INTERGOVERNMENTAL FISCAL TRANSFER SYSTEM IN FEDERATION OF BOSNIA AND

HERZEGOVINA – EQUALIZATION ASPECTS

ByHalko Basaric

Submitted toCentral European UniversityDepartment of Public Policy

In partial fulfillment of the requirements for the degree of Master of Arts in Public Policy

Supervisors: Professor Robert Ebel and Profesor Adrian Ionescu

Budapest, Hungary

2006

ABSTRACT

i

Decentralization as a policy option has become, together with other public sector

reforms, a hallmark of the transition process in the Central and Eastern Europe.

Decentralization accounts for improved transparency and accountability, well designed it can

enhance resource efficiency as well as serve as key to the accomplishment of a nation’s

broader economic goals of macro-economic stability, privatization and maintenance of the

social safety net. But, for decentralization to “work” it is essential that there is a fiscally

strong central government that can set standards which serve the purpose of equalizing the

service delivery and living conditions of the citizens.

This thesis explores present system of intergovernmental relations in Federation of

Bosnia and Herzegovina (FBiH), an entity comprising State of Bosnia and Herzegovina

(BiH). It argues that those arrangements create huge vertical and horizontal imbalance

between cantons and municipalities, which represents a potential threat to building a new

nation. Peculiar intergovernmental relations, which favor intermediate level-cantons, left

FBiH Government without fiscal power for equalization, and are direct consequence of strong

centrifugal forces that emerged in the course and after the four-year conflict. For the first time

after the Dayton Peace Accord, and after introduction of VAT on the state level, FBiH

Government has introduced a formula based public revenues allocation system, thus replacing

the previous derivation based one. The thesis examines new system, suitability of its design to

Bosnian fiscal federalism and test whether the formula corrects growing fiscal imbalances. It

finds that new transfer system is comparatively better solution for transition country such as

BiH, provided that other reforms are implemented simultaneously.

TABLE OF CONTENTS

ii

Introduction.........................................................................................................................1Why Decentralize?..................................................................................................1Classis Theoretical Frameworks..............................................................................3Fiscal Decentralization............................................................................................5Intergovernmental Transfer System in Federation of Bosnia and Herzegovina.....8

Chapter 1: Legal Framework for Intergovernmental Fiscal Relations..............................121.1 Tradition of Self-Governance..........................................................................121.2 Legal Framework.............................................................................................131.3 Local Governance in Federation of Bosnia and Herzegovina.........................151.4 Local Governance in Republika Srpska..........................................................18

Chapter 2: Functions of Local Governments in Federation of Bosnia and Herzegovina. 202.1 Functions Exclusively Assigned to Federal Level..........................................262.2 Functions Assigned to Cantons and Functions Delegated to Municipal Level..........................................................................................262.3 Concurrent Responsibilities.............................................................................282.4 Functions Assigned Exclusively to Municipalities.........................................28

Chapter 3: Revenues Sources............................................................................................303.1 Reasons for Decentralizing Revenues.............................................................303.2 Situation Before VAT Introduction.................................................................323.3 Tax System Reform-A Policy Window for Fiscal Equalization.....................36

Chapter 4: New Intergovernmental Transfer System in Federation of Bosnia and Herzegovina............................................................38

4.1 Theoretical Framework...................................................................................384.2 New Formula Based Allocation Mechansm in FBiH......................................434.3 How Good New System Performs?.................................................................46

Conclusion.........................................................................................................................48

Appendices........................................................................................................................51

Bibliography......................................................................................................................56

iii

I would like to express my sincere gratitude to team implementing USAID’s Governance Accountability (GAP) Project in Bosnia and Herzegovina, especially to Policy Advisor Mrs.

Jasmina Djikic. I would like also to thank my supervisors Robert Ebel and Adrian Ionescu for their guidance and support.

iv

INTRODUCTION

Why decentralize?

Decentralization is the word that has marked the transition process through which

countries of Central and Eastern Europe have been going after the fall of communist regimes.

Decentralization, however, has been on the agenda of many countries already for decades,

developed and developing ones, unitary and with more complex governmental structure, rich

and poor. Some 95% of the democracies around the world have elected some sort of sub-

national governments (World Bank Development Report 2000).

Countries decide to decentralize for a variety of reasons. Some see decentralization as

the mean to boost the economic growth and as part of effort to improve the governance in the

country (Ebel and Yilmaz 2002, Vazquez-Martinez and McNab 2005). In other countries,

central government is “forced” to devolve some power to the regional government in face of

strong centrifugal or even secessionist forces in the society, and to achieve political stability

(Bird and Ebel, 2005). Macroeconomic stability and equity are also of concerns for the

decentralization, though the results are not unambiguous, especially for the countries with

weak governance (Ebel and Yilmaz 2002, Anwar Shah 1997). Centralization has its benefits

as well. Roy Bahl (1999) for example, finds that transition countries are more sensitive to the

swings on the international markets, which are transmitted and magnified in the domestic

public sector. In this case stabilization of economy is more easily achieved if the central

government is not locked into fixed arrangements with the subnational governments.

According to World Bank (2001) there are three imperatives behind this phenomenon:

- Political Imperative. After years spent under monolithic political system, people in

transition countries ask for more autonomy and self-governance. The decision-making power

1

vested by the legislation to the locally elected officials, increase accountability and citizens

participation;

- Economic Imperative. In socialism, the size of the public sector in providing goods

and services for citizen was immense. For transition countries there was not other option but

to reduce public sector share in the national economy. Decentralization of provision of goods

and services is one way to achieve that goal;

- Service Delivery Imperative. This imperative re-affirms Oates’ Decentralization

Theorem (Oates 1972): Provision of public services is done more efficiently by the local

government than if it is centrally provided, for local governments can more accurately

determine citizen’s preferences and diversify the level of service, provide that there is no

existence of economies of scale and externalities for delivering those services. Under the old

system, majority of the services were either directly delivered by the central government, or

the provision was centrally regulated.

Decentralization entails transfer authority and responsibility for public functions to

subordinate or quasi-independent governments organization of the private sector

(Rondinneli). Following types of decentralization can be identified:

- Political decentralization. It aims to give citizens and their elected representatives

more power in public decision making. It is often associated with pluralistic politics and

representative government, but it can also support democratization by giving citizens or their

representatives more influence in formulating and implementing policies. Local politicians

better know the need and desires of their constituents and citizens can better know their

political representative. Final result should be more political stability and better governance

- Administrative decentralization seeks to redistribute authority, responsibility, and

financial resources for providing public services among different levels of government. It is

the transfer of responsibility for planning, financing, and managing certain public functions

2

from the central government and its agencies to field units of government agencies,

subordinate units or levels of government, semi-autonomous public authorities or

corporations, or areawide, regional, or functional authorities.

- Fiscal decentralization. Financial responsibility is a core component of

decentralization. If local governments and private organizations are to carry out decentralized

functions effectively, they must have adequate revenues—raised locally or transferred from

the central government—as well as the authority to make expenditure decisions. Fiscal

decentralization include devolving some of the political decision-making power to the local

level, which is not often the case with administrative decentralization.

Classic Theoretical Frameworks

The theoretical foundation for the public finance analysis was laid in the works of

three distinguished scholars: Charles Tiebout (1956), Richard Musgrave (1959) and Wallace

Oates (1972). Tiebout’s “Pure Theory of Local Expenses” is focused on the local public

finance, and it seeks to prove that there is a “market-type” solution the determining the level

of public expenditure on the local level. In Tiebout’s model, local households seek a

community of residence that provides a fiscal bundle closely approximating their demands for

local services. Local voter reveal their preferences by “voting by feet”, which generates

Pareto-efficient outcome, similar to that in private markets, under the condition that there is

large number of local communities providing local services at the tax price equal to the

marginal cost.

Musgrave’s “The Theory of Public Finance” is a departure point for any analysis of

the public finance. It splits the public sector into three branches:

- Macroeconomic Stabilization-which is assigned to the central government. Overall

managing of the economy and monetary policy should be prerogatives of the central

government. Regional government because of their openness and possibility of interregional

3

competition cannot perform this function. This implies at the same time that central

government must some tax handle suitable for achieving that task of stabilization.

- Distribution-which is also assigned to the central government. Central government can

more easily redistribute the national income or wealth, as it usually has broad-based taxes

available for this. Also, the mobility of the taxpayer from the jurisdictions with unfavorable

taxes to those with less redistributive policies weakens the sub-national government potential

for redistribution. On the other hand, the fact sub-national governments are in most countries

responsible for providing the social assistance gives, and therefore can increase efficiency of

the programs, provides some space for redistributive role of sub-national governments.

- Allocation Function. This function in Musgrave’ tripartite division explains how

different levels of government provide mix of private and public goods to satisfy the needs of

the consumers. Some goods are most efficiently delivered at the national level (e.g. defense)

while other which have local character can be better suited to meet the needs of the local

community (fire protection, water supply). Centralized provision of these service would deny

their local character, thus resulting in welfare losses and inefficiencies.

The third function is closely related to Oates “Decentralization Theorem”: “For a

public good–the consumption of which is defined over geographical subsets of the total

population, and for which the costs of providing each level of output of the good in each

jurisdiction are the same for the central or for the respective local government–it will always

be more efficient (or at least as efficient) for local governments to provide the Pareto-efficient

levels of output for their respective jurisdictions than for the central government to provide

any specified and uniform level of output across all jurisdictions” (Oates 1972). Briefly, the

Theorem states that under the given conditions, diversification of the outputs for different

communities as demanded by the residents will increase welfare. The underlying condition of

the theorem emphasizes the importance of the optimal size of the jurisdiction for providing a

4

particular public service. The question of mobility of tax payers in this case is closely related

to vertical and horizontal fiscal equity in the country as well as minimum service provision

(Oates, 1977).

Fiscal Decentralization

Countries around the world are decentralizing their governmental structure in an effort

improve the overall efficiency of the economy by devolving service delivery and taxation to

the lower tiers of government. Proportion of countries with some form of democratic

government rose from 28% in 1974 to 61% in 1988 partly as a response to the ongoing

globalization and integration of national economies with increasing confinement of the

powers of nation-states that it entails (Dabla-Norris 2006).

In some countries fiscal decentralization is seen as the way for increasing the

accountability and legitimacy of the government, as fiscal decentralization strengthen the

voice of local communities and opens the space for citizens’ participation in decision-making

process. Locally elected politicians can focus better tailor the bundle of local services for the

citizens, while the same time transfer of assigning of own-revenues increases the

accountability on the local level. In transition countries of Central and Eastern Europe,

decades of state dominance created a strong impetus for decentralization initiatives (Bird,

Ebel and Wallich 1995). Therefore, those countries musth “catch-up” with the decades-long

practice of western countries.

The set of government that is closest to the citizens can adjust budgets to the local

preferences, and therefore enhance the efficient utilization of the scarce resources (Ebel and

Yilmaz). Local tax bases are more accessible to the sub-national governments, provided it

function well. Decentralization leads to enhancing democratic values at the local level by

allowing citizens to take active role and assume responsibility for the local government

5

actions. In the transition countries, fiscal decentralization can also play role in privatization of

the state owned enterprises and poverty alleviation (Bird, Ebel and Wallich 1995)

In any of the cases, decentralization is about intergovernmental arrangements amongst

the different types of governments, regardless of how country’s public sector is arranged-

there are at least two tiers of government even in the unitary states, which makes inevitable to

have some sort of intergovernmental fiscal relations.

The process of decentralization must give answer to the following questions (Bird and

Vallincourt 1998):

1. Who does what?-the question of expenditure assignment

2. Who levies what taxies-the question of revenues assignment

3. How is the imbalance between the revenues and expenditures for sub-national

governments that results from the answers from the first two questions to be resolved?-the

question of vertical imbalance

4. To what extent should fiscal institutions attempt to justify for the differences in needs and

capacities among different governmental units at the same level of government?-the

question of horizontal imbalance, or equalization.

How one value decentralization depends on the point of view: bottom-up

decentralization generally stresses political values-improved governance in the sense of

increased responsiveness and citizen’s participation, as well as allocative efficiency. This

approach emphasizes political dimension of the decentralization, as it is said the better use of

local knowledge and more efficient and equitable service delivery would presumably result

with more support for the government; top-down approach however is characterized by the

central government’s pursuing some national policy through fiscal decentralization. This can

6

be for example “deficit shifting” to the sub-national governments or desire of the central

government to achieve more equity or more of the social welfare.

Decentralization lies on the continuum which depends on the degree of independent

decision-making exercised at the local level (Bird and Villancourt 1998):

- Decocentration means the dispersion of responsibilities within a central government to

regional offices or local administrative units;

- Delegation refers to a situation in which local governments acts as agent for the central

government, executing certain function on its behalf;

- Devolution refers to a situation in which not only implementation, but also authority to

decide what is done is in the hands of local government.

Before we move to discuss transfer system as the component of fiscal decentralization

and its importance in Federation of Bosnia and Herzegovina, it is important to mention valid

alternatives. Fiscal decentralization is not a panacea for organizational issues in a country.

Vito Tanzi (Tanzi 2001) identifies at least two alternatives to the fiscal decentralization:

improving current politics and reducing the role of public sector. Often demands for

decentralization are result of the citizen’s perception of central government policies as unjust,

inefficient or inadequate. Badly designed regional development policy or corrupt central

government will create pressures for more decentralized policy decision-making. If the central

government does not change its policy, it is likely that the pressures for more efficient and

equitable government will intensify trough decentralization. Reducing the role of the public

sector is another alternative to the decentralization, because in Tanzi’s word “the growth of

government in recent decades-which has been accompanied by a huge expansion of public

sector activities in many countries-must have been partly responsible for increased demands

for decentralization witnessed in recent years”. Therefore, according to him at least part of

7

the government activities could be privatized, depending on governments efficiency or

privatization and government’s sectoral specific policies.

Intergovernmental Transfer System in Federation of Bosnia and Herzegovina

It is very rare situation that revenues assigned to the sub-national governments are

fully matched with the expenditures needs of that level. In those cases, intergovernmental

fiscal transfers are used to close the gap between the expenditure responsibilities and

revenues-raising capacity, or in other words to close vertical fiscal gap. At the same time,

resources available to the sub-national can vary significantly, or the expenditures needs for

providing standardized basket of service can be different, thus effectively creating a

horizontal fiscal imbalance (Bird and Tarasov 2002 ). Fiscal transfers can also be used to

influence spending patters of the sub-national governments or to stimulate local fiscal efforts

(Ebel, Bird and Wallich 1995).

First objective of this paper, elaborated in Chapter 1 through III is to examine

intergovernmental fiscal system in Federation of Bosnia and Herzegovina (FBiH), claiming

that it creates huge vertical and horizontal imbalances between the cantons and municipalities.

This will be done by analyzing expenditures and revenues assignments, as well as legal

framework for intergovernmental relations. A unique character of the fiscal federalism in BiH

has not created conditions for development of the transfers system in the post-conflict period.

Republika Srpska (RS) and FBiH, two entities comprising Bosnia and Herzegovina, are

ethnically based administrative units with excessive revenue-raising powers assigned, while

major spending responsibilities were kept either at the entity level (in RS), or as it case in

FBiH also at the sub-entity cantonal level. Smaller BiH entity (RS) has a unitary form, with

strong central government and no intermediate level of authorial. Until the recent tax reform,

8

State of Bosnia and Herzegovina had very expenditure responsibilities assigned to it, and thus

is limited in its capacity to influence the fiscal sector.

Vertical and horizontal imbalances in FBiH which are result of non-existing transfer

system have reached worrying proportions. Measure of the Minimum of GDP per capita to

FBiH average, shows that Canton Una-Sana has only 66,7% of the GDP pc of the average.

Another measure, Maximum to Minimum Ratio (MMR) of GDP per capita for ten cantons

comprising FBiH reveals that the ratio between the poorest and richest canton is 2,8:1. At he

same time, municipalities in FBiH in 2003 constituted only 11% of all public spending in the

entity, or 3.2% of FBiHs GDP. This very low for in comparison with RS and shows that

municipalities in FBiH are radically underfunded (GAP, 2005).

One of the main problems of FBiH fiscal system is its fragmentation to in fact ten

small cantonal systems, which compounded with the inability of FBiH to actively pursue any

equalization policy, represents a huge social, political and economic problem.

Decentralization inevitably has a result some degree of inequality between the richer and

poorer jurisdictions, or in other words “equal treatment of equals” (Buchanen 1950) can only

be achieved by using intergovernmental fiscal transfers. Equalization requires a consensus

that social citizenship and solidarity among all citizens apply with equal force nationwide as

opposed to being region-specific (Boadway 2004).

In FBiH this was not the case. UNDDP (Human Development Report for BiH 2005)

states that there are huge differences in the level of education expenditure amongst

administrative units (cantons).These differences work against the uniform application of

human rights and the modernization agenda. The same report further notes in the health care

sector, administrative duplication is strongest in the Federation where, in addition to the

cantonal authorities, the Federal Ministry also exercises a coordinating role.

9

So called Net Fiscal Benefit approach to equalization (Bird and Waillincourt 2004)

state that tax structure need to be as neutral as possible, to avoid distortion of the resources

allocation, which should be done only on the economic basis. If this is not so, then fiscal

transfer are used to restore the neutrality of the tax structure so that resources are distributed

according to their productivity. Though the mobility of capital and people is low in BiH, this

type of distortion is present, especially between the cantons with urban amalgamation (for

example Sarajevo or Mostar) and those scarcely populated and remote cantons (Livno or

Gorazde).

As of January 2006, Bosnia and Herzegovina introduced a Value Added Tax (VAT) as

part of the reform of national tax system. Collection of indirect taxes and customs is

centralized on the state level, under the newly formed Indirect Taxation Authority (ITA) . Tax

policy has been also transferred to the state level, thus effectively leaving entities with very

little power in the fiscal domain. The revenues collected to the ITA’a Single Account are,

however, ceded to the entity level, based on the entity share of the processed VAT invoices,

subject of much political debate.

This solution enabled FBiH entity to introduce for the first time in post-conflict era

revenues-sharing mechanism incorporating some elements of fiscal equalization. Unlike with

the previous solution when each canton had its own revenue-sharing law, new Federal Law on

Distribution of Public Revenues (Official Gazette 22/06) employs distribution formula to

allocate the revenues between cantons and municipalities, with gradual introduction over the

period of six years. A myriad of cantonal revenues-sharing laws has been abolished and

replaced with the a single law, which despite the fact that FBiH is no longer in control of

indirect taxes, still allows for active policy role to be played. As second objective, this paper

will examine new system of intergovernmental transfers in FBiH, specifically focusing on its

10

equalization aspects. In Chapter IV we will analyze new system, both in the light of

theoretical literature and its practical suitability for FBiH conditions. This will be

complemented by testing how newly introduced formula-based transfer treats sub-national

governments with respect to equalization. We will calculate per capita amount for the last

year of transition period, and then we will use statistical method of regression to determine its

correlation with fiscal capacity and expenditure needs of cantons and municipalities.

11

CHAPTER 1: LEGAL FRAMEWORK FOR INTERGOVERNMENTAL FISCAL RELATIONS

1.1 The Tradition of Self-Governance

Former Socialist Federative Republic of Yugoslavia had a long and unique tradition of

local self-governance. The so called ‘self-management’ under which all enterprises operated

as well as the other institution of the Yugoslav socialist system, has been developed after the

WWII when Tito rejected soviet model, and chose to take Yugoslavia on a separate road to

socialism. The self-management was first introduced in factories and enterprises, but later it

became a hallmark of the Yugoslav socialism as it integrated into the socio-economic life of

the whole country.

Self-management allowed workers in factories and enterprises to decide on all aspects

of managing through the “workers councils”. The councils for example, had the power to

decide on investments, managers or profit sharing. On the level of municipalities, the workers

were represented in one of the two municipal councils. The other council was popularly

elected. These councils had wide responsibilities, from deciding on local fees, rents and

apartment allocation, to the taxation of the companies and deciding on their business plans.

Local voters controlled the work of councils through the regular “voters meetings” and

consultations, and the council would be disbanded if one-third of local residents voted for

non-confidence. With the constitutional changes in 1974, Yugoslav Republics gained

additional powers, and units of local self-governance-municipalities, were transferred more

responsibilities. Though this kind of Yugoslav “direct democracy” was taking place in a one-

party system, constituent nations in Bosnia and Herzegovina-Bosniacs, Croats and Serbs have

learnt basic skills of participation in local self-governance. This tradition has been kept until

today, though the war has severely influenced its functioning in the Daytonian Bosnia and

Herzegovina.

12

1.2 Legal Framework

Bosnia and Herzegovina emerged after the war as a heavily decentralized country.

Peace talks held in Dayton in November 1995, were naturally more directed toward reaching

the peace, than to find optimal solution for the fiscal federalism in the new country. Very

strong secessionist forces at that time in Bosnia and Herzegovina were driven by a simple

maxim: to legalize what has been won in the war, and if conditions permit, to join the

neighboring states. Therefore, there was not too much of desire or opportunity to create a state

functioning on the principles of economic efficiency and social equity of its citizens. The

Dayton Peace Accord (DPA) marked the beginning of the new era for the centuries old

country.

In the new Constitution, Bosnia and Herzegovina was envisaged as decentralized

asymmetric federation, composed of two entities. The smaller entity, Republika Srpska (RS)

has 49% of the pre-war territory of BiH. Bigger entity, Federation of Bosnia and Herzegovina

(FBiH) through negotiations was allocated 51% of the territory1. A small town of Brcko in

northern Bosnia was a subject of dispute of parties. In 2000, an international mediator decided

to give Brcko the status of special administrative district in Bosnia and Herzegovina,

effectively creating a third territorial unit.

1 In fact, Federation of Bosnia and Herzegovina had been formed 18 months before the Dayton Peace Accord (DPA) was signed. Two Bosnian “constituent nations” Croats and Bosniacs agreed in 1994 in Washington to establish a Federation. Since at that time it was not possible to predict whether Bosnia and Herzegovina will continue to exist as a state, in other word whether Bosnian Serbs will secede into the separate entity, FBiH entity was envisaged as the fully functioning state, with executive, judicial and legislative authorities military forces and other attributes. Eventually, DPA has ended the war by recognizing the second entity, but the previously established structure remained unchanged.

13

Both entities are administratively, politically, and until the introduction of State VAT

fiscally, independent from the central government. They have very distinct administrative

organizations. While Republika Srpska has kept a form similar to the socialist BiH, with two

levels of government, an intermediate level-cantonal was introduced in the Federation. This

was necessitated by the deep mistrust between the ethnic groups, product of war, but also it

was caused by till lively memory of Yugoslav interregional equalization policy, for which

especially Croats had aversion (Fox and Wallich 1997).

Figure 1. Governmental Structure of Bosnia and Herzegovina as from Dayton

Central Government of Bosnia and Herzegovina (Council of Ministers-CoM) had very

narrowly defined functions in the new Constitution. Those included competencies over

monetary policy, foreign policy, immigration and asylum, and debt service (though in the case

of the last, only entities were entitled to borrow). The assignment of responsibilities was

negatively defined, so that all competencies not assigned to the state level, are the function of

the entities. These quasi-federal arrangements reflected the Bosnian reality in 1995, and were

the only way for maintaining the sovereignty of the country, at the same bringing three

nations of a ‘divided house’ to live together. Weak central government, however, proved to be

a one of the main ‘system errors’ of the new structure. Macroeconomic stabilization, fiscal

Council of Ministers

Federation of Bosnia and Herzegovina

Republika Srpska

Municipalities (79) Municipalities (56)

Cantons (10)

14

consolidation, privatization of state property, were some of the critical areas in which the lack

of coordination was often translated into bad policy solution, obstruction or even irrational

rejection.

The role that the ethnic division has played in the public sector should not be

neglected. Even now, after ten years of reforms and enormous efforts invested into creating

efficient institutional framework, the ethnic questions remain dominant and firmly

entrenched, not only in the public sector, but unfortunately in the everyday life as well. As

Bird (2003) says “fiscal rationality may seem irrelevant when faced with symbolic meaning”.

The same can be said for any other policy domain.

Central authorities on the state level do not have any competencies over the territorial

organization of entities or internal administrative structure. Because of very different extent of

decentralization of the entities, they will be analyzed separately.

1.3 Local governance Federation of Bosnia and Herzegovina

The BiH bigger entity in fact has all characteristics of federally arranged state. There

are ten cantons, which comprise the FBiH, and fiscally strong Federal government, though

with the recent changes in the VAT introduction, the fiscal power has been reduced. Cantons

are created as ‘mini states’ with their own constitution, government and parliamentary

representation, elected through an open and universal election. They are a result of the efforts

to reconcile the centrifugal forces in the society, and promote political stability in the ‘divided

house’ by constitutionally guaranteeing division of power between the entity level and

cantons. The topic of assignment of responsibilities is more elaborated in Chapter 2, but just

to briefly mention that cantons have exclusive competencies for example over health, security

(police), education or urban planning. At the time when the FBiH was conceived, assigning

of significant share of power to the sub-entity level was the only viable option for maintaining

15

the integrity of the country. Along with the decentralization of functions to the cantons, FBH

has introduced a number of consociational mechanisms, such as vital national interest

procedure in the Federal Parliament, to ensure checks and balances in the system between the

constituent people.

Table 1: Basic information for cantons in FBiH

Canton Populat

ionArea

(km2)Ethnic

composition

1. Una-Sana 288.935 4.125 Bosniack majority

2. Posavina 41.436 324,6 Croat majority

3. Tuzla 496.885 2.649,0 Bosniack majority

4. Zenica-Doboj 401.401 3.343,3 Bosniack majority

5. Bosansko-Podrinjski 34.508 504,6 Bosniack majority

6. Central Bosnia 256.175 3.189 Croat majority

7. Herzegovina-Neretva 228.649 4.401 Mixed canton

8. West Herzegovina 83.570 1.362,2 Croat majority

9. Sarajevo 410.031 1.276,9 Bosniack majority

10. Livno 82.402 4934,1 Croat majority

Total Federation 2.323.9922 26.110,5

As it can be seen from the table, federal cantons vary significantly in their size and

population. Two of the cantons are without ethnic majority, while others have a dominant

ethnic group. The size, population and urban-rural division not only between cantons, but also

within the cantons, bear impact on the provision of the public goods. The lack of horizontal

cooperation between cantons is often translated into duplication or underprovision of services.

Local self-government is mentioned in the Constitution of FBiH in section VI, which

lists the main competencies of cantons and municipalities, where it is stipulated that

municipalities are the basic unit of self-government. Further, this matter is regulated on the

2 About 512.00 of citizens of FBiH are still residing abroad as refugees, and they are excluded from this figure.

16

basic level in Federal Law on Self-Governance. This framework law affirms the underlying

notion of subsidiarity from the European Charter on Local Self-Governance, and it stipulates

that municipalities are independent within their own scope of competencies. But the Law does

not regulate this matter further, as it transfers the local self-governance to the cantonal level.

Effectively, this has created ten different systems of local self-governance in FBiH, though

cantons have to operate within the framework law. For illustration, suffice it here to say that

often cantons violate the autonomy of municipalities by unilaterally changing the revenues

sharing formulae or delegated responsibilities to municipalities. For a highly decentralized

country, at least to the intermediate level, it comes as surprise that no government at any level

has a ministry or governmental agency tasked with the affairs of local self-government. The

plethora of sub-national laws regulating this area, with poor coordination, especially in the

early days of FBiH, put the municipalities in an unfavorable position. The situation had

somewhat improved with strengthening of associations of municipalities and cities, but also

partially because of the growing awareness of the importance of local level not only in

delivering the services efficiently, but what is more important, in building trust between the

people. Another factor that might play a key-role that is examined in this thesis is the new

system of intergovernmental transfers, introduced together with the VAT on the state level.

At the moment, there are 79 municipalities and two cities-Mostar and Sarajevo, in the

Federation of BiH. Municipalities differ significantly in size and population, varying from

small detached parts of the pre-war municipalities which formed new ethnically congruent

units, to the big urban centers. Cities are consisted of several urban municipalities, which

joined can enjoy the economy of scale and externalities in service delivering.

Local councils are elected on open, direct election, with universal suffrage. They

cannot be dissolved by the higher authority, but legal acts such as municipal Statutes, can be

17

challenged by the cantonal authorities before the Constitutional Court of Federation, only if

such acts are against the Constitutions. Also, cantonal laws restrict the scope of supervision

only to the tasks delegated by the cantonal authorities to the local level, and in accordance

with the European Charter on Local Self-Government, the intervention of the cantonal

authorities should be kept in proportion to the importance of interests which it protects. Chief

local executives (mayors and heads of municipalities) have been directly elected since 2004 in

both entities. Municipalities are also free to associate and collaborate with other

municipalities in the country and abroad.

The decentralization of FBiH has stopped at the level of cantons. Although this

intermediate level does have advantage of the economy of scale in providing some services

and internalizing the externalities, this new level of authority is not historically or legally

rooted in the Bosnian society. It is an unnatural division imposed to keep the most of

resources at the local level under the control of technocratic parties. Cantons very often

encroach on the powers of the municipal level, which makes it difficult to develop the

accountability at grassroots level. Municipalities have different position in the

intergovernmental system, depending in which canton they belong. In general, there was no

political will to strengthen the position of municipalities, as the political reality dictated the

development not only of intergovernmental relations in FBiH, but in the whole as well.

1.4 Local Governance in Republika Srpska

The smaller BiH entity, Republika Srpska has not existed before the war as a territorial

unit comprising Bosnia and Herzegovina. It is a product of the nationalistic expanding politics

that erupted after the dissolution of Yugoslavia, joined with secession forces in Bosnian

society. It has a unitary form with strong central government and no intermediate level of

authority. In Republika Srpska, there was no need for decentralization polices because of the

18

high degree of ethnic homogeneity achieved by pursuing the politics of “one nation on one

land” during the war.

The question of local self-governance is addressed in the Constitution of RS in several

articles, and is given high importance as for example Art 5 proclaims local self-

government is one of the fundaments the constitutional arrangement of

the Republic of Srpska is based on. Furthermore Art 66 stipulates that self-

government constitutes a basic socioeconomic relations of RS , and that

the borders of its autonomy are protected. Despite these provisions, the

RS Constitution does not explicitly guarantee the independence of self-

government units, and treats them more as territorial arrangements than

a genuine devolution of power.

The Law on Local Self-Governance, however, compared to the federal level is much

more detailed in description of competencies to the municipalities. It also lists a range of legal

measures for protection of the autonomy on the local level. Similar to FBiH, the Law does not

distinguish between rural and urban municipalities, but it makes distinction between

municipalities and urban amalgamation-cities.

The same conclusion for the position of municipalities in the intergovernmental

system in FBiH can be drawn for Republika Srpska. But in case of this smaller BiH entity, the

reason for centralization of power on the entity and cantonal level is not the unfinished

decentralization as in the case of FBiH. This centralization is a deliberate policy of

strengthening the “statehood” of Republika Srpska by concentrating power at the entity level.

The prevalence of the ethnic criterion and national exclusivity were decisive in shaping the

politico-administrative structure of RS, which has profound consequences on the public sector

functioning in that entity.

19

CHAPTER 2: FUNCTIONS OF LOCAL GOVERNMENTS IN FEDERATION OF BOSNIA AND HERZEGOVINA

The assignment of functions in countries with complex internal structure, consisting of

several tiers of government, is extensively discussed in the theoretical literature. Canons of

fiscal federalism do not leave much uncertainty about the proper aligning of the expenditures

between the central governments and subnational governments, regardless whether political

arrangement in the country is of unitary or federal nature. The casting of the three “branches”,

allocation, stabilization and distribution between has been already discussed at the

introductory part. Here, I will add few important considerations not mentioned previously.

Because of openness and small size of the local economy, efforts aimed at boosting

local economy can actually end in some other jurisdictions. Local governments might simply

have too small capacity to absorb new investments. Another important aspect is the tendency

of local governments to “free ride” on the publicly provided goods with spillovers from other

jurisdictions, and to engage in strategic games. Debt and borrowing policy of the central

government and subnational governments is often in the focus of intergovernmental relations.

The imprudent and narrow-sighted borrowing policy of the local governments can lead into

destabilization of economy, if followed by the bail-out policy of central government.

However, central government very often shifts the burden of deficit to the local governments

20

by assigning them competencies which are not followed by appropriate revenues. The

redistribution function on the local level is possible only if there is limited mobility of the

poor, the rich and businesses between jurisdictions. The redistributive role of the local

governments is becoming increasingly more important with decentralization of the provision

of services.

The most important criterion in assigning the expenditure functions to the local

governments is efficiency. This criterion requires that government assigns the functions to the

lowest level of government, in which the benefit area of the service, decision-makers on local

goods and jurisdictions of tax payers’ are the same (McLure and Vazquez). What criteria

should be used in determining the optimal size of the jurisdictions? Oates (1972) for example

suggests the following criteria: individual preferences; economies of scale; spillovers or

externalities; congestion costs and information and decision costs. In a decentralized setting

with heterogeneous preferences, local government can reduce welfare losses occurring when

the provision of public goods is centralized. Although it would be optimal from the point of

view of fiscal federalism to have as many local authorities as the number of public services

provided, the costs of participating in the decision making process would be too high.

Consequently, there is a good argument reducing the number of local authorities to the

acceptable level. Economies of scale in providing some public goods can also have impact on

the size of the local jurisdictions. Lower cost per unit of bigger jurisdiction can bring lower

prices to the consumer and therefore increase the consumer’s welfare. Provision of goods by

the local governments can spill-over the boundaries causing underprovision of services and

free rider behavior of the beneficiary jurisdiction. There are three possible solutions to the

problem: compensatory subsidies, expanding the jurisdiction boundaries to match the real

scope of provision of services or exclusion of free-riders from accessing the services.

21

Economists do have a key-role to play in describing and evaluating the economic

criteria, but the weights attached to each of those criteria in the final addition are prominently

those of politics (Dafllon, forthcoming). After all, the success of decentralization can be

measured by the extent it fulfills the tasks set by the government. But very often the existing

political tiers of government are not the “optimal” ones according to the efficiency criteria.

Here, economic arguments are not central to the debate, because the debate is about

distribution of power between existing political tiers.

Those de facto existing political tiers and centrifugal forces which emerged during the

conflict in BiH, have set the limits within which new function assignment scheme could have

been developed. The three nations had different views about the functions of which should be

assigned to the different tiers of government. The result was the state with extremely weak

central governments, and asymmetric decentralization of two entitles of which it consists.

Table 2: Assignment of expenditures in Bosnia and Herzegovina

Expenditure Assignment

State

EntitiesFederation of BiH Republika Srpska

Federation Canton Municipalities Republic Municipalities

Immigration, Refugees and AsylumPoliciesImmigration,Monetary policy?

Immigration refugeesand asylum policiesImmigrationMonetary Policy

Refugees and displaces persons

Justice, International Seciruty and Police

Internationl and Inter-entity Criminal Law enforcment

Internal Affairs/Justice and Police. Transfer of Police to State level pending successful negotiations

Police

Internal Affairs/Justice and Police. Transfer of Police to State level pending successful negotiations

Local Police

22

Communications

State Regulatory Agency

Inter-entity regulations

Regional TV and radio

Intra-entity regulation

Public Transportation

Regulates National Trans. System and Inter-entity

Intra-entityRegulates intra-canton

Intra-entity

Socila Safety Net and Welfare

Refugee Programs

Pension System

Social Assistance-joint

Social Assistance joint

Pension SystemSocial Assistance

Energy

State Tranmission Regulation

Intra-Entity Regulation

Intra-entity regulation

Roads Highway Roads

Regional Roads

Local RoadsHighway Roads

Local roads

DefenseReformed to State level

Education

Universities

Teachers salaries

Secondary and primary education

Pre-school education

Primary schools partially maintenance

School buses

Universities

Teachers salaries

School buildings

Pre-school

Primary school partially maintenance

Health Care

Federal Health Insurance Fund

Primary and Secondary Health Care

Health Insurance Funds

Ambulance services

Primary and secondary Health Care

Republic health Insurance Fund

Ambulance services

Sanitation 100% 100%

Water and Pubic Utilites 100% 100%

Housing and Spatial Plannin

Housing Policy

City planning Housing policy City planning

Sweage Treatmen 100% 100%

Culture, Museums, Culture, sport, Museums Culture, sports,

23

Sport, Recreation, Parks, Street Lighting

Theatersparks, street lighting

Theatersparks, street lighting

Fire protection

Joint responsibilities

Joint responsibilities

Joint Responsibilities

Joint Responsibilities

Libraries 100% 100%

Environmental and use of natural resources

Concurrent responsibility Concurrent responsibility

State Government had very few expenditure functions assigned to it, and it was

without any tax handle at its disposal before introduction of VAT. All functions which are

not explicitly assigned to the state level belong to the entities. Until the recent reforms of the

defense system, when the competencies were shifted to the state level, the central government

practically did not have any major spending responsibilities. This is reflected in the share of

the subnational governments in the overall public expenditure.

Table3: BiH Consolidated Budget 2000, estimate. Source: World BankRevenues Expenditures

Million KM

% of Total

% of Entity

Million KM

% of Total

% of Entity

Total 4695 100 5233 100State 25 0.5 69 1.3Federation, o/w:

3452 73.5 100 3713 71.0 100

Federal 892 19.0 25.9 937 17.9 25.2Cantons 1174 25.0 34.0 1272 24.3 34.2Municipalities 231 4.9 6.7 294 5.6 7.9Public Funds 1154 24.6 33.4 1211 23.1 32.6Repubika Srpska o/w:

1199 25.5 100.0 1428 27.3 100.0

Central Government

784 16.7 65.3 862 16.5 60.4

Municipalities 107 2.3 8.9 115 2.2 8.0Public Funds 309 6.6 25.7 451 8.6 31.6Brcko 20 0.4 22 0.4

24

The decentralization in FBiH has stopped at the cantonal level. Only 7.9% of the entity

aggregate budget is executed on the municipal level in Federation. However, if we exclude

budgets of municipalities and measure federal and cantonal budgets as a share of the

consolidated budget, one might conclude that Bosnia and Herzegovina is one of the most

decentralized countries in Europe. If we for example take only the Federation and compare it

with other federal countries, the share of municipal expenditure still remains to be extremely

low.

Figure 2:

FBiH comparative share of municipalities FBiH Comparative intermediate Gov. federation countries. Source: World Bank Expenditure. Source World Bank

Sw

itzer

land

Un

ited

Sta

ted

Ca

nada

Ger

man

yS

outh

Afr

ica

Spa

in

Bra

zil

FB

iHA

ustr

alia

Mex

ico

0

5

10

15

20

25 FB

iHIn

dia

Aus

tral

iaC

anad

aA

rgen

tina

Sw

itzer

land

Bra

zil

Ger

man

yU

nite

d S

tate

sS

outh

Afri

caS

pain

Mex

ico

0

10

20

30

40

50

60

70

In the Federation of BiH, all functions not directly assigned to the entity level are the

competencies of the cantons. This level of authority had a broad range of competencies

assigned to it, which reflects Bosnian ethnic post-war reality. In designing cantonal borders,

the notion of subsidiarity from the European Charter on Local Self-Governance was one of

the elements used. But the element with the most weight attached to it certainly was the ethnic

criterion. Similar to this, assignment of functions to the cantons also followed the pattern of

25

preserving the resources on territory in the hands of the ethnic group. Other factors such as

the socio-economic cleavages in the form of the need for distinction of mother language,

religious factors (although this is overlapping with the ethnic factors) and the still fresh

memory of the unsuccessful former Yugoslav redistribution mechanisms, have all had an

appealing effect in assigning functions the cantons. The absence of any horizontal equalizing

policy can be attributed to the bad memory of the former state fiscal system. The functions

assigned to the authorities can be divided into exclusive, concurrent and overlapping. We will

use this division to examine how functions are assigned in Federation of BiH.

2.1 Functions exclusively assigned to Federation level

The Federation of BiH entity in fact was created as a typical federal state, with checks

and balances mechanisms and some elements of the consociational federalism. Until recently,

it had exclusive competencies over the defense and fiscal policy (including tax administration

and tax policy), but with recent establishing of the state ministry of defense and introduction

of VAT on the state level, it has lost part of its competencies. Still, the Federal government

has retained exclusive competencies over the economic policies, production and distribution

of energy, highways and railways. Extra budgetary funds such as Pension Fund and Road

Fund are constituted on federal level. Categorical social assistance for war veterans, disabled

and their families is also part of the federal government functions.

26

2.2 Functions assigned to canton and functions delegated to municipalities

Cantons have a wide range of functions assigned to them. It includes health care,

education policies implementation, cultural affairs, housing, social assistance, tourism and

local businesses. Cantons can delegate some of the functions to municipalities, and in case

that a municipality has ethnic majority other than that of the canton, it is constitutionally

mandated to do so. Each canton has adopted a Law on Local Self-Governance, which further

regulates municipal responsibilities and sharing of competencies between these two levels.

Figure 3: Average cantonal expenditure by functional category (2002-2003). Source: USAID GAP

41%

22%

13%

11%

9%

2% 2% 0%

Education

Public Order and Security

General Public Services

Social Welfare

other

Housing

Culture

Defense

The broader definition of the subsidiarity principle was used in assigning the

expenditure responsibilities to cantons. Too strong a federal government would not have

enabled for the specific ethnic interest to be realized, and it bore a negative message from the

past socialist redistributive policies. In this sense, the border of the cantons and new

municipalities which emerged has been not taking into account economies of scale in

providing services and spillover effect. In most cases, the boundaries between cantons were

drawn to accommodate the ethnic criteria of keeping the control over the resources and

providing the services to the “proper” ethnicity.

27

Health care funds exist on the cantonal level, though the spillover effect is

significantly present increasing the tax burden of the local tax payers. This was partially offset

by establishing the Federal Health Care Solidarity Fund, which has the function of pooling the

risk of the cantonal funds and introduces the economy of scale in the health care system of the

Federation.

Though it has been constituently guaranteed that cantons can shift responsibilities to

the federal level, this has not occurred so far. In the area of education, cantons have retained

all the competencies: educational policy in general, academic curricula and research, and

teacher’s salaries, in all three levels of education. Some cantons (Canton 10, Herzegovina and

West Herzegovina) have responsibilities for school maintenance and pupils transportation.

Present here is also the inability of cantons to internalize the spillover effect in

providing the educational services, especially on tertiary level. This has caused

underprovision of services and misallocation of the public funds, because the cantons cannot

restrain students from other jurisdictions in capturing part of the benefit. The solution would

be to shift the financing responsibility to the upper level, either federal or the state one.

Social protection is also responsibility of the cantons, though federal government

provides “categorical” social assistance and unemployment benefit. The delivery of the

service is delegated to the municipal level, while financing is provided by cantons. This

fragmented system of social protection caused huge disparities in the level of social

assistance. Better-off cantons (e.g. Sarajevo) with higher taxable base and revenue raising

capacity provide their citizens with significantly higher social assistance (for orphans, elderly

and displaced persons and refugees) than the worst of cantons (Cantons 10, Gorazde or

Posavina). As a result of the unfunded mandate, the arrears of unpaid entitlements in those

cantons have been already accumulating for years. Though mobility of population is relatively

28

low in BiH, cantons can not provide the redistributive effect which Federation would have as

a bigger territorial unit. Therefore, as in the case of education, financing of the social

assistance should be done by the federal level, but actual delivery of services needs to be

retained in municipalities.

2.3 Concurrent responsibilities

The Federal Constitution provides that the Federation and cantons can exercise joint

functions in certain areas. The Federal government has concurrent responsibilities with

cantons and municipalities in the health policy implementation, education, internal affairs,

environmental protection and social protection, but all functions not explicitly assigned to the

federation are the responsibility of cantons.

2.4 Functions assigned exclusively to municipalities

As it is provided by the Constitutions and Laws on Local Self-Government,

municipalities have exclusive responsibilities in urban planning, provision of local utilities

(delivered by the off-budget municipal enterprises), pre-school education, students

transportation, libraries, parks and museums and other responsibilities delegated by cantons.

Regardless of legal provisions, municipalities in FBiH have been severely restricted as

decision-makers in any of these fields, and it is even more pronounced by the weak own-

revenues assigned to the municipalities. The municipal level not only in the FBiH, but in the

whole country as well, has been seized by the post-war political setting, in which the

prevailing criteria were of ethnic character, not the economic efficiency or citizen’s welfare.

29

30

CHAPTER 3: REVENUE SOURCES

3.1 Reasons for decentralizing revenues.

The question of decentralization of the revenue assignment is much less clear-cut than

for the expenditures. One of the arguments mostly used is that of accountability, which means

that subnational governments will be more responsible to their constituency. This will result

with more efficient and prudent use of the locally raised money (Boadway and Shah,

forthcoming). The central government can use its assigned fiscal weight to affect the

subnational governments spending priorities. Another important consideration is the

uncertainty which local governments face when the central government decides unilaterally to

change the funding of subnational governments. This will affect the predictability of the

revenue stream for the SNU and their long term budgeting capabilities.

On the other side, decentralization of the taxing powers can seriously compromise the

efficiency and the economic unity of the country. Decentralization can distort allocation of

labor, capital and businesses, and it can lead to wasteful tax competition between subnational

governments. Different redistribution standards between the states can significantly

jeopardize the national equity objectives. Administrative costs of tax collection can also

increase because of reporting to several layers of authorities. Modern federations assume great

spending responsibilities, for which they need broad-based tax handle. If the states are left

with narrow tax base, they might be tempted to use it too excessively to the detriment of the

whole system.

31

Proper assignment of taxes needs to follow certain rules, if it is to be successful. In

modern federations, governments employ a mix of broad-based taxes, that spread the burden

of tax raising across large number of taxes, which keep any tax rate low. In the federal

economy, there is also a question of how those taxes should be allocated among the levels of

governments. If, for example, the constituent land is given limited tax sources, then the tax

rate in that state might be set too high from a point of view of a desired tax mix .

There are several principles which need to be observed when deciding on the

assignment of tax to lower governments (Boadway and Shah, forthcoming):

- Efficiency of the common internal market may be jeopardized by the different taxing

of the same transaction in constituent states. This can lead to distortion of business decision,

which would be based only on tax differences. Firm operating in more than one jurisdiction

faces increased administrative costs when working with two different tax structures, and

taxation of mobile factors.

- National equity is usually understood as an intention to treat all citizens equally for the

redistributive purposes. This is often seen as the task of the central government, as only the

central government can ensure that citizens in all regions are treated equally. But the taste of

the states can differ in that regard. If the redistribution is viewed as the central government,

then the instruments suitable for redistributing, such as income and wealth taxes should be

under the central government control.

- Administrative and Compliance Costs can be reduced by joint collection of state and

federal taxes, both on the sides of the government and taxpayers. The tax system should be as

simple as it can be, compatible with achieving other goals of the system.

32

- Fiscal and Political Accountability principle states that fiscal accountability is greater,

more jurisdictions are required to finance their expenditures. Revenues sources should be

matched with the expenditure needs as closely as possible, as the accountability is fostered by

the close scrutiny of the electorate.

33

These principles have been significantly compromised in case of revenue assignment

in Bosnia and Herzegovina. The Central Government (CoM) practically did not have any

taxation power, until the recent introduction of the VAT on the state level. In the peace

agreement, CoM was given only some minor fees (such as passport and diplomatic), without

any own source revenues. Financing of the State expenditures, until the VAT introduction,

was done by transfers from the entity level. FBiH contributed 2/3 to the State budget, while

Republika Srpska’s contribution was 1/3. Asymmetric and decentralized solution for revenue

assignments left the CoM without any real tax leverage. This had as a consequence huge

disparities in the revenues of the subnational governments, most notably between cantons and

between municipalities. The Council of Ministers or the Federal government did not have any

active policy aimed at decreasing those disparities, before the reform of the taxation system.

3.2 Situation before the introduction of VAT

34

Prior to the introduction of the VAT, a revenue sharing in FBiH was regulated by the

Law on Allocation of Public Revenues. Federation Entity was exclusively assigned custom

and excise revenues, and corporate income tax revenues from the companies owned by the

Federal government. The customs policy is harmonized across entities by the CoM. However ,

customs entirely accrue to the entity governments. The importance of customs is declining

however, due to the free-trade agreement with neighboring countries. Corporate income tax

is assigned both to the Federation and to the cantons, but tax rate and tax base is regulated by

the Federal government. Each canton, with the exception of Sarajevo, has passed a Law on

Revenues Sharing which regulates sharing of revenues for respective canton and its

municipalities. The most important shared tax was Sales tax, a general turnover tax, which is

assigned 100% to cantons, and shared with the respective municipalities. Personal income

tax (PIT), in a form of the wage tax also accrued to the cantons, but the rate and the tax base

are determined by the Federal government. It is again based on the derivation principle,

ensuring that the tax stays in the canton/municipality of origin, or better to say of place of

work. Property and property transfer tax is shared between the cantons and municipalities,

with 80% going to the canton and 20% to the municipality in most of the cantons. The tax is

regulated by cantons, but the potential of this tax is still not fully understood by the

subnational authorities. Appendix A contains sharing ratios for all taxes before introduction of

the VAT.

Figure 3: Structure of cantonal revenues for 2003. Source: GAP

35

Municipal (or local) taxes are the taxes over which the lowest tier of governments

has the right to regulate tax base and /or a tax rate. Local taxes have not been used adequately

to finance municipal expenditures or to re-allocate resources on the local level. Municipalities

prefer to rely on the shared tax, which does not allow for development of a genuine local tax

effort. Very few of these taxes are regulated exclusively by cities/municipalities, except for

the user charges which are collected and appropriated by the off-budget companies. On the

one side, these charges and taxes on the level are extremely fragmented, and on the other the

potential of the property tax as the most buoyant and stable local tax is still underutilized not

only in the FBiH, but in the whole BiH as well (GAP 2005).

Figure 4: The Structure of Municipal Revenues by Canton. Source: GAP

36

In 2003, municipal revenues from all sources amounted to 268 million of KM of all public

spending in FBiH. More importantly, municipal revenues in 2003 represented only 3.2% of

FBIH’s GDP, which was 8.340 million KM (see figure 2 in Chapter II for comparative world

overview). This all suggests that municipalities in the Federation are under-funded, and that

system is in serious vertical imbalance (GAP 2005).

Horizontal fiscal imbalances are also huge in the Federation. There are serious

disparities in per capita revenues between municipalities and cantons. This does not come as a

surprise, having in mind that that share of the taxes assigned to municipalities differs

significantly from canton to canton. Fiscal capacity of the cantons varies, which compounded

with the derivation based taxation further exacerbates these disparities. None of the cantons

(perhaps with exception of Canton Tuzla) has introduced the mechanism for fiscal

equalization. The magnitude of disparities between the cantons is illustrated in the next chart.

By far the richest canton (Sarajevo) has three time bigger per capita revenue than the poorest

canton (Central Bosnia).

Figure 5: Total per capita revenues of cantons in 2003. Source-GAP

7%

14%

34%

4%4%

21%

13%

3%

4%1%

23%

1%

30%

24%

8%

8%

7%

13%

29%

4%6%

28%

11%

2%

9%

14%

25%

3%

9%

28%

12%

0%

12%

5%

32%

5%5%

13%

27%

2%

8%

10%

24%

4%

6%

35%

9%

4%

16%

17%

11%

1%

26%

23%

4%2%

11%

2%

40%

3%

33%

0%0%

11%

0%

26%

0%2%

23%

14%

34%

1%

7%2%

41%

1%

23%

19%

4%2%

8%

15%

21%

3%

16%

22%

14%

2%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9 10 ALL

other

grants

fees & charges

asset income

citizens taxes

sales taxes

property taxes

wage taxes

37

408 489385 396

497359

478 513

1,168

339

113

210

148 11492

90

255 191

173

157

0

200

400

600

800

1,000

1,200

1,400

1,600

1 2 3 4 5 6 7 8 9 10

total cantons total m unicipalities

Even in the case of Tuzla Canton, where the five poorest municipalities were assigned larger

share of revenues, the disparities still remain to be large. Disparities caused by the poor fiscal

capacity of the municipalities cannot be eliminated unless there is a proactive equalization

policy. But the Federation has no available tax revenues that could be used for this purpose.

Figure 6: Estimates of Disparities in Per Capita Municipal Revenues: Canton 2, 3 & 5. Source: GAP

25 28 51 44 55 51 55 50 75 65 70132

287

77 64 78 67 72 5319 15

4913 2120 21 30 25 46 69 38 61 57

64

151

48 44

131194

105

1714 37

187 4

11 28 17 2815 4 19 15 21

1

46

12 31

19

93

32

11151

284

31

050

100150

200250300

350400

450500

Other

Non-Tax

Tax

38

3.3 Tax System Reform-A Policy Window for Fiscal Equalization

In 2003, CoM has initiated a major reform of the taxation system, which was one of

the requirements for accomplishing the tasks from the European Union Feasibility Study for

Bosnia and Herzegovina. New Agency-the Indirect Taxation Authority (ITA) was established

on the state level, with the competences of collecting all indirect taxes in both entities and

Brcko District. A framework Law on System of Indirect Taxation (LSIT) and Law on Value

Added Tax (VAT) were passed, that were harmonized with the EU Sixth Directive. Fiscal

Council-a new body for intergovernmental coordination of fiscal and budgetary police was

established in 2005, members of which are the Entity Ministers of Finance, State Ministers

and their deputies, as well as Entity and State Prime Ministers.

In 2005, ITA started with collecting indirect taxes (the sales tax, customs, excise and

road fee) to the Single Account, in accordance with the Law on Single Account. Starting from

January 2006, the sales tax was replaced with 17 percent of VAT, and all indirect taxes in the

country now flow in to the Single Account.

Figure 7: New Structure of Indirect Taxation System in Bosnia and Herzegovina

Indirect Taxation Authority (ITA)

SINGLE ACCOUNT

ITA Steereing Board

Fiscal Council of BiH

Council of Ministers

FBiH Budget Republika Srpska Budget

Brcko DistrictBudget

39

LSIT does not oblige ITA to trace the origin of the revenues below the entity level.

Therefore, as of 1 January 2006, the derivation principle can no longer be applied for

allocation of revenues in the FBiH. The state has a priority in being financed from the Single

Account revenues, but there is a need for developing a mechanism for allocation of what is

left to the sub-entity governments. These revenues represent a huge portion of the sub-entity

income, amounting up to 70% of the cantonal income and 21% of municipal.

The sharing ratio between the entities and the Brcko District is defined in the LSIT,

but it has been the subject of much of political debate recently. It depends on the number of

the processed VAT invoices from entities, a procedure which the smaller BiH entity sees as

biased and unjust, as according to them, invoices from the FBiH are given priority. The

amount to be distributed to the entities can vary in the future, depending on this highly

politicized decision. Consequently, the federation, cantons and municipalities could face

uncertainties in the amount of income from the Single Account. This is partially offset by the

improved tax collection and fiscal compliance, which is inherent to the VAT. Still, there is an

urgent need to reduce the presence of politics and resolve this issue on the expert level.

ITA has justified its existence by collecting, 1.55 bill. KM of indirect taxes and

customs in the first five months of 2006, which is comparably much higher than the amount

of 1.39 bill KM collected by the three separate tax and custom authorities. The plan for this

year is to collect 3.5 bill. KM of indirect taxes and customs, which is realistic, having in mind

that the tax rate of 17% has been set higher than it would be the neutral tax rate replacing the

sales tax, and the fact that the Law on VAT does not envisage exemption or zero rate, except

for exports.

40

CHAPTER 4: NEW INTERGOVERNMENTAL TRANSFER SYSTEM IN FEDERATION OF BOSNIA AND HERZEGOVINA

4.1 Theoretical Framework

As it was already discussed in chapter II, vertical fiscal imbalances FBiH are

significant. Absence of any kind of information about the municipal expenditures and very

few available for the cantons, make the task of quantifying the imbalance very difficult. Still,

it can be said that the municipalities are uderfunded, while the most buoyant taxes are

assigned to the cantonal level. Horizontal disparities between cantons and between

municipalities also pose a serious obstacle to the economic convergence and creation of a

single economic space in BiH.

41

The intergovernmental transfer system can be introduced for four good reasons (Bahl

1999). First is the question of vertical imbalances, when the subnational governments are

allocated responsibilities in providing services, but the most buoyant revenues are kept at the

central level. There are two solutions to this problem: either sub-national governments are

given more revenue-raising powers, or there are revenues transfers to fill the gap. The

question of precise measuring of the vertical imbalance is not easy to resolve, especially in the

conditions of constrained data availability. New transfer system in FBiH has taken historic

(year 2005) information on cantonal and municipal revenues as starting point, trying to

increase their share in the public revenues through new transfer system. Second, transfer

system can be used for fiscal equalization between the sub-national jurisdictions. Horizontal

inequalities are natural occurrence, and they reflect the diversity of the resources and

economic potential of the subnational governments units (SGU). Mobility of the people and

capital from less productive to more productive regions increase efficiency, but the fiscal

system should be designed to have minimal impact on the allocation of factor of production.

Third, equalization policy of the government can also be implemented by assigning the own-

source revenues to the SGU, but more typically it is done through transfer system. Fourth,

externalities occur when the SGU decided to underspend on the service with significant

spillover effect. Transfers can be used to correct these anomalies, and at the same time central

government can influence spending patterns of the SNU.

There are following types of intergovernmental transfers (Bird and Tarasov 2002):

- Revenues sharing. Under this arrangement, SNU can keep certain percentage of the

tax collected within their territories. The revenue is shared based on its “derivation”, and fixed

percentage is returned to the jurisdiction of origin, or it can be divided strictly of per capita

basis.

42

- Unconditional (general purpose) block grants have no strings attached to it. Because

they are unconditional, SGU have substantial policy discretion over spending it. Grants used

for equalization purposes are most frequently unconditional grants.

- Matching grants are conditional grants which require a specific contribution by the

SGU in a particular expenditure area. Matching grants can be designed very specifically or for

the whole sector; rate of matching of central government can vary depending on the central

government desire to influence spending in certain area; and it can be open-ended or capped.

43

- Cost reimbursement is a scheme through which central government reimburse local

governments for certain approved costs. These types of transfers are recommended when

central government relies on local government for implementation of certain types of

programs.

- Ad-hoc transfers are the least desirable type of transfers. The distribution of grant is

left to the discretion of the central government or one of its agencies. This approach opens the

door for the political manipulation and buying favors for the grant allocation. It hardly ever

results with the grants allocated on the sound fiscal principles.

Under the shared revenue scheme for grant distribution, each SGU receives a share of

tax colleted within its boundaries. This type of grant allocation scheme was operating on the

cantonal level in FBiH before introduction of VAT. Revenue sharing distribution scheme are

inherently not equalizing, as richer municipalities and cantons were receiving proportional

larger amounts of tax. Another problem with this approach is that it discourages autonomy

and accountability on the local level, as the tax being centrally collected and tax rate and base

centrally determined, local population would not hold local officials responsible for quality of

services provided. This was very much case in FBiH municipalities and cantons. Though

there were not too many centrally determined expenditure mandates for municipalities, local

officials did not use this freedom in spending the revenues to increase accountability and

improve quality of living, except in rare cases. On the other side, revenue sharing is stable and

predictable source of income for SGU. It allows them to plan and forecast future revenues

with certainty, but on the other side, it is more difficult for the central government, as it was

the case in FBiH Government, to influence the fiscal area. FBiH Government did not have tax

capacity to equalize between cantons and municipalities, as the all major taxes were either

assigned or shared with the cantons.

44

Another common approach used in distribution of pool of revenues is formula based

transfers. Using formula for allocating grants between the SGU generally improves the

transparency and reduces discretion in deciding about grant allocation, as all know exact

criteria according to which distributions are made. Flexibility of formula allows for changes

as the needs for public expenditure change. Four main components of the formula are:

elements of the formula; data necessary to calculate the formula; cost associated with

administering the system, and finally conditionality, should be designed to reflect objectives

of the grant program.

45

Grant allocation formula might in the first place reflect difference in expenditure

needs across the SGU. Different regions in the country differ in their cost of providing

standardized basket of public services, because of differences in needs (for example percent

of school aged children or retired), incidence of poverty or unemployment or geographical

conditions. Countries use various indicators of expenditure needs. Lagged expenditure values

are the simplest and in most cases available data for quantifying expenditure needs. Using

historic expenditures obviously has the risk of not incorporating possible changes in

expenditure patterns or inflation over time. Expenditure needs can also be measured bottom-

up, by calculating costs of current expenditure obligations of SGU. However, this approach is

data-heavy and requires well elaborated procedure how to cost expenditure of SGU. Third and

more widely used approach is estimation of some type of weighted index of relative

expenditure need. These indexes try to capture some of the factors which determine the cost

differences in delivering standard basket of services. Factor could be demographics, to

include special needs of young and elderly, or other factors such as illiteracy or

unemployment. One of the problems with the index-based approach is the political pressure

for inclusion (exclusion) or weighting of some of the factors that might be beneficial (or

harmful) for specific SGU. The weights which ought to represent relative contribution of each

factor to the overall measure of need can be easily calculated using some statistical method.

However, these weights are often determined on ad-hoc basis, subject to political pressure.

Formula should include limited number of factors, as inclusion of too many factors reduces

transparency and increases costs of collecting data. Data used in formula should come from a

official agency, and should be updates regularly, every year of two.

46

Revenue capacity of the SGU government can be defined as the potential revenue that

could be raised from the tax base assigned to the SGU, if and average level of effort is applied

to the tax base. Two SGU with the same fiscal capacity may collect different tax amount as

the result of applying different tax rate or because of different enforcement efforts invested in

collecting revenues. Finally, taxpayers compliance, ceteris paribus, also can influence level of

collected tax. Using actually collected revenues might provide SGU with negative incentives,

as the SGU which collects more revenues would receive less of transfers. Different proxies of

revenue capacity have been used in countries around the world. Level of collected revenues or

lagged revenues from the previous year is used very often, but the actual revenues can

provide the SGU with perverse incentive to decrease tax effort. However, in majority of the

transitional and developing countries, where the SGU have limited tax autonomy and bulk of

their come from tax sharing, local fiscal capacity can be safely measured by actual or

forecasted level of shared revenues and transfers (Boex and Martinez-Vazquez 2004).

Another widely used measure for fiscal capacity is Gross Regional Product (GRP) which is

more comprehensive measure than the level of per capita level of personal income, because it

includes incomes generated within regions irrespective of worker or producer residence.

In the absence of clear and unambiguous measures for quantifying expenditure needs

and fiscal capacity, second best approach is to multiple factors as elements in the formula. In

this case, pool of revenues is allocated in proportion to some allocation factors that together

represent a proxy of SGU fiscal capacity or fiscal need. In this approach, a number of factors

are selected, and then each of them is assigned weight, determining hwo imprant is each of

them in final allocation. SNU unit will receive its share in proportion to that factor.

Particularly poor or depressed region might still need additional transfers, as the formula

might not capture all extreme conditions in these SGU. For those SGU, it will be necessary to

include special weights for equalizing their fiscal capacity and expenditure needs.

47

4.2 New Formula Based Allocation Mechanism in FBiH

Government of FBiH has passed a new Law on Allocation of Public Revenues

(Official Gazette 22/06), using the for the first time formula for distribution of revenues from

to sub-national level. Both entities and Brcko district will receive revenues from ITA’s Single

Account according to the processed invoices from their territories, and after the revenues are

transferred to the state budget for financing CoM functions. Of this amount, FBiH will

distribute:

36.2% for financing function of entity level;

51.48% for financing of cantonal functions,

8.42% for financing municipal level functions,

3,9% for Road Fund.

The formula is based on a number of weighted factors which try to capture differences

in expenditure needs and revenue capacity of cantons and municipalities.

Table 4: Weights for cantons and municipalities used in allocation formula

Weights Area Population Elementary School Children

Secondary School Children

Development Index

Cantons 0.06 0.57 0.24 0.13Municipalities 0.05 0.68 0.20 0.07

48

In addition to this, formula contains a weight for special spending needs of Sarajevo

canton, which is 2. Two cantons, Gorazde and Livno, which had the smallest revenue from

the sales tax in derivation-based system, are also given a special weight of 1.8. Formula has a

factor of 1.2 for municipalities with over 60.000 of inhabitants, and 1.5 for the municipalities

covering the costs of elementary schools. Index of development actually measures fiscal

capacity factor of the municipality. It is calculated as the average of the collected sales tax and

income tax for respective municipality for 2005. Weights for the fiscal capacity are in the

range between 1.8 for municipalities with below 20% of the FBiH average collection of those

taxes, to 1.2 for municipalities with below 80% of average. All data used in formula are

collected by Statistical Agency of FBiH, and are the best possible data obtainable at the

moment. It is, however, unclear what method was used in calculation of particular weights

attached to formula factors. At the moment it is not possible to obtain any reliable information

on municipal expenditures which could be used for calculation of weights, either using

regression or simply as historic information.

New Law envisages six years phase-in period for implementation of the system. This

solution is fully understandable having in mind magnitude of current horizontal fiscal

disparities on one side, and on the other level of investments and long-term expenditure in

better-off cantons and municipalities. Phasing-in period combines percentages of sales tax

collected prior to the introduction of new system, and gradual introduction of percentages

according to new formula:

Cantons Municipalities

Year 1 (2006): Cs = (0.9*HCs + 0.1*NCs) Ms = (0.9*HMs + 0.1*NMs)

Year 2 (2007): Cs = (0.7*HCs + 0.3*NCs) Ms = (0.7*MCs + 0.3*MCs)

Year 3 (2008): Cs = (0.5*HCs + 0.5 *NCs) Ms = (0.5*MCs + 0.5 *NMs)

49

Year 4 (2009): Cs = (0.3*HCs + 0.7*NCs) Ms = (0.3*HMs + 0.7*NMs)

Year 5 (2010): Cs = (0.1*HCs + 0.9*NCs) Ms = (0.1*HMs + 0.9*NMs)

Year 6 (2011): Cs = (NCs) Ms = (NMs)

Cs, Ms – Percentage of the share in the pool of indirect taxes in the FBiH for canton and

municipality, respectively;

HCs, HMs – Historical share in the sales tax in FBiH of canton and municipality,

respectively, and;

50

NCs, NMs – Percentage of share the according to new system for cantons and municipalities,

respectively.

The pool of indirect taxes available for distribution in FBiH is determined in

percentage of the indirect taxes, in the first place VAT, collected to the Single Account of

ITA. ITA’s estimation is that in 2005 about 3.5 bill. KM of indirect taxes will be collected3.

State budget forecasts 566 mil of indirect taxes to be transferred for 2006 from Single

Account for financing central government functions and service of foreign debt. This leaves

amount of 2,934 bil. KM to be allocated between FBiH, RS and Brcko District

Distribution of revenues between entities and Brcko District has been subject of much

debate recently, when officials from RS claimed that distribution mechanism is unjust and this

entity was losing millions of KM every month. However, if we take percentages for June

2006, which allocate 66.77% to FBiH, 29.76% to RS and 3.46 to Brcko District, and apply

them to the amount forecasted amount for 2006, then amount available for distribution in

FBiH is 1.95 bil KM. Applying percentages for allocation between FBiH level, cantons and

municipalities, in 2006 cantons will receive 1.008 bil KM and municipal level 164.9 mil KM.

Comparing to 2005 level, municipalities will receive 72% more and cantons 2% more of

indirect taxes.

At the end of phase-in period, municipalities will have at its disposal significant

share of the public revenues in FBiH. This will call for evaluation of the assignment of

functions to municipalities, and possible re-aligning of functions between tiers of

governments in FBiH. Any changes in intergovernmental relations should reflect the role of

municipalities as the closest tier of government to the citizens, but also potential threat of low

capacity on the local level to implement newly assigned functions.

3 ITA press statement from June 5, 2006.

51

3,95

4.3 How good new system performs?

To test how well new formula based transfer system does the job, we have run simple

LSE regressions to determine the correlations between the dependent variable - amount of

transfer per capita for cantons and municipalities, and several indicators revenue capacity and

expenditure needs – independent variables. As indicator of revenues capacity, we have

selected GDP per capita and PIT per capita of cantons and municipalities, and as indicator of

expenditure needs, we have used population. Sarajevo municipalities were given as one sum,

and we have reduced number of observations from 79 to 71, taking all Sarajevo municipalities

as one. The results of regression are given below, while Appendix contains calculated per

capita amounts of transfers to cantons and municipalities for 2006 and 2007 (t-statistics is

given in parenthesis):

Cantons:

Transfer2006 = 44.11 + 0.118*GDPpc

Transfer2007 = 96.52 + 0,102*GDPpc

Municipalities:

Transfer2006 = 32.67 + 0.575*PITpc

Transfer2007 = 43.78 + 0.423*PITpc

We have not been able to find any statistically significant correlation between per

capita transfer and population at either 5% or 10% of significance.

We have also tested how system will function at the end of phase-in period. We have

for that purpose calculated per capita amounts in 2011, and correlated them with GDpc, PITpc

and population. Here are the results:

(3.42)

(3.95)

3,74

(1.65)

52

Cantons:

Transfer2011 = 267.3 8+ 0.053*GDPpc

Municipalities:

Transfer2001 = 82.98 – 0.107*PITpc

Again, coefficients for population were not significant at 10%, but what can be

concluded for municipalities is that system in the final year achieves horizontal equalization.

Transfer per capita have negative correlation with the revenues capacity indicator-PIT, which

shows that, at the end of transition period, formula will provide municipalities will bigger

revenue raising capacity will receive smaller per capita amount of transfer. This trend is

visible throughout six year period for municipalities, but formula for cantons even after six

years period still does not equalize based on revenues capacity.

(-1.23)

(3.82)

53

CONCLUSION

New system for allocation of public revenues in FBiH has marked a milestone in the

intergovernmental relations in FBiH. After ten years of stalemate, without any major reform

processes of fiscal system, newly introduced formula based mechanisms has finally changed

the relations between the municipalities, cantons and FBiH level. The reform was more than

necessary having in mind that post-Dayton relations between the tiers of government in FBiH

have not been adapted to the changing socio-economic situations in the country.

The disparities between the cantons, but also on the intra-cantonal level, without any

active policy for equalization pursued by the FBiH government or cantons, have threatened to

compromise social equity of the FBiH citizens and economic convergence of FBiH. But, this

was a deliberate policy or “system error” of the Dayton: On the one side, FBiH Government

did not have sufficient fiscal power nor was there a political will to equalize between

obviously growing disparities, but on the other cantons were by Dayton design put in a

position of fiscally strongest. This was not by chance: in the fiscal sphere, the primary task for

cantons was to retain ethnic control over the financial resources. Until recent introduction of

VAT, State of BiH has been completely deprived of any role in fiscal sector.

54

New system for allocation of public revenues incorporates both expenditure needs and

fiscal capacity indicators. For the first time, derivation based system will be replaced with

unbiased and simple formula, which will enable municipalities and cantons for more stable

budgeting, and hopefully better local governance. The formula uses simple and verifiable

data for calculation of transfers and it allows for easy modifications, which almost certainly

will take place in the following years. Formula captures the expenditure needs differences

across the municipalities and cantons, as well as fiscal capacity differences – in the last year

of transition period, transfers per capita vary negatively with PIT per capita indicator on

municipal level. This is the indicator that formula is performing well in correcting horizontal

imbalances.

55

Though the formula represents somewhat an enigma for cantonal and municipal

authorities, at the end of six years phase-in period, but gradually even before, poorer local

governments in FBiH will have at their disposal significantly bigger revenues than comparing

to the present system. This is partially helped by the increased compliance and tax collection,

inherent to VAT. In that sense, it will be necessary to accompany present reforms of fiscal

sector with parallel projects of capacity building on the local level. The first and foremost task

will be to improve financial management in municipalities and cantons, train local officials

project preparation and management, improve their accountability by opening channel for

communication with citizens and for citizens active role in local government affairs, and to

introduce program and performance budgeting and multi-year budgeting. The increased flow

of revenues for local governments might provide incentive for realigning of the

intergovernmental relations in FBIH, by assigning additional responsibilities to

municipalities. If this happens, and this is purely political issue, it will be necessary to direct

reform so that municipalities are not assigned only mandated functions, but that there should

be genuine devolution of power to the local level. Of special importance will be regulation of

how municipalities and cantons access capital markets. The conditions are already created

which enable some municipalities to borrow even now, and this will become even more

realistic as the situation with local finances improves. This requires that new law is adopted in

the course of the fiscal reform, allowing local governments access to capital markets, but at

the same time issuing upper tiers of government from bailing-out municipalities and cantons.

56

If the role of local governments becomes more important, there will be a need for

central data collection, monitoring and analysis of local governance performance. New

ministry, either on entity level, or if political reality permits on the state level, will need to be

established to coordinate efforts in this area and to harmonize different policy solution in

entities. It is not, of course, possible to predict future political developments in BiH, but we

believe that in the same way it was uncompressible that there would be some model of fiscal

equalization introduced in FBiH, in the same way we hope that at some point in the near

future there will be a fiscal equalization scheme for whole BiH. Only in this way growing

disparities between two BiH entities will be possible to eliminate, and through that process

additionally strengthen integrative forces in the society. We hope that politicians, decision-

makers but common people as well have learnt a lesson in the case of FBiH, and that we will

not have to wait for a long time for this to realize. This study will then need to be expanded to

include results for FBiH after the system has been in function for some time, but also to

analyze possible new BiH-wide fiscal equalization scheme.

57

Appendix A: Shared tax between cantons and municipalities before VAT introduction

58

Appendix B: Cantons

Distribution of transfer for 2006. Distribution of transfer for 2007.

Cantons % of FBiHDistribution Pool

Per capita amount ofnew transfer, in KM

Una-Sana 9.499 331.55Posavina 1.84 447.83Tuzla 17.095 346.97Zenica 12.538 315.013Gorazde 1.714 500.92Central Bosnia 8.017 315.613Neretva-Herzegovina 8.90 392.55Wesz Herzegovina 3.426 413.44Sarajevo 34.768 855.15Livno 2.203 269.62

Distribution of transfers for 2011: New formula applied 100%. Independent variables used for regression.

Cantons % of FBiHDistribution Pool

Per capita amount ofnew transfer, in KM

Una-Sana 10.733 374.62Posavina 1.404 341.68Tuzla 17.782 360.91Zenica 14.627 367.50Gorazde 2.122 620.10Central Bosnia 9.603 378.04Neretva-Herzegovina 8.525 376.03West Herzegovina 3.205 386.74Sarajevo 28.110 691.39Livno 3.890 476.10

Canton GDP per capita, in KM PopulationUna-Sana 2120.8 288,935Posavina 2862.4 41,436

Cantons% of FBiH

Distribution PoolPer capita amount ofnew transfer, in KM

Una-Sana 9.799 342.03Posavina 1.743 424.23

Tuzla 17.247 350.10Zenica 12.972 325.92

Gorazde 1.830 534.82Central Bosnia 8.37 329.52

Herzegovina-Neretva 8.821 389.00West Herzegovina 3.385 408.50

Sarajevo 33.275 818,43Livno 2.578 315.51

59

Tuzla 2749 496,885Zenica 3019.9 401,401Gorazde 2430 34,508Central Bosnia 2851 256,175Neretva-Herzegovina 4376 228,649West Herzegovina 3011.2 83,570Sarajevo 5957.8 410,031Livno 2411.6 82,402

60

Appendix C: MunicipalitiesDistribution of transfers for 2006, 2007 and 2011

Municipality

% of FBiH Distribution Pool for2006

Per capitaamount of transfer for2006, in KM

% of FBiH Distribution Pool for2007

Per capitaamount of transfer2007, in KM

% of FBiH Distribution Pool for20011

Per capitaamount of transfer 20011, in KM

Bihac 4.922 133.37 4.432 120.09 2.723 73.76Bosanska Krupa 0.934 54.90 1.015 59.66 1.295 76.10Bosanski Petrovac 0.259 53.27 0.303 62.32 0.456 93.73Buzim 0.423 39.65 0.509 47.72 0.812 76.01Cazin 2.334 62.43 2.481 66.36 2.989 79.96Kljuc 0.495 40.89 0.572 47.26 0.841 69.50Sanski Most 1.52 54.51 1.613 57.84 1.932 69.29Velika Kladusa 1.473 51.98 1.596 56.32 2.026 71.49Domaljevac-Samac 0.261 95.99 0.243 89.37 0.183 67.23Odzak 0.682 69.60 0.674 68.78 0.647 66.00Orasje 1.212 96.17 1.109 88.00 0.750 59.50Banovici 0.996 64.38 1.012 65.41 1.067 68.95Celic 0.636 72.85 0.628 71.94 0.601 68.82Doboj-Istok 0.727 118.78 0.66 107.83 0.426 69.62Gracanica 2.269 72.74 2.24 71.82 2.148 68.88Gradacac 1.825 65.19 1.842 65.80 1.891 67.55Kalesija 0.789 37.28 0.956 45.17 1.533 72.43Kladanj 0.541 58.51 0.576 62.29 0.697 75.33Lukavac 1.731 55.39 1.823 58.33 2.151 68.83Sapna 0.468 55.40 0.504 59.66 0.623 73.73Srebrenik 1.476 59.49 1.536 61.90 1.738 70.02Teocak 0.612 136.53 0.547 122.03 0.318 71.00Tuzla 9.848 123.65 8.834 110.92 5.298 66.53Zivinice 2.013 61.61 2.093 64.06 2.379 72.82Breza 0.263 29.30 0.342 38.11 0.617 68.74Doboj-Jug 0.187 69.91 0.181 67.67 0.162 60.74Kakanj 1.222 46.61 1.344 51.26 1.842 70.28Maglaj 0.444 31.10 0.566 39.64 0.987 69.12Olovo 0.225 28.94 0.312 40.13 0.619 79.60Tesanj 1.463 50.79 1.59 55.20 2.032 70.53Usora 0.192 44.76 0.211 49.19 0.282 65.64Vares 0.202 29.48 0.273 39.85 0.529 77.24Visoko 1.263 52.15 1.313 54.22 1.487 61.40Zavidovici 0.71 30.74 0.938 40.61 1.755 76.00Zenica 4.013 51.86 4.304 55.62 5.324 68.80Zepce 1.03 54.84 1.105 58.83 1.364 72.65Foca 0.034 30.35 0.051 45.52 0.108 96.61Gorazde 0.979 51.16 1.043 54.51 1.268 66.26Pale 0.019 28.57 0.028 42.10 0.062 93.42Bugojno 0.733 32.30 0.921 40.58 1.578 69.53Busovaca 0.328 33.80 0.412 42.46 0.707 72.90Dobretici 0.009 22.43 0.015 37.38 0.033 83.34Donji Vakuf 0.348 40.60 0.419 48.88 0.666 77.74Fojnica 0.274 36.55 0.338 45.09 0.565 75.34Gonji Vakuf 0.377 31.96 0.494 41.88 0.902 76.50Jajce 0.545 36.92 0.66 44.71 1.057 71.60Kiseljak 0.893 70.87 0.869 68.96 0.787 62.46

61

Municipality

% of FBiH Distribution Pool for2006

Per capitaamount of transfer for2006, in KM

% of FBiH Distribution Pool for2007

Per capitaamount of transfer2007, in KM

% of FBiH Distribution Pool for20011

Per capitaamount of transfer 20011, in KM

Kresevo 0.18 51.87 0.197 56.77 0.255 73.50Novi Travnik 0.494 32.92 0.627 41.78 1.093 72.85Travnik 1.916 56.85 2.013 59.73 2.344 69.55Vitez 1.696 112.32 1.533 101.53 0.957 63.37Capljina 0.941 65.55 0.976 67.99 1.088 75.80Citluk 0.945 94.79 0.897 89.98 0.727 72.93Jablanica 0.331 45.91 0.387 53.68 0.577 80.08Konjic 0.875 49.51 1.018 57.60 1.513 85.58Mostar 8.35 123.80 7.634 113.18 5.124 75.97Neum 0.322 112.36 0.301 105.04 0.227 79.29Prozor 0.253 24.82 0.383 37.58 0.835 81.94Ravno 0.016 19.55 0.035 42.76 0.112 136.48Stolac 0.208 25.73 0.295 36.49 0.587 72.66Grude 1.691 178.06 1.465 154.26 0.666 70.14Ljubuski 1.436 100.02 1.344 93.61 1.013 70.55Posusje 1.035 105.75 0.997 101.87 0.856 87.44Siroki Brijeg 2.111 124.02 1.917 112.62 1.232 72.37Sarajevo Centar

18.99 76.39 18.357 73.85 16.139 64.93

HadziciIlidzaIlijasNovi GradNovo SarajevoStari GradTrnovoVogoscaBosansko Grahovo 0.072 56.15 0.106 82.67 177.08Drvar 0.347 48.85 0.396 55.74 80.12Glamoc 0.175 58.40 0.223 74.42 130.13Kupres 0.22 102.95 0.229 107.16 123.08Livno 1.755 89.30 1.718 87.42 80.60Tomislavgrad 1.442 85.92 1.425 84.91 81.33

Table continued

62

Independent variables used for regression:

Municipality Population PIT per capitaBihac 60,876 102.97Bosanska Krupa 28,065 35.68Bosanski Petrovac 8,020 43.23Buzim 17,596 25.06Cazin 61,673 35.82Kljuc 19,966 27.84Sanski Most 45,998 33.63Velika Kladusa 46,741 25.16Domaljevac-Samac 4,485 44.42Odzak 16,163 44.52Orasje 20,788 77.33Banovici 25,519 64.14Celic 14,400 18.20Doboj-Istok 10,096 48.99Gracanica 51,450 41.27Gradacac 46,179 52.13Kalesija 34,911 29.51Kladanj 15,252 65.51Lukavac 51,551 48.56Sapna 13,934 16.55Srebrenik 40,929 49.95Teocak 7,394 35.13Tuzla 131,374 113.92Zivinice 53,896 56.77Breza 14,804 52.41Doboj-Jug 4,412 55.31Kakanj 43,245 75.43Maglaj 23,550 38.97Olovo 12,823 55.80Tesanj 47,514 48.54Usora 7,076 30.60Vares 11,301 57.66Visoko 39,948 66.57Zavidovici 38,102 35.85Zenica 127,646 95.07Zepce 30,980 33.95Foca 1,848 45.66Gorazde 31,563 70.32Pale 1,097 52.27Bugojno 37,434 39.88Busovaca 16,005 52.42Dobretici 662 33.64Donji Vakuf 14,140 39.25Fojnica 12,365 93.64Gonji Vakuf 19,459 39.13Jajce 24,351 46.30Kiseljak 20,786 57.40Kresevo 5,724 53.43Novi Travnik 24,753 48.84Travnik 55,590 69.37Vitez 24,906 62.07Capljina 23,678 65.90Citluk 16,444 73.13

63

Jablanica 11,892 70.74Konjic 29,154 49.43Mostar 111,259 126.34Neum 4,727 80.97Prozor 16,811 27.30Ravno 1,350 25.72Stolac 13,334 33.58Grude 15,665 74.00Ljubuski 23,683 63.78Posusje 16,144 51.49Siroki Brijeg 28,078 66.36Sarajevo Centar 70,294 368.27Hadzici 21,958 76.16Ilidza 48,291 129.81Ilijas 15,462 59.29Novi Grad 119,883 148.53Novo Sarajevo 73,381 239.10Stari Grad 38,000 243.31Trnovo 2,187 38.20Vogosca 20,575 112.80Bosansko Grahovo 2,115 51.82Drvar 11,718 27.54Glamoc 4,943 48.45Kupres 3,525 57.77Livno 32,418 48.27Tomislavgrad 27,683 33.28

64

BIBLIOGRAPHY

ACKERMAN S.R., 1974. Does Federalism Matter? Political Choice in a Federal Republic. The Journal of Political Economy, 89(1), 152-165.

ADAMOVICH, I.B. and HOSP G., 2001. Fiscal Federalism for Emerging Democracies: Lessons from Switzerland?. Paper prepared for the Sixth Annual Meeting of the Latin American and Caribbean Economic Association (LACEA). Fribourg: Swiss University of Fribourg.

BAHL, R., 1999. Intergovernmental Transfers in Developing and Transitional Countries. Principles and Practice. Intergovernmental Fiscal Relations in Central and Eastern Europe: A Sourcebook and Reference Guide [CD-ROM]. Urban Institute and World Bank.

BAHL, R. and WALLACE, S., 2004. Intergovernmental Transfers: The Vertical Sharing Dimension. WP 04-19. Andrew Young School of Policy Studies. Georgia: Georgia State University.

BIRD, M.R., 2000. Intergovernmental Fiscal Relations: Universals Principles, Local Applications. WP 00-02. Andrew Young School of Policy Studies. Georgia: Georgia State University.

BIRD, M.R., 2003. Asymmetric Fiscal Decentralization: Glue or Solvent. WP 03-09. Andrew Young School of Policy Studies. Georgia: Georgia State University.

BIRD, M.R. and EBEL R.D., 2005. Fiscal Federalism and National Unity. Washington, DC: Urban Institute.

BIRD, M. R., EBEL R.D. and WALLICH C. I., eds., 1995. Decentralization of the Socialist State: Intergovernmental Finance in Transition Countries. Washington, DC: World Bank.

BIRD, M.R. and TARASOV, A., 2002. Closing the Gap: Fiscal Imbalances and Intergovernmental Transfers in Developed Federations. WP 02-02. Andrew Young School of Policy Studies. Georgia: Georgia State University.

BIRD, M.R. and VAILLANCOURT, F., eds., 1998. Fiscal Decentralization in Developing Countries. Cambridge: Cambridge University Press.

BAJO, A. and BRONIC, M., 2005. Fiscal Decentralization in Croatia-Problems of Fiscal Equalization. Occasional Paper No. 25. Zagreb: Institute for Public Finance.

BIRD, M.R. and VAILLNCOURT, F., 2004. Expenditure-Based Equalization Transfers. WP 04-10. Andrew Young School of Policy Studies. Georgia: Georgia State University.

BLONDEL, J., 1995. Comparative Government. New York: Harvester Wheatsheaf.

BOADWAY, R., 2004. The Theory and Practice of Equalization. CESifo Economic Studies. 50 (1), 211–254.

65

BOADWAY, R. and SHAH, A., (forthcoming). Fiscal Federalism: Principles and Practices. Washington, DC: World Bank.

BOEX, J. and MARTINEZ-VAZQUEZ, J., The Design of Equalization Grants: Theory and Applications. Andrew Young School of Policy Studies. Georgia: Georgia State University.

BOEX, J. and MARTINEZ-VAZQUEZ, J., 2004. Designing Intergovernmental Equalization Transfers with Incomplete Data: Concepts, Practices and Lessons. WP 04-21. Andrew Young School of Policy Studies. Georgia: Georgia State University.

BOEX J., VAZQUEZ-MARTINEZ J. and TIMOFEEV A., 2004. Subnational Government Structure and Intergovernmental Fiscal Relations: An Overlooked Dimension of Decentralization. WP 04-01. Andrew Young School of Policy Studies. Georgia: Georgia State University.

BRAUN, D., 2003. Fiscal Policies in Federal States. Aldershot: Ashgate.

BRANLEY, G., 1990. Equalization Grants and Local Expenditure Needs. Aldershot: Avebury.

BUCHANAN, J.M., 1950. Federalism and Fiscal Equity. The American Economic Review, 40 (4), 583-599.

BURGESS, M., 2005. Comparative Federalism in Theory and Practice. New York: Routledge.

BURGESS, M. and GAGNON, A.G., eds., 1993. Comparative federalism and federation: competing traditions and future directions. New York: Harvester Wheatsheaf.

CORDIERA-GUERRA, S. and HOFMAN B., 2004. Ensuring Inter-regional Equity and Poverty Reduction. WP 04-11. Andrew Young School of Policy Studies. Georgia: Georgia State University.

DAFFLON, B., forthcoming. The Assignment of Functions to Decentralized Government: From Theory to Practice. In: AHMAD, E. and BROSIO, G., eds., (forthcoming). Handbook on Fiscal Federalism. Cheltenham: Edward Elgar.

DAFFLON, B., 2002. Local public Finance in Europe: Balancing the budget and controlling debt. Northampton, MA: Edward Elgar.

DE WILLIERS, B., 1994. Evaluating Federal Systems. Cape Town: M. Nijhoff.

DUCHACEK, I.D., 1970. Comparative Federalism: The Territorial Dimension of State. New York: Holt, Rinehart and Winston.

EBEL, R.D. and YILMAZ, S., 2002. Concept of Fiscal Decentralization and Worldwide Overview. Washington DC: Urban Institute.

ELAZAR, D., 1987. Exploring Federalism.Tuscaloosa: University of Alabama Press.

ELAZAR, D., 1995. Federalism: An Overview. Pretoria: HSRC Publishers.

66

FOX W. and WALLICH, C., 1997. Fiscal Federalism in Bosnia and Herzegovina: The Dayton Challenge. Working Paper. Washington DC: World Bank.

FOX W. and WALLICH, C., Fiscal Federalism in Bosnia and Herzegovina: Subsidiarity and Solidarity in a Three Nation State. In: EBEL, R. D. and BIRD M. R., eds., (forthcoming). Subsidiarity and Solidarity: The Role of Intergovernmental Relations in Maintaining an Effective State. Cheltenham: Edward Elgar.

GAP (Governance Accountability Project), 2005. Policy Briefing Book. Sarajevo: USAID-SIDA.

Government of FBiH, 1995. Law on the Local Self-Governance. Sarajevo: Official Gazette No. 6/95.

Government of FBiH, 1996. Law on Allocation of Public Revenues in FBiH. Sarajevo: Official Gazette No. 26/96.

Government of RS, 1999. Law on Local Self-Governance. Banja Luka: Official Gazette No. 35/99.

Council of Minister of BiH, Law on the System of Indirect Taxation. Sarajevo: Official Gazette No. 44/03.

Council of Ministers of BiH, 1995. Law on Value Added Tax. Sarajevo: Offical Gazette No. 09/05.

Government of FBiH, 2006. Law on Allocation of Public Revenues in FBiH. Official Gazette No. 22/06.

Government of FBiH, 2006. Instruction for Calculation of Shares of Local Self-Government Units, Cantons and Road Funds, in the Revenues from Indirect Taxes. Sarajevo: Official Gazette No. 29/06.

HORVATH, M. T., ed. Decentralization: Experiments and Reforms. Budapest: OSI.

INMAN, R.P. and RUBINFELD D.L., 1997. Rethinking Federalism. The Journal of Economic Perspectives, 11(4), 43-64.

KANDEVA, E., ed., 2001. Stabilization of Local Governments. Budapest: LGI.

LANG, N.R., 1975. The Dialectics of Decentralization: Economic Reform and Regional Inequality in Yugoslavia. World Politics, 27(3), 309-335.

LIJPHART, A., 1999. Patterns of democracy: government forms and performance in thirty-six countries. New Haven: Yale University Press.

MARTINEZ-VAZQUEZ, J. and McNAB R., 1997. Fiscal Decentralization, Economic Growth, and Democratic Governance. WP 97-7. Andrew Young School of Policy Studies. Georgia: Georgia State University.

67

MARTINEZ-VAZQUEZ, J. and ALM J., eds., 2003. Public Finance in Developing and Transitional Countries: Essays in Honor of Richard Bird. Northampton: Edward Elgar Pub.

MARTINEZ-VAZQUEZ, J. and McNAB, R., 2005. Fiscal Decentralization, Macroeconomic Stability and Growth. WP 05-06. Andrew Young School of Policy Studies. Georgia: Georgia State University.

McLURE, C.E. and MARTINEZ-VAZQUEZ, J. The Assignment of Revenues and Expenditures in Intergovernmental Fiscal Relations. Washington DC: World Bank.

MILICEVIC, N., 2005. State and Problems of Local Self-Government in Bosnia and Herzegovina. Zagreb: Friedrich Ebert Stiftung.

MUSGRAVE A.R., 1965. Essays in Fiscal Federalism. Westport: Greenwood Press.

MUSGRAVE A.R., and MUSGRAVE B.P., 1989. Public finance in theory and practice. New York: McGraw-Hill.

MUSGREVE, A.R., 1997. Devolution, Grants and Fiscal Competition. The Journal of Economic Perspectives, 11(4), 65-72.

NORRIS-DABLA, E., 2006. The Challenge of Fiscal Decentralization in Transition Countries. Comparative Economic Studies, 48 (1), 100-131.

OATES, W. E., 1972. Fiscal Federalism. New York: Harcourt Brace Jovanovich.

OATES, W.E., ed., 1977. The Political Economy of Fiscal Federalism. Lexington, MA: Heath-Lexington.

OATES, W.E., 1981. On Local Finance and Tiebout Model. The American Economic Review, 71(2), 93-98.

OATES, W.E., 1991. An Essay on Fiscal Federalism. Journal of Economic Literature, 37(3), 1120-1149.

OATES, W. E., 1991. Studies in fiscal federalism. Aldershot, Hants: E. Elgar.

PETCHEY, J. and LEVTCHENKOVA, S., 2004. Fiscal Capacity Equalization and Economic Efficiency. WP 04-15. Andrew Young School of Policy Studies. Georgia: Georgia State University.

PETERI, G., ed., 2002. Mastering Decentralization and Public Administration Reforms in Central and Eastern Europe. Budapest: LGI.

RIKER, W.H., 1964. Federalism: Origin, operation, significance. Boston: Little Brown.

RONDINELLI, D. What is Decentralization? Washington, DC: World Bank Institute.

SEARLE, B., 2004. Revenue Sharing, Natural Resources and Fiscal Equalization. WP 04-16. Andrew Young School of Policy Studies. Georgia: Georgia State University.

68

SEARL B. and HULL C., 2004. The Impact of Equalization on Service Delivery. WP 04-12. Andrew Young School of Policy Studies. Georgia: Georgia State University.

SHAH, A., 1996. A Fiscal Need Approach to Equalization. Canadian Public Policy, 22(2), 99-115.

SHAH, A., 1997. Fiscal Federalism and Economic Governance: For Better or for Worse? Paper prepared for International Conference on Decentralization, Intergovernmental Fiscal Relations and Macroeconomic Governance, Brasilia.

SHAH, A., 2004. Fiscal Decentralization in Developing and Transition Economies: Progress, problems, and the Promise. WPS 3282. Washington DC: World Bank.

SHANKAR, R. and SHAH, A., 2001. Bridging the Economic Divide Within Nations. WPS 2717. Washington DC: World Bank.

SLUKHAI, S. ,ed., 2003. Dilemmas and Compromises: Fiscal Equalization on Transition Countries. Budapest: LGI

SMITH B.C., 1985. Decentralization: The Territorial Dimension of the State. London: Allen&Ubwin.

SMITH, G., ed. 1995. Federalism: The multiethnic challenge. London: Longman.

SPAHN, P.B., 2004. Intergovernmental Transfers: Funding Rules and Mechanisms. WP 04-17. Andrew Young School of Policy Studies. Georgia: Georgia State University.

STROSCHEIN, S., 2003. What Belgium Can Teach Bosnia: The Uses of Autonomy in ‘Divided House’ States. Journal on Ethnopolitcs and Minority Issues in Europe, 3.

TANZI, V., 2000. On Fiscal Federalism: Issues to Worry About. Conference on Fiscal Federalism. Washington, D.C: International Monetary Fund.

TANZI, V., 2001. Pitfall on Road to Fiscal Decentralization. Washington DC: Carnegie Endowment.

TIEBOUT, C.M., 1956. A Pure Theory of Local Expenditures. The Journal of Political Economy, 64(5), 416-424.

TOMAS, R,. Macroeconomic Implications of Fiscal Decentralization. Banja Luka: School of Economics.

UNDP, 2005. National Human Development Report: Better Local Governance in Bosnia and Herzegovina. UNDP: Sarajevo.

WATTS, R.L., 1999. Comparing Federal Systems. Kingston, Ont.: McGill-Queen's University Press.

WALLACE, S., 2001. Fiscal Architecture and the Analysis of Public Expenditure Needs and Revenue Capacity. WP 01-11 Andrew Young School of Policy Studies. Georgia: Georgia State University.

69

World Bank, 1999. The World Development Report 1999/2000. Oxford: Oxford University Press.

The World Bank, 2001. Decentralization in Developing Countries: Challenges and Road Ahead. Washington, DC: World Bank.

World Bank., 2002. Bosnia and Herzegovina: From Aid Dependence to Fiscal Self-Reliance. Report No 24297-BiH. Washington DC: World Bank.

70