government's stabilization and redistributive role. Thus a
conscious effort is needed to restore to the federal government its rightful role in these areas. In most
countries, though, it is the all too pervasive and intrusive role of the national government that is coming
under increasing scrutiny. The national government's role in national defense and security is well
appreciated but its over-extension to purely local functions such as pothole repair and rat control is being
questioned. A review of the theory and practice of expenditure assignment suggests that invariably the
assignment problems arise not from the constitutional assignment being at variance with the theory but
from de-facto assignment being at variance with the de-jure responsibilities. Thus it appears that major
reforms are possible with administrative orders ar,d without resorting to constitutional amendments. In
tax assignment, problems are of a similar nature and primatily arise from overlapping and uncoordinated
administration of certain taxes, especially sales and excises. These problems, again, in many instances
can be overcome by simply fine tuning the existing assignmenlts rather than attempting a major redesign.
For example, to avoid problems associated with overlapping administration, tax base sharing with the
higher level government determining the tax base and lower level governments levying supplementary
(piggyback) rates on a uniform base requires more serious attention than has been accorded to this
alternative in the past. Of course in countries making a transition to market economy, major restructuring
is inevitable.
Issues in Intergovernmental Transfers
The fiscal federalism literature has stressed that for each and every objective specified by the
grantor, there is an associated specific design of a grant program. Thus the consistency of grant design
with its objectives is a must for its effectiveness in reaching the desired objectives. For example, general
non-matching transfers or tax base or revenue sharing mechanisms are needed to deal with the fiscal gap;
general non-matching equalization transfers to address differential net fiscal benefits or horizontal fiscal
imbalances; open-ended matching transfers, with the matching rate determined by benefit-spillout ratio,
for benefit-spillout compensation; conditional non-matching transfers to ensure minimum standards of
services across the nation; and conditional open-ended matching transfers to stimulate public expenditures
on areas of high national importance but low local priority. A review of these transfers in
developing/transitio, economies s-:ggests that the above conceptual guidance in grant design for the most
part continues to be ignored.
TIhe most striking observation to come from this review is that inspite of the high national pricrity
attached to limiting inter-regional fiscal disparities in most countries, not a single developing/transition
economy has adopted a program of equalization to a specified standard along the lines of the
representative tax system approach inspite of the apparent simplicity and practicability of its design. More
sophisticated but understandably less successful attempts at equalization because of obvious design flaws
have been adopted in Brazil, India and Nigeria. The formulae adopted by these countries lack an explicit
equalization standard and fail to address regional equity objectives in a satisfactory manner. Most
countries do not even attempt to go this far.
As elaborated upon earlier, specific purpose transfers, if properly structured, can support
important policy objectives in a federation. These objectives include: benefit spillover compensation;
bridging fiscal gaps; ensuring minimum standards of public services across the nation; fulfillment of the
redistributive function of the federal government; creation of common internal market; reduction in net
fiscal benefits across jurisdictions and achievement of stabilization objectives. Grant objectives would pre-
determine grant design. In a typicai developing country, almost without exception, there are incredibly
large number of specific purpose programs. For many of these programs, program objectives are either
not specified or specified vaguely, and in some instances reflection upon grant objectives is done after
the release of funds. There are some obvious reasons for this state of affairs. Various governments can
exercise complete discretion over these funds without having any accountability. Enhanced flexibility is
being achieved at the cost of transparency, objectivity and accountability. Thus specific purpose grant
programs are frequently used as a means of pork-barrel politics rather than in the pursuit of key national
objectives. Some specific purpose grant programs are structured in such a way as to provide perverse
economic incentives. For example, several developing countries provide transfers to cover deficits or
public sector wages at subnational levels. Such grants - contrary to the intentions of the grantor - would
encourage lower own tax effort and thereby higher deficits and higher spending on public sector wages
to qualify for higher grants. A review of these grant programs should be high on any agenda for public
sector reform.
Federal-local and state-local transfers in most developing countries, are in need of a major
restructuring. In none of the countries reviewed here, do these transfers pay any special attention to the
fiscal capacity or revenue potential of local governments. Allocation of funds is usually on an adhoc and
discretionary basis - negating transparency, predictability and autonomy objectives. Major recent
increases in revenue sharing funds from the centre to municipalities in Brazil and Mexico have resulted
in reduced own tax effort at the local level. This result has been attributed to a lack of recognition of
fiscal capacity (tax bases for municipal sources) and an inappropriate design of fiscal effort component
in the allocation formulae. Further, federal government in a large country does not usually have the
administrative capacity to closely monitor finances of individual municipalities and therefore, the case for
direct federal transfers to local governments is quite weak. Such transfers should naturally be placed in
the domain of state governments. Further, it would not be too difficult for state governments to
restructure their transfers to local governments in an objective fashion in view of the easy access to their
economic and fiscal data. A recognition of municipal taxable capacity in these formulae would also assist
in state level monitoring of the utilization of local revenue bases and thereby assist in corrective action
on a timely basis. Much useful guidance on restructuring these transfers is available from the developed
country practices.
Local borrowing in the credit market for capital expenditures remains a major unsettled issue in
most developing countries. Local governments are generally not permitted to borrow in the credit
markets, and therefore, they rely exclusively on higher-level capital transfers for undertaking capital
investments. This is an area where there is potential for setting up autonomous bodies to supervise and
assist local borrowing for capital projects.
In conclusion, there is a universal recognition now that the way taxing and spending amthorities
are delineated and the manner in which intergovernmental transfers are structured have important bearings
on the efficiency and equity of public services provision. Fortunately, much useful guidance in the design
of intergovermnental fiscal relations in developing/transition is available from the theoretical and practical
literature on fiscal federalism. It is also apparent from a review of the developing country literature that
very few developing countries have paid any serious attention to this guidance in the design of their
transfers. Immediate attention must, therefore, focus on making this guidance accessible to policy makers
in developing countries and also to adaptations of this guidance to suit individual developing country
circumstances. This paper takes a small step in this direction. Significant work lies ahead.
1.0 INTRODUCTION
Tle way taxing and spending authorities are delineated and the manner in which
intergovernmental transfers are structured in a country have come to be recognized in inodern public
finance as of fundamental importance in the efficient and equitable provision of public services (see for
example, Bahl and Linr (1990), Bird (1990) and Oates (1990)) . In view of this recognition, both the
theoretical and empirical literature on fiscal federalism ("fiscal federalism " is defined here broadly as
a public sector with more than one fiscal tier and any constitutional distinction between a unitary and a
federal state is for the most part ignored in the followirg analysis) in developed countries is rich and
varied and unlike other branches of economics, offers specific guidance for nations attempting to reform
and/or restructure their fiscal relations. Unfortunately, much of this literature, remains largely
unaccessible to practitioners of federalism both within and outside various layers of governments in
developing and transition economies. This paper attempts to distil principal lessons of practical policy
nature emerging from this literature and reflects upon their relevance for the reform of intergovernmental
fiscal arrangements in developing/transition economies. Thus the principal objective ,of this paper is to
provide a broad framework in assessing intergovernmental fiscal arrangements in individual countries and
also to develop general guidelines in helping developing nations chart their own courses of actions in their
reforms. With the major exception of centrally planned economies in transition to market economies, one
does .'ot expect that these arrangements will be fully restructured in any developing economy, yet as the
following discussion suggests major economic gains may be possible in some countries by simply fine
tuning the existing structure of intergovernmental transfers without reassigning expenditure and taxing
responsibilities.
The rest of the paper is organized as follows:
Section 1 is devoted to a discussion of the tax and expenditure assignment issues and tax base
and revenue sharing mechanisms. The conceptual basis of expenditure assignment is discussed first and
any special problems of this nature observed in selected countries are highlighted. This is followed by
a review of the theory of tax assignment accompanied by a discussion of tax assignment issues of current
relevanee. Tax Base or revenue sharing concepts are then introduced but a discussion of such mechanisms
adopted in various countries is postponed to Section 2.
Section 2 is concerned with the theory and practice of intergovernmental transfers. It provides
a taxonomy of grant programs based on their objectives. Further, economic rationale of intergovernmental
transfers is spelled out. Issues and options in the design of an equalization program are debated. Centre-
state-local revenue sharing mechanisms and transfers are reviewed and special problems in the design of
state-local transfers are highlighted.
Section 3 provides evidence on vertical and horizontal fiscal imbalances irn selected countries
and Section 4 prov;des an impressionistic view of fiscal arrangements ir. selected countries. Finally, in
Section 5, conclusions from various sections of the paper are brought together in summary form and some
general lessois for their reform are presented.
1.1 EXPENDITURE ASSIGNMENT: THEORY AND PRACTICE
This section reviews the conceptual basis of expenditure assignment and the assignment
practices in selected countries.
1.11 Conceptual Basis of Expenditure Assignment
Allocation of oxpenditure and tax functions to various member units is the most fundamental
issue in a federation. Fiscal fec,eralism literature argues that expenditure assignment must precede tax
assignment. This is because tax assignment would in general be guded by expenditure requirements at
different levels and these cannot be worked out in advance of expenditure assigment. This literature also
provides broad guidance in delineating expenditure responsibilities. For example the so called
'decentralization theorem' advanced by Wallace Oates states that "each public service should be provided
- 2 -
by the jurisdiction having control over the minimum geographic area that would internalize benefits and
costs of such provision" (Oates (1972), p.55).
The above theory provides strong rationale for decentralized decision making on efficiency,
accountability, manageability, and autonomy principles. It is argued that:
1. Local provision allows governments to cater better to the tastes and needs of local
residents whereas central provision often results in more uniform provision;
2. Decision making is closer to the people for whom the services are intended. This
induces more responsiveness to local concerns as well as more fiscal responsibility and
efficiency of provision especially where financing of services are decentralized as well;
3. Eliminates multiple layers of jurisdiction; and
4. Enhances inter-jurisdictional competition and innovations in the provision of public
services.
A decentralized system thus ensures consistency of level and mix of public services with
voters' preferences as well as provides incentives for efficient provision of such services.' The following
economic considerations nevertheless warrant some degree of centralization or compensatory grants in
the provision of services for which such factors are relevant.
a. Geographical Variations of Preferences with Imperfect Mobility: Note that the Tiebout
mechanism of voting with one's feet requires small open area and perfect mobility. In
the presence of imperfect mobility, a decentralized mechanism would not ensure
matching of public services with local preferences.2
b. Spatial Externalities: Spatial externalities arise when benefits and/or costs of public
services are realized by non-residents. In the case of benefit-spillout the jurisdiction
providing the service does not consider the proportion of benefits of a public service
accruing to non-residents and therefore would under-provide such a service. 'Me
- 3 -
reverse result is obtained in the case of cost spillouts i.e. where the public service could
be financed by exporting taxes to other jurisdictions. There are also public services
whose benefits are considered national in scope such as defense and foreign affairs. As
a corollary, these services would be best provided by the government at the highest
level i.e. by the federal government.
c. Economies of Scale: Certain services require service areas larger than a local
jurisdiction for cost effective provision. Examples include transportation services,
water and sewerage in a metropolitan area.
d. Administrative and Compliance costs: It is generally agreed that a centralized
administration leads to lower administrative and compliance costs associated with
financing of public services.
There are also certain policy functions which can be carried best by the federal goverment.
For example:
(i) Stabilization Policy: A stabilization policy cannot be carried out effectively by a local
jurisdiction. Local pursuit of such a policy would lead to much of the gains being
lost to outside jurisdictions. A monetary policy has little scope of being carried at
a local level.
Note that the above guidelines for a centralized fiscal policy have only limited
relevance for countries with decentralized constitutions as a centralized fiscal policy
was inconsistent with a decentralized constitution. In any event, decentralized fiscal
policies have worked well in highly decentralized federations like Canada,
Switzerland and the USA. In the area of monetary policy, on the other hand, a
decentralized alternative does not exist and further that the proposition that the
monetary authority should be independent of any level of government conflicts with
- 4 -
a parliamentary system of government. In both Switzerland and Canada, monetary
policy function is delegated by the federal government to an independent central bank
whereas fiscal policy is a sharsd responsibility for all levels of government. The
federal governments in these countries use their powers of the purse (transfers) and
moral suasion through joint meetings to induce a coordinated approach. The Swiss
practice of allocating a portion of the profits (seigniorage) of the central bank to
cantons promotes a sense of wider ownership of the monetary authority and could
possibly be useful for Canada and other countries to follow. Recent constitutional
accords in Soviet Union have recognized monetary policy to be the sole responsibility
of the Centre and fiscal policy, a joint responsibility for all levels. This view could
be implemented by having an independent central bank and fiscal policy flexibility
for the centre could be assured by restructuring tax assignments as well as fiscal
policy coordination through regular meetings of the centre and republics.
(ii) Redistribution Policies: Effective redistribution is possible only through programs
which are national in scope. A local jurisdiction that attempts to carry out
redistributive policies is likely to drive out the rich. The City of New York faced
a fiscal crisis as a result of redistributive programs carried out in the early eighties
and again in 1991. Brown and Oates (1987) provide empirical evidence suggesting
that the migratory responses of the poor to redistributive policies, have discouraged
the adoption of decentralized measures to assist the poor in the U.S.
The above discussion suggests that the federal government be given exclusive authority in
carrying out stabilization policies and providing public services whose benefits are national in scope. The
federal government also has a redistributive role which could be exercised through the tax and transfer
system or through joint provision of public services such as education and health which primarily serve
- 5 -
as "transfers in kind" (see Boadway (1980), (1990)). The federal government also may be assigned a role
in providing compensatory grants for spillout of benefits from state level provision of services. A similar
role for each state is in order for spillout of benefits from local provision of services within their
jurisdictions. All other services would be best provided by the local governments with federal and state
govemments having some role in defining minimum standards.
Table 1.1 provides a summary view of broad guidance provided by economic theory and
discussed above. As shown by this Table, reasonably clear expenditure assignment emerges from the
basic economic principles. The following paragraphs provide a commentary on the expenditure
assignment in selected countries based on these guidelines.
1.12 Expenditure Assignment Issues in Selected Countries
Table 1.2 provides a summary view of 'de-jure' expenditure assignment in selected countries.
This table while informative does not capture fully the nature of the assignment problem created by de-
facto assumption of responsibilities being at variance with the constitutional assignment. Expenditure
assigment issues in developing/transition economies cover the whole spectrum, on one extreme in
Yugoslavia, the federal government has a limited role in national economy beyond national defense and
at the other extreme federal role is all too pervasive in countries such as Argentina, India (see Table 1.3),
Mexico and Pakistan. Brazil and Nigeria fall in the middle of this spectrum and the problem here is
uncoordinated roles of federal and subnational governments in various functions. In the following,
principal expenditure assignment issues currently under discussion are summarized.
Brazil: The 1988 Constitution has made a reasonably clear assignment of public service provision and
revenue raising responsibilities to federal, state and municipal levels in Brazil (see Shah (1990) and
Bomfim and Shah (1991)). Purely local functions such as intracity transport, zoning, preventive health
-6-
Table 1.1
CONCEPTUAL BASIS OF EXPENDITURE ASSIGNMENT
Service ProvisionExoenditure Cateaory Responsibility of Service Comment
Defence F F Benefits/costs national in scopeForeign Affairs F FInternationl Trade F FEnvironment F S,LCurrency, banking F FInterstate commerce F F -
Immigration F F U
Unemployment Insurance F FAirlines/Railways F F -
Industry and Agriculture F,S,L S,L Significant interstate spilloversEducation F,S,L S,L Transfers in kindHealth F,S,L S,L Transfers in kindSocial Welfare F,S,L S,L Transfers in kindPolice S,L S,L Primarily local benefitsHighways F,S,L S,L Some roads with significant
interstate spillovers, othersprimarily local
Natural Resources F,S,L S,L Promotes a common market
Table 1.2
EXPNSDITUfE SSWGENT IN SELECTED COUTRIES
Defe. Fore. Int'l Envir. Cur. IntSt Immig Unempt Air Ind. Educ. Health Soc. Police Hgwys Uatu. ResidAff. Trade Bank Trade Insur. Rail Agric. Uelf Resour
Argentina:(1) F F F F F F,S F,S,L F,S,L F,S F,S F,S S(2) F F F F F F,S F,S,L F,S.L F,S F,S F,S SBangladesh:(1) F F F F L L L L L L L(2) F F F F L L L L L L LBraztl:(1) F F F F,S F F F F F F,S F F,S F,S F,S F F,S S(2) F F F F,S F F F F F F,S F,S,L F,S,L F,S F,S F,S F,S,L SChina:(1) F F F F F F F,S,L F,S,L S,L(2) F F F F F
Czechoslovakia:(1) F F F F F F S L L S F(2) F F F F F F S L L S,L FIndia:(1) F F F F,S F F F F,S F F,S F,S S F,S S F F,S F(2) F F F F,S F F F F,S F F,S F,S,L S,L F,S S F F,S FIndonesia:
oo (1) F F F F F,S,L F,S,L F,S,L1 (2) F F f F F,S,L F,S,L F.S,LJapan:(1) F F F L F,.L F,L F,L L L(2) F F f L f,L F.L F,L L LMalaysia:(1) F F F L F F F F F,S F F,S F,S F F S F(2) F F F L F F F F F,S F F,S F,S F F S FMexico:(1) F F F F F F F F F F,S F.S S,L S(2) F F f F F F F F F F,S F,S S,L SNigeria:(1) F F F F,S F F F F F,S F,S S,L F F,S,L F,S,L F(2)Pakistan:(1) F F F F F,S F F,S F,S,L F,S,L F,S S F,S F S(2) F F F F F F,S S,L S,L F,S S F,S F SPhilippines:(1) F F F F F F F(2) F F F F F S,t S,L S,L
.... (Continued on next page)
Tabte 1.2: EJnditure Assigpint In Setected Catatries (Page 2 of 2)
Defe. Fore. Int'l Envir. Cur. IntSt Imig !kwmpl Air Ind. Educ. Health Soc. Police Ngwys Natu. ResidAff. Trade Bank Trade Itnur. Rail Agric. Welf Resour
Soviet Uinion:(1) F F F F F F(2) F F F F F F,S,L F,S,L F,S,L F,S,L F
Thai land:(1) F F F F L L L L(2) F F F F L L L L
Yugoslavia:(1) F F F F F F F S S S F F(2) F F F F F F S,L S,L S,L
Notes:F = Federal/Central GovernmentS = State/Provincial/Republic/Departmental GoverruentL = Local Government(1) = Service Responsibility(2) = Provision of service* Allocation of Expenditure responsibility between republics and local governments is still an open question. For
row (2), the table reflects actual distribution of main expenditures between different budgets in 1989.
Sources: The data presented in this table was drawn from a large number of sources. Only the major sources for each country are listed below. For acomplete list of the articles and documents used, please refer to the references list at the end of this paper.
Bangladesh: ESCWP (1989); Brazil: Bomfim and Shah (1991), Shah (1990); China: ESCAP (1989); Colombia: Bird (1984), Direccion General de lpuestoslacionales (1988): Banco de La Republica (19 ); Czechoslovakia: Prust et al. (1990); India: Indian Finance Commission (1989); Indonesia: ESCAP(1989); Japan: ESCAP (1989); Malaysia: ESCAP (1990); Nigeria: Ashwe (1986a), Gandhi (1982); Pakistan: Akhtar (1990); Papua New Guinea: ESCAP(1989); Philippines: ESCAP (1989); Soviet Union: Alexashenko (1990); Thailand: ESCAP (1989); Yugoslavia: Graham (1984)
Table 1.3
EXPENDITURES BY FUNCnONS BY LEVEL OF GOVERNMENT
Function National Subnational Total
ARGENTINA (1987)Gen Pub Ser & Pub Order 0.35 0.65 1.00Defense 1.00 0.00 1.00Education 0.34 0.66 1.00Health 0.30 0.70 1.00Soc. Sec. & WeHare, and 0.78 0.22 1.00Housing & Commun. Amenitles
Recr, Cuir, Relig Affrs. 0.58 0.42 1.0OEcon. Affrs & Services (a) 0.75 0.25 1.00
INDIA (1986)Gen Pub Ser & Pub Order 0.37 0.63 1.00Defense 0.99 0.01 1.00Education 0.09 0.91 1.00Health 0.30 0.70 1.00Soc. Sec. & Welfare 0.00 1.00 1.00Housing & Commun. Amenitles 0.48 0.52 1.00Recr, Cuir, Relig Affrs. 0.00 1.00 1.00Econ. Affrs & Services 0.47 0.53 1.00
YUGOSLAVIA (1987)Gen Pub Ser & Pub Order 0.11 0.89 1.00Defense 1.00 0.00 1.00Education 0.00 1.00 1.00Health 0.00 1.00 1.00Soc. Sec. & Welare, and 0.08 0.92 1.00Housing & Commun. Amenities
Recr, Cuir, Relig Affrs. 0.02 0.00 1.00Econ. Affrs & Services (a) 0.65 0.35 1.00
SOVIET UNION (1987)Defense 1.00 0.00 1.00Educatlon 0.09 0.91 1.00Heafth 0.10 0.90 1.00Soc. Sec. & Welare, and 0.47 0.53 1.00Housing & Commun. Amenities
Recr, Culture 0.28 0.72 1.00Econ. Affrs & Services (a) 0.65 0.35 1.00
Notes: (a) Economic Affairs & Services Include: Fuel & Energy,Agriculture, Forestly, Fishing, Hunting, Mining,Manufacturing, Construction, and Transportation & Communic.
Source: Soviet Union: Alexashenko (1990)Others: IMF-Govemment Finance Yearbook 1989
- 10 -
care and elementary education has been assigned to the municipal level exclusively. The responsibil.ty
for public services that are national in scope such as defense and foreign affairs etc. has been entrusted
to the federal level and the remaining functions are designated as shared responsibilities of the federal and
state levels with the federal government setting the norms and the state government responsible for the
delivery of services. Unfortunately, the defacto assignment or practice in Brazil is at substantial variance
with the de jure assignment and the federal government's direct involvement in purely local functions is
quite pervasive. This involvement is likely to disappear in the near future in view of the fiscal squeeze
introduced by the new fiscal arrangements upon the federal government.
Mexico: Mexico is a highly centralized federation. Over 80% of public expenditures are controlled by
the central government. In addition to the usual functions of a central government namely defense, justice,
external affairs, commerce and finance, the federal government in Mexico assumes responsibilities for
functions which are allocated to other levels of government in other federations such as health and
education. There is some evidence to suggest that while education expenditures in Mexico were broadly
distributive, the same was not true of health expenditures and per capita health spending was positively
correlated with per capita income of the state (see Boadway (1990)). States are responsible for public
transport and infrastructure expenditures only. It is expected that Mexico will decentralize or relax
federal authority over a number of functions including health and education in the coming years.
Yugoslavia: In Yugoslavia, defense, general administration and social insurance of disabled veterans and
military personnel accounted for 94.2% of total central government expenditures and the rest primarily
as transfers to less developed regions in 1988 (see Bogoev(1990)). Therefbre, the central government is
totally devoid of any macro stabilization and redistributive roles. These latter roles are being carried out
by the republics in an uncoordinated fashion. Thus, in Yugoslavia, the challenge for reformers is to
secure a more active role in macroeconomic mangement and redistribution for the Central government.
This is not an easy challenge given the current state of regional disharmony in the country.
- 11 -
Sovjet lnion: In the Soviet Union, until i990, the Centre or the Union government exercised complete
control over all public expenditures. Nearly half (50.8% of total in 1990; c Sichev and Ilyin (1990))
of total public expenditures were undertaken by the Republics wit;i express approval and clearance from
the Union Finance Ministry. Beginning in 1991, this is all changing. Beginning that year, the Union will
retain exclusive authority for defense, state security agencies , the Office of Public Prosecutor and the
regulation of money, credit, finance and insurance activities. All other public expenditures will be shared
responsibilities and the republics will no longer need any central clearance to plan and execute their
budgets.
1.ni : In areas of concurrent assignment, the federal government excercises strong control over states'
priorities through the "carrot" (grants) and "stick"(withholding of federal grants and approvals for state
capital expenditures) approach followed by the Ministry of Planning (see Gulati and George (1985)).
Pakistan: Federal control over concurrent and provincial responsibilities is quite strong in Pakistan. This
control is excercised through the plan review and approval process, through transfers and through the
federal civil service officials which hold key administrative positions at provincial and local levels.
1.2 TAX ASSIGNMENT : THEORY AND PRACTICE
This section outlines the general conclusions of the theoretical literature on tax assigmment
and also provides an overview of such practices in selected countries.
1.21 The Theory of Tax Assignment
The division of revenue sources among federal and lower levels of government is being
referred here as the 'tax assignment problem'. Once expenditure assignment has been agreed upon, tax
assignment and design of transfers become critical elements in providing reasonable matching of
expenditure needs with revenue means for various levels of government. Although tax assignment could
- 12 -
be undertaken independently of expenditure assignment-a practice which is quite common in developing
countries, yet the tradeoffs between the advantages of a centralized tax administration and decentralized
provision of public services become more apparent when tax assignment takes into consideration pre-
determined expenditure assignment. In such a situation over-dependence of lower levels of government
on intergovernmental transfers with potentially distortionary effects on expenditure priorities could be
avoided. Furthermore, in those grey areas where theoretical guidance on tax assignment is unclear,
expenditure assignment can provide a powerful argument for assigning taxing responsibility to the
government with greater need for additional revenues. Musgrave ( 1983) uses equity (consistency of
revenue means with expenditure needs) and efficiency ( minimizing resource cost) criteria and suggests
the following broad principles in tax assignment;
i. Progressive redistributive taxes should be central;
ii. Taxes suitable for economic stabilization should be central; lower level taxes should be
cyclically stable;
iii. Tax bases distributed highly unequally between jurisdictions should be centralized;
iv. Taxes on mobile factors of production are best administered at the centre;
v. Residence based taxes such as sales of consumption goods to consumers or excises are
suited for states;
vi. Taxes on completely immobile factors are best suited for local level;
vii. Benefit taxes and user charges might be appropriately used at all levels.
Based on these principles, reasonably clear guidelines for assignment of revenue sources to
various levels of government emerge. Table 1.4 provides a summary view of such assignment. The
table suggests that for certain taxes such as resource taxes or a value-added tax (VAT), base and rate
determination and collection and administration could be assigned to different levels of government. By
following this approach both inter-jurisdictional equity and efficiency of tax administration and
- 13 -
Table 1.4
CONCEPTUAL BASIS GF TAX ASSIGNMIENT
Determination of Tax CollectionTax Base Rate & Administration Comments
Customs F F F Inernational TradeIncome Tax F F,S F Redistnbutive, mobile factorEstates and Gifts F F,S F RedistnbutiveCorporate Tax F F,S F Mobile factorResource Tax F F,S F Unequally distributedRetail Sales S S S Higher compliance cost
F S F Harmonized, lower compliance cost.
VAT F F,S F,SC Border tax adjustments possible.Excises S S S Residence based taxesProperty Tax S L L Completely immobile factor, benefit tax.User charges F,S,L F,S,L F,S,L Payment for services received.
Notes: F: FederalS: State/ProvinceL: Municipal/LocalSC: The Council of States
compliance could be achieved. It should be noted that the theory contravenes the advice sometimes
offered by international agencies to developing countries that local taxes on wage and capital income
should be instituted. With factor mobility, bases for such taxes would be subject to erosion. Also, such
a regime encourages tax competition among various jurisdictions.
1.22 Tax Assignment Issues in Selected Countries
Table 1.5 presents an overview of tax assignment in selected countries. The table provides
interesting comparisons on the assignment of major taxes in selected countries. It is interesting to note
that customs duties which would clearly be a central levy, based on the theory, is a share' tax in
Malaysia and in China, the tax is administered by local governments on behalf of the Centre. Resource
taxes are central levies except in India. Sales taxes are either central levies or shared taxes except in
Nigeria. The value-added tax in Brazil is a state tax whereas the same is a central levy in Mexico but
administered by states. Property tax is a truly local tax only in Nigeria, Colombia and Mexico. Residual
power for taxation lies with the central government in India and with state governments in Brazil. In the
following paragraphs, special current issues in tax assignment in selected countries are reviewed.
Brazil: The constitutional assignment of taxes in Brazil has created some difficulties. These are primarily
in the area of sales taxes. All three levels of govermnent have partially overlapping responsibilities in tax
policy development and administration for value-added type taxes. The federal government is responsible
for a manufacturer level sales tax called by the name of Industrial Product Tax ([PI). The municipalities
can levy a value added tax on services (ISS). The state- have the mandate of a general value added tax
(ICMS) whose base encompasses IPI and ISS. This multiplicity in administration raises the administration
and compliance costs of these taxes. Further, Brazil is the only country in the world with a state level
general value added tax. This has the potential of evolving into a tax with multiple rates on non-uniform
- 15 -
Table 1.5
TAX ASSIGUIhT IN SELECTED CDUTRIES
Customs Income Estates Corpora. Resourc. Sales VAT Excises Property Fees Residual& Gifts
Argentina:Base F S S S f L S S,L SRate F S S S F L S StL SAdo F S S S,L F L S S,L S
Bangladesh:Base F F F F F F L L LRate F F F F F F L t LAdm F F F F F F L L L
BraziL:Base F F S,L F F F,S,L S F,L SRate F F,S S,L F F F,S,L S F.L SAdm F F S,L F F F,S,L S F,L S
China:Base F F F F F F F F FRate F F F F F F F F FAds S,L S,L S,L S,L S,L S,L S,L S,L S,L
Colombia:Base F F F F F F F,S LRate F F F F F F F,S LAds F F F S,L F F F,L,S L
Czechoslovakia:Base F F F FRate F F F FAdm F L F,S
India:Base F F,S F F S F,S F.S,L F,S,L F,S,L FRate F F,S F F S F,S F,S,L F,S,L F,S,L FAds F F,S F F S F,S F,S,L F.S,L F,S,L F
Indonesia:Base F F F F F F FRate F F F F F F FAce F F F F F F F
Japan:Base F F F p F,L LRate F F F L F,L LAdm F F F L F,L L
... Continued on the next page
If
Tdale 1.5 - Tax Assiwgent in Selected Countries (Page 2 of 2)
Custons Income Estates Corpora. Resourc. Sales VAT Excises Property Fees Residaal& Gifts
Nalaysia:Base F,S F F F,S F F,S S,L S,LRate F,S F F F,S F F.S S,L S.LAda F,S F F F,S F F,S S,L S,L
Nexico:Base F F F F F F F F L F,S,LRate F F F F F F F F L F,S,LAdo F F F F F F S F L F,S,L
Nigeria:Base F F S F F S F,S L S,LRate F F S F F S F,S L S,LAdm F F,S S F F S F,S L S,L
Pakistan:Base F F,S F F F F F,S,L S,L S,LRate F F,S F F F F F,S,L S,L S,LAd. F F,S F F F F F,S,L S,L S,L
Papua New Guinea:Base F F F F S F.S L S FRate F F F F S F,S L S FAd. F F F F S F,S L S Fshilippines:Base F F F F F F F F FRate F F F F F F F F FAd. F F F F F S S S,L F
Soviet Union (from 1991):Base F F F F FRate F F f F FAdu 0 F S,L F,S,L L S,L
Thailand:Base F F,L LRate F F.L LAds F F,L L
Notes:F FederallCentral GovernmentS State/Provincial/Republic/Departmental GovernmentL = Local Goverroent* Shows distribution of tax proceeds.
Sources: The data presented in this tabte was drawn fron a large xrlaber of sources. Only the major sources for each country are listed below. For acorplete list of the articles and documents used, please refer to the references list at the end of this paper.Bangladesh: ESCAP (1989); Brazil: Bomfim and Shah (1991), Shah (1990); China: ESCAP (1989); Colombia: Bird (1984), Direccion General de ImpuestosNacionales (1988); Banco de La Repuilica (19 ); Czechoslovakia: Prust et al. (1990); India: Indian Finance Commission (1989); Indonesia: ESCAP(1989); Japan: ESCAP (1989); Malaysia: ESCAP (1990); Nigeria: Ashwe (1986a), Gandhi (1982); Pakistan: Akhtar (1990); Papua New Guinea: ESCAP(1989); Philippines: ESCAP (1989); Soviet Union: Alexashenko (1990); Thailand: ESCAP (1989); Yugoslavia: Graham (1984);
bases in the long run. Already inter-state tax crediting issues remain by and large unresolved and are fast
becoming a source of major concern for the Council of States. A second source of difficulty concerns the
assignment of the rural property tax. This tax is more suitable for administration at the state-local level
but is currently being administered at the federal level.
1j0!: In India major difficulties in tax assignment are encountered in the area of sales, excises and
related (e.g. octroi) taxes. Sales taxes are assigned to the state level, excises are administered by the
Centre and proceeds shared with the states and octroi is a local tax on intermunicipal trade. Sales taxes
are administered on narrow bases with multiple rates by various states. The number of rates vary from
6 in Orissa to 19 in Bihar and Gujarat (see Rao and Chelliah (1990), p.25). Some states consider sales
tax an important element of their redistributive policies. As a measure to reform the existing structure
of sales taxation, a broad based value-added tax to be administered by the Centre has sometimes been
proposed. In view of the strong oppposition by states, such a measure could not be implemented. The
states have also expressed dissatisfaction with the central administration of excises as their flexibility to
vary revenue and rates from this source have been eliminated. Further, the Centre finds it tempting to
raise additional revenues from administered prices rather than excises as the proceeds from the latter have
to shared with the state governments. The octroi tax is a tax on intermunicipal trade and offers varying
degree of tax exporting potential. Further, it is a source of significant revenues for local governmnets.
Thus it remains a popular tax at the local level inspite of its anti-trade bias.
Nigiia: In Nigeria, the Federal Government has vacated the personal income tax field to the States. The
administration of this tax by 22 states individually create significant problems in tax administration and
compliance. Tax enforcement practices vary by individual states. This special arrangement severely limits
the redistributive role of the federal goverment. The role of income tax as a redistributive tool is also
significantly weakened.
-18 -
Sgyiet Union: In the Soviet Union, all taxes are administered at the local level and then revenues
returned to Republics and the Centre. The Centre retains the full proceeds of taxes on international trade
and shares other taxes in varying proportions with the republics. In 1990, republics received 47.5% of
the share of profit payments, 76.5% of the tumover tax, 95% of income taxes on state form cooperatives
and public organizations, 54.2% of poll taxes, 100% of excises and other local taxes and 89.3% of state
social insurance contributions. From 1991, the Centre/Union Government will piggyback republics on
a profit tax with a republican rate of 23% and Union rate of 22% for a total general rate of 45% on
profits. In addition, it will retain sole access to customs and other taxes on international trade. Turnover
tax will be shared by the Union with Republics in accordance with a formula to be established by the
Union Ministry of Finance. Republican governments, on the other hand, will have shared access to the
profits tax and sole access to income taxes on persons and enterprises and excises and local levies (see
Sichev and Ilyin (1990)). It is expected that as republics seek greater autonomy from the Union, tax
assignment would become a major issue and Union access to taxes administered by the republics will be
curtailed and therefore, the Union may seek a centrally administered tax for its sole use. One suggestion
has been to replace the existing turnover tax by a general value added tax administered by the Union.
uXggslavia: In Yugoslavia, federal government has access to three revenue sources only: customs duties,
sales tax and federal revenue stamps. These three sources of revenue account for 61% of federal budget
and the remaining 39% is derived from grants from lower levels. Thus the federal government in
Yugoslavia does not have any access to progressive income taxes (see Bogoev (1990)).
Mexico: In Mexico, lower level governments have access to only a limited set of revenue sources namely
fees, charges and property taxes.
Pakistan: Pakistan also does not allow direct access to major revenue sources to lower levels of
governments but instead revenues from major sources are shared using population as the principal
criterion.
- 19 -
1.3 TAX BASE AND REVENUE SHARING MECHANISMS
Tax base and/or revenue sharing mechanism are frequently used to address fiscal imbalances
or a mismatch of revenue means and expenditure needs arising from constitutional assignment of taxes
and expenditures to different levels of governments. The tax base sharing refers to two or more levels
of government levying own rates on a common base. Tax base determination usually rests with the higher
level government and lower levels of government levy supplementary (piggyback) rates on the same base.
Tax collection is by one level of government, more commonly the federal/central government in market
economies, and the local government (sometimes termed as "riding the co-tails") in centrally planned
economies. Proceeds are then automatically shared either downwards or upwards based upon revenue
yields. Tax base sharing is quite commnon in developed countries and almost non-existent in developing
countries.
A second alternative to address the vertical fiscal imbalances is to structure some sort of
revanue sharing mechanism. By this mechanism, one level of government is given an unconditional access
to a specified share of revenues collected by another level of government. Often such agreements specify
means of sharing specified revenues among the federal government and multiple governments at lower
levels. As a result criteria specified for allocation is often complex and quite frequently conditions are
also imposed on the eligibility and use of such funds. Needless to say that such conditionality runs
counter to underlying rationale of unconditionality of such funds. Revenue sharing mechanisms are quite
common in developing countries and often they attempt to address multiple objectives namely revenue
sharing, equalization, regional development, earmarking etc. These mechanism also vary a great deal
across countries. In Brazil, India and Nigeria, complex grant allocation formulae employing a number
of factors such as population, per capita income, school enrollments, backwardness index (India),
"minimuum responsibilities"(Nigeria) are used. On the other hand, quite simple criteria are used in other
- 20 -
countries. For example, Mexico and Pakistan use population and derivation and Malaysia and China
primarily use derivation as the basis for revenue allocation.
As the existing revenue sharing practices in developing/transiition economies attempt to
achieve multiple objectives beyond tax sharing, it is best to postpone their discussion to a later section
(Section 2) after having reviewed the economic rationale and practical design issues for intergovernmental
transfers.
2.0 INTERGOVERNMENTAL TRANSFERS: THEORY AND PRACTICE
This section provides an overview of the theory and practice of intergovernmental transfers.
2.1 Intergovernmental Transfers: A Review of Theoretical Underpinnings
This section provides a brief introduction to the voluminous theoretical literature on the
economics of intergovernmental transfers.
2.11 Intergovernmental Transfers: A Conceptual Perspective
In attempting an economic evaluation of intergovernmental transfers in developing/transition
economies, it is instructive first to review economic rationale of transfers in a federation. This review
will serve as a framework for passing qualitative judgements on the design of existing transfers. From
the theory one can predict the general direction of influence a grant may have on the recipient's
behaviour. Of course, the magnitude of this influence can only be determined by an empirical analysis
which is beyond the scope of this paper (see Shah (1979), (1989a) for an econometric model which
distinguishes the income and price effects of grants empirically). Some knowledge of the general
direction of grants is useful both in designing grant programs to meet specific objectives as well as to
- 21 -
evaluate the existing structure of transfers. In the following subsections, major types of grant programs
are discussed and their rationales are also reviewed.'
2.111 Grant Types
Intergovernmental transfers, for the purpose of economic analysis, can be broadly classified
into two categories: non-matching and selective matching (see also McMillan, Shah and Gillen (1980)).
A discussion of these two categories follows:
2.1111 Non-mrtching Transfers
Non-matching funds may be either selective or general (conditional or non-conditional).
Selective non-matching transfers offer a given amount of funds without any local matching required
provided it is spent for a particular purpose. The effect of this upon a lower level government's
(hereafter referred as local government) budget constraint is shown in Figure 1. The post-grant budget
line (ACD) is the pre-grant budget line (AB) shifted right by the amount of the transfer (AC). Since OE
(equal to AC) of the assisted activity is free good from the local perspective, at least OE will be acquired
and perhaps, but not necessarily, more.
Selective non-matching grants are best suited as a means of subsidizing activities to which
the higher level government (say the Federal Government) assigns a high priority but are given low
priority by local governments. Such a case would occur if a program generated a very high degree of
spillovers up to some level of provision (ike OE) after which the external benefits abruptly terminate.
Although, there is no evidence to suggest that these features characterize state and local provision of
education, health, regional developoment and agriculture in Brazil, yet almost all the funding through
convenios (negotiated transfers) is of the selective non-matching variety.
- 22 -
If the non-matching grant is unconditional or general, then no constraints are put upon how
it is to be spent. Tlus unlike the previous case no minimum expenditure in any area is expected. In this
case, the recipient's budget line is shifted upwards and to the right throughout by the amount of the grant
(AC=BD) and the new budget line becomes CD as opposed to AB (see Figure 2). The grant monies can
be spent on any combination, public goods or services and/or to provide tax relief to residents. Since
general non-matching assistance does not modify relative prices and must not be spent for a particular
activity, it is the least stimulative of local expenditures. Typically, local expenditure will only increase
by less than $0.50 for each additional $1.00 of unconditional assistance with the remaining funds going
towards tax relief (i.e. made available to local residents to use for private goods and services). It has
been noted that the portion of grants retained for greater local expenditure exceeds local government's
own revenue relative to resident's incomes. This is referred to as the "flypaper effect" (Gramlich
(1977)); that is grant money tends to stick where it first lands. The implication is that for political and
bureaucratic reasons grants to local governments tend to result in greater local expenditure than would
result if the same transfers were made directly to local residents (see McMillan, Shah and Gillen (1980)).
T'he federal tax transfers to states and municipalities in Brazil are examples of grants of this
variety. There is some evidence to suggest that such transfers are inducing municipalities to underutilize
their own tax bases.
Given an amount of available assistance, recipients of grants prefer unconditional non-
matching transfers. This is because these grants provide them with the maximum flexibility to pursue
their own objectives as they augment resources without influencing their pattern of spending. Hence, the
recipient is able to maximize his own welfare. The grantor, however, may be prepared to sacrifice some
of the recipient's satisfaction in order to ensure that the funds are directed towards expenditures on which
he places a priority. This is particularly so when federal objectives of affording fiscal assistance to states
and municipalities is implemented through the programs of many different departments (e.g. planning,
- 23 -
OtherPublic Goods(NCzS)
A --- C
-X dPublic Service
0 (~~~~~~~~~~~~~NCz$)
FIGURE 1 : CONDITIONAL NON-MATCHING GRANT
OtherPublic Goods Ci(NCzS)
A
B D Assisted Public Service(NCzS)
FIGURE 2: UNCONDITIONAL NON-MATCHING GRANT
- 24 -
health, education etc.) rather than through the Ministry of Finance which does not have a dominant
influence in any specific area. In this case, the different departments do not want their program funds
to be shifted , or even seem to be shifted, by local governments towards expenditures in other areas. In
this situation, conditional (selective) non-matching "block" grants are attractive. Such funding of state-
local assistance can ensure that the funds are spent in a department's area of interest (e.g. health care)
yet need not distort local priorities among alternative activities or induce inefficient allocations in that
expenditure area. This would appear to in part be the objectives pursued through the non-matching health
transfers programs in Brazil, although as presently structured, these programs may not be attaining these
objectives in the most cost effective manner.
2.1112 Selective Matching Transfers
Selective matching grants are those which require that the funds be spent for a specific
purpose (i.e. conditional) and that the recipient matches the funds to some degree (e.g. 50:50, 66.3:33.7
etc.). They are also called cost-sharing programs. Consider a 25 percent subsidy program for
transportation. The effect of such a program on local government budget constraint is shown in Figure
3. Without this transportation subsidy, AB indicates the combinations of transportation and other public
goods and services a city can acquire with a budget of OA (equal to OB). Introducing a federal subsidy
for transportation amounting to 25 percent of transportation expenditures (or $3.00 of local funds for each
$1.00 of grant), the budget line of attainable combinations becomes AC. At any level of other goods and
services, the community can obtain one-third more transportation services than previously.
If the community chose combination M prior to the grant, it will likely select a combination
such as N afterwards. At N, more transportation is acquired. This results from two effects. One due
to the subsidy the community has more resources (i.e. higher income) and some of those go to acquiring
more transportation services. This is the income effect. The income effect is reinforced by the price or
- 25 -
substitution effect. Since the subsidy reduces the cost (or price) per unit to the community of
transportation services, they acquire more (even if total resources were no greater). Hence, both the
price and income effects of the subsidy stimulate the acquisition of more transportation services.
Although the grant is for transportation, more of other public goods and services may also
be acquired, although the relative price of other goods increase (i.e. they become more expensive) and
the price effect works against them. If the positive income effect is sufficiently large, it will dominate
and a part of the grant's effect will be greater level of consumption of other goods and services. Most
studies tend to find that for grants of this kind, expenditures in the specified area increase by less than
the amount of the grant with remainder shifted towards other public output and tax relief. For example,
$1.00 of grant tends to increase expenditures in the assisted area by $0.80. The other $0.20 is diverted
to finance other public sen ices (see McMillan, Shah and Gillen (1979) and Shah (1979), (1989)).
Open ended matching (no limit on available assistance of matching provisions are met) grants
are well suited for correcting inefficiencies in public goods provision arising from benefit spillovers.
Benefit spillovers , or externalities, occur when services provided and financed by a local government
also benefit members of other local governments without their contributing to finance their provision.
As members of the providing government perceive all the costs but obtain only a portion of the benefits,
they will tend to under-provide the goods from the social viewpoint. If compensation arranged through
negotiation among the affected communities is not feasible, the situation can be corrected by a senior
government subsidizing the provision of the public service. The extent of spillover should determine the
degree of subsidy (matching ratio for grant).
While matching grants can correct for inefficiencies arising due to spillovers, they do not cope
with problems arising from uneven or inadequate fiscal capacities among state-local governments. Hence,
local governments with ample resources, e.g. the State of Sao Paulo in Brazil, can afford to meet the
matching requirements and so can acquire a substantial amount of assistance. However, other
- 26 -
govermnents such as the States of Para and Ceara in Brazil with more limited fiscal capacities may be
unable to devote as much to match federal funds and so fail to obtain as much assistance althouxgh their
requirements may be no less and, indeed, their needs may even be greater. Other forms of assistance
are needed to equalize fiscal capacities.
Closed ended matching (funds are provided upto a limit only) provisions are usually preferred
by grantors as they then are better able to control their budgets. The effect of closed ended matching
grants on the local budget constraint is shown in Figure 4. AD is the original budget line with when no
subsidies are available. However, when assistance for say education is available at say 66.7 percent rate
up to a limit, the budget line is ACD. Initially, costs are shared on a one-thrd:two-third basis upto a
level of OF at which the subsidy limit of CD'(=CE) is reached. Expenditures for education beyond OF
receive no subsidy so the slope of the budget line is 1:1 rather than 1:3 along the subsidized segment,
AC.
Although one would not expect closed ended grants to stimlate expenditures ov the
subsidized activity more than open-ended grants, empirical literature typically finds them to be more
stimulating. Gramlich (1977) and Shah ((1979),(1989a)) report that the estimated response to an
additional $1.00 of this kind of grant is typically $1.50. Institutional factors may explain the rather large
response. Gramlich notes that this type of grant is frequendy used to encourage spending in areas with
elastic (i.e. responsive to income and price effects) demands, the grants are large relative to normal
spending by recipients in these areas, and/or the granting governments take measures to discourage the
reduction of recipients' expenditures on these aided activities.
One might reasonably ask why the extensive use of selective closed ended matching grants
by developed countries when they seem ill-designed to solve problems and inefficiencies in public goods
provision. The answer seems to be that correcting for inefficiencies is not the sole and perhaps not even
the primary objective sought with their use. Rather they are employed as a means of assisting local
- 27 -
OtherPublicGoods(NCz~)
AX
< \ ~25% subsidy
0 8 C Transportation ( S)
N~~~~~~~~~~~~~~(cs
figure 3: OPENS ENDED MIATCHING GRANTS
.4
OatherPublicGo>ods
04CM~~~~~~~~~~~~4
A~~~~~
c~~
-~~~~~~~ 28"
N '"sN
'\ \NF B Education t ~~~~NCzS)
Flgure 3: COSED4 ENDED M&TCHING GWATS
Other ~ ~ ~ ~ ~~~ 2
governments financially while promoting expenditures on activities given priority by the grantor. The
selective aspects or conditions on the spending ensures that the funds are directed towards a desirable
activity in the grantor's view. The local matching or cost sharing component affords the grantor a degree
of control over and requires a degree of accountability by the recipient. Finally, of course, the cost to
the granting government is known.
The conditional closed end matching grant has certain advantages from the grantor's
perspective but there are also disadvantages. While they may result in a significant transfer of resources
providing financial assistance, it may distort output and cause inefficiencies in doing so. The reason for
this is that this aid is often only available for a few activities which results in excessive spending in those
areas while others are underfinanced. The common argument is that local priorities are distorted. Also,
it is typical that capital outlays are subsidized while operating costs are not. This results in selecting
alternatives which are too capital intensive to be least cost. A summary view of the impact of selective
transfers is presented in Table 2.1. This table suggests that open ended selective matching grants are the
most suitable vehicles in inducing increase in expenditure on the assisted function by lower level
governments. General non-matching transfers, on the other hand, would be preferred if the objective was
simply to enhance welfare of local residents.
2.12 The Economic Rationale for Intergovernmental Transfers
In the preceding sections grants were discussed according to the different arrangements under
which they are provided, i.e., selective and general non-matching and open and closed ended selective
matching. In that discussion, the underlying rationale for grants was introduced but not elaborated upon.
In order to re-emphasize this fundamental rationale and its implications for grant structure, the basic
reasons for transfers are reiterated and summarized. Gramlich (1977), Boadway ((1980),(1990)),
McMillan, Shah and Gillen (1980) and Shah ((1979),(1983b)) among others note that there are econonic
- 29 -
Table 2.1 1
THE CONCEPTUAL IWACT Of CONDITIONAL GRANTS
Type of Crant Incoag Effect Price (aubstitution) Total Effect OAIOG Rook by Objective FunctienEff et
a; A U *1 A U e A U Increase inExenditur Wlf.re
A. SELECTIVE MATCHINC
; Open-ended t t t t t I tt tt tU ) I 3
ii. Close--nded
binding constrelnt t t t t t tt It 2 I 2 or 3 4
non-binding constraint t t t - _ _ t t t 1 3 2
0 SELECTIVE NN-MATCHING t t t - - - t t t 1 3 2
C. GENERAL NON-MATCHING t t - - - t t < 3 1
Notes: * = Assisted subfunctionA = Assisted FunctionU = Unnssisted functions (service.)C = Grant
Source: Anwar Shah (1979, 1985)
and political justifications for grants. Economic justifications include efficiency (spillovers, common
market arguments, differential net fiscal benefits), equity (fiscal gap, differential net fiscal benefits,
redistribution) and to a minor degree stabilization objectives. An interesting aspect of the theory of grants
is that efficiency and equity objectives are often complementary. Boadway ((1980),(1990)) suggests that
application of efficiency and equity principles results in four main economic reasons for grants. These
are discussed below:
(i) Inter-jurisdictional Spillovers: Intergovernmental transfers can be used to increase the
efficiency with which public goods and services are provided. Their major contribution
is to correct for inefficiencies arising in the presence of interjurisdictional spillovers.
Spillovers usually occur because the benefits of a locally provided good or service itself
spills beyond the local jurisdiction to benefit those not contributing to the costs (e.g.,
benefits from control of air and water pollution, and locally educated students who
relocate) and because non-residents come to the locality and enjoy the public services
provided (e.g. parks, cultural, recreational and transportation facilities, state universities,
state welfare and health care system). In the presence of spillovers, state-local
governments consider own benefits only and under-provide public services. While other
approaches such as redrawing of jurisdictional boundaries and/or separate jurisdiction for
each service may also be used (McMillan (1975)), intergovernmental transfers are a major
and often the most practical means of alleviating the inefficiencies arising from spillovers
of this kind. Open ended conditional matching grants which modify relative prices are the
kind of transfers that is most appropriate for implementing these corrections. The extent
of cost sharing by the higher level of government should be consistent with the degree of
spillover.
- 31 -
(ii) Fiscal Gap: A mismatch between revenue means and expenditure needs at various levels
creates a structural imbalance resulting in revenue shortfall usually for a lower level
government. This imbalance is characterized as a fiscal gap. Four often cited reasons for
this imbalance are:(a) inappropriate expenditure and tax assignment;(b) limited and or
unproductive tax bases available to a lower levels of government so that tax rates would
have to be inefficiently high (e.g. in Mexico and Pakistan);(c):regional tax competition i.e.
state and local governments fearful of losing capital, labour and business activity to other
jurisdictions do not fully exploit business tax potentials and thus provide lower levels of
public services ; and (d): federal government crowds out tax room for state and local
public sector.
To correct for problems associated with (a) and (8) above joint occupancy of some tax
fields or decentralization of some taxes are advocated. Alternately unconditional grants
and/or revenue sharing based on origin principle are also appropriate solutions to deal with
this problem. Higher revenue effort by the federal government and unconditional grants
are required to deal with tax competition issues discussed in (c) above. Finally, to deal
with (d) above some form of tax abatement by the federal government is necessary to
provide more tax room for lower levels in fields jointly occupied by the two or three levels
of government.
(iii) Minimum Standards of Services: A third justification for intergovernmental transfers is
based on an efficiency-cum-equity case for ensuring common minimum standards across
jurisdictions in a federation. For certain services, expenditure assignment to state and local
government may be based on efficiency of public service provision as well as
responsiveness to local needs and concerns. This may conflict with the national equity and
efficiency objectives (see Boadway (1980), (1990)).
- 32 -
Musgrave (1976) has argued that the redistributive role of the public sector is best
performed by the federal government. Mobility of factors within a federation severely
limits the redistributive role of local governments. The New York city provides a prime
example in this case. Redistributive policies adopted by the city in the 1970s created a
major fiscal crisis and the federal government had to intervene and reverse these policies
to restore the financial health of the city. Hence, the case for the assignment of
redistributive function to the federal government is quite strong on theoretical grants.
Several of the public services assigned to state and local governments on efficiency or
accountability grounds are strongly redistributive in character. Social insurance, health,
education and welfare are prime examples of such services. Consider the case of health
and education services. These are quasi-private goods and strictly from a technological
efficiency point of view would be best provided by the private sector. Indeed in the U.S.
health care is by and large treated as a private good. International development finance
institutions and agencies, in recent years, have marshalled a great deal of resources to
advocate private provision of health and education services in developing countries based
on this view of economic efficiency. Needless to argue that such a viewpoint completely
ignores information asymmetries such as moral hazard and adverse selection associated
with private provision of such services. Fiscal federalism literature has argued that
informational inefficiencies alone do not provide a convincing case for the public provision
of health care and education. This literature points out that most governments treat health
care as a fundamental public responsibility and strive to provide these services on a
uniform basis. This is because these services are viewed as "redistributions in-kind".
Thus a case for public provision of these services primarily rests on equity objectives.
Consider the case of health services. Incidence of disease is directly correlated with the
- 33 -
incidence of poverty and by corollary inversely associated with economic well-being. Thus
public finance and provision of health care enhances the redistributive role of the public
sector. Similarly public provision of education, by improving access of the poor, serves
to further equality of opportunity goals. Relative importance of expenditures on health,
education and social services further suggests that the redistribution effected by the tax
system or direct cash transfers pale in comparison with the in-kind redistribution made
possible by these public services.
In a federal system, lower level provision of such services while desirable from efficiency,
preference matching and accountability perspectives, create certain difficulties in the
fulfillment of federal equity objectives. Factor mobility and tax competition influences
create strong incentives for lower level governments to under-provide such services as well
as restrict access of such services to those most in need such as the poor or the old in view
of their greater susceptibility to disease and thereby posing potentially greater risks for cost
curtailment. These perverse incentives can be alleviated by conditional selective non-
matching grants from the federal government. Such grants would not effect local
governments incentives for cost efficiency while ensuring compliance with federally
specified standards for access and level of services.
A second justification for common minimum standards for public services in a federation
is based on economic efficiency considerations. Common minimum standards would help
reduce interregional barriers to factor and goods mobility and thereby contribute to
efficiency gains. Establishment of some minimum standards of social services will
encourage labour mobility and of infrastructure capital will enhance factors and goods
mobility. Boadway has emphasized that harmonization of expenditures would improve
gains from interregional trade and help foster a common internal market.
- 34 -
Common minimum standards for public services across different states can be encouraged
through conditional non-matching or conditional closed ended matching programs. A
conditional non-matching program is to be preferred due to its non-obtrusive nature. State
governments would be free to spend grant monies as they chose provided cerain minimum
standards of service and access are met. The higher level government will simply monitor
compliance with these standards. Note that conditional non-matching grants in pursuit of
common minimum standards serve both efficiency and equity objectives.
(iv) Differential Net Fiscal Benefits Across States
Differential net fiscal benefits across various states arise due to a number of reasons:
a. Some states are better endowed in natural resources and therefore have better access to
an enlarged revenue base than others;
b. Some states or localities have relatively higher incomes and therefore greater ability to
raise revenues from existing bases; and
c. Some states or localities have inherited higher cost disability factors such as lacking the
minimum threshold for scale economies or difficult terrain factors etc. or higher need
factors such as greater proportion of young, old and the poor.
The presence of differential net fiscal benefits encourages fiscally induced migration.
Labour and capital may move to areas with positive net fiscal benefits based on fiscal
considerations alone. In the process, some negative externalities imposed on the
jurisdictions they leave and the jurisdictions they enter are ignored. A fisherman from the
State of Para, Brazil (a resource poor state) may migrate to the State of Minas Gerais,
Brazil (a resource rich state) , although he may not have any gainful employment
opportunity there. Overall result of such fiscally Induced migration would be that too
many of the factors will move to resource rich areas creating in its wake social and
- 35 -
economic problems and thereby serious inefficiencies and inequities. Inefficiency arises
as factor movement takes place in response to fiscal considerations alone. Inequity is
caused by identical persons in various states being treated differently by the public sector
as a whole. National welfare is reduced by the externalities imposed by the fiscally
induced migration.
Fiscal equalization grants to eliminate/reduce differential net fiscal benefits across states
can enhance both the efficiency and equity of a federal system. An ideal form of such
transfers would be an interstate revenue pool which provides both negative and positive
equalization grants to member states such that net transfers equal zero. Thus the prograin
by design will be self financing. Such a grant system must be unconditional and must not
reward strategic behaviour to enhance positive grant entitlement or minimize negative
transfer by member states. Thus grant design must incorporate factors over which states
have little control. Three grant programs that have endured and are broadly consistent
with the above guidelines are the West German, Canadian and the Austalian systems of
equalization transfers. The West German system is a fraternal system of equalization
among the German states. The federal government simply acts as an observer and
occasionally as a mediator. The Canadian and Australian system are not self financing and
instead are federal programs. The Canadian system attempts to augment the fiscal capacity
of member provinces up to the national average capacity. The system measures the fiscal
capacity of a state by the revenue that could be raised in that state if the state government
employed all of the standard revenue sources at the nationwide average intensity of use.
The Australian system worries about expenditure needs as well (see Shah (1983c),
(1984a)). A further but infrequently mentioned objective of these transfers is to advance
stabilization policies of the federal government as discussed below:
- 36 -
(v) Stabilization Objectives: Intergovernmental trasfers can also be turned to assist in
achieving economic stabilization objectives. Grants could increase in periods of slack
economic activity to encourage local expenditure and diminish during the upswing of the
economic cycle. Capital grants would be a suitable instrument for this purpose. Care
must be exercised in ensuring that funds are available for operating expenditures associated
with such initiatives.
2.13 Criteria For the Design of Intergovernmental Fiscal Arrangements
The above discussion outlined economic objectives of various types of grant program. These
objectives provide guidance as to the type of assistance that is suitable to meet those objectives. In the
following, some general principles of grant design which further assist in grant design and in ranking
various grant design options, are outlined.
Autonomy: The subnational government should have complete independence and flexibility in setting
their own priorities. They should not be constrained by the categorical structure of programs and
uncertainty associated with decision making at the Centre. Tax base sharing where subnational
governments are allowed to piggyback and introduce their own tax rates on central bases, formula based
revenue sharing or block grants should be consistent with this objective.
Rgvonue Adequacy: The subnational governments should have adequate revenues to discharge designated
responsibilities.
E5quiy: Allocated funds should vary positively with fiscal need factors and inversely with taxable
capacity of each province.
Predictability: The grant mechanism should ensure predictability of subnational governments shares. This
objective could be achieved by publishing five year projections of funding availability.
- 37 -
Efcf=ny: The grant design should be neutral as to subnational government choices about allocation of
resources to different sectors or different t,pes of activity. The current system of transfers In some
countries to finance public sector wages at lower levels e.g. in Sri Lanka (see Shah, 1990) contravenes
this criterion.
Simplichiy: The subnational governments allocation should be based on objective factors on which
individual units should have little control and the formula should be relatively easy to comprehend so that
"grantsmanship" is not rewarded. Plan assistance in Pakistan and India appears to encourage
"grantsmanship" by the recipients.
Incentiye: The proposed design should provide incentives for sound fiscal management and discourage
inefficient practices. Thus there should be no specific transfers to finance the deficits of subnational
governments. The current system of central transfers to finance subnational govenment deficits in India,
Pakistan, Sri Lanka and Soviet Union and state transfers for the same purpose in Malaysia clearly violate
this criterion and should be revamped.
Safeguard of Grantor's Objectives: The grant design should ensure that certain well defined objectives
of the grantor are properly adhered to by the grant recipients. This could be accomplished by proper
monitoring, joint review of progress and by providing technical assistance. Alternately a selective
matching transfer program would have to be structured.
It is quite obvious that various criteria specified above could be in conflict with each other
and therefore, a grantor would have to assign priorities to various factors in comparing various policy
alternatives.
In the following, various aspects of design of equalization grants are discussed.
- 38 -
2.14 The Design of Fiscal Equalization Grants: Issues and Options
An equalization program in addition to safeguarding national objectives of providing certain
minimum levels of public services across the nation can foster a greater sense of participation In the
federation of member states and therefore is often viewed as a glue that holds a federation together.
Economic literature has long recognized that equalization is justified on horizontal equity grounds and
in recent years, that under certain conditions it could promote economic efficiency. The following
paragraphs deal with some fundamental issues in designing an equalization program.
In principle, a properly designed fiscal equalization transfers program would correct
distortions caused by fiscally induced migration. Such a program would equalize net fiscal benefits across
states and thereby promote economic efficiency. To measure net fiscal benefits reasonable approximation
of costs and benefits of public services provision in various states is essential. This requires developing
measures of differential revenue raising abilities and the costs of provision of public services of the
member states. Equalization of net fiscal benefits could then be attempted by adopting a standard of
equalization and establishing the means of financing these transfers. These and related issues of
unconditionality; tax effort; stabilization effects; and employment of strategy are discussed in the
following subsections.
Measurement of Fiscal Capacity
The estimation of fiscal capacity, i.e., the ability of governmental units to raise revenues from
their own sources, is difficult both conceptually and empirically. The alternative measures of fiscal
capacity are unlikely to show approximately the same results. Of a large variety of such measures that
are available the two most prominent measures are discussed below.
- 39 -
MacroJIdicato
Various income or output measures serve as indicators of ability to bear tax burdens by the
residents of a state. Among the better known measures are:
(i) Personal Income: The personal income of a state is the sum of all incomes
received by the residents of a state. It is not a satisfactory measure of overall
fiscal capacity as it is a measure of ability to bear tax burdens but a highly
imperfect and partial measure of ability to impose them.
(ii) Personal Disposable Income: Personal disposable income is defined as
personal income less direct taxes. This concept of income shares the
weakness of personal income as a measure of fiscal capacity.
(iii) State Gross Domestic Product: It represents the total value of goods and
services produced within a state. It also is an imperfect guide to the ability
of a state government to raise taxes as it is conceivable that a significant
proportion of income may accrue to non-resident owners of factors of
production. Even if these conceptual problems are ignored, significant errors
in estimating provincial gross domestic product in Brazil are encountered to
make the measure unacceptable for an equalization program.
We already noted that the equalization of net fiscal benefits across states is required from the
standpoint of economic efficiency. The estimation of these net benefits is best done by a comparative
analysis of taxing and spending behavior of state and local governmental units. Various income concepts
do not relate to the taxing practices of the states but merely indicate what they potentially have available
for taxation. Fortunately, a representative tax system approach is well suited for such a task. This
approach is discussed below.
- 40 -
The Representative Ta System (RTS)
This system measures the fiscal capacity of a state by the revenue that could be raised in that
state if the state government employed all of the standard sources at the nationwide average intensity of
use.
To estimate equalization entitlements based on a representative tax system approach,
information on both the tax bases and tax rates for each state is required. For most revenue sources this
information is usually readily available. Then a decision has to be made as to the standard of equaiization
i.e. whether fiscal capacity of the have-not states should be brought up to the median, arithmetic mean
or some other norm based on all states data. As an example, consider arithmetic mean of all states as
a standard. Then equalization entitlement for a state say (x) for resource source i could be determined
as follows:
E' - (POP),, [{(PCTB), x tj
- {(PCTB)z x 4t}J
where Ei = equalization entitlement of state from revenue source i.
POP = Population
PCTB' = Per capita tax base of revenue source i.
= national average tax rate of revenue source i.
subscript na = national average
subscript x = State x
The equalization entitlement for a state from a particular revenue source could be negative, positive or
zero. These figures would then have to be summed up for all revenue sources considered for equalization
and the overall sum would indicate whether a state would receive a positive or a negative entitlement
from the interstate revenue sharing pool.
- 41 -
It should be noted that the data on tax bases and tax collections required for the
implementation of a RTS are usually published on a regular basis by various levels of government. Thus
the RTS does not impose any new data requirements and could be implemented using the existing data
in many developing countries.
Measurement of Expenditure Needs
Economic theory suggests that an ideal equalization transfers program should also take into
consideration the expenditure side of the provincial-local budgetary operations. Many economists have
argued for taking expenditure needs and differential unit costs of provision of public services into
account. Several countries follow this approach. The following paragraphs examine this issue.
We already noted that the case for equalization rests on differential net fiscal benefits across
states. These differentials could arise either due to differences in revenue raising capacity and/or due to
differences in the cost of provision of public services. Consider two states with the same revenue raising
capacity where the residents have identical tastes for provincial public services but the cost of providing
them differs due to supply factors. For example, differences in the degree of urbanization, population
density and age distribution among states will have significant effects upon the relative costs of public
services. The degree of urbanization can effect the costs of salaries and wages, land, and construction,
as well as particular services such as pollution control, public transit, police and fire protection and sIhe
provision of utilities. Population density will effect the costs of providing public utilities and will also
effect the costs of highways. Age distribution will influence the need for school rooms, hospitals,
recreational facilities etc.
These differential costs are likely to cause substantial variations across the two jurisdictions
in the level and mix of public goods provided, resulting in differential net fiscal benefits. A seng all
- 42 -
for equalization can, therefore. be established on both efficiency and qily grounds to compensate for
cost differentials that give rise to differential net fiscal benefits.
The fiscal federalism literature, in general, treats differential costs as synonymous with
differential needs but it must be noted that some cost differences may arise due to deliberate policy
decisions of the provincial governments and thus do not constitute need. Compensation for unavoidable
cost variations resulting from differences in the costs of inputs and from dissimilar input-output
relationships which might arise because of distance from sources of supply and geographic features can
be justified on equity grounds. Equalization grants should offset such inherent disabilities but should
disregard cost differences due to differences in efficiency with which resources are used.
Expenditure need is more difficult to define and derive than a measure of fiscal capacity. The
difficulties involved in measuring expenditure need are substantially higher than those encountered in
using a representative tax system to measure fiscal capacity. They include defining an equalization
standard, determining differential costs due to differing input-output relationships, nature of service areas,
composition of population and isolating need/cost differentials due to differential tastes or policy decisions
as distinct from inherent cost disabilities. A further concern would be the susceptibility of the grant
shares based on need factors to strategic ibehavior on the part of the recipient states. The experience of
Australia, West Germany, Switzerland wlth federal unconditional transfers, of the U.S. with highway
grants and of the Canadian states with provincial-municipal transfers, indicates that these concerns can
be addressed and expenditure need incorporated in formula grants in a manner acceptable to both the
donor(s) and the recipients.
Some empirical questions are resolved easily. For example, to avoid problems associated with
subjective standards such as 'minimum service levels' or 'reasonable levels of services', expenditure need
could be defined as "the cost. of supplying average performance levels for the existing mix of provincial -
local programs". Relative expenditure needs could then be determined empirically either using direct
- 43 -
Imputation methods or by adopting a simpler approach using a representative expenditure system. The
latter approach is preferred for its objectivity and ease of computation. Furthermore, it enables the
analyst to derive expenditure need measures based on actual observed behavior of the provincial-local
governments under study rather than basing it on ad hoc value judgements. The relative weights to be
assigned to various need factors in the representative expenditure system could be determined by
econometric analysis. This method requires specification of determinants for each service category to
be analyzed. These determinants would include relevant fiscal capacity and public services need
variables. The estimating equation so specified would then yield quantitative estimates as to the
independent influence of each specified factor in determining the spending level for that category of public
service. This information could also be analyzed further to determine as to what each state would have
actually spent if it had average fiscal capacity and average tastes but actual need factors.
More specifically, the formula for equalization entitlement on account of expenditure
classification i for state x could be stated as follows:
EE'1 = (POP). ((PCSE)'X - (PCSEY)'
where
EE' = Equalizataion entitlement on account of expenditure classification i for
state x
PoP. = Population of state x
(PCSE)'I - Per Capita standardized expenditure by state x on expenditure
classification (i). This is the estimated expenditure which a state
would have spent if it had national average flscal capacity but its
actual need factors.
- 44 -
(PCSE)'E, = Nadonal average per capita standardized expenditure on expenditure
classification i. 'iis is the estimated expenditure for all states based
on national average values of fiscal capacity and need factors.
Equalization entitlement on account of a particular expenditure classification could be positive, negative
or zero. These entitlements would have to be summed up for all expenditure categories considered for
equalization.
Overall entitlement of a state based on a comprehensive system of equalization would be
determined by summation of its separate entitlements from the Representative Tax System (RTS) and the
Representative Expenditure System (RES). Only the states with positive entitlements would be eligible
to receive transfers in equivalent or some fraction of the total amount (the fraction to be determined by
the centre depending upon the availability of funds) from the centre.
A phased approach to a comprehensive equalization may be well advised. Initially, a
representative tax system could be implemented for a five year period. 'hen depending upon this
experience a representative expenditure system could be brought in to complement the RTS in the next
five years. A joint Federal - States Fiscal Arrangements Sub-Committee may be instituted to monitor
the working of the system closely.
The Equalization Standard
Equalization of net fiscal benefits requires that we adopt an explicit standard of equalization.
The specified standard would be the level to which each state would be entitled to be raised to enable it
to provide public sector net benefits per household comparable with other states. Simplicity dictates
choosing either the arithmetic mean or the median of the governmental units involved as the standard.
Arithmatic mean provides a good representation of the data in the absence of extreme values. In the
event that the sample values have wide range, the median or the arithmatic mean after elimination of
- 45 -
extreme values would provide a better representation of the sample. Mean is to be preferred over the
median, however, for ease of computation.
Costs and Finaclng
An Ideal fiscal equalization program would be self-financing. The member governments are
assessed both positive and negative entitlements which sum to zero. The federal government merely acts
as conduit for such a policy. If such an interstatal revenue sharing pool in practice would create,
administrative difficulties, then the equalization program could be financed out of general federal revenues
which are in part derived from the equalization receiving states.
Other Considerations
So far the discussion has focussed on the basic elements of an equalization program. Several
related aspects of equalization transfers are considered in the following subsections.
nconditonflty
There is a general consensus in the academic literature that an equalization system should enable
state governments to provide a standard bundle of public services if the government imposes a standard
level of taxes on the bawes at its disposal. The state governments (or certainly their citizens) should,
however, be permitted to substitute lower rates of taxation for higher level of services and vice versa.
As such the equalization payments should be in the nature of unconditional grants having only income
effects. Service areas in which there appears a good reason to actually set minimum national standards
are better handled b, Wonditional grants and shared cost programs. It should be noted that by raising a
state's fiscal capacity, the unconditional equalization grants enable the poor states to more easily
participate in the shared cost programs.
- 46 -
TaxIEffort
Incorporating tax effort into the formula for determining equalization would involve making the
equalization entitlement a function of the ratio of actual tax collections in a state to its own tax base.
Potentially non-recipient states may wish to see such a factor incorporated into the program to prevent
states with a positive fiscal deficiency in an area from collecting equalization payments even if they may
not levy a tax in the area. Potendally recipient states may wish to see tax effort incorporated because
without it, extra tax effort on their part will be relatively unproductive compared to a wealthy state.
Problems exist with incorporating tax effort into the program. First, the inclusion of tax effort
will cause the program to depart from its unconditional nature. A state should be free to substitute grant
funds for revenue from own sources. Similarly, if a state raises taxes in order to provide a bundle of
services that is higher than the standard it should not receive equalization for doing so, e.g. other states
should not have to pay most of the cost if a state decides to paint its roads. Incorporating tax effort
would tie the federal government to expenditure philosophies of the various states. A problem also arises
in that some states do not have tax bases in all areas. Incorporating tax effort may also encourage the
employment of strategy by a state. Another major problem associated with the inclusion of tax effort in
the formula is that in view of the differential abilities of the states to export taxes, the measurement of
tax effort would be crude. Inclusion of tax effort in the formula could also result in increase in taxes on
the poor states. In view of the above considerations, it appears that a program of equalization payments
would not be improved by including tax effort.
Stabilization Effects
If the equalization payments were to be based upon relative measures of fiscal capacity, they
will be expected to have a stabilizing effect upon state revenues. The level of payments will move in a
direction opposite to that in which the states own revenue raising capacity moves. Maximum stabilization
- 47 -
of state-local revenues will occur when the payments are based on all revenue sources, a national average
standard of equalization is used, cyclical fluctuations in provincial economies are small and the time lag
in calculating the grants is relatively short. When any large component of the total base is quite volatile,
such as natural resource revenues, the destabilizing effects can be quite large and some sort of averaging
formula would have to be employed to ease the difficulties associated with provincial budgeting in the
face of uncertainty.
Strategy
Strategy refers basically to the actions that provincial governments can take to affect the level
of payments they receive. A scheme that enables a state to employ strategy must be considered
undesirable because in general such extra payments received may not have any relation with actual
disparities. For example, a program employing tax effort could enable states to raise their payments by
imposing heavy taxes in areas in which they have a below average base. This problem, however, is much
less serious in practice than it might appear as the room for additional taxation from sources in which the
potentially "have-not" states are not well-endowed would be extremely limited.
Concluding Remarks
Economic theory provides a strong rationale for fiscal decentralization as it promotes efficient
provision of public services by promoting a better match of these services with citizens preferences; by
minimizing the cost of political decision making and by encouraging political accountability and by
addressing regional and local concerns.
- 48 -
2.2 Intergovernmental Transfers In Praltice: An Overview
Based on the discussion on economic rationale for grants, one expects to see that the following types of
grants instituted in developing countries.
i) General non-matching transfers or tax base sharing or revenue sharing mechanisms to deal
with the fiscal gap problem. Revenue sharing mechanisms are indeed in place in a number
of countries (see Table 2.2 for details) but tax base sharing has generally not been tried.
General unconditional transfers are also not very popular and instead 'deficit" grants have
been tried in a number of countries including India, Pakistan, Malaysia and Soviet Union
(see Table 2.3 for details).
ii) Equalization transfers to reduce/eliminate differential net fiscal benefits (to deal with
horizontal fiscal imbalances). Inspite of serious horizontal fiscal imbalances observed in a
large number of countries, one cannot find a single example of an explicit equalization
program. Equalization objectives are nevertheless implicitly attempted through the general
revenue sharing mechanism used in India, Pakistan, Brazil, Mexico, Nigeria, and Colombia.
These mechanisms often combine diverse and sometimes conflicting objectives such as
revenue sharing and equalization into one and the same formula and significantly fall short
on individual objectives (see Table 2.2). The formulae lack an explicit equalization standard
and therefore fail to address regional equity objectives in a satisfactory manner ( Shah
(1990)) provides supporting empirical evidence from Brazil).
- 49 -
Table 2.2
FDERALICUTEAL REVENUE 511ING N1ECEAEISMS IN SELECTED COCN5RES
COUNTRY MeCHAUlSMS DISTRIBUTION CRITERIA COMMENTS
BrazilStates:
States Participation Pre-specified share of federal taxes A proposal to extend the formulasVund (1PE): 21.52 of transferred to this pool. Council of determining the shares of individualthe proceeds of federal States determine state shares based on states to include land areataxes on Income and area, population and per capita income (fiscal need), intersLbte tradeIndustrial Products (PCY). (spillover factor), and ratio
of own revenues to expendituresStates In the Northern, Pormula that Includes: (fiscal effort) is underNortheastern, and Center- ° Population consideration.western regions (less Inverse of Per Capita Incomedeveloped areas): Share of eacb& state i in FPE is Merits of FPE:852 of PPE (pop 1 *ypc1 ) /13T, (pop*ypc5
*) Transparency
* Predictability*Autonomy
where: Redistribitive (Some Equalization)pop1 - state I's population factor Fiscal Need (Population)ypc1 - state I's Income per capita Fiscal Capacity (PCY)
factor Basic minimum grant assuredN - total number of states° Po is given by the *t of national Flaws of FPE:population represented by each * Income as a measure of fiscal capacitystate"(2Npop); its value for each imperfect measure of ability to raisestate ranges from 2.0 (if 2Npop a 2) revenues at state levelto 10 (if ZNpop > 10) Significant measurement errors, DatedI Ypc depends On theratio of per Lacks explicit standard of equalizationcapita income of all states to per States with slmilar fiscal capacitycapita income of the state' (Rypc); receive wodely differing entitlements.its value ranges from .4 (if Para and Acre with PCY NCZ $266 (9 vs 49Rypc s .00045) to 2.5 (if Rypc > .022) p.c.) Roraima with PCY NCZ $286 NCZ $123
States Coulcil finds formula resultsStates In the Southern, Same as for less developed areas. unacceptable.and Southeastern regions:152 of PPE
Tax on Industrial Products Compensation to exporting states(102) for loss of revenues from state
VAT (ICMS) on account of federal exportincentives.
Payroll Tax (66.72) Derivation (Earmarked for education)
Taxes on hydroeletricity Originand on minerals (45Z)
... Continued on the next page
Table 2.2: FederallCentral Revenue Sharing Bechanisms in Selected Countries (Page 2 of 10)
COUNTRY IECHANLIS9S DISTRIBUTION CRITSEIA COSENTS
Local Covern ents:
Ihmicipalities Parti- Municipal Participation Fund (PPM) cipation Fund (FPs): Pre-specified share of federal taxes(Federal Taxes on populationIn-cse sd on and 8tate Per capita incomendustriAl products, 2/3rd plas of municipal revenues22.52 of the proceedsof each)
State capitals and Formula slmilar to that for FPE Merits:large =anicipalities that the population factor is tgven Population (Fiscal Need)(more than 400,000 by 02 of total population in this Recognizes Threshold in Populationinhabitants) category represented by each slze(122 of ppM) municipality* (MtpopL). State Capital and larger munc. treated
This factor ranges from 1.0 (if ZNpopLs1) separatelyto 5.0 (if 2NpopL > 5). Predictability
CDll1 and nudism sized ° Population factor Transparency and Autonomyouaicipalitles This factor is given by the municipality°s FPM Drawbacks:(Population less than population (P). No municipal iccome tax but municipal400,000) It ranges from .6 (Ps16,188) fiscal capacity measured by state PCY -(822 of PPH) to 4.0 (P > 156,216) a poor indicator of local fiscal
capacity. Also does not distinguishMunicipal Participation Same criteria adopted for state capitals poor and rich municipalities within eachReserve Fundt large and and large mtnicipalitles' share discussed state.mdlum sized nunicipalities above. Inequitable allocation
Per capita transfers to Minas Gerais(Population greater than higher than Para.156,216, except state Generosity of federal funding dlecouragescapitals (62 of FPM) local fiscal effort and contrlbutes toreduced financial accountability.Rural Property Tax Derivation rOwn Revenue Effort declining in recent(502 to Municipalities) years.Payroll deductions of Derivationmunicipal employees (1002)
Tax on Cold (702) Origin
2.32 of revenues from Origincrude oil production
Taxes on hydroelatricity Originand on minerals (502)
... Continued on the next page
Table 2.2: Federal/Central Revenue Sharing Iechanisms in Selected Countries (Page 3 of 10)
COUNTRY NECHISNS DISTRIBUTION CRITERIA CKETIM
Chia"Provinces: Although all rates and tax bases
are set by she center, most taxesInacome tax and adjustment Derivation are collected by the localtax of locally-owned enter- authorities and then upwardlypri=es, income tax from distributed according to thecollectIvely owned enter- determinations from the center.ptries, rural market trading Subnational governments havetaz on private traders, and little formal power; their budgetstaxea on housing, vehicles, are determined as part of aurban maintenance and con- consolidated central-provincial-struction, individual local budget designed by theincome, wage bowls, self- center. Provinces, howeveremployed entrepreneurs, approve the budgets and financial*laugher, cattle trading plane of localities.ad contract. (1002 retainedby aubnational governments)
Sales tax collected from Derivationenterprises owned by the
inisatry of Power, SINOPEC,and the China onferrousNetaLs Company. (30S tosubnatlonal gavernments)
Tax on sales, profits, For each province, the centernatural resources, con- determines a share of the taxstruction, foreign joint- revenues it collects that it mayventures, energy, salt, retain. This determination is madetransportt on the basis of a combination of:
o Derivation° Formula° Negotiation and ad hoc decisions
The last factor has become of increas-ing importance in recent years.The 1988 formula was:
M- Iiw lL6S8 R83-ThBR- MUM 31%6-T88
where:RR-proportion of shared taxes retainedby subnational governmentL86-=local fixed revenue in 1986WS86-shared revenue in 1986T88-the 13 taxes assigned to localgovernments in 1988R83-allowable expenditures in 1983
/Actual revenues collected in 1983L86 corresponds to the first 2 sourcesof revenue listed above.
Continued on the next page
Table 2.2: FederallCentral Revenue Sharing Hechanims in Selected Countries (Page 4 of 10)
COUNTRY MECHANISMS DISTRIBUTION CRITERIA C("UENT5
ColombiaDepartments:
Tax Allowance ° Equal shares (302) These transfers are paid to the Regional(15X of Central Govern- Population (702) Education Funds and the Sectional Healthment's "ordinary (Earmarked for primary education, Services. These are close ended non-revenues*, i.e., "all 742, and healch, 262) matching conditional grants. In practicecurrent revenues not the actual allocations often, andlegally earmarked for arbitrarily dlvers from the ones justspecific purposes.") specified. (See Bird, 1984, p. 148).
Beer Tax Estimated beer consumption(402 of the proceeds)
Tobacco Tax Tobacco consumption (the BogotaDistrict's share is limited to 202of the revenues collected in Bogotaand the Department of Cundinamarca).
Sale Tax Transfer ° Equal shares (302) Departments must transfer 402 of these(302 of the proceeds): ° Population (702) revenues to their -mnicipalities.
Czechoslovakia
Republics:
Payroll Tax Earmarked for social security(1002)
Turnover Tax MA(Shared between the Republicsand the Federal Govt.)
Local Covernments:(National Comimttees)
Individual Income Tax(1002 of the proceeds)
lndiaThe "Finance Comnissionu meets at
States and Territories "Contribution (102). (as measured least once in every five yearsFederal Income Tax by the assessment of Income Tax to make recommendations regarding(852 to States, for the years 1985-86 to 1987-88. the distribution of shared taxes1.42 to Union Territories) ° "Distance" of income/capita between center and states, the
relative to state with higher per allocation across states, thecapita income X population (452) principles governing the
allocation of grants-in-aid.
... Continued on the next page
Table 2.2: Federal/Central Revenue Sharing Mechanisms In Selected Countries (Page 5 of 10)
COUNTRY tECHiIISMS DISTRIBUTION CRITERIA CoCITS
° Population (22.52),* Backwardnessw (11.251)
(as measured by population ofScheduled Castes and ScheduledTribes and the number of agricul-tural laborers--1981 Census)° 3 luverse of per capita income Xpopulation (11.252)
Union Excise Duties Population (252) The Finance Comiasion uses 1971(45Z to States) ° Lncome Adjusted Total population data in the revenue(All union excises Populatlon (IATP) (12.52) abaring fonmalas vlth theexcept the additional (IATP- pop7l*ypc where: exception of the Additionalduties of excises and pop71- populatlon in 1971 Excise duties which uses 1981the earmarked cesses.0 ypc-inverse of the average per population (see belov). The datedcapita income of state for data on population is usedthe trienium 1982-83 to 1984-85.) purportedly to encourage populationI 3Backwardnese (12.52), defined control activities by states
above.o *Distance", defined above,
(33.52)° *Projected budget deficits
(16.52)
Additional Excises in ° State Domestic Product (50S) Adopted as a proxy for consu Lptionlieu of Sales Tax ° State population (502) of the gods previously covered by(Include excises on the sales tax given that theVI textiles, tobacco, end -ualityof the actual consumptlonsugar) data seems to be questioned byi ~~~~~~~~~~~~~~~~~~~~~~~~~~~~the Commission.Grant in lieu of Tax In proportion to the averageon Railway Paasenger of the noa-suburban passengerFares (Rs. 150 crore earnings ir ;tate relative toannually for 1990-95) national earnings (1984-88)Estate Duty on property Derivationother than land and wealthtax on agricultural property(1002 to states)
IndonesiaProvinces:
Royalties on oil and gas sales, * Derivationroyalties on forestry andmining activities, Property tax,motor vehicle tax. (1002)Tax on land and Buildings ° Deriv iion(162 to Provinces)
... Continued on the next page
Table 2.2s Federal/Central Revenue Sharing Mechanisms In Selected Countries (Page 6 of 10)
COUNTRY MECHAMISMS DISTRIBUTION CRITERIA coImENTS
Local paverriments:
Tax on slaughtering animals, Derivationstreet tax, development tax 1Iforeign people tax, radio tax.(1002)
Tax on land and Buildings o Derivation(652 to local governuents)
JapanLocal Allocation Tax: The largest share of the IAT is322 of the total of distributed by formula that includes:personal income tax, Basic Financial Need (BFN)corporation tax, and Basic Financial Revenue (BPR) whereliquor tax. the two factors above are themselves
documented in by formulae by the LATLaw and the Ministry of Home Affairs.
Local Transfer Tax Formlae that includes:Ln ° Total area of public roads.
Malaysia
lmport and excise duties on oil Derivation Revenue sharinx In Malaysia is not(302 to States) intended to address horizontal
imbalances. This suggests that aseparate explicit equalization to
Export duty on tin (102 to States) Derivation standard programs is needed.
Export duties on other mn Ierals Derivation
Ezport duties on timber Derivation (exclusively for the The states of Sabah and Sarawak wereand other forest products states of Sabah and Sarawak) granted special privileges as a condition
Export duty on mineral Sbortfall of state's royalty on minerals for joining the federation.products export from 102 of export duty payable.
(Exclusively for Sabah and Sarawak)
MexicoStates:
General Fund: 502 to states and the Federal A welcome change towards more18.12 of federal District oa the basis of Population equitable revenue sharlngrevenues subject (Starting from an initial 102 in introduced in 1990.to sharing and rising to 502 in 1994)
... Continued on the next page
Table 2.2: Federal/Central Revenue Sharing Mechanisms in Selected Countries (Page 7 of 10)
COUNTRY ECHANISMS DISTRIBUTION CRITERIA coMENTS
° The division of the remaining Arbitrary and overemphasis on502 (starting from an lnitial 902 historical shares. It would resultln 1990 and declining to 502 in in an utnecessarily long1994) is made on the basis of transition to the originhistoric states' shares adjusted (derivation) principle. Rewardsfor the increase in states, shares higher tax efforr on local taxes.of federal administered excisetaxes on gasoline, motor vehicles,alcohol, and tobacco, and thestate-local administered water andsever charges and property taxes.
Contingency Fund: To compensate states who lose from6.52 of shareable federal revenues. the change in sharing arrangements.(Will merge with the General Fundin 1997)
Municipal Fund: The share of municipalities in Small and ineffective re-2X of the revenue sharing funds. each state from the fund varies distribution.(It is intended to be passed through in lnverse proportion to thethrough states to municipalities.) shares from the General Fund.Import taxes and Petroleum Sbared with petroleum refining andexport duties exporting cities (DerivationU'principle)
0' Federal VAT States can retain increased This is intended to reward higherrevenues received through state tax enforcement effort at theaudits for their own use. state level.
NigeriaFederation Account (FA): See below States collect and retain theMost revenues collected federal personal income sax. Thisby the federal govermnent. complicates taxation of personal(Corporate income tax, income in Nigeria.capital gains tax, stampduties, import duties,excise duties miningrents and royalties, andradio and TV licenses)
EStates Joint Account (SJA) 95Z of SJA distributed according A suggestion has been made to31.5 x of FA to: include a health indicator that° Minimum responsibilities of would serve ac a socialgoverment (402). development factor (in addition(proportion of recurrent expendi- to primary school enrollment).ture to total federal revenue of The current scheme also ignoresthe state with the smallest budget the different costs of providingin the period 1976-77 to 1979-80.) public services in the different states.Population (40?)Social development factor (152)
of which:
... Continued on the next page
Table 2.2: FederallCentral Revenue Sharing Mechanisms In Selected Countries (Page B of 10)
cOUIRTR MECARNSims DISTRIBUTION CRITERIA CONHENTS
° Direct school enrollment (11.25:)l Inverse school enrollment (3.75X)
o Internal revenue effort (52).(Proportion of own revenues to totallocal expenditures of a state gover-nment.)
52 of SJA distributed to mineralproducing states on the basis ofderivation.
Local Covernments: 102 of FA Equal shares (252) The current scheme does not considerPopulation (752) capacities of local governments.
Snecial Fund: 3.52 of VA Federal Capital Territory (2.52of VA)0 States, according to ecologicalproblems (12 of FA)
Pakistan
Provinces:
802 of the proceeds from Provincial population No account of Provincial variationfederal tax on income in per capita tax bases. Lacks ancorporations, sales. explicit equalization standard.
Excise duty and profits generated Origin So far, the federal government hasfrom natural gas vell-heads and transferred only the receipts fromand hydroelectric power stations. natural gas to the provinces.
Papua Nev Cuinea
Provinces:
Mnliemmm Unconditional ° Formula:Grants' (MUG): HUG= min(A+AB,A+AC)Income and corporate tax, where:customs and excise duties. A- Amount of money spent on the activity
in 1976177B- 2 increase in the cost of llving inthe preceding fiscal yearC= 2 change in the total of the paymentto the Consolidated Revenue Fund (CRF)for the year of the grant ascompared vith the total for thepreceding year from:
... Continued on the next page
Table 2.2s Federal/dentral Revenue Sbaring Mechanlsms ln Selected Countries (Page 9 of 10)
COUNTRY mEKCAISms DISTRIBUTION CRITERIA ComNEWES
(0) the shared taxes(ii) Mineral Resource Stabilization Fund(iII) Hon-repayable grants (conditionaland unconditional from foreign government)
When substantial increases in thedistributable revenues are realized,extra funding, above MUG grants canbe authorized. Inter-provincialsharing of these funds is determinedby the National Finance Comslision(NFC) on the basis of:
° Population.
Vehicle reglstration and ° Derivationlicensing fees (collectedby center and handed over toprovinces)
Royaltiest minerals and ° Derivationpetroleum, natural gas, (These reduce the amount of derivationtimber, fish and private Grant that the province is entitled to.)oo ~~~~hydroelectricity generation.
Philippines
National Internal revenue o Population (702) All local governments are requiredtaxes whlch were collected Land area (202) to spend 202 of their internalduring the third 0 Equal shares (102) revenue allotment to development(201 to subnational governments) preceding projects approved by theDepartment of Local Governments.Mandato7, local contributions tocertain funds are as follows:° Aid to Hospitals (5 to 72 oflocal government income)° National police (182)and other smaller mandatorycontributions and expenses.
Tax sharing ignores tax effort andhas tended to magnify horizontalImbalances by transferring morefunds to richer states than to poorer.
Tax on petroleum products Derivation(Z shared with subnationalgovernments: KA)
... Continued on the next page
Table 2.2: Federal/Central Revenue Sharing Mechanisms In Selected Countrles (Page 10 of 10)
COUNTRY NEC8RUISMS DISTRIBUTION CRITERIA CfKENTS
Soviet Unlon
Remblies & Local Goverruents:
Participate on the proceeds In acccrdance with pre-1991of the taxes on enterprise structure of revenues and expendi-profits, consumption expen- tures of the budgets of the threeditures. levels of government.,
ThaliVbeicle tax 502 to municipalities in equal Municipalities are the closest(1002 of the proceeds) shares form of local self-government in
° 252 to Sanitary Districto In Thailand. 3 types of municipalitiesequal shares * Tambon
° 252 split evetly among Changwat e Town: (at least 10,000 peopleAdministration Organization. with the same population density
as city municipalities andRice export tax o 502 of distrlbutable pool to necessary financial resources or(952 of the proceeds) Sanitary DistTlets (S.D.s) on be the seat of the provincial
the basis of population. government)o 502 to aunicpalitiest the e City (at least 50,000 IaMb. and3an8kok Metropolitan Authority an averape population density of(B.N.A.), the Pattaya city. 3,0001).
The Provincial Government has no directinfluence over municipal government; it hasonly a monitoring role.
Surcharges on central Collected by the center andtaxes on business, liquor, returned to localities on thenon-alcoholic beverage, basis on derivation.entertainment, petroleumproduct, cement, andg mbling. Other special forms of local gover nent arethe S.D.s, the C.A.O.s,
the B.N.A., and the Pattaya City.S SD.s are small concentrationsof population not large enoughto constitute a municipality.
e C.A.O.s provide services torural areas.
Sources: The d2ata presented in this table were drawn from a large number of sources. Major major sources for each country are listedbelow. For a complete list of the articles and documents used, please refer to the references list at the end of this paper.
Bangladesh: ESCAP (1989); Brazil: Bomfim and Shah (1991), Shah (1990); China: ESCAP (1989); Colombia: Bird (1984), Banco deLa BepublIca (1990); Czechoslovakia: Prust et al. (1990); India: Indian Finance Commission (1989); Indonesia: ESCAP(1989); Japan: ESCAP (1989); Malaysia: ESCAP (1990); Mexico. Boadway (1990); Nigeria: Ashwe (1986a and b); Pakistan:hkhtar (1990) Papua New Guineat ESCAP (1989); Philippines: ESCAP (1989); Soviet Union: Alexashenko (1990); Thailand: ESCAP(1989); Yugoslavia: Graban (1984)
Table 2.3
FEDERALICENTRAL TRAISFERS TO LOWER LEVELS OF GOVER_
COUNTRY MECHAB1SMS DISTRIJUTION CRITERIA COMMENTS
BangladeshAnnual Upazila Development Fund
Rural Works Prograrme a Area (113) Earmarked for capital expenditureso Population (213) on physical infrastructure.
Hatching Bloc) Grants ° Population (202)for Collaborative Projects o Area (202)
° Backwardness (302)o Work progress (302)
Food for Work Programme ° In-kind (wheat) transfersto needy population.
General Purpose Grants(includes budget deficitgrants)
BrazilStates
"Statutory" Transfers Allocations to the Federal Convenlos: All selective non-matching(required to comply with District and to nerly created project grantsapeciflc ordlnary laws) states, royalties paid to "> Exp. prior. of grantor and recipient
states for the extraction of differentoil In their jurisdiction. Project reviwe and approval process
provides incentives for recipients to putConvenlos and "Special Negotiated transfers not regulated forward their best and mutual interestInvestment Funds" by any law. Based on negotiatlons projects and use the funds to finance
with each ludlvidual state or services of interest only to recipients.municipality. Incredibly large number of programs with
unspecified/vague objectives. Vehiclesfor pork-barrel politics e.g. President
Transfers made though Arbritary allocation criteria. Sarney's home state and Sao Paulofederal agencies benefitted disproportionately.
* Foreign debt relief grantsLocal GovernmentsSUDS: An important exception but still
Convenios, "Special Same as above. here Fed heavily involved in programlnvestment Funds", administration. Allocation to unitsand transfers through based on history of past health expendi-federal agencies. rures and capital projects.
... Continued on the next page
Table 2.3: federal/Central Transfers to Lower Levels of Covernment (Page 2 of 7)
COUNTRY MECHANISMS DISTRIBUTION CRITERIA COCMENTS
Difficulties: Current and Potentill1. Availability and Predictability ofFunds: Finaneed out of social securitycontributions. Range of SC benefitsexpanded by the Constitution - also taxsharing squeeze. Fiscal strains likely toconstrain federal funds for SUDS.2. Degree of Federal Involvements Fed.has the responsibillty of setting normsand distribution of funds but 60S directby Fed. and rest effectively controlled.3. Private provides receive preferentialtreatment. Shorter lag and CPI adjustment.Transfer lag 45 days vs 70 for publicsector.Education: Convenlo Unicoprimary education finance and in-kindtransfers through school lunches and textbooks.
primary education local function withno significant spillovers.
ChinaC% Province:
(Special purpose non- Discretionary grantsmatching grants)
ColombiaDepartments:
National Road Fund ane other No formulae that provide for uniformearmarked and budgetary transfers and predictable allocations.
Czechoslovakia
Federal-to-Republic NA 52S of the total revenues of the CzechRepublic and 62X of those of Slovakiacorrespond to federal transfers.
IndiaFinance Comlission:(i) Unconditional Projected budget deficits
for the next 5 year period
(ii) Conditional Disaster Relief
Continued on the next page
Table 2.3: Federal/Central Transfers to Lower Levels of Gaoernment (Page 3 of 7)
COUNTRY NECE&NISNS DISTRIBUIION CRITERIA COMENDTS
Non-Statutory Transfers Discretionary transfers account for(Transfers made outside over 50! of all transfers and oftenthe Finance Comisaion) lack objective allocation criteria.
Mi) Planning Comission (For schemes approved by(Plan Assistance) the Commis ion)o Population (602)e Tax Effort (102),° Backwardness (per capitaincome relative tonational average) (202)
° Special problems (10)(1i) Other Conditional Often these are matching grantsFederal Discretionary (See comment for Non-StatutoryGrants Transfers above)
IndonesiaSpecific Purpose Grants: (1986/87s 86.82 of all grants)- Autonomous Subsidies (SDO) To pay wages and salaries of(66.92 of all grants) subnational goverment employees- Soil Conservation Program NA(1.1 of all grants)
- Development of Primary NASchool Program(0.62 of all grants)
- Health Services Program NA(2.9! of all grants)
- Road and Bridge NAConstruction Program(3.3! of all grants)
Block Grants:(13.22 of all grants)
- Village Development ° number of villager (a village IsProgram a subdivision of a mwdncipality)(2.5 of all grants)
- Municipal Development o Rp 1450 per capitaProgram °p 170 million minimum subsidy(5.62 of all grants) per municipality
- Provlacial Development Equal sharesProgram(7.1! of all grants)
... Continued on the next page
Table 2.32 Federal/Central Transfers to Lower Levels of Covernment (Page 4 of 7)
COUNTRY MECHANISMS DISTRIBUTION CRITERIA COHENTS
JapanNational Treasury Disbursements to local public bodiesObligatory Share. to part or all of specific local
expenditures on functions under theresponsibility of the Central govern-ment or under the concurrent responsi-bility of both levels.
National Treasury Disbursements to execute specificGrants-in-Aid services or to render ftnancial
assistance.
MalaysiaStates:
Capitation Grant 0 Adult population(in 1982: M$15 per capita forfirst 50,000 of population andthen dropping to H$4 forpopulation greater 200,000.)
Revenue Growth Grant Population (502)I 10 poorer states, on the
basis of growth in per capitaState Domestic Product (502)
State Road Grant ° Mileage of state roadswhich meets federal standards.Average cost of maintalingmile of road.
Stage of Economic 2evelopment The criteria are not weLlGrants, Commodity export deflned (adult populationgrant, and other specific and relative backwardnessgrants for approved schemes. are common factors)
State Reserve Fund (to assist states experiencing "Deficit Grants" provide perverseextraordinary problems) incentives.Occasionally used to helpstates balance their budgets.
Special Giant to the states 402 of Incremental stateof Sabah and Saravak revenues since 1963.
Local Governments:
Operating Grant from Adult population,federal government local revenue generation(annual)
... Continued on the next page
Table 2.3: Federal/Central Transfers to Lower Levels of Government (Page 5 of 7)
COUNTRY MECHANISMS DISTRIBUTION CRITERIA CIMENT
HexicoFederal-to-StatesPrograma Normal (512 of Allocation criteria Criticized as being contrary tototal sources of state are internal to each the spirit of decentralizationfunds). These are programs federal ministry. and as not being able to reflectdirected by central agencies. local priorities.
Convenios Unicos de Desar- No clear allncation Grants have been provided to coverrollo: federal matching criteria. provincial deficits. The budgetaryspecific purpose grants situation in Mexico has caused a(awarded awnually) substantial cut in these programs,delaying the flow of funds andcausing a stop-go execution ofprojects.
NigeriaStates:
Conditional Grants for No clear criteria.approved projects on In some cases, the prevailingagriculture, education, criteria are:health, water supply.
0 Population° Equal shares
41 S Local Governments:
Pension Fund for LocalGovernments, and fundstraining local officials(Contributions from all3 tiers of governDent)
PakistanProvinces:
Grants-in-Aid:
(i) Fixed Annual To supplement the budgets ofSubventions: the disadvantaged provinces.
Rs 100 million to NWFP and Rs50 million to Baluchistan.)
(11) Specific Transfers: To finance "provincial outlays "The interprovincial distributionon behalf of the Center". these grants has witnessed erraticThese grants are specific for fluctuations.-"development, highways, naturalcalamities, etc.
... Continued on the next page
Table 2.3: FederallCantral Transfers to LAser Levels of Government (Page 6 of 7)
CNTRY HECANISMS DISTRIBUTION CRITERIA CWHETS
(iil) Grants for Budgetary * Revenue deficits Incentives for fiscal mismanagement.Support.
(iv) Development Grant for * Excess of education expenditures Incentive for higher educationEducation over 1983 levels. expenditures.
(v) Discretionary Grants Arbitrary
Special Development Program To supplement provinces' efforts in(SDP) the development of selected backward
regions. Assistance is availableonly for the projects spproved by theCenter.
Provincial Annual Development 102 euglly divided between NWFP and The size of the funds to be distributedPlans (PADP). and Baluchistan; fluctuates in view of the resource
* 902 to all provinces. The primary constraints and priorities of theinterprovincial distribution center.criterion for this 902 is:
* Provincial population.
Local Governmentes Complex negotiations with each localgovernment.
0'
Papua New Guinea
Derivation Crants: Derivation. The following formula was01.252 of the value derived The total derivation grant is the proposed in 1989:from the province of above amount minus the value of PDER - R*E*(1+RE)goods exported during the royalties paid to the province where:preceding fiscal year* during the year of calculation. PDER- present derivationIf this is a negative value, R= 1252the grant is zero. E- FOB value of province
exportsRE= rate of growth of
provincial exports.
Conditional, Staffing and Medium term development planOther Grants:
(i) Public InvestmentProgre (PIP):
(it) Staffing Grants: Equal amount payment to allprovinces for public servantscarrying out national functionsat the provincial level. In-creased by a constant amountevery year.
... Continued on the next page
Tablo 2.3: Udarel/Cantra1 Transfera to Lower Levels of Govewament (Page 7 of 7)
COUNTRY UEAUISE DISTRIBUTION CRITERIA COMENTS
PhillppinesCentral Goverment Ad hoc criteria. Grants have not been automaticallyGrants released in accordance with the
Law. They do not take into accountfiscal effort.Because they are distributed on anad hoc basis, they hamper longterm planning by recipients.
Increased degree of decensralizationis quite high on the present politicalagenda.
ThailandGrants (Subsidies): To cover services administered Usually constitutes the bulk of the local
by the local governments in revenues. It is difficult however tolieu of the center. to access the real magnitude of these
transfers given that many are made throughMI) General Grants: ° Population (60 baht per person) central government enterprises.
(IL) Grants for Local To support compulsory This is evidenced by the trend in role ofeducationz elementary education the local governments since 1975 coupleda. expenditures. with an expansion of central governmententerprises.
(Ili) Specific purpose For public infrastructure construction This downvard trend is also observable fromgrants: relief from disasters and other over the low yields of local taxes arising fromdisasters and other over which the Center too many exemptions, pervasive evasions,puts a higher priority. low rates, low elascicity of its revenues.)
Soviet UnionUnion-to-Reuwblics
Assistance to backward "A Recently, many republics have refused torepublics, contribute to these centralized funds.(with contributions Moreover the differences in standard offrom other republics) living among the "have" and 'have-not"republics (the latter includes mainly theCentral Asian republics) have increasedrather than declined.
Sources: The data presented in th's table were drawn from a large number of sources. Only the major sources for each country listedbelow. For a complete list of the articles and documents used, please refer to the references list at the and of this paper.Bangladesh: ESCAP (1989); Brazil: Bomfim and Shah (1991), Shah (1990); China: ESCAP (1989); Colombia: Bird (1984), Banco de LaRepublica (1990); Czechoslovakiat Prust et al. (1990); India: Indian Finance Commission (1989); Indonesia: ESCAP (1989);Japan: ESCAP (1989); Malaysia: ESCAP (1990); Mexico: Boadway (1990); Nigeria: Ashwe (1986a and b); Pakistan: Akhtar (1990);Papu- New Guinea: ESCAP (19:9); Philippines: ESCAP (1989), Soviet Union: Alexashenko (1990); Thailand: ESCAP (1989);Yugoslavia: Grahan (1984)
iii) Benefit-Spillover Compensation Grants: This as discussed earlier calls for selective open-
ended matching transfers. Although benefit-cost spillout is a serious factor in a number of
countries e.g. in India both the tax and services spillouts are considered serious, such
transfers have not yet been implemented in any developing country.
iv) Conditional Non-matching (Equal Per Capita Transfers) or to Ensure Minimum Standards
of Service Across the Country: There are very few examples of such transfers in vogue in
developing countries. The so-called 'capitation grant" to Malaysian states comes close to the
concept of such a transfer.
v) Conditional Open-ended Matching Transfers to Encourage Ceitain Expenditures: India,
Pakistan and Malaysia use conditional closed-end matching transfers. The literature does not
provide any example of open-ended matching transfers.
2.21 A Critical Look at Revenue Sharing Practices In Developing/Transition Econonies
In the following, revenue sharing mechanisms in selected countries are reviewed.
Brazil: One of the main instrument for federal-state revenue sharing is the State Participation Fund
(FPE). The Federal Government transfers a pre-specified share of cartain federal taxes to this pool. The
Council of States the determines state shares based on a formula that incorporates population and per
capita income as its main components. A proposal currendy under discussion would extend this list of
components to include land area, interstate trade and fiscal effort factors. In recent years, formula
determined shares have been found unacceptable to the Council and therefore, it had to resort to a
compromise allocation based on an arbitrary adjustment to formula shares. The principal merits of this
program are the consistency of its design with transparency, predictability and local autonomy objectives.
The program further pays some attention to fiscal equalization objectives. The program, nevertheless, has
many design flaws which inhibit achievement of its objectives. For example, state per capita income is
- 67 -
included as a measure of fiscal capacity. It is an imperfect guide to the ability of a state government to
raise taxes as it is conceivable that a significant proportion of income may accrue to non-resident owners
of factors of production. Further, only a small proportion of total state revenues are raised from income
taxes alone. This measure is also subject to implementation difficulties in Brazil. Estimates of state per
capita income are subject to significant errors and are available with a long lag. For example, currently
only 1980 estimates are available. These difficulties significantly diminish the usefulness of per capita
income for use in a program of fundamental importance in federal-state fiscal relations. The FPE further
combines diverse and sometimes conflicting objectives such as revenue sharing and fiscal equalization at
the state level into a single formula in a multiplicative manner and therefore, significantly falls short on
individual objectives. The program is redistributive in its overall impact but consistency of individual state
shares with the formula objectives is not assured and states with similar fiscal capacity receive widely
different entitlements. Since the formula lacks an explicit equalization standard, it also fails to address
regional equity objectives in a satisfactory manner. These failings explainwhy the Council of States finds
it easier to strike a political compromise rather than accept the formula results.
The program to channel federal reventAe sharing monies to municipalities is called the Municipal
Participation Fund (FPM). This program considers municipal population and state per capita income in
the determination of shares of individual municipalities. This program has two major drawbacks. First,
the formula used for this program fails to incorporate differential fiscal capacity of the Brazilian
municipalities in a meaningful way and therefore formula application does not result in a fair and
equitable distribution of funds among individual municipalities. As there is no local income tax in Brazil
(and none is called for due to capital and labour mobility), per capita income is a poor indicator of a local
government's ability to raise revenues. Further, in each state rich municipalities coexist with poor
municipalities but state per capita income, by definition, would fail to make a distinction between the two
classes. Second, this program, is observed to discourage local fiscal effort by meeting nearly two-third
- 68 -
of municipal revenue requirements from the federal revenue sources. Such overwhelming dependence of
municipal govermnents on outside revenues creates a dichotomy between spending and revenue raising
decisions and contributes to reduced financial accountability at the local level.
Mexico: In 1990, Mexico restructured its assistance to states and municipalities. Revenue sharing with
states is through the so-called General Fund (18.1% of federal shareable revenues) and Contingency Fund
(6.5% of federal shreable revenues till 1997 only). The new allocation criteria for the General Fund uses
population (50% weight) and state historical shares (50% weight) adjusted by annual increases In federally
administered excises on petroleum, motor vehicles, alcohol and tobacco and locally administered water
charges and property charges. The Contingency Fund is designed to compensate states that lose with this
restructuring. The Municipal Fund (2% of federal shareable revenues) provides states pass-through funds
intended for final distribution to their municipalities and uses inverse of the allocation for the General
Fund. The use of population for distribution of 50% of the General Fund revenues is a welcome change
as it will enhance autonomy and equity objectives. The reliance on adjusted historical shares for the
distribution of remaining funds perpetuates anomolies created by higher petroleum revenues accruing to
certain states in the early 1980s. This component will clearly favour oil-rich states. The Municipal Fund
makes only a minor contribution to rectify this problem.
Nigeria: Nigeria channels 45% of federal revenues to revenue sharing funds with states and
municipalities. 95% of revenue sharing with states uses "minimum responsibilities", population, primary
school enrollment and internal revenue effort as formula factors and the remaining 5% is distributed to
mineral producing states on the basis of origin. Transfers to municipalities are based on equal shares
(25% to recognize minimum needs) and population (75%). Several aspects require further attention in
fine tuning existing revenue sharing arrangements. These include equalization to a standard and instability
associated with resource revenues. The former could be addressed by adopting some form of the
- 69 -
Representative Tax System and latter by establishing an Oil Fund managed jointty by the federal and state
governments.
lafil: A significant proportion of revenues are returned using population, and some measure of income
relative to the average and therefore some degree of implicit equalization is attempted by the formulae.
Since the formulae embody factors to get a handle on multiple objectives, the extent to which each of the
objectives is accomplished requires further analysis. The formulae do not pay any special attention to
fiscal capacity (revenue means) of individual states in grant determination.
Pakistan: Revenue sharing is based on population and derivation factors. Equalization to a standard by
considering revenue means of the provinces has not yet been tried.
Pajua New Quinea: Minimum unconditional grants are based on a base year expenditures (1976n7) and
some revenues are shared using the derivation principle. It is not clear why a base year expenditures
should be consistent with the priorities and economic-demographic dyanamics prevalent two-decades later.
Philinine: Population, land area, equal shares and derivation factors are used in revenue sharing. While
the formula factors are objective and reasonable, one expects that horizontal imbalances would not be
corrected by these factors alone and some explicit consideration of revenue raising potential of subnational
governments would have to incorporated into the formulae.
Soviet Union: Starting in 1991, the Union Govemment has established tax base sharing arrangements
with the Republics. According to these arrangements, Union will piggyback on a profits tax to be
administered by the Republics. Revenues from the turnover tax (possibly a Union VAT in future) would
also be shared according to a formula still to be worked out.
China: In China revenues are transfered upwards according to the dictates of the Centre. Arrangements
vary by province based on individual negotiations of the Centre with each province. This system is legally
known as the "Provincial Overall Contracting System". At the present time, the Centre has entered into
- 70 -
five year contracts (due to expire in 1993) with 14 out of 29 provinces to receive fixed sums of revenues
annually and with the remaining 15 provinces to provide them with fixed annual subsidies (see Wu and
Lou (1991)). Note that the above system of 'fixed amounts' sharing up is intended to encourage higher
tax efforts at lower levels.
2.22 Federal Transfers to States and MunidpaUdtes - A Review
Specific purpose transfers support important policy objectives in a federation. These objectives,
as discussed earlier, include: benefit spillover compensation; bridging fiscal gap; ensuring minmum
standards of public services across the nation; fulfilment of the redistributive function of the federal
government; creation of a common internal market; reduction in net fiscal benefits across jurisdictions
and achievement of economic stabilization objectives. Grant objectives would predetermine grant design.
Table 2.3 shows that specific purpose transfers usually lack any specific criterion and grant
funds are distributed in an adhoc manner purely at the discretion of tMe centre. A prominent example of
such adhoc transfers are the "convenios" (negotiated transfers) in Brazil. Thus the practice of
intergovernmental transfers is at substantial variance with the economic principles enunciated above and
significant opportunities exist for the reform of these arrangements in developing/transition economies.
In the following, these transfers in selected countries are reviewed.
Brazi: The federal and state governments in Brazil have an incredibly large number of specific purpose
programs. For many of these programs, program objectives are either not specified or specified vaguely
and in some instances reflection upon grant objectives is done after the release of funds. In more recent
years, federal specific purpose transfers have served not as a means of safeguarding federal objectives
but increasingly as a vehicle for pork-barrel politics. There are only a handful of programs with some
- 71 -
desirable features. One such program is for unified and decentralized system of health care provision
known as SUDS. Federal financing is provided to achieve certain minimum standards of health care
across the nation. The Intent of the program is for the federal government to specify policies and state
and local governmeents to implement federally mandated programs. In practice, however, the federal
government continues to be heavily involved in program administration as well and therefore, the
decentralization objectives have as yet not been fully achieved. In the coming years, fiscal pressures of
the new fiscal arrangements on the federal government are likely to constrain the federal funding for
SUDS. Finally, the existing program gives preferential treatment to private contractors over state and
local government agencies.
Bangladesh: Bangladesh offers a number of closed end matching and non-matching grants for upgrading
infrastructure. Grants allocation is based upon verifiable indicators of general assistance. Thus these
grants may be serving to provide general budgetary support to lower level governments rather than
providing any special incentive for higher expenditures for infrastructure especially when matching rates
are small and non-binding. Bangladesh also provides budget deficit grants which create incentives for
running higher and higher deficits.
India: India also provides budget deficit grants. Specific purpose plan grants attempt to provide higher
assistance to relatively less well-off states as well as to encourage tax effort at subnational levels but long
drawn out review and approval processes may be working against some of these goals.
Indonesia: In Indonesia, 67% of central transfers go to pay wages and salaries o. subnational government
employees and thereby creating an implicit incentive for padding the rolls of the public sector at lower
levels.
Malaysia: Most of the transfers in Malaysia are based upon objective criteria except for 'deficit grants'
through the state reserve fund. The latter grants are however granted in only exceptional circumstances.
- 72 -
Mexico: Specific purpose transfers in the past have been mired into political controvex.y and debate in
-'ew of the lack of any transparent criteria.
Nigeria: Nigeria appears to have a mixed record on the design of transfers. Some specific purpose grants
to states follow objective criteria and the federal government simply sets standards of service to be
achived while other programs lack any transparency in the allocation of funds.
Pakistan: In Pakistan, federal transfers have worked as vehicles for federal bureaucratic control over
provincial expenditure priorities. Most of the central transfers are based on adhockery rather than proper
consideration of central objectives and fiscal needs and relative fiscal capacities at the provincial level.
Examples include deficit grants which provide an incentive for provinces to run higher deficit to have a
greater claim on central resources and education grants that finance provincial expenditures above their
1983 levels - again an incentive for higher expenditures. Most of the plan grants are capital grants with
no provision for financing maintenance expenditures. As a consequence, the grant structure encourages
capital intensive technology and subsequently results in quality deterioration due to lack of funds for
proper upkeep. The central grants are highly unpredictable in nature and discourage long term planning
at lower levels. Economic criteria e.g. efficiency, equity, spillover compensation, autonomy etc. are not
usually recognized in the grant programs currently in place. Grantmanship, on the other hand, is
rewarded.
In China, Colombia, Phillipines and Thailand, specific purpose grants lack any transparency of objectives
and allocation criteria. In the Soviet Union, such transfers to low income republics have become a
contentious issue in Union-Republics relations.
- 73 -
2.23 Spedal Issues in State-Municipal Transfers
The same economic principles govern state-municipal fiscal relations as those enunciated earlier
for centre-state fiscal relations. Since local governments in many countries are simply handmaiden of their
state govermments, they are often subjected to a greater depree of interference and control. Further, the
degree of local government dependence on state transfers is usually greater than the dependence of states
on central transfers. 1P a developed country, local governments typically account for over 20% of
consolidated general government expenditures and finance less than 30% of their expenditures from
higher level transfers. Property taxes is the mainstay of local government revenues in most advanced
nations except Nordic countries where local governments rely heavily on local income taxes. In
developing countries, on the other hand, local governments account for less than 10% of consolidated
general government expenditures and derive more than two-third of their revenues from higher level
transfers. An increase in revenue sharing transfers in some instances have contributed to reduced local
tax effort. For example, during the early eighties, Mexico more than doubled its transfers to municipal
governments as well as gave them exclusive access to property tax revenues. It is observed that nearly
half of these transfers were directed to increased local expenditures and the rest were used as a tax relief
to municipal residents. As a result, municipal reliance on own revenues declined from 75% of total
expenditures in 1980 to 40% of the same in 1984 (see McMillan (1989)). Increased federal transfers to
municipal governments in 1989/90 has also led to reduced local ta; effort in Brazil (see Shah (1990)).
Prop .ty tax, in many instances, is a state responsibility and proceeds shared with local governments and
sometimes also with the central government. This tax is generally not a very productive revenue source
as its base has often been eroded by exemptions and dated assessments. In principle, it should be easy
for state governments to structure their transfers to local governments in an objective fashion in view of
easy access to their economic data. In practice, in developing countries, state transfers to local
- 74 -
governments are based on adhockery and discretion. Only a few countries, e.g. Brazil, India, and Nigeria
have made serious attempts to structure at least part of their assistance in a non-discretionary fashion (see
Table 2.4 for details).
Local borrowing in the credit market for capital expenditures remains a major unsettled issue
in most developing countries. Local governments are generally not permitted to borrow in the credit
markets and therefore rely exclusively on higher level capital transfers for undertaking capital
investments. This is an area where there is a potential for setting up autonomous bodies to supervise and
assist local borrowing for capital projects. State governments could set up a "municipal finance
corporation' or a "municipal loan council" to provide technical assistance in project selection and
appraisal and to assist in securing financing. These loans could be secured on prefered terms in view of
state loan guarantees.
In the following, brief comments on state-local transfers in Brazil and Mexico are offered.
Information on such transfers for many countries is scanty and therefore, is not amenable to even rough
analysis.
.azil: In Brazil, state-municipal transfers have two important components. One such component is the
constitutionally mandated state-municipal revenue sharing arrangements or state-municipal tax transfers.
The distribution of such transfers for the most part follows the origin principle. 75% of state value-added
tax (ICMS) revenues are distributed in proportion to the value added in each municipality and for the
remaining 25% states have been given discretion in incorporating fiscal need factors as they see fit.
Population and area are the two most commonly used need factors. Some states have also used fiscal
effort as a special factor. A major criticism of the existing arrangements is that fiscal equalization by
varying a proportion of funds inversely with fiscal capacity (tax bases for
- 75 -
Table 2.4
STATE TRANSFERS TO LOCAL GOVERMIENTS IN SELECTED COUNTRIES
COUHTRY MECHANISMS DISTRIBUTION CRITERIA COMMENIS
Brazil Revenue Sharing
State VAT (ICQS) Formula Major Criticism: Municipal tax bases(252 to Municipalitiea) A municipality's share of the hardly enter in the formulaedistributable pool is given by: Fiscal equalization by varying a
(VAIVAS)*p + X*(l-p) proportion of funds with fiscalwhere: capacity (municipal tax bases)VA= Municipality's Value Added not recognized.
average of past two years) Fiacal effort component if reco nized,VAS. State's V.A. poorly structured and benefits larerpr 2 of distributable pool to municipalities without regard to their
be distributed according to fiscal effort.municipality's contributionto state's V.A. (by federallaw p must be at least 752)
X a other factors that mayenter the criteria fordistribution States are givencomplete discretion over what,if anything to include in X,e.g., the state of Para uses:° ratio of municipality to statepopulation (72)
° ratio of municipality to statetax receipts (92)
cratio of municipality to statearea (2Z)
o equal shares (72)V.A. is defined as the value ofoutflow of goods + value of servicesrendered within each municipality -value of inflow of goods.
Motor Vehicle Registration Derivation(502 to municipalities)
Share of Federal Tax on Same as VATIndustrial Products accruingto states as compensationfor loss of VAT revenues onaccount of exports:252 to municipalities(Pass-through revenues)
Grants Discretionary Specific Purpose Transfers: Opaqueand arbitrary - large numbers defyanalysis.
... Continued on the next page
Table 2.4t State Transfers to Local Governments in Selected Countries (Page 2 of 4)
COUNTRY NECHABISMS DISTRIBUTION CRITERIA COMMENTS
Colombia Revenue Sharing:
Departments' Share o populationof national salestax collections(40S to municipalities)
Grants Discretionary
China Revenue Sharing:Each province determines theof tax collections chat will beretained by each of its localgovernments.
"Grants: Discretionary
CzechoslovakiaNA Transfers to National Coamittees
from the Republics covered 502 ofthe local expenditures until 1989.
-4
India Revenue Sharing:(1) Urban Authorities: Derivation The shares of the state tares
Motor Vehicles going to municipalities varyacross states.
Entertainment Derivation
tIl) Rural AuthoritiesLand Revenue Rural Population
Entertainment Derivation
Grants: Specific purpose transfers Follow no uniform distribution(Some states use popula- criteria. Autonomy of localtion and local tax effort government units is greatlyas distribution criteria.) undermined by the nature of
these transfers.
MalaysiaGrants Estimated Budget deficits.
Other Discretionary In addition to the above, local Lack a clear allocation criterion.Grants governments may apply for annual
grants to federal and stategovernments.
... Continued on the next page
I
Table 2.4s State Transfers to Local Governments ln Selected Countries (Page 3 of 4)
COUNTRY NECHANISMS DISTRIBUTION CRITERIA COMMENTS
Mexico Revenue Sharing:
At least 20X of what Criteria are specified The criterila followed in someis received from revenue by the states individually. are arbitrary (often lacksharing (General and Several states use the objective indicators likeComplementary Funds) derivation principle, others population, need, and tax-follow U.S. revenue sharing effort). The fact that theformulae. criteria have to be approved
annually by each state legialaturemakes it hard for manicipalitiesto project revenues for thefollowing 5 to 10 years.Complementary funds are directedto be distributed according totax collection (derivation) andand the thmnicipal Fund on a re-distributive basis. Howeverseveral states fall to follow aclear objective distributioncriteria.
Nigeria Revenue Sharing:
102 of internally generated Criteria vary for each state. Local governments receive too0 ~~~~revenues of states little fiscal revenues. Thecurrent scheme does notaddress horizontal imbalancesproperly.Grants:
Pension Fund for Local NAGovernments, and fundstraining local officials(Contributions from all3 tiers of government)
Conditional Grants for approved No clear criteria. In some casesroj ects on agriculture, education, population and equal share areeaLth, water supply. the prevailing crlteria.
Supplementary speciflc grants number of pupilsfor primary school constructionand for teachers' salaries.
Pakistan Revenue Sharing:
Property tax (502 of proceeds) Derivation
Grants: negotiated with each(Matching grants for Province- local government unitapproved projects)
... Continued on the next page
Table 2.4: State Transfere to Local Governments in Selected Countries (Page 4 of 4)
COUNTRY MECHANISMS DISTRIBUTION CRITElRIA CCRiENTS
Soviet Union
Revenue Sharing
Local goverments: The Center allows apro-determined share ofspecified revenues tobe retained by localgovernments.
Sources: The data presented in this table was drawn from a large number of sources. Only the major sources for each country are listedbelow. For a complete list of the articles and documents used, please refer to the references list at the end of this paper.
Brazil: Bomfi. and Shah (1991), Shah (1990); China: ESCAP (1989); Colombia: Babl and Linn (forthcoming); Bird (1984). Bancode La Republica (1990); Czechoslovakia: Prust et al. (1990); India: Indian Finance Commission (1989); Malaysia: ESCAP (1990);Mexico: Boadway (1990); Nigeria: Ashwe (1986a and b); Pakistan: Akhtar (1990); Papua New Guinea: ESCAP (1989); Soviet Union:Alexashenko (1990); Yugoslavia: Graham (1984);
-J'0
municipal sources) is not recognized in the formulae currently in vogue. In fact, municipal tax bases
hardly enter into any consideration. Even the fiscal effort component is usually poorly designed and
serves to benefit larger municipalities only without any due regard for their fiscal effort. A second
component of state transfers to municipalities is specific purpose or negotiated transfers. Most states have
a large number of convenios usually in thousands to provide project assistance. Sheer numbers of these
transfers defy any analysis. Anecdotal evidence suggests that political considerations dominate in the
distribution of grant funds.
Mexico: In Mexico, several states use the derivation principle while others follow the now defunct U.S.
revenue sharing formulae which attempt to vary grant funds directly with population and tax effort and
inversely with relative index of per capita income. Municipal fiscal capacity does not enter into
consideration of these formulae at the present time. In some states, grant allocation is quite arbitrary and
in many states criteria for allocation of grants is approved by state legislatures annually making it difficult
for municipalities to carry out a long term projection of revenues and expenditures.
2.23 The Relevance of Developed Country Experiences
Developed countries offer important examples of grant programs that do recognize some of the
economic principles enunciaid earlier. German experience suggests that a well thought out revenue
sharing system can obviate th) need for many specific purpose transfers. Canadian experience with federal
equalization transfers based on the representative tax system approach suggests that an objective
equalization program which assists member units in establishing minimum standards of basic services can
endure and work as a glue that helps keep the federation together. Australian experience with equalization
is also instructive but much more difficult to replicate elsewhere. U.S. experience with road transportation
assistance holds important lessons in structuring specific purpose transfers. The program used objective
- 80 -
indicators of need in the allocation of funds among states and established matching provisions to induce
local participation. Canadian federal transfers for health and post-secondary education recognize the
'redistribution in-kind" nature of these public services and provide per capita transfers to provinces
conditional on universal access to these services. Switzerland provides spillover compensation and
equalization transfers to its cantons (see Dafflon (1990)). Canadian provinces, U.K. grants to local
governmnents and Victoria Grants Commission, Australia, provide alternate elaborate models of transfers
to local governments (see Shah (1983a)) for formulae details). Many of these ideas can be implemented
without much difficulty in a developing country. In short, there is plenty of conceptual and practical
guidance in the literature on the design of these transfers. Each developing country's specific
circumstances would require drawing upon this guidance but tailoring and adapting it to suit its own
circumstances. It appears from the developing country literature that very few developing countries have
devoted serious attention to the design of these transfers and therefore monumemtal and important work
lies ahead.
- 81. -
3.0 Vertical and Horizontal Fiscal Imbalances: Some Evidence
Vertical fiscal imbalance refers to the mismatch between revenue means and expenditure needs
at various levels and the horizontal imbalance refers to inconsistency between revenue raising ability and
fiscal needs of governments at the same level in a federation. Some degree of mismatch between revenue
means and expenditure needs at various levels is inevitable in all federations. Efficiency in tax
administration for certain revenues requires central administration and this in itself contributes to the
vertical imbalance problem. Thus after expenditure and tax assignment have been completed, revenue
sharing and transfers are frequendy used to correct for any imbalances that result from assignment of
responsibilities. However, revenue sharing and transfer mechanisms due to difficulties in design or due
to conflicting claims of relative needs by various levels of government may not fully resolve this issue.
It would be interesting to see how various countries compared with each other in terms of these
imbalances.
Vertical Imbalances: Table 3.1 presents some rudimentary evidence on vertical imbalances in
selected countries. The table shows that out of the fifteen countries for which data on revenue and
expenditure shares at the national and subnational levels are presented, only in Germany and Yugoslavia
fiscal arrangements eliminate vertical imbalances. For the remaining thirteen countries, seven report a
fiscal deficiency at the centre and six at the subnational level. It is interesting to see that the fiscal
deficiency at the national level does not appear to show any relationship to the degree of central control
over subnational governments. It would be instructive to develop an aggregate measure of vertical
imbalance which incorporates some measure of national control. One such highly imprecise measure of
vertical balance- termed as the coefficient of vertical imbalance and proposed
- 82 -
Table 3.1
VERIWAL IMBALANCES IN SELECTED COUNTRiES
Level of Revenue Expenditure Surplus/Country Govemment Share (*) Share (*) Deficiencv
Argentina:(1987) National 0.86 0.64 0.22
Subnational 0.14 0.36 -0.22All levels 1.00 1.00 0.00
Australia:(1987) Nationai 0.77 0.76 0.01
Subnatlonal 0.23 0.24 .0.01All levels 1.00 1.00 0.00
Brazil:National 0.37 0.43 -0.07Subnational 0.64 0.57 .' 07States 0.41 0.43 .02Municipalitles 0.23 0.14 0.09All levels 1.00 1.00 0.00
Canada:(1989) National 0.48 0.50 -0.02
Subnatonai 0.52 0.50 0.02All levels 1.00 1.00 0.00
China:(1988) National 0.364 0.359 0.005
Subnatonal 0.636 0.641 -0.005All levels 1.00 1.00 0.00
Colombia:(1983) National 0.78 0.85 -0.07
Subnatlonal 0.22 0.15 0.07All levels 1.00 1.00 0.00
India.(1986) National 0.71 0.70 0.01
Subnatlonal 0.29 0.30 -0.01All lves 1.00 1.00 0.00
Malaysia:(1987) National 0.81 0.84 -0.02
Subnatlonal 0.19 0.16 0.02Al levels 1.00 1.00 0.00
Mexico:(1984) National 0.81 0.86 -0.05
Subnatlonal 0.19 0.14 0.05All levels 1.00 1.00 0.00
Nigeria:(1988) National 0.91 0.74 0.17
Subnational 0.10 0.26 -0.17All levels 1.00 1.00 0.00
Pakistan:(1988) National 0.78 0.71 0.06
Subnatlonal 0.23 0.29 -0.06All levels 1.00 1.00 0.00
Continued on the next page
- 83 -
Tais .1: Veutcal Inbalances in Sebcted Couutis (age 2 d 2)
Level of Revenue Expenditure Surplus/Country Govemment Share (*) Share (*) Deficiencv
Soviet Union:;989) National 0.47 0.52 .0.05
Subnatlonal 0.53 0.48 0.05All levels 1.00 1.00 0.00
United States:(1987) National 0.59 0.65 .0.06
Subnatlonal 0.41 0.35 0.06All levels 1.00 1.00 0.00
West Germany(1988) National 0.64 0.64 .0.00
Subnational 0.36 0.36 0.00All levels 1.00 1.00 0.00
Yugoslavia(1987) National 0.25 0.25 0.00
Subnatlonal 0.75 0.75 -0.00All levels 1.00 1.00 0.00
Sources: Pakistan: Akhtar (1990)Soviet Union: Alexashenko (1990)Brazil: Shah (1990)Nigeria: World Bank (December 1987)AJI others: IMF - Govemment Finance StatstHcs Yearbook 1989
- 84 -
by Hunter (1977) attempts to measure the degree of control excercised by the federal government over
lower levels of governments. The way this measure i8 constructed (see Table 3.2) suggests that a
coefficient of zero would indicue absolute federal control over state and local governments and a
coefficient of one would indicate that lower levels of governments are absolutely autonomous in their
decision making. Note that while a high value on this coefficient Is desirable, a value of one has never
been a goal in any federation. A value closer to but certainly less than one would also be consistent with
the assignment principles enunciated earlier. Table 3.2 presents these calculations for a few selected
countries. According to this table, central/federal control is pre-eminent in some countries like Australia,
India, Colombia and Pakistan and whereas in other countries such as Brazil and Yugoslavia, federal
influence over local priorities is quite limited. A discussion of the Brazilian case is particularly instructive
in this case. In Brazil, constitutional tranisers attempt to address this issue. With the new tax assignment
and transfers, federal goverments historical position vis-a-vis state and local governments has
significantly deteriorated. States now command one of the most dynamic revenue base (ICMS) and
municipalities are guaranteed a large share of federal and state revenue collections. While a precise
calculation of the magnitude of the squeeze on the big brother put by the new fiscal arrangements must
await more careful analysis, Table 3.1 presents some rough estimates to outline the broad picture of
vertical imbalance that characterizes Brazil of today. According to these calculations, federal and state
governments' revenue means significantly fall short of their expenditure needs. The opposite situation
holds for municipal goverments. The table shows that federal government would be (if it is already not)
in dire straits if it continued to follow in future as in past years a broader interpretation of its
responsibilities. State level governments as a whole face some difficulties now but these may not persist
in the long run in view of expected growth of ICMS revenues. Municipal governments in Brazil, on the
other hand, should be the envy of all governments in developing (or even advanced nations) world.
- 85 -
Table 3.2
COEMCIINTS OF VERTICAL IMBALANCE FOR SELECTEDCOUNTkfh3
Count-r Period CoeMfcient
Australia 1987 0.43India 1982-86 0.45Colombia 1979.83 0.50Pakistan 1987-88 0.53Malaysia 1984.88 0.65Canada 1988 0.79West Germany 1988 0.79United States 1987 0.88Brazil 1988 0.89Yugoslavia 1983-87 0.93
Notes:Coefficient of Vertical Imbalance - 1 ((B)+(C) )/(D)
where:(A): Total Revenue & Grants(B): Transfers= Grants from other levels of Nat. Govt.(C): Net Borrowing= -( (A)-(D) )(D): Expend. = Total Expenditures & Lending - Repayments
Sources: Brazil: Shah (1990)Pakistan: ESCAP (1989)Others: IMF.GFS (1989)
- 86 -
Horizontal Imbalances: Empirical evidence on horizontal flscal imbalances is scarce but
nevertheless horizontal fiscal imbalances appears to be a much more serious problem in developing
nations than vertical imbalances. For example, in Brazil, per capita income in Rondonia, one of the
poorest in the nation, is only 12% of per capita income in Sao Paulo, the richest in the Union. Per capita
own revenues In the former state are only 20% of the same in the latter (see Shah (1990)). In China, per
capita tax collections vary from a low of Rmb 40 in Tibet to a high of RMb 1,492 in Shanghai (Wu and
Lou (1991)). In Sri Lanka, local revenues in per capita terms range from rupees 66 in the North Central
province to rupees 237 in the Western province (see Shah (1990), Table 1). The above fiscal disparities
are not unique to Brazil, China and Sri Lanka and exist in most other developing nations as well. What
is remarkable is that as yet not a single country has attempted to deal with these disparities by equalizing
to a standard.
4.0 An Impressionistic Overview of Intergovernmental Fiscal Relations in Selected Countries
Table 4.1 presents a bird's eye-view of selected fiscal systems. Of the countries reviewed here,
Canada and West Gemany offer two altemative neat models of a federation. The former emphasizes
diversity in public services with minimum standards achieved by tax harmonization and transfers. The
latter emphasizes uniformity in public services achieved through rational expenditure assignment and tax
sharing arrangements and transfers. Smaller developing countries like Sri Lanka could benefit from the
German model whereas large and diverse countries like Brazil, India, Mexico and Pakistan have much
to learn from the Canadian model.
- 87 -
Table 4.1
AN IXPPZSSIONISTICE EVALUATION OF S2LZCSED FISCAL SSTEKS
Tax Senaration Tax Overla nine Tax Sharin"Selected Indicators Australla Nexico Canada Unlted States West Germany Niaeria Brazil IndL.
National Unity Strong Strong Fairly Strong Strong Weak Strong Weakstrong
State Influence on Fed- Fairly Weak Strong Fairly weak Strong Strong Strong Weakeral Policymakers strong
State Government Con- Strong Weak Fairly Fairly weak Strong Strong Strong Strongstitutional Status strong deJure; verystrong defacto
Actual State Control of Strong Strong Strong Varies from Strong Strong Weak StrongLocal Government fairly strongto fairly weak
Range of Local Goverment Limited Llmited Fairly Extensive Llmited Limited Extensive LimitedResponsibilities extensive
Local Government Influence Weak Weak Fairly Fairly strong Weak Weak Strong Weakoo on State Policy-makers strongLocal Government Influence Weak Weak Fairly Fairly strong Weak Weak Strong Weakon State Policy strong
Local Government Influence Weak Weak Weak Fairly strong Weak Weak Very strong Weakon Federal P?licy
The Character of Fiscal Two-tiered; Three-tired; Two-tiered; Three-tiered Two-tiered Three-tiered Three-tiered Tvo-tieredFederalism centralined centralized decentralized unstructured integrated decentralized decentralized centralizedFederal-State Inter- Important; Important Important; Important; Unimportant Important Important importantgovernmental Tansfers emphasis on emphasis on Emphasis on Emphasis onconditlonal unconditional conditional tax sharinggrants grants grantsFederal/Interstate Equali- Very strong; Weak Strong; Very weak Strong; Weak Weak Weak tozation Performance revenue and revenue dis- revenue and to fair Fairexpenditure parities some expend-disparities reduced sub- iture dis-reduced stantially paritiessubstantially reduced
substantiallyState Tax Performance Fairly weak Weak Strong Fairly strong Fairly strong Weak Strong WeakLocal Government Fiscal Fairly strong Weak Fairly Fairly strong Weak Weak Weak WeakIndependence strong
... Continued on the next page
Table 4.1: An *lpressionietiv- Evaluation of Selected Fiscal Systems (Page 2 of 2)
Tcx segaratlon _ Tax Overlaplin Tax SharinaSelected Indicators Aust-Ral nioCadU Canada UniteS braves Wetst etmany ieer_a Ir=aziia
Equalization Fornla Fed. -state Impllcit & Fed. -State lmpliclt and Explicit and Implicit Implicit and ImlicitExplicit piecemeal Fiscal piecemeal complex and piecemeal andEqualization piecemeal piecemealState tax base conforvlty Yes No Yes no Yes Yes No NoState tax rate uniformity Yes No No No Yes Yes Yes NoSingle tax collection and Yes no Yes Yes Yes Yes No Noad-lnisltration
State-Local wevennes more No Yes Yes Yes Yes no Ho Noor less match responsi-bility
Sources Some data for this table are extracted from a Table prepared by John Shannon, Washington, D.C. ACIR, 1980 entitled'Rating Federal Systems - An Impressionistic Evaluation).
OD
5.0 Blueprints and Pathways For Fine Tuning, Restructuring and Reform
In this section, major results of previous sections are brought together to provide an overview
of the state of fiscal systems in developing/transition economies and to discern any common themes that
emerge in the reform of intergovernmental fiscal relations in developing/transition economies.
the Assignment Issues
The assignment problems in developing/transition countries cover the whole spectrum. In some
countries like Yugoslavia, there is a growing recognition that the past decentralization effort may have
gone too far and circumvented the federal government's stabilization -nd redistributive role. Thus a
conscious effort is needed to restore to the federal government its rightful role in these areas. In most
countries, though, it is the all too pervasive and intrusive role of the national government that is coming
under increasing scrutiny. The national government's role in national defense and security is well
appreciated but its over-extension to purely local functions such as pothole repair and rat control is being
questioned. A review of the theory and practice of expenditure assignment suggests that invariably the
assigmnent problems arise not from the con tcitutional assignment being at variance with the theory but
from de-facto assignment being at variance with the de-jure responsiblities. Thus it appears that major
reforms are possible with administrative orders and without resorting to constitutional amendments. In
tax assignment, problems are of a similar nature and primarily arise from overlapping and uncoordinated
administration of certain taxes, especially sales and excises. These problems, again, in many instances
can be overcome by simply fine tuning the existing assignments rather than attempting a major redesign.
For example, to avoid problems associated with overlapping administration, tax base sharing with the
higher level goverment determining the tax base and lower level governments levying supplementary
(piggyback) rates on a uniform base requires more serious attention than has been accorded to this
- 90 -
alternative in the past. Of course in countries making a transition to market economy, major restructudng
is inevitable.
Issues In Intergovernmental Tansfers
The fiscal federalism literature has stressed that for each and every objective specified by the
grantor, there is an associated specific design of a grant program. Thus the consistency of grant design
with its objectives is a must for its effectiveness in reaching the desired objectives. For example, general
non-matching transfers or tax base or revenue sharing mechanisms are needed to deal with the fiscal gap;
general non-matching equalization transfers to address differential net fiscal benefits or horizontal fiscal
imbalances; open-ended matching transfers, with the matching rate determined by benefit-spillout ratio,
for benefit-spillout compensation; conditional non-matching transfers to ensure minimum standards of
services across the nation; and conditional open-ended matching transfers to stimulate public expenditures
on areas of high national importance but low local priority. A review of these transfers in
developing/transition economies suggests that the above conceptual guidance in grant design for the most
part continues to be ignored.
The most striking observation to come from this review is that inspite of the high national
priority attached to limiting inter-regional fiscal disparities in most countries, not a single
developing/transition economy has adopted a program of equalization to a specified standard along the
lines of the representative tax system approach inspite of the appare t simplicity and practicability of its
design. More sophisticated but understandably less successful attempts at equalization because of obvious
design flaws have been adopted in Brazil, India and Nigeria. The formulae adopted by these countries
lack an explicit equalization standard and fail to address regional equity objectives in a satisfactory
manner. Most countries do not even attempt to go this far.
- 91 -
As elaborated upon earlier, specific purpose transfers, if properly structured, can support
important policy objectives in a federation. These objectives include: benefit spillover compensation;
bridging fiscal gaps; ensuring minimum standards of public services across the nation; fulfillment of the
redistributive function of the federal government; creation of common internal market; reduction in net
fiscal benefits across jurisdictions and achievement of stabilization objectives. Grant objectives would pre-
determine grant design. In a typical developing country, almost without exception, there are incredibj.
large number of specific purpose programs. For many of these programs, program objectives are either
not specified or specified vaguely, and in some instances reflection upon grant objectives is done after
the release of funds. There are some obvious reasons for this state of affairs. Various governments can
exercise complete discretion over these funds without having any accountability. Enhanced flexibility is
being achieved at the cost of transparency, objectivity and accountability. Thus specific purpose grant
programs are frequently used as a means of pork-barrel politics rather than in the pursuit of key national
objectives. Some specific purpose grant programs are structured in such a way as to provide perverse
economic incentives. For example, several developing countries provide transfers to cover deficits or
public sector wages at subnatiou4i levels. Such grants - contrary to the intentions of the grantor - would
encourage lower own tax effort and thereby higher deficits and higher spending on public sector wages
to qualify for higher grants. A review of these grant programs should be high on any agenda for public
sector reform.
Federal-local and state-local transfers in most developing countries, are in need of a major
restructuring. In none of the countries reviewed here, do these transfers pay any special attention to the
fiscal capacity or revenue potential of local governments. Allocation of 'Amds is usually on an adhoc and
discretionary basis - negating transparency, predictability and autonomy objectives. Major recent
increases in revenue sharing funds from the centre to municipalities in Brazil and Mexico have resulted
in reduced own tax effort at the local level. This result has been attributed to a lack of recognition of
- 92 -
fiscal capacity (tax bases for municipal sources) and an inappropriate design of fiscal effort component
in the allocation formulae. Further, federal govermment in a large country does not usually have the
administrative capacity to closely monitor finances of individual municipalities and therefore, the case for
direct federal transfers to local governments is quite weak. Such transfers should naturally be placed in
the domain of state governments. Further, it would not be too difficult for state governments to
restructure their transfers to local governments in an objective fashion in view of the easy access to their
economic and fiscal data. A recognition of municipal taxable capacity in these formulae would also assist
in state level monitoring of the utilization of local revenue bases and thereby assist in corrective action
on a timely basis. Much useful guidance on restructuring these transfers is available from the developed
country practices.
Local borrowing in the credit market for capital expenditures remains a major unsettled issue
in most developing countries. Local governments are generally not permitted to borrow in the credit
markets, and therefore, they rely exclusively on higher-level capital transfers for undertaking capital
investments. This is an area where there is potential for setting up autonomous bodies to supervise and
assist local borrowing for capital projects.
In conclusion, there is a universal recognition now that the way taxing and spending authorities
are delineated and the manner in which intergovernmental transfers are structured have important bearings
on the efficiency and equity of public services provision. Fortunately, much useful guidance in the design
of intergoverunental fiscal relations in developing/transition is available from the theoretical and practical
literature on fiscal federalism. It is also apparent from a review of the developing country literature that
very few developing countries have paid any serious attention to this guidance in the design of their
transfers. Immediate attention must, therefore, focus on making this guidance accessible to policy makers
in developing countries and also to adaptations of this guidance to suit individual developing country
circumstances. This paper takes a small step in this direction. Significant work lies ahead.
- 93 -
NOTES
l.The implementation of above principles require operation ofvoting with feet mechanism. This brings us to the Tieboutliterature which suggests that voting with feet will lead tojurisdiction formations creati g a market analogue to publicservice provision. Wallace oates had earlier suggested thatallocative efficiency questions associated with voting with one'sfeet could be settled by examining tax and benefit capitalizationand that the existence of capitalization implies allocativeefficiency. This conclusion has been rejected by Jan Brueckner((1979),(1982)) and Anwar Shah ((1983),(1988),(1989),(1990)).Brueckner's test is based on the theoretical result that a non-positive relationship between public services and residentialproperty values is a definite indication of over-provision of localpublic goods beyond optimal levels in the case of typical mixedcommunities i.e., communities having substantial business property.An alternative test proposed by Shah uses the criterion that, whenthe level of public spending is optimal, a balanced budget changein local spending and residential property taxation should leaveresidential property values unaltered. Thus a positive impact of abalanced budget change would indicate under-provision and anegative impact over-provision of public services.
2.Oates (1990) has argued that even in the case of complete absenceof mobility, substantial welfare gains could arise from varyinglocal outputs with local tastes and costs (p.7).
3. This section draws heavily upon McMillan, Shah and Gillen (1980)and Boadway (1990).
- 94 -
Appendix A
Munidpal-Local Fiscal Needs and State-Munidpal Unoondillonal Trasfers
In AustraUa, Canada and the USA
State-municipal transfers address essentially the same objectives as the federal-state transfers namely,
spi1lover compensation, fiscal gap and equalization. But in many countries municipal-local units are
handmaiden of state governments (as in Canada and Ausria) and therefore, both uncondidonal and
conditional transfeis are used to encourage own fiscal effort by these units. In the following, we review
two broad types of unconditional transfers. The first type of transfers are patterned after the now defunct
U.S. revenue sharing system of the seventies (expired in 1986). The hallmark of these formulae are that
the total amount of grant funds are arbitrarily decided by the higher level government and then formula
factors are used to distribute these funds in a transparent manner among the recipients.
a. U.S. Revenue Sharing - Mill Rate Formula
A = (PCYA/PCYi)* (Ti/PCYi) *(POPi)
Si = AI(Sum(A))
b. U.S. Revenue Sharing - Expenditure Formula
B =(PCYA/PCYi)*(PCEXi/PCEXA)*(POPi)
Si= B/(Sum(B))
where PCYA= State average per capita income
PCYi = Per capita income of municipality i
Ti = Local own source revenues
POP= Population of municipality i.
PCEXA= Average state per capita expenditures
-9S -
PCEXi= Ps' capita expenditure in municipality i.
Note that grant funds according to formula A vary inversely with fiscal capacity (defined by per capita
income) and directil, with tax effort. According to formula B grant funds are expected to vary inversely
with fiscal capacity but directly with per capita expenditure- an indicator of expenditure need.
Canadian Approaches to Provincial-Municipal Unconditional Transfers:
Table A. 1 from Shah (1983a) provides a summary view of the equalization criteria used by the Canadian
provinces to determine the distribution of unconditional grants to their municipal governments. Most
provinces attempt to vary grant funds inversely with revenue means (measured as per capita equalized
assessment) and directly with expenditure needs and only in a few provinces with tax effort. Most
interesting approach to municipal revenue sharing is adopted by the Province of Saskatchewan.
Saskatchewan uses the following formula for grant distribution:
G = X + (Y*POP) + (RE-RR) *F
where G = the grant to municipality i
X= a basic lumpsum grant which is the same for all municipalities regardless of size or
urban/rural distinction
Y = Per capita grant
RE= recognized expenditures= A hypothetical estimated expenaiture representative of that
incurred by municipalities in the same population size i.e. an average of total expenditure made by all
municipalities with comparable populations (a step function estimated by regression analysis).
- 96 -
Table A.1
CANADA: BASIS OF PROVINCIAL UNCONDITIONAL ASSISTANCE TO LOCAL GOVERNMENTS
TaxableCapacity Tax Effort Expenditure
Province Factors Factors Needs Factors
Newfoundland loss of revenue with property tax - populationrespect to exemptions revenues and per capitaprovided to old age water nd assistancepensioners sewer rates - road mileage
Prince EdwardsIslards ---------------- see New Brunswick ----.-------------------------
Nova Scotia property assessment -- - dwelling unitsstandardizedexpenditureper dwellingunit bymunicipa ityclass.popultionsize andurbanVrural).
New Brunswick property assessment -- shareable(per capita and per expendituresroad kilometer)
Quebec property assessment taxes fromlocal sources
Ontario property assessment previous year's populationnot levy -poputation
density-Location*municipal grouping
Manitoba -- - population- urban
population
Saskatchewan property assessment -- - population- expenditure by
populationclass adurban/ruralcategory
Alberta property assessment total tax - populationrevenues growth in
excess of 5Xper anus
British property assessment - populationColurbia - expenditure
Source: Shah (1983a)
RR= Recognized local revenues= Average Effective Mill rate applied to equalized
assessment of the municipality.
F = availability of funds factor.
The above formula suggests that the Province of Saskatchewan has given a great deal of thought to grant
assessment methodology. Yet the fiscal equalization methodology adopted by the Province is not nearly
as comprehensive as the approach adopted by State of Victoria (Australia) Grants Commission. The
Victoria Grants Commission assesses revenue needs on the basis of municipal fiscal capacity by class of
assessed properties. Expenditure needs are determined by assessing differential costs and disabilities over
the full range of municipal functions (see Shah (1983a) for details).
9B -
Appendix B:
The Practice of Federalism in Selected Developed Countries
A: USA has a 3-tier system with states as the weakest link traditionally in the system. In
federal-local fiscal relations states are often by-passed. The intrusive role of the federal government has
largely been the result of urban and racial problems of the 1960s and dominance of state legislatures by
rural interests. Tax and expenditure assignment in the U.S. is not consistent with the economic principles
enunciated earlier. Other than taxes on international trade, exclusively reserved for the federal
government and property taxes for state and local levels, all other tax fields are open to all levels of
govermnent. Federal, state and local governments have overlapping and uncoordinated personal and
corporate income tax administration. Expenditure assignment is also not clearly delineated. Defense,
foreign iffairs and space administration, foreign and interstate commerce, the postal service, coinage,
weights and measures, patents and copy rights and crimes against the United States are reserved for the
federal government. In housing, education, transportation, and social welfare, all three levels are
involved to varying degrees. Federal government gets involved in such local functions as fire protection,
pothole repair, rat control, urban transit, local libraries and rauseums, and zoning regulations as a result
of pork-barrel politics. Federal government often exercises strong control over local priorities through
carrot (specific purpose transfers-in early 1980s there were 492 federal programs) or stick (court ordered
racial integration of school pupils and teachers leading to a decay in schooling in inner cities; highway
speed limits; withholding of federal highway funds from states not raising the drinking age to 21). The
hallmark of the U.S. federal system is diversity, a "fend for yourself federalism" and a 'jungle for tax
administration". The efficiency costs of such a system are large which only an advanced nation like the
U.S. could afford. Major progress to reform this system was made during Carter and Reagan years.
Interestingly, U.S. reforms in recent years have also moved it in the direction of a Canadian style two-
- 99-
tiered system.
AUS)RALIA: Australia has a two-tiered hiZhly centralized systenm. The centre emphasizes
uniformity of public services across the nation and uses conditional grants to achieve that purpose. Tax
administration and collection is primarily central (80% of revenues). Local governments are handmaiden
of states but are given reasonable autonomy in local service delivery. The Commonwealth has sole
responsibility in defense, trade, immigration, external affairs, social security and employment. States are
responsible for education, health and social services, transport, railways, electricity and water. The
federal government nevertheless exercises strong influence in these areas through conditional transfers.
In tax assignment, customs and excises are reserved for the Centre and concurrent responsibilities are
assigred in all other areas. One half of customs proceeds are mandated for states. The Uniform Taxation
Act of A942 eliminated any role for states in income taxes and subsequent court rulings closed sales and
excise taxation fields to states. State-Local governments are responsible for 50% of the total outlay of
the public sector but raise only 17% of revenues.
CANADA: Canada has a two-tiered highly decentralized system. In 1988, 59% of total
expenditures were undertaken at the state-local level. Tav and expenditure assignment are transparent.
Tax assignment is overlapping but harmonized. Expenditure assignment is as follows:
Federal: money, banking, trade, airlines, railways, foreign affairs, defense, unemployment
insurance.
Federal-Provincial: Pensions, immigration, agriculture, industry.
Provincial: Education, health, social welfare, police, natural resources and highways.
Equalization to a standard by using the representative tax system approach is the hallmark of federal
equalization transfers and some provincial-local transfers.
GRMANY: The Upper House of the Parliament is called the Council of States
(BUNDESRAI). State ministers or their deputies are represented on this council and vote at the direction
-100-
of their governments. This provides a check to any centralizing tendency in the federation. The
expenditure assignment is as follows:
Federal: defense, foreign affairs, immigration, railways, air transport, post office.
Concurrent: public welfare, regulation of commerce, industry, banking, insurance and labour
relations, promotion of social responsibility, public roads and shipping. Note that all
concurrent responsibilities are carried out by states (Laender).
States: Education, culture and residual powers.
Tax Assignment: Federal government has exclusive authority over customs and federal
monopolies (alcohol etc.) and priority over remaining taxes. Taxes are primarily collected
by the Centre and then shared with the states and local governments based on agreed
percentages.
The German equalization program has three distinct components:
1. Distribution of 75% of state shares of value added on a population basis and the remaining 25% to
states with below average tax reciepts to enable them to attain 95% of the federal average, excluding
Berlin;
2. Contributions from financially strong to financially weak states according to an equalization formula;
and
3. Federal supplementary allocations to states with below average per capita revenue yields.
Interstate equalization entitlements for the component #2 above are worked out as follows:
First, the tax capacity of each state is calculated by the addition of revenues from (i) state taxes, (ii) the
state's share of the joint taxes according to local yields, (iii) half of the property and trade taxes of the
municipalities, also according to local yields and uniform assessments. Deductions are then made for any
special burdens (extraordinary expenditures) facing a particular state. In this way the adjusted tax capacity
of each state is determined. Comparisons of the adjusted tax capacity of each state are then made with
- 101 -
the average tax capacity per capita of all states. When the average tax capacity is multiplied by the
population of each state, the result is the so-called equalization yardstick for each state. In calculating the
equalization yardstick, consideration is given by the way of an allowance for population density, to higher
tax yields of the city states and to the size of municipalities. lTus in so far as strong states also tend to
be states with relatively higher population densities Oarge cities) - and this is in fact the general pattern-
the intensity of the financial settlement has been somewhat reduced. Finally, the financial settlement
yardstick is calculated for each state as the difference between its adjusted tax capacity and its equalization
yardstick. Symbolically, consider the three step process as follows:
Step 1: Fiscal capacity
ATCi=TCi-Si
Step 2: Fiscal Need
NEEDi= SUM(TC)/SUM(POP). PDC . POPi
where PDC= population density adjustment factor
Step 3: Equalization Contribution
Ei = ATCi - NEEDi
if > 0 contribution to the pool
< 0 receipts from the pool.
Thus the emphasis is on a fraternal settlement as opposed to the paternal setlement in vogue in Canada.
The Federal Government in Germany simply acts as a broker to ensure that the agreed upon rules are
followed (see also Bird (1986) and Hunter (1977)).
- 102 -
B -REES
Afonso, J.R.R. (1989, May), "Despesas Federais com Transferencias Inter-governmentais: Uma Revisaode Concentos, Estatisticas e Diagnostico." INPES - Relatorio Interno #12. 102.103.
Afonso, J.R.R. (1989, October). "Mecanismos Institucionais das Relacoes Intergovernamentais noBrasil." mimeo. 21-A - 21-H.
Afonso, J.R.R. (1988), "Despesas do Governo Federal com Transferenciasde Recursos aos Estadoc e aosMunicipios na Decada de 80". Processed.
and Fernando Rezende (1990). 'Increased Decentralization and Macroeconomic Policy inBrazil". Processed.
and Jose Serra (1989). "Financas Publicas Municipais: Evolucao, Reforma Constitutional ePerspectivas". Revista de Administracai Municipal. Vol. 36, No. 193, p. 3242.
Akhtar, Shamshad (1990) "Pakistan: A Study of Provincial Finances". Processed.Alexashenko, Sergey (1990). A Reform of the System of Fiscal Relations in the Soviet Federation:
Possible Solutions and the Main Complexities. Commission for Economic Reforms, USSRCouncil of Ministers, Processed.
Andrade, Thompson A. (1989). "Endividamento Municipal: Analise da Situacao Financeira de QuatroCapitais Estaduais." Revista Brasileira de Economic, Vol. 43, No. 1.
Ashwe, Chiichii (1986a) "Fiscal Federalism in Nigeria". Centre for Research on Federal FinancialRelations, The Australian University, Canberra, Australia
Ashwe, Chiichii (1986b) "A Critique of Nigeria's Existing Revenue Sharing Scheme" Center forResearch on Federal Financial Relations, The Australian University, Canberra, Australia
Bahl, R. and J. Linn (forthcoming) Urban Public Finance in Developing Countries. Washington, D.C.:The World Bank.
Bahl, R. and Johannes Linn (1990). Fiscal Decentralization and Intergovernmental Transfer in LessDeveloped Countries. Processed.
Bahl, R. and Johannes Linn (1983). The Assignment of Local Government Revenues in DevelopingCountries. In Charles McLure, Jr. Editor (1983).
Bird, Richard (1990). Intergovernmental Finance and Local Taxation in Developing Countries: SomeBasic Considerations For Reformers. Public Administration and Development, Vol. 10, 277-288.
Bird, Richard (1986). Federal Finances in Comparative Perspective. Canadian Tax Foundation,Toronto.
Bird, Richard (1984) "Intergovernmental Finance in Colombia: Final Report of the Mission onIntergovernmental Finance". Harvard University
Boadway, Robin (1990). The Reform of Federal-State Transfers in Mexico. Processed.Boadway, Robin (1980). Intergovernmental Fiscal Transfers in Canada. Canadian Tax Foundation,
Toronto, Canada.Bogoev, K. (1990). Central Government in a Highly Decentralized Federation: The Case of Yugoslavia.
Processed.Bomfim, Antiilio N. and Anwar Shah (1991). "Macroeconomic Management and the Divisionof Powers
in Brazil: Perspectives for the Nineties". The World Bank, PRE Working Paper Series No. 567(January)
Brazil, Camara dos Deputados (1989, November). "Projeto de Lei Complementar No. 104-A, de 1989."Brasilia, DF: Centro Grafico do Senado Federal.
Brazil, Camara dos Deputados (1989, November). "Projeto de Lei Complementar No. 177-A, de 1989."Brasilia, DF: Centro Grafico do Senado Federal.
- 103 -
Brazil, Senado Federal (1989, September). "Redacao Final do Projeto de Lei do Senado, No. 45, de1989." mimeo.
Brazil, Senado Federal (1989, July). "Projeto de Lei do Senado No. 165, de 1989 - Complementar."Brasilia, DF: Centro Graflco do Senado Federal.
Brazil, Senado Federal (1989, March). "Projeto de Lei do Senado No. 45, de 1989." Brasilia, DF:Centro Grafico do Senado Federal.
Brazil, Tribunal de Contas da Uniao (1988). 'Relatorio e Parecer sobre as Contas do Governo daRepublica." Brasilia, DF: Centro Grafico do Senado Federal. 69-70.
Brazil, American Chamber of Commerce. Coutiin of the Federal Reublic-of Brazil (1988). SaoPaulo, SP.
Brazil, Secretaria do Tesouro Nacional, Ministerio da Fazenda (1988). Balancos Gerais da Uniao.Brasilia, DF: Imprensa Nacional.
Brueckner, Jan K. (1982). A Test for Allocative Efficiency in the Local Public Sector. Jourmal oPublic Economics 19: 311-331.
. (1979). Property Values, Local Public Expenditure and Economic Efficiency.Journal of Public Economics 11 (April): 22345.
Bulutoglu, Kenan (1976). Fiscal Decentralization: A Survey of Normative and Positive Contributions.FinanznMchix, N.F. 35 Heft 1, 1-34.
Castro, Firmo de (1989). "Justificacao ao Projeto de Lei Complementar No. 104-A/89 da Camara dosDeputados." Brasilia, DF: Centro Grafico do Senado Federal.
Colombia, Banco de La Republica (1990) "Las Finanzas Departamentales y Municipales en el Contextodel Sector Publico Colombiano: Una Resefia Historica"..
Colombia, Direccion General de Impuestos Nacionales (1988) "Estadisticas: Tributacion DepartamentalY Municipal"
Courchene, Thomas J. (1986) Economic Management and the Division of Powers. Toronto: Universityof Toronto Press.
Dafflon, Bernard (1990). Intergovernmental Equalization in Switzerland. Processed.ESCAP - Economic and Social Commission for Asia and the Pacific. (1989) "Fiscal Decentralization,
Resource Mobilization and Effective Use of National Resources for Development: The Case ofBangladesh" (ESCAP/NIPFP Seminar on Fiscal Decentralization, Resource Mobilization andEffective Use of National Resources for Development. September 1989)
(1989) "Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of China". Processed
_____ (1989) "Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of Indonesia". Processed
_____ (1989) "Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of India". Processed.
(1989) "Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of Japan". Processed.
(1989) "Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of Malaysia". Processed.
(1989) "Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of Pakistan". Processed
(1989) "Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of Papua New Guinea". Processed.
(1989) 'Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of the Philippines". Processed
- 104 -
(1989) "Fiscal Decentralization, Resource Mobilization and Effective Use of National Resourcesfor Development: The Case of Thailand". Processed.
Gandhi, V. (1983). "Tax Assignment and Revenue Sharing in Brazil, India, and Malaysia." In CharlesE. McLure, Ed. (1983) Tax Assignment in Fedexal Couptries, ANU Press, Australia
Graham, Lawrence (1984) "Yugoslav and Brazilian Experience with Federalism". Technical PapersSeries No. 41. Institute of Latin American Studies, The University of Texas at Austin.
Gramlich, E. (1977). Intergovernmental Grants: A Review of the Empirical Literature. In W.E. Oates(ed.) The Political Economy of Fiscal Federalism. Lexington, D.C. Heath, Mass. USA. pp.219-239.
Gulati, I.S. and K.K. George, (1978). Interstate Redistribution Through the Budget. Economic andPolitical Weekl, Vol. XlII, No.11.
Hunter, J.S.H. (1977). Federalism and Fiscal B ce. ANU, Canberra, Australia.India, Second Report of the Ninth Finance Commission (for 1990-95). New Delhi, Dec. 1989Khalilzadeh-Shirazi, J. and Anwar Shah (forthcoming). Tax Policy in Developing Countries.
Washington, D.C. The World Bank.King, David (1984). Fiscal Tiers: The Economics of Multi-I evel Governmen. London: George Allen
and Unwin.Kirchgassner, Gebhard and Werner Pommerehne (1990). Evolution of Public Finance as a Functionof
Federal Structures: A Comparison Between Switzerland and the Federal Republic of Germany.Knight, Peter and Roberta Waxman (1991). Fiscal Federalism in the Soviet Union. Processed.Kuhn, Thomas and Horst Hanusch (1990). Vertical and Horizontal Equity and the Grants to
Communities in the Federal Republic of Germany. Processed.Linn, J. et al. (1984) "Urban Finances in Bogota, Coiombia" World Bank Discussion Paper.Longo, Carlos A. (1987). "A Distribuicao dos Gastes e Receitas Publicas entre Niveis de Governo: Um
Enfoque Economico". Revista de Financas Publicas No. 369, pp 16-29.Lordello de Melo, Diogo (1988). "Resource Mobilization Strategies for Urban Development in Brazil".
Processed.Mahar, Dennis and Dillinger, W.R. (1983). Financing State and Local Governments in Brazil. The
World Bank Staff Working Paper No. 612.McGuire, Martin (1978). A Method For Estimating the Effect of a Subsidy on the Receiver's Resource
Constraint: With Application to U.S. Local Governments, 1964-71. Journal of Public Economics.August, 25-44.
McLure (1990). The Tax Assignment Problem in Mexico. Processed.McLure, C.E. (Ed.) (1983). Tax Assignment in Fedeal Countries. ANU Press, Canberra, Australia.McMillan, Melville (1989). Municipal Finance in Mexico. Processed.McMillan, M., Anwar Shah and David Gillen (1980). The Impact of Proincial-Municipal
Transportation Subsidies. A Report prepared for the Alberta Transportation Minister. August1980.
McMillan, Melville (1975). Towards the More Optimal Provision of Local Public Goods: Internalizationof Benefits and Grants. Public Finane QuarOter, Vol. 3, July 1975, 229-261.
Musgrave, Richard (1983). Who Should Tax, Where and What ? In Charles Mclure, Jr. Editor, (1983).Musgrave, Richard (1976). Public Finance: Theory and Practice.Musgrave, Richard (1965). Essays in Fiscal Federalism. Washington, D.C.: Brookings Institution.Oates, Wallace (1972). Fiscal Feeralism. Harcourt Brace Jovanovich, New York, USA.Oates, Wallace (1990). Public Finance With Several Levels of Government: Theories and Reflections.
Processed.
- 105 -
Olowu, Dele (1980) "Local Governments and Intergovernmental Fiscal Relations in Nigeria and Brazil:A Comparative Study" Philip2ine Journal of Public Administration. XXIV 2
Owens, Jeffrey and John Norregaard (1990). The Role of Lower Levels of Government: TheExperienceof Selected OECD Countries.
Pasha, Hafiz and Aisha Ghaus (1986) "Analysis of Provincial Finances in Pakistan". Applied EconomicsResearch Centre, University of Karachi, Pakistan. Processed.
Prust, Jim et al. (1990) "The Czech and Slovak Federal Republic: An Economy in Transitio..International Monetary Fund, Washington, DC
Rao, M.G. and R.J. Chelliah (1990). Survey of Research on Fiscal Federalism in India. Processed.Shah, Anwar. (1991a). Intergovernmental Finance in a Comparative Perspective. Processed
(Preliminary draft).Shah, Anwar (199lb). A Fiscal Need Approach to Equalization for Canada. Paper presented to the
Canadian Economic Association Meetings, Kingston, Ontario, Canada, June 1991.Shah, Anwar (1990a). "The New Fiscal Federalism in Brazil' The World Bank PRE Working Paper
Series 557, December.Shah, Anwar (1990b). "Empirical Tests for Allocative Efficiency in the Local Public Sector". Public
Economics Division, World Bank. Processed.Shah, Anwar (1990c). Intergovernmental Fiscal Relations in Sri Lanka. Paper presented at the 46th
Congress of the International Institute of Public Finance, Brussels, August 27-30, 1990.Shah, Anwar (1989a). A Linear Expenditure System Estimation of Local Fiscal Response to Provincial
Transportation Grants. Kentucky, Journal of Economics and Business, pp.150-168.Shah, Anwar (1989b). A Capitalization Approach to Fiscal Incidence at the Local Level. Land
Econgmics, Vol. 65, No. 4, November 1989, pp. 359-375.Shah, Anwar (1988a). An Overview of Fiscal Decentralization Issues in Sri Lanka. The World Bank,
processed.Shah, Anwar (1988b). Capitalization and The Theory of Local Public Finance: An Interpretive Essay.
Journal of Economic Surveys, Vol. 2, No. 3, pp.209-243.Shah, Anwar (1988c). Dynamics of Public Infrastructure and Private Sector Productivity and
Profitability. Forthcoming, The Review of Economics and Statistics, Harvard University.Shah, Anwar (1984a). Alternative Approaches to The Measurement of Expenditure Needs of Canadian
Provinces. Finance Canada Discussion Paper, March 1984.Shah, Anwar (1984b). An Empirical Analysis of Public Transit Subsidies in Canada. Finance Canada
Discussion Paper, June 1984. Also published in Roy Brown (ed.). Quantity and Quality inEconomic Analysis. Vol. II (1986), University Press of America.
Shah, Anwar (1984c). A Preliminary Analysis of Expenditure Need Compensation in FiscalEqualization. Finance Canada Discussion Paper, November 1984.
Shah, Anwar (1983a). Municipal Fiscal Needs and Provincial-Municipal Unconditional Transfers inCanada. Discussion Paper, Finance Canada, Ottawa, March 1983.
Shah, Anwar (1983b). Efficiency, Equity and Fiscal Equalization Grants: Issues and Alternatives.Finance Canada Discussion Paper, October 1983.
Shah, Anwar (1983c). The New Federalism in Australia. Finance Canada Discussion Paper, February1983.
Shah, Anwar (1981). The Reform of Provincial-Municipal Transfers in Alberta. A Report prepared forthe Minister of Municipal Affairs, Alberta.
Shah, Anwar (1979). Provincial Transportation Grants to Alberta Cities: Structure, Evaluation and aProposal for An alternate Design. Alberta Transportation, Edmonton, Canada. An abridgedversion published in Roy C. Brown (editor), Ouantity and Ouality in Economic Research,Volume I (1985), New 'York and London: University Press of America, pp. 59-108.
- 106 -
Sichev, N. and V. Ilyin (1990). Management and Interrelations of Budgets in a Federative State: TheCase of the USSR. Processed.
Singh, S.K. (1990). Fiscal Federalism in India in the Context of Planned Development. Processed.Tanzi, Vito (1978). "Inflation, Real Tax Revenue, and the Case of Inflationary Finance: Theory with
an Application to Argentina". IMF Staff EaC . Vol. 25, pp. 417-51.Wallich, Christine (1982) "Revenue Sharing". State Fces in India-Vol 1. World Bank Staff Working
Papers 523.World Bank (1986). Brazil: Finance of Pimary Education. A World Bank Country Study.Wu, Jing-lian and Ji-wei Lou (1991). China: Intergovernmental Fiscal Relations and Macroeconomic
Management. Processed.
- 107 -
PRE Working Paper Series
ContactAuth Da for paer
WPS702 Should Price Reform Proceed Sweder van Wijnbergen June 1991 M. StroudeGradually or in a "Big Bang?" 38831
WPS703 The Political Economy of Fiscal Sebastian Edwards June 1991 A. BhallaPolicy and Inflation in Developing Guido Tabellini 37699Countries: An Empirical Analysis
WPS704 Costs and Financs of Higher Rosemary Bellew June 1991 C. CristobalEducation in Pakistan Joseph DeStefano 33640
WPS705 What Causes Differences in Abby Rubin Riddell June 1991 C. CristobalAchievement in Zimbabwe's Levi Martin Nyagura 33640Secondary Schools?
WPS706 Successful Nutrition Programs in Eileen Kennedy June 1991 0. NadoraAfrica: What Makes Them Work? 31091
WPS707 Population, Health, and Nutrition: Population, Health, June 1991 0. NadoraFiscal 1990 Sector Review and Nutrition Division, 31091
Population and HumanResources Department
WPS708 Nongovernmental Organizations and Jocelyn DeJong June 1991 0. NadoraHealth Delivery in Sub-Saharan Africa 31091
WPS709 An Empirical Macroeconomic Model Luis Serven June 1991 S. Jonnakutyfor Policy Design: The Case of Chile Andres Solimano 39074
WPS710 Urban Property Tax Reform: William Dillinger June 1991 V. DavidGuidelines and Recommendations 33734
WPS711 Financial Reforrm in Socialist Millard Long June 1991 M. RaggambiEconomies in Transition Silvia B. Sagar 37657
WPS712 Foreign Direct Investment in Thomas L. Brewer June 1991 S. King-WatsonDeveloping Countries: Patterns, 31047Policies, and Prospects
WPS713 The Determination of Wages in Simon Commander June 1991 0. Del CidSocialist Economies: Some Karsten Staehr 39050Microfou ndations
WPS714 Women in Forestry in India Ravinder Kaur July 1991 A. Sloan35108
WPS715 Promoting Girl's and Women's Rosemary Bellew July 1991 C. CristobalEducation: Lessons from the Past Elizabeth M. King 33640
WPS716 Financing Training: Issues and Christopher Dougherty July 1991 C. CristobalOptions Jee-Peng Tan 33640
PRE Working Paper Series
ContactAutho DA for paer
WPS717 Does Financial Liberalization Really Jacques Morisset July 1991 S. King-WatsonImprove Private Investment in 31047Developing Countries?
WPS718 Impact of Investment Policies on Andrea Gubitz July 1991 S. King-WatsonGerman Direct Investment in Developing 31047Countries: An Empirical Investigation
WPS719 How Trade and Economic Policies Ramon Lopez July 1991 M. GunasekaraAffect Agriculture: A Framework for Ridwan Ali 32261Analysis Applied to Tanzania and Bjorn LarsenMalawi
WPS720 The Outlook for Commercial Bank Ellen Johnson Sirleaf July 1991 S. King-WatsonLending to Sub-Saharan Africa Francis Nyirjesy 31047
WPS721 The Demand for Money in Developing Patricio Arrau July 1991 S. King-WatsonCountries: Assessing the Role Jose De Gregorio 31047of Financial Innovation Carmen Reinhart
Peter Wickham
WPS722 Is Rice Becoming an Inferior Good? Merlinda D. Ingco July 1991 P. KokilaFood Demand in the Philippines 33716
WPS723 Improving Women's and Children's Olayinka Abosede July 1991 0. NadoraNutrition in Sub-Saharan Africa: Judith S. McGuire 31091An Issues Paper
WPS724 Fiscal Issues in Adjustment: Riccardo Faini July 1991 D. BallantyneAn Introduction Jaime de Melo 37947
WPS725 How Structure of Production Indermit S. Gill July 1991 A. SloanDetermines the Demand for Human Shahidur R. Khandker 35108Capital
WPS726 Perspectives on the Design of Anwar Shah July 1991 A. BhalIaIntergovernmental Fiscal Relations 37699
WPS727 The Effects of Debt Subsidies on Mansoor Dailami July 1991 A. Bruce-KonuahCorporate Investment Behavior E. Han Kim 80356
Top Related