Fiscal Decentralization Indicators for 2006-2014 Fiscal Decentralization Indicators for South-East Europe: Fifth Edition March, 2016 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 CIP - Каталогизација во публикација Национална и универзитетска библиотека “Св. Климент Охридски”, Скопје 352:336.14(4-12)”2006/14”(047) FISCAL decentralization indicators for South-East Europe : report : 2006-2014 / [autrhors Levitas Anthony ... [и др.]. - Skopje : Network of Associations of Local Authorities of South-East Europe NALAS, 2016. - 130 стр. : граф. прикази ; 21 см Фусноти кон текстот ISBN 978-608-4680-06-2 1. Levitas, Anthony [автор] а) Локална самоуправа - Фискална децентрализација - Југоисточна Европа - 2006-2014 - Извештаи COBISS.MK-ID 100890890 (NALAS) Network of Associations of Local Authorities of South-East Europe CONTENT Table of Figures The Report in Brief Introduction I. Data, Terms, and Methodological Issues II. Overview of Local Governments in South-East Europe Number and types of sub-sovereign governments The Average Population of Municipal Governments The Dynamics of the Gross Domestic Product III. Basic Indicators of Fiscal Decentralization Local Governments Revenues in South-East Europe The Local Fiscal Autonomy and the Basic Composition of Local Revenues The Composition of Own-Revenues and the Property Tax The Composition of Expenditures and Investment Spending Local Government Borrowing IV. Gender Mainstreaming in Fiscal Decentralization Indicators for Gender Responsive Budgeting in SEE Findings and Recommendations V. Country Reviews of Fiscal Decentralization Trends and Developments Albania Bosnia and Herzegovina - Federation of Bosnia and Herzegovina Bosnia Herzegovina – Republic of Srpska Bulgaria Croatia Kosovo Macedonia Moldova Montenegro Romania Serbia Slovenia Turkey REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 3 7 11 13 15 15 17 18 20 22 28 32 36 40 43 43 45 47 48 54 60 66 72 78 84 92 98 106 114 122 128 This publication is a collaborative effort of the NALAS Task Force on Fiscal Decentralization (TF FD). Publisher: Network of Associations of Local Authorities of South East Europe (NALAS) Authors: Levitas Anthony, International Expert in Intergovernmental Fiscal Relations, and: Alkan Adil representing the Marmara Municipalities Union, Coku Iva representing the Association of Albanian Municipalities, Comsa Radu representing the Association of Communes of Romania, Coric Dijana representing the Association of Towns and Municipalities of Republic of Srpska, Djukic Zana, representing the Union of Municipalities of Montenegro, Elezi Shiret representing the Association of the Units of Local Self-Governments of the Republic of Macedonia, Furdui Carolina representing the Congress of Local Authorities of Moldova, Kabil Mustafa representing the Marmara Municipalities Union, Kovacevic Dzenita representing the Association of Municipalities and Cities of the Federation of Bosnia and Herzegovina, Milunovic Vilma representing the Association of Towns and Municipalities of Slovenia, Naic Dunja representing the Standing Conference of Towns and Municipalities of Serbia, Runtic Dario representing the Association of Cities in the Republic of Croatia, Sadikaj Osman, representing the Association of Kosovo Municipalities, Savov Emil, representing the National Association of Municipalities in the Republic of Bulgaria, Stafa Elton representing the Association of Albanian Communes and Qesku Arben representing the Association of Albanian Communes. The section Gender Mainstreaming in Fiscal Decentralization was analyzed by the experts on Gender Responsive Budgeting of the UN Women project “Promoting Gender Responsive Policies in South East Europe and Republic of Moldova, phase II”, based on data collected by the members of the NALAS TF FD and Fabris Vesna representing the Association of Municipalities in the Republic of Croatia, Misovic Branislav representing the Association of Towns and Municipalities of Republic of Srpska, and Lunkasu Irina representing the Congress of Local Authorities of Moldova. Publication Coordinator: Ilijeva – Acevska Natasha, NALAS Program Officer With the support of: Zajazi Kelmend, NALAS Executive Director NALAS Secretariat NALAS Knowledge Management Assistants GIZ project “Institutional Strengthening of NALAS” UN Women project “Promoting Gender Responsive Policies in South East Europe and Republic of Moldova, phase II” Swiss Agency for Development and Cooperation Design by: Brigada Design, Skopje TABLE OF FIGURES: Table 1: Numbers and Types of Sub-Sovereign Governments Table 2: Local Government Social Sector Functions* 15 21 Figure 1: – Financial lows of the Romanian equalization system 107 Chart 1: Average Population of 1st Tier Local Governments Chart 2: Percentage of Population Living in Capital Cities Chart 3: GDP Per Capita in 2006 & 2014 and GDP Growth 2006-14* Chart 4: GDP Growth in the Crisis of 2008-9; between 2009 and 2013; and in 2014 Chart 5: General and Local Government Revenue as a Percentage of GDP in 2014 Chart 6: Local Revenues as a share of Total Public Revenues & GDP (2014) Chart 7: Public Sector Size and the Extent of Decentralization as Measured by the Devolution of Social Sector Functions* Chart 8: Local Government Revenue as a Share of GDP in 2006, 2009, 2014 Chart 9: Consolidated Public and Local Government Revenue (EUR Per Capita-2014) Chart 10: Composition of Local Revenue in 2006 Chart 11: Composition of Local Revenue in 2014 Chart 12: Composition of Local Government Revenue 2006 Chart 13: Composition of Local Government Revenue2014 Chart 14: Growth/Decline of Local Government Revenue between 2013 and 2014 Chart 15: Composition of Own-Revenues in 2006 Chart 16: Composition of Own-Revenues in 2014 Chart 17: Absolute Growth of Own Revenues 2006-2014 Chart 18: Absolute Growth of Own Revenues and their per capita Yield in EUR 2014 Chart 19: Property Tax as % of GDP and Total Local Revenue in 2006 Chart 20: Property Tax as % of GDP and Total Local Revenue in 2014* Chart 21: Composition of Local Government Expenditure in 2014 Chart 22: Local Government Investment in 2006, 2009, 2014 (EUR per capita) Chart 23: Total Public Investment by Level of Government as shares of GDP (average 2006-2014) Chart 24: Investment Spending in Million EUR 2014 Chart 25: Public Debt and Budget Deicits in SEE Region in 2014 Chart 26: Public Debt by Level of Government as Share of GDP Chart 27: Increase in Local Government Debt 2006-2014 (EUR per capita) Chart 28: Existence of a Law on Gender Equality/Equal Opportunities Chart 29: Existence of Policies for Gender Responsive Budgeting at Central/Local Level Chart 30: Budget Circular (central/local) includes gender provisions Chart 31: Embedded Practice of GRB Analysis as an Essential Step in Identifying Impacts of National and Local Policies REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 17 18 19 19 22 23 24 26 27 28 28 29 29 31 32 32 33 33 35 35 36 37 38 39 40 41 42 44 44 45 45 3 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 4 Chart 32 Albania: Local Government Revenue as a Share of GDP and Total Public Revenue 2006-2014 Chart 33 Albania: Revenue Fluctuations of the General Government and Local Governments 2006-2014 Chart 34 Albania: Composition of Local Government Revenue 2006-2014 Chart 35 Albania: Composition of Revenues 2006-2014 (EUR Per Capita) Chart 36 Albania: Composition of Local Government Expenditures 2006-2014 Chart 37 Albania: Investment, Wages, and Property Tax as Share of GDP 2006-2014 Chart 38 FBiH (of BiH): Local Government Revenue as Share of GDP and Total Public Revenue 2006-2014 Chart 39 FBiH (of BiH): Annual Fluctuations in the Revenues of the General Government and Local Governments Chart 40 FBiH (of BiH): The Composition of Local Govrnment Revenue 2006-2014 (EUR Per Capita) Chart 41 FBiH (of BiH): The Composition of Local Government Expenditure 2006-2014 Chart 42 FBiH (of BiH): Local Government Wages, Investment, Property Taxes as % GDP 2006-2014 Chart 43 RS (of BiH): Local Government Revenue as Share of GDP and Total Public Revenue 2006-2014 Chart 44 RS (of BiH): Annual Fluctuations in the Revenues of the General Government and Local Governments Chart 45 RS (of BiH): Composition of Local Government Revenues 2006-2014 Chart 46 RS (of BiH): Composition of Expenditures in 2006-2014 Chart 47 RS (of BiH): Investment, Wages, and Property Tax as Share of GDP 2006-2014 Chart 48 Bulgaria: Local Government Revenue as Share of GDP and Total Public Revenue in 2006-2014 Chart 49 Bulgaria: Fluctuations in the Revenues of the General Government and Local Governments Chart 50 Bulgaria: Composition of Local Government Revenue 2006-2014 Chart 51 Bulgaria: Public of Investment by Level of Government and as a % GDP 2006-2014 Chart 52 Bulgaria: Local Government Wages, Investment, Property Taxes and Outstanding Debt as % GDP 2006-2014 Chart 53 Croatia: Local Government Revenue as a Share of GDP and Total Public Revenue 2006-2014 Chart 54 Croatia: Fluctuations in the Revenue of the General Government and Local Governments Chart 55 Croatia: Composition of Local Government Revenue 2006-2014 Chart 56 Croatia: Local Government Revenue 2006-2014 EUR per capita Chart 57 Croatia: Composition of Local Government Expenditure 2006-2014 Chart 58 Croatia: Local Government Wages, Investment, Property Taxes and Outstanding Debt as % GDP 2006-2014 Chart 59 Croatia: Shares of Public of Investment by Level of Government and as % GDP 2006-2014 Chart 60 Kosovo: Local Government Revenue as Share of GDP and Total Public Revenue 2006-2014 Chart 61 Kosovo: Fluctuations in the Revenue of the General Government and Local Governments Chart 62 Kosovo: Composition of Revenues 2009-2014 Chart 63 Kosovo: Composition of Own Revenue 2009-2014 (EUR per capita) Chart 64 Kosovo: Composition of Expenditures in 2008-2014EUR) Chart 65 Kosovo: Investment, Wages & Property Tax as a Share of GDP 2006-2014 Chart 66 Macedonia: Local Government Revenue as a Share of GDP and Total Public Revenue in 2006-2014 Chart 67 Macedonia: Composition of Local Government Revenues 2006-2014 Chart 68 Macedonia: Composition of Own Revenues 2006-2014 (mln EUR) Chart 69 Macedonia: Composition of Expenditures in 2006-2014 Chart 70 Macedonia: Investment, Wages, & Property Tax as Shares of GDP 2006-2014 Chart 71 Macedonia: Public Investment by Level of Government and as a share of GDP 2006-2014 Chart 72 Moldova: Local Government Revenue as a Share of GDP and Total Public Revenue 2006-2014 Chart 73 Moldova: Composition of Local Government Revenue 2006-2014 (NALAS) Network of Associations of Local Authorities of South-East Europe 50 50 51 52 52 53 56 56 57 58 59 62 62 63 64 65 69 70 70 71 71 74 74 75 75 76 76 77 80 80 81 82 82 83 87 88 88 89 90 91 94 95 (NALAS) Network of Associations of Local Authorities of South-East Europe Chart 74 Moldova: Composition of Local Government Revenue 2006-2014 (EUR per capita) Chart 75 Moldova: Composition of Local Government Expenditures in 2006-2014 Chart 76 Moldova: Investment, Wages, and Property Tax as Shares of GDP 2006-2013 Chart 77 Montenegro: Local Government Revenue as Share of GDP and Total Public Revenue 2006-2014 Chart 78 Montenegro: Composition of Local Government Revenues 2006-2014 Chart 79 Montenegro: Composition of Own Revenues 2006-2014 Chart 80 Montenegro: Composition of Expenditures in 2006-2014 (EUR per capita) Chart 81 Montenegro: Investment, Wages, Debt Service and Property Tax as Share of GDP 2006-2014 Chart 82 Montenegro: Public Investment by Level of Government and as a Share of GDP 2006-2014 Chart 83 Romania: Local Government Revenue as Share of GDP and Total Public Revenue in 2006-2014 Chart 84 Romania: Composition of Revenue 2006-2014 Chart 85 Romania: Composition of Local Government Own Revenue 2006-2014 (mln EUR) Chart 86 Romania: Composition of Local Government Expenditures in 2006-2014 (EUR per capita) Chart 87 Romania: Investment, Wages, and Property Tax as Shares of GDP 2006-2014 Chart 88 Romania: Public Investment by Level of Government and as a Share of GDP 2006-2014 Chart 89 Serbia: Local Government Revenue as a Share of GDP and Total Public Revenue 2006-2014 Chart 90 Serbia: Fluctuations in the Revenues of the General Government and Local Governments Chart 91 Serbia: Composition of Local Government Revenue 2006-2014 Chart 92 Serbia: Composition of Own Revenues 2006-2014 (mln EUR) Chart 93 Serbia: Composition of Local Government Expenditure 2006-2014 Chart 94 Serbia: Investment, Wages, and Property Tax as Shares of GDP 2006-2014 Chart 95 Slovenia: Local Government Revenue as a Share of GDP and Total Public Revenue in 2006-2014 Chart 96 Slovenia: Fluctuations in the Revenue of the General Government and Local Governments Chart 97 Slovenia: Composition of Local Government Revenue 2006-2014 Chart 98 Slovenia: Composition of Own Revenue 2006-2014 Chart 99 Slovenia: Composition of Expenditure in 2006-2014 Chart 100 Slovenia: Investment, Wages, Property Tax as Share of GDP 2006-2012 Chart 101 Turkey: Local Government Revenue as a Share of GDP and Total Public Revenue in 2006-2014 Chart 102 Turkey: Composition of Local Government Revenue 2006-2014 Chart 103 Turkey: Composition of Local Revenue 2006-2014 (EUR per capita) Chart 104 Turkey: Composition of Expenditure in 2006-2012 Chart 105 Turkey: Investment, Wages, Outstanding Debt and Property Tax as Shares of GDP 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 96 96 97 102 103 103 104 104 105 110 111 111 112 113 113 117 117 118 119 120 121 124 125 125 126 126 127 130 131 132 132 133 5 (NALAS) Network of Associations of Local Authorities of South-East Europe THE REPORT IN BRIEF: O ver the last 20 years, eforts to decentralize public services to democratically-elected local governments has been a common theme across SouthEast Europe. Progress however has been uneven, and there are few places in the region where local government revenues or expenditures approach EU averages, either as percentages of GDP or of total public revenue. While it is always diicult to judge the adequacy of local government revenues relative to their expenditure needs, there seems little question that in many places in the region municipalities are underfunded, and that central governments are not giving them a fair share of the overall iscal pie. Underfunding is particularly obvious in Albania and the Federation of Bosnia and Herzegovina (BiH) despite the fact that in both, municipalities have few social sector responsibilities. Municipalities in Macedonia and Bulgaria also appear to be signiicantly under-resourced, though here the underfunding of basic municipal services is intertwined with the underfunding of primary and secondary education which in both countries has been devolved to local governments. Romania and Kosovo1 are at the other end of the spectrum. Here, local governments pay for all pre-university schooling, as well as for much of primary healthcare. But their revenues are closer to the average for the EU, both in relationship to total public revenue and the GDP. On paper, the situation is similar in Moldova, but central and regional control over municipal budgets makes the indicators misleading. Elsewhere in the region, the degree to which local governments are underfunded is less obvious. Equally importantly, we do not have the data that would allow us to assess how much local is concentrated in a few wealthy jurisdictions, and in capital cities in particular. Without this data, it is impossible to make reasonable judgements about the horizontal equity of the region’s intergovernmental inance systems, and about how much radical disparities in budget income among local governments is being masked by apparently healthy macro-economic indicators or intensiied by poor ones. 1 “This designation is without prejudice to positions on status, and is in line with UNSC 1244 and the ICJ Opinion on the Kosovo declaration of independence.” REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 7 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 It is however clear that in many places economic activity and political power is concentrated in capital cities which contain disproportionate shares of the population. The concentration of wealth and power in large capital cities thus makes it likely that in many places important issues of redistribution are not being adequately addressed. Surprisingly, decentralization has gone furthest in places where higher level governments have trouble collecting taxes and the overall public sector is relatively small. The correlation between small public sectors and the decentralization social sector functions to local governments –particularly primary and secondary education— suggests that in some places national governments have sought to relieve themselves of the burden of administering services they feel they can’t aford to adequately inance themselves. 8 Grants, Conditional Grants, and Shared Taxes, particularly shared personal income tax. This level of dependency on central government transfers is, however not unusual. Indeed, it is in line with the average for OECD member-states. Moreover, “transfer dependency” in the region increases as social sector functions are devolved to municipalities, a trend that is also in line with experiences elsewhere. The reason for this is simple, but often overlooked: Local governments become more dependent on transfers as social sector functions are devolved to them because there are not enough robust tax bases that can be reasonably assigned to them. It is important to appreciate this “decentralization paradox” because too often advocates of decentralization measure its success by the degree to which local governments “inance themselves.” The global downturn of 2008-2009, hit much of the region very hard. Central governments often responded to the iscal pressures of the crisis by making ad hoc adjustments in transfer systems that compounded the negative efects of the recession on municipal budgets. In some places, however, the fall in global economic activity had relatively little impact on the domestic economies of the region or this impact was delayed. The paradox also has important policy implications: Instead, of trying to make local governments “iscally autonomous”, reforms should focus on developing the tools, habits, and institutions that allow national and local oicials to constructively adjust their mutual dependence to changing circumstances. Needless to say, the quality of the institutions habits and tools necessary for constructive intergovernmental dialogue varies substantially across the region, but in general remains weak and need of support. With a few exceptions, economic growth since 2009 has been slow, though in most places local inances improved in 2010 and 2011. Signiicant growth however, has yet to return to the municipalities of Slovenia, Croatia, Serbia, and the Republic of Srpska. It is also worth noting that at least in Montenegro and Slovenia some local governments tried to borrow their way out of the crisis and are now having dificulty paying of debt. In a number of places, local governments also seem to be burdened by signiicant payment arrears, though data on this problem is scarce. In much of South-East Europe, municipalities derive signiicant amounts of own-revenue from quasi-iscal instruments imposed on real-estate transactions, new investment, and business operations. Central governments in a number of places have started to constrain these practices in order to improve the “business enabling environment”. As legitimate as these eforts may be, they are compounding the inancial problems of local governments in a number of places and should be accompanied by eforts to replace the lost revenue. With the notable exception of Montenegro, municipalities in South-East Europe derive only about 35% of their revenue from sources over which they have some control. The rest comes from some combination of Unconditional With the exception of Croatia and the Federation of Bosnia Herzegovina, the Property Tax has been decentralized throughout the region. In most places, municipalities have substantially improved the yield of the tax. Nonetheless, (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe and with the exception of Montenegro, it still generates revenue equal to well under 1% of GDP, which is the average for the EU. It is thus unrealistic to expect the Property Tax to yield anything like the revenue it does in North America (2-3% of GDP). And while achieving EU norms is certainly desirable, it alone will neither solve the region’s problem with underfunding nor radically increase the “iscal autonomy” of the region’s municipalities. Instead, eforts to enhance the revenue raising capacities of local governments in many places should focus on transforming the Personal Income Tax from a Shared Tax into a tax over which local governments have some rate-setting powers. This can be done by giving them the right to impose a surcharge on the rate set by the central government, as is already practiced in Montenegro and Croatia. Or “PIT space” can be divided between the national government and local governments, as is currently being considered in Bulgaria. In most of the region, local governments are spending higher shares of expenditure on investment than their counterparts in the EU, despite receiving signiicantly lower shares of total public revenue. This suggests that municipalities in South-East Europe are working hard to make-up for the infrastructure deicits they inherited from the past. Since 2009, however, investment rates have fallen signiicantly in most of the region, and in a number of places are holding their own only because of the inlux of EU structural funds. Scarce investment funds also tend to be spent on payas-you-build road projects and not on pay-as-you-use environmental facilities. This is because planning roads is simpler; construction can be delayed if money runs out; tangible beneits can be delivered within a single election cycle; and because in much of the region municipal borrowing remains a marginal phenomenon. likely to restrict the access of local governments to credit in order to reduce the consolidated public debt and/or to preserve borrowing space for their central governments. In these countries, eforts should be made to reserve some debt space for municipalities. More generally, however, the adequacy and predictability of local government revenues will have to be improved if municipalities are to have the resources against which to prudently incur debt. Part of the answer here is to increase the own-revenue raising powers of municipalities by strengthening property taxation and/or by introducing local PIT surcharges. And part of the answer lies in enhancing and stabilizing transfer systems, eforts that almost everywhere pay particular attention to questions of horizontal equity. Finally, and perhaps most importantly, national and local government oicials need to recognize that decentralization actually intensiies the need for continuous, informed and substantive intergovernmental dialogue, and that like it or not, the fate of national and local governments are linked together at the hip. In most of South East Europe there is a Law on Gender Equality in place and in almost all cases the provisions include requirement for integrate gender in planning and budgeting processes. This suggests a high political commitment among decision-makers which not only opens an important venue for integrating gender perspectives into decentralization processes, but also places an obligation on local governments for integrating a gender perspective into all areas of decentralized competence. Gender responsive budgeting tools are currently applied only in Kosovo, Republika Srpska (BiH) and Turkey. The extent of application or possible form of institutionalization at central/local level in South East Europe require further and more in-depth research. In Albania, Croatia, Serbia and Slovenia, the consolidated debt of the General Government now exceeds the limits set by the Maastricht Treaty. Here, Ministries of Finance are REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 9 (NALAS) Network of Associations of Local Authorities of South-East Europe INTRODUCTION T his report has been prepared by the Fiscal Decentralization Task Force of the Network of Associations of Local Authorities of South-East Europe (NALAS). It is the ifth edition of an ongoing efort to provide policy-makers and analysts with reliable comparative data on municipal inances and intergovernmental iscal relations in SouthEast Europe2. The irst edition was published in March 2011 and covered the years 2006-2010. This edition covers the period 20062014. As before, the report tries to both capture regional trends, and major developments in particular countries/ entities. This year, the report also includes short descriptions of property tax systems in the region, partly based on the indings of the NALAS Second Summer School of Local Governments and Intergovernmental Fiscal Relations which held in Ohrid, in August 2015 and focused on property taxation in SEE. A novelty in this edition is chapter on gender mainstreaming in iscal decentralization. 2 The report is divided into ive sections. The irst reviews the data used in the report and discusses some basic methodological issues. The second begins with a presentation of the structure and functions of municipal governments in the region. The third section examines selected indicators of macro-economic performance and iscal decentralization. The fourth section introduces an overview of the gender mainstreaming in iscal decentralization. The ifth section focuses on the evolution of intergovernmental inances in each NALAS’ member country or entity and describes their property tax systems. The report has been used by member associations to argue for policy changes at home. It has also provided input for the design of the monitoring system of the regional strategy South East Europe. See 2020. http://rcc.int/pages/62/south-east-europe-2020-strategy REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 11 (NALAS) Network of Associations of Local Authorities of South-East Europe Data, Terms, and Methodological Issues T I he data used in the report has been provided by NALAS members and comes from their respective Ministries of Finance, Central Banks and Statistical Agencies. The data was checked for consistency and compared, where possible, with similar data from EuroStat —the statistical agency of the European Union— and other sources. What Municipal Governments Do: Throughout South-East Europe, municipalities and communes bear primary responsibility for maintaining and improving local public infrastructure. This includes local roads, bridges, and parks, as well as water supply and sewage treatment, garbage collection and disposal, public lighting, local public transport, and district heating. Comparing intergovernmental inance systems, however is never straight forward because of diferences in how subsovereign governments are organized, what they do, and how they get (or don’t get) the money to pay for what they do. In the following, we discuss how the report addresses some of the methodological issues involved in making reasonable comparisons with imperfect data. In a number of countries/entities, however, local governments are responsible for delivering important social sector services, particularly in education, but also in some places, healthcare. The degree to which local governments are responsible for social sector services has a profound efect on their “iscal weight” everywhere. It is thus important when reading the report to remember what social sector services local governments are providing in diferent places. We discuss these issues in greater detail in the next section. But in many of the report’s Charts and Tables, places in which local governments are responsible for paying teachers’ wages –the single weightiest function devolved to them—are indicated with an asterisk (*). Levels of Government: The report’s primary object of analysis are irst-tier local governments, meaning democraticallyelected municipal or communal authorities. They constitute the most important level of sub-sovereign government in the region and in the report are collectively referred to as municipalities. Democratically-elected regional governments however are important in the Federation of Bosnia Hercegovina, Moldova, Turkey and Romania. In the report, the revenue and expenditure of regional governments is included in the data presented for local governments in Romania and Moldova, but is excluded for the Federation of Bosnia Hercegovina and Turkey. Population: In general, the population numbers used in the report are from the most recently conducted censuses. In Albania, Bosnia and Herzegovina, Kosovo and Macedonia however, the results of recently conducted censuses have been abandoned or remain unoicial for political reasons. In these places, we have used either older census data or the data which the Ministry of Finance is using calculate grants and transfers. Since there has been a profound demographic REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 13 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 decline in most of the region, the use of older census igures signiicantly inlates the actual number of citizens residing in a given country or entity. GDP: We have used the GDP igures calculated by the respective Ministries of Finance of each country or entity according to the production method. Where we converted GDP into EUR igures for comparative purposes we have used the average annual exchange rates provided by the relevant Central Banks. 14 Consolidated Public Revenue of the General Government: To compare the relative importance of local governments across settings we have generally used revenues— and not expenditures— as a share of the consolidated inances of the General Government. This is because data on revenues data tends to be more consistent than data on expenditures at the subnational level. By General Government Revenue we mean the total revenues of the national government and its agencies, including the revenues of of-budget (social security) funds and those of subnational governments. For local governments we have excluded proceeds from borrowing, but included income from asset sales and carry-overs from previous years. General Grants: In most of South-East Europe, local governments receive freely disposable (unconditional) General Grants from their central governments. In some places, the size of the relevant grant pools are deined by law as percentages of national taxes. Because these funds are allocated by formula we consider them Grants, despite the fact that in some places they are popularly referred to as shared taxes. Unless otherwise indicated, we use the term Shared Taxes only for national taxes that are shared with local governments on an origin basis. Conditional and Block Grants: Throughout South-East Europe, local governments receive grants from higher level governments which they can only be use for particular purposes. We refer to grants that must be spent on speciic projects or programs as Conditional Grants. Grants that are designed to help local governments fund a particular function (such as primary education), but which they are free to spend across (NALAS) Network of Associations of Local Authorities of South-East Europe that function as they see it, we refer to as Block Grants. In many places however, the “block” function of Block Grants is limited due to other centrally imposed constraints on local spending. In the extreme, some “Block Grants” (particularly for primary and secondary education) make local governments little more than the payroll agents of the national government. Shared Taxes: In most of the region, local governments are entitled to shares of national taxes generated in their jurisdictions (origin-based tax sharing). The most important shared tax is usually the Personal Income Tax (PIT), which is also usually accounted for oicially as a Shared Tax. The Property Transfer Tax is also often shared (100%) with local governments but is usually misclassiied as an own-revenue. In a few places, the recurrent property tax is shared between levels of government and in Romania a small fraction of the Corporate Income Tax is shared with regional governments. Own-Source Revenues: As in much of the world, data on local own-revenue is often poorly maintained and classiied. Own-revenues include locally imposed taxes; income from the sale or rental of municipal assets; ines, penalties, and interest; local user fees and charges; and fees for permits, licenses, and the issuance of oicial documents. Typically, the most important local tax is the Property Tax, though it is often not the single-largest source of own-revenue. Importantly, Montenegrin and Croatian municipalities can impose local surcharges on personal income tax, powers that are being considered in other countries/entities. In many places, the regulation of local fees and charges is weak, allowing local governments to use them as quasi-taxes. Particularly important in this respect are three fees inherited from the (Yugoslavian) past: the Land Development Fee, the Land Use Fee, and the Business Registration Fee (or Sign Tax). In most of the region however, the Land Development and Business Registration fees are being phased-out in the name of improving the local “business enabling environment”, while the Land Use Fee is being eliminated or constrained with the introduction or expansion of the Property Tax3. 3 These fees go under diferent names in diferent inheritor states of the former Yugoslavia. (NALAS) Network of Associations of Local Authorities of South-East Europe Overview of Local Governments in South-East Europe Number and types of sub-sovereign governments T able 1 presents the numbers and types of sub-sovereign governments in NALAS- member countries or entities. Bosnia and Herzegovina (BiH) is the most complicated and has four-plus levels of government: 1) The state of BiH 2) Two entities: Republic of Srpska (RS of BIH) and the Federation II of Bosnia-Herzegovina (FBiH of BiH) —plus the Brcko District; 3) Cantons in FBiH (BiH); and 4) municipalities in both entities, 80 in FBiH and 63 in RS. In FBiH, the entity level government is small and the cantons receive the lion’s share of public revenues and provide lion’s share of public services, at the cost of both the entity government and local governments. The inancial data used in the report for local governments in FBiH does not include the revenues or expenditures of Cantons. Table 1 Numbers and Types of Sub-Sovereign Governments4 Levels of SubNALAS Member Sovereign Government AAM/AAC 2 Albania 3 Bosnia Herzegovina 2 FBiH SOGFBIH ALVRS 1 RS NAMRB 1 Bulgaria UORH 2 Croatia AKM 1 Kosovo ZELS 1 Macedonia Moldova CALM 3 Montenegro 1 2 Counties; Municipalities/Communes Serbia Slovenia UMMo FALR, ACoR STCM SOG Types of Sub-Sovereign Government Counties; Municipalities/Communes Entities; Cantons; Municipalities Cantons; Municipalities Municipalities Municipalities/Communes Counties; Municipalities/Communes Municipalities Municipalities Autonomous Province; Raions/Regions; Municipalities/Communes Municipalities 2 1 Turkey MMU 4 Autonomous Provinces; Municipalities Municipalities Special Provincial Administrations; Metropolitan Municipalities; District Municipalities ; Village Administrations Romania Number of Municipalities 373 143 80 63 264 556 38 81 4 In 2015, Albania consolidated 373 municipalities and communes into 61 municipalities and the RS (of BiH) added a 64th municipality. 5 The inancial data refers to 21 municipalities 898 235 3,181 145 212 1,395 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 15 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Albania and Croatia both have democratically elected county level governments. In Albania, the qarks play a very limited role while in Croatia zupanije are more important, though both are small compared to the municipal sector. The situation in Moldova is more ambiguous. Moldova has three levels of sub-sovereign government: 1) The autonomous province of Gaugazia 2) raions or regions, and 3) communes and municipalities. Raion heads are indirectly elected by raion councils but operate under strong central inluence. They also exercise signiicant control over the budgets of municipalities and communes. This blurs the distinction between 1st and 2nd-tier governments in Moldova, as well as the distinction between local governments and the territorial arms of the national government. Because education and other social sector functions are still at the raion level, Moldova appears to be a highly decentralized small state but in fact remains quite centralized. 16 Romania has two levels of sub-sovereign government, communes and municipalities on the one hand and counties or judets on the other. Judets play a more important than their counterparts in Albania or Croatia, particularly because of their role in healthcare. Nonetheless, communes and municipalities are the iscally weightier level of government. In the report, the local revenue and expenditure data for Albania, Croatia, Romania, and Moldova includes both communes and municipalities, and 2nd-tier local governments at the county or regional level. (NALAS) Network of Associations of Local Authorities of South-East Europe Serbia has two levels of sub-sovereign government: 1) provincial and 2) municipal. The inancial data in the report is only for municipalities. Turkey has four levels of sub-sovereign government: 1) Special Provincial Administrations (SPAs) 2) Metropolitan Municipalities 3) District municipalities and 4) Village Administrations. Both types of municipalities are considered 1st tier local governments, but they have diferent functions. Recently, the boundaries of many Metropolitan Municipalities were expanded to near provincial proportions, with the larger towns within the expanded jurisdiction becoming District Municipalities (with diminished authority). The 51 democratically-elected SPAs function alongside the territorial arms of the national government at the regional level and deliver a few public services (e.g. water) primarily in rural areas. The data in the report includes the revenue and expenditures of SPAs. (NALAS) Network of Associations of Local Authorities of South-East Europe The Average Population of Municipal Governments The average population of municipal governments difers signiicantly across South-East Europe. As can be seen from Chart 1, Moldova has the smallest municipal governments, averaging less than 4,000 inhabitants. Municipalities in Romania, Croatia, Albania and Slovenia are also relatively small, averaging less than 10,000 inhabitants6. Nonetheless, the average size of municipalities in the region is signiicantly larger than the average for the EU. Chart 1 Thus, while jurisdictional fragmentation in some parts of South-East Europe may present obstacles to decentralization (the high administrative costs, weak tax bases, and human capital shortages associated with small local governments) it is hard to argue that size alone accounts for the limited progress the region as a whole has made towards EU-levels of iscal decentralization. Indeed, the average size of municipalities in many parts of the region is quite high. Average Population of 1st Tier Local Governments 17 6 In 2015, Albania reduced the number of its local governments from 373 to 61, increasing the average population of municipalities to over 45,000. Bulgaria also added a municipality in 2015. The averages in Chart 1 are for the status quo in 2014. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 A more plausible causal force working against decentralization is the relatively high percentage of the total population living in capital cities. As can be seen from Chart 2, in most NALAS-member countries/entities much higher shares of the population reside in capital cities than is the norm for the EU. The oversized importance of capital cities in the region skews economic activity towards a single metropolitan area. This creates technical and political obstacles to decentralization. Technically, it is diicult to assign local governments robust own-revenues or to create eicient equalization mechanisms when a disproportionate share of the tax base “originates” in a single city. And politically, the struggle of ruling parties to control both the national government and the capital city often complicates eforts to redistribute public revenues to poorer local governments. Chart 2 The Dynamics of the Gross Domestic Product Chart 3 presents GDP per capita for all NALAS countries and entities in 2006 and 2014, as well as their cumulative growth rates for the period. There is considerable variation across the group in both relative wealth and GDP growth. Moldova has the lowest per capita income in both 2006 and 2014, and is almost eleven times poorer than Slovenia, itself a 30% poorer than the EU average. Nonetheless, Moldova grew the fastest over the period while Slovenia and Croatia –the wealthiest of the group—have grown slowest. Some of the variation in economic performance can be explained by the diferent ways the countries and entities of the region experienced the economic crisis of 2008-2009. As can be seen from Chart 4, Slovenia and Croatia were hit hardest by the crisis and have taken the longest to recover. Indeed, Croatia and Serbia have yet to return to growth. Percentage of Population Living in Capital Cities 18 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe The crisis hit Moldova, Romania, Montenegro, Turkey and Bulgaria all pretty hard. But Moldova and Turkey rebounded swiftly while the recovery in Montenegro, Romania and Bulgaria has been slower. Interestingly, growth was slow Chart 3 but essentially positive for the entire period in Macedonia, Albania and Kosovo, suggesting their limited integration with the world economy and shocks. GDP Per Capita in 2006 & 2014 and GDP Growth 2006-14* 19 *Eurostat Chart 4 GDP Growth in the Crisis of 2008-9; between 2009 and 2013; and in 2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 III Basic Indicators of Fiscal Decentralization T he most straight-forward indicators of the relative importance of local governments in a country’s governance structure are local revenues and expenditures as shares of total public revenues and expenditures, and as a percentage of GDP. Their signiicance, however, depends on both the functions that local governments are responsible for, and the overall size of a country’s public sector. 20 If the total public sector is small, it is unlikely that local government revenues or expenditures will represent a large share of GDP. They may, however, represent a substantial share of total public revenues or expenditures. This would suggest that all levels of government have trouble collecting taxes, but that local governments are considered important and that the national government is trying to provide them with adequate inancing. If, however, the public sector is large, and local government revenues and expenditures are small, both as share of GDP and total public revenues and expenditures, then this suggests that local governments have not been assigned signiicant responsibilities and/or the national government is underfunding them. To make reasonable judgements about the role of local governments in a given polity it is important to know what functions they have been assigned, and in particular whether they pay the wages of teachers, doctors or other social sector employees. (NALAS) Network of Associations of Local Authorities of South-East Europe This is because the wage costs associated with education, health and to a lesser extent, social welfare services are so big that they inevitably change the nature and intergovernmental relations. For example, most OECD countries spend 12 to 20% of all public revenue or 3 to 6% of GDP on pretertiary education, of which between 60 and 80% goes to wages7. As a result, assigning important social sector functions to local governments fundamentally alters the nature of intergovernmental iscal relations. In short, if the full costs of running schools or hospitals are devolved to local governments, then they must be given large grants by the national government because there is no way that these services can be inanced for locally raised revenues. Equally important, they cannot reasonably be inanced by shared taxes. This is because bases of the robust tax that might be generate the needed revenue –Personal and Corporate Income Tax—are highly skewed towards a limited number of jurisdictions, but the services that need to be inanced are everywhere. Worse, the costs of providing many of those services actually go up in the poorest places –think small schools in rural setting. Table 2 summarizes the social sector functions assigned to local governments in the region. As can be seen from the Table in Kosovo, Romania, Macedonia, Bulgaria and Moldova, and Romania local governments pay the full costs or pre-tertiary education. In Kosovo, local governments also maintain pri7 See Education at Glance, OECD Paris 2013, pp 193, 218, 240-48. (NALAS) Network of Associations of Local Authorities of South-East Europe mary health care clinics and pay the wages of some doctors and nurses. Similarly in Romania, local governments pay for most of the costs of primary and secondary health care. By all rights, local governments in these countries should have higher revenues and expenditures as shares of both GDP, and of total public revenues and expenditures. Local governments should also be receiving very large shares of their revenue from Conditional Transfers because without them they cannot pay for the schools, hospitals, and other social sector institutions that they have been tasked with managing. Conversely, local governments in Albania, FBiH (of BiH), Montenegro, and Turkey do not pay the wages of any social sector employees. Indeed, local governments in Montenegro and Turkey have no responsibilities in either education or health, not even maintaining facilities. Here, local government revenues as shares of both GDP and of total public revenues should be lower, as should the share of conditional transfers in their budgets. Table 2 Local Government Social Sector Functions* Preschools Primary Schools Secondary Schools Primary Health Buildings Wages Buildings Wages Buildings Wages Buildings Wages Kosovo XX XX XX XX XX XX XX XX Romania XX XX XX XX XX XX XX XX Macedonia XX XX XX XX XX XX Bulgaria XX XX XX XX XX XX Moldova XX XX XX XX XX XX Serbia XX XX XX Slovenia XX XX XX Croatia XX XX XX XX Albania XX XX XX XX FBIH (BIH) XX XX RS (BIH) XX XX XX XX Secondary Health Buildings Wages XX XX 21 XX XX Montenegro Turkey *In some places some social sector functions are provided by 2nd tier local governments but they are included here for those members of the group for which our financial data covers both levels. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Local Governments Revenues in South-East Europe Chart 5 shows the revenues of the General Government – total public revenues— for each NALAS-member country or entity, as well as the average for the EU and the region (SEE) as percentage of GDP. Local revenue is distinguished from other General Government revenue to indicate the relative size of the local sector in the total public sector. As can be seen from Chart 5, local government revenue in Albania is lowest as a share of GDP. But it also has the smallest General Government. The EU28 is at the other end of the spectrum, with both General and Local Government revenue highest as shares of GDP. Everyone else is in the middle. So on average, the countries of the EU have both larger public sectors and have decentralized more revenue to local governments than their counterparts in South-East Europe. 22 Chart 5 But within the region there is also a lot of variation. Albania, Kosovo, Macedonia, Romania and Bulgaria all have public sectors that generate less than 30% of GDP in revenue. This suggests they all have problems with tax collection. Meanwhile, Montenegro, Serbia, BiH, Croatia and Slovenia have public sectors that approach 40% of GDP, suggesting a greater capacity to tax. There is also fair amount of variation in the relative size of the local government sector, a variation which is more important for our purposes here. Chart 6 suggests what this variation means with respect to iscal decentralization in the region. It plots local government revenue as a share of GDP against local public revenue as share of total public revenue. As we can see from the Chart, the NALAS members whose local government sectors most closely resemble those of the EU28 as both percentages of GDP and total public revenue are Moldova (MD), Romania (RO), and Kosovo (RKS). They are followed by Bulgaria (BG), Macedonia (MKD), Croatia (HR), with the rest of the pack further away from the average for the EU. General and Local Government Revenue as a Percentage of GDP in 2014 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe At one level this makes sense: As is often the case in the EU, local governments in Kosovo, Romania, Moldova, Macedonia, and Bulgaria are all responsible for pre-tertiary education. It is thus not surprising that their local governments represent larger shares of the total public sector than those of their counterparts elsewhere in the SEE, or that their local governments require larger shares of their respective GDPs to inance these social sector responsibilities. Chart 6 At the same time however, it is a little curious that four out of the ive most decentralized countries or entities in the SEE are also four of the ive whose public sector revenues are equal to less than 35% of the GDP. Moreover, the ifth is Moldova, whose public sector is also small and equal to well under 40% of GDP. Local Revenues as a share of Total Public Revenues & GDP (2014) 23 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 In other words, within the SEE there seems to be a correlation between the extent of decentralization and countries with smaller public sectors, a correlation that probably does not hold up within the EU itself.8 Chart 7 below illustrates this correlation by plotting the overall size of the public sector in the SEE against a scale of decentralization based not on inancial data but on the degree to which social sector functions have been devolved to local governments (based on Table 2 above). Chart 7 As can be seen from the Chart, all countries or entities that have devolved responsibility for paying the wages of primary and secondary school teachers to local governments are in the quadrant in which the total size of the public sector is less than 40% of GDP. Albania and Turkey also have public sectors less than 40% of the GDP. But they have devolved no social sector wage responsibilities to local governments. Hence they fall in the quadrant for countries with small public sectors and where decentralization is limited to the provision of basic urban services. Public Sector Size and the Extent of Decentralization as Measured by the Devolution of Social Sector Functions* 24 8 With the exception of Switzerland, many of the most decentralized countries in Europe –particularly the Nordic ones— have public sectors signiicantly larger than the EU average. (NALAS) Network of Associations of Local Authorities of South-East Europe * The scale is based on Table 2 and has been created by assigning one point for the maintenance of the physical facilities of each type social sector institution that local governments pay for, and 2 points for the wages of each type of social sector function that local governments pay for. (NALAS) Network of Associations of Local Authorities of South-East Europe Meanwhile, all other members of the group fall in the quadrant with public sectors’ larger than 40% of GDP, but in which the devolution of social sector functions to local governments remains. Finally, and perhaps, most importantly, no member of the group is in the quadrant reserved for both high levels of decentralization and larger public sectors, the quadrant where most of the EU’s decentralized unitary states would be found. What is driving this apparent correlation between “small states” and high levels of decentralization is not clear. But to the degree that the correlation is real, it is worth pointing out that its seems to be the size and not relative wealth that is the more important factor: While Moldova and Kosovo are certainly among the poorest of the group when measured by GDP per capita (Chart 3), Macedonia is in the middle of the spectrum, and Bulgaria and Romania are at the higher end. One way to interpret this inding is that in the region countries that collect less taxes (and hence have smaller public sectors) are more inclined to decentralize social sector functions. This of course begs the question whether small public sectors are caused by the inability of central governments to collect taxes (weak state capacity) or their unwillingness to do so (policy preferences). But either way it seems hard to reject the idea that at least some of the decentralization in the region is being fueled by national governments ofloading social sector functions on local governments because they understand that they don’t have the revenues to fund these service at adequate levels. Suggesting, that this sort of of-loading is going on, however, is very diferent from making judgements about whether it is a “good” or “bad” thing: It is entirely possible that in some resource constrained contexts –and within limits— local governments do a better job delivering social sector services than central ones. And it is even more possible that in less resource constrained environments local governments could do a better job delivering public services than central governments are actually doing. But let us return to the ive highly decentralized countries in our group as measured by the fact that they have fully devolved very costly education functions to local governments (Table 2). As can be seen in Chart 6, the local share of total public revenues in Kosovo, Romania and Moldova is line with EU norms in 2014, despite the fact that their public sectors are comparatively very small (Chart 5). This suggests that despite their small public sectors, central governments here are treating their local governments reasonably fairly –call it “well-balanced dumping”. But the situation in Macedonia and Bulgaria is a little different. Here, local governments receive a substantially smaller share of total public revenues in comparison to both their highly decentralized counterparts in the SEE and the EU. As such, the central governments of Bulgaria and Macedonia seem to be underfunding local governments in general and their social sector functions in particular relative to the size of their public sectors. Indeed, they seem to be cases of what might called “unbalanced of-loading.”9 Unfortunately, however, “unbalanced of-loading” need not be limited to places where local governments have been assigned signiicant social sector functions. It can also be a feature of places where local governments remain responsible for “only” basic urban services. Here, Albania, with the smallest local government sector in terms of both GDP and total public revenues (2.8 & 9.4% respectively) stands out, followed closely by FBiH (of BiH) (4.0 & 9.7%) The situation is less clear in Turkey, Slovenia, RS (of BiH), Serbia, Montenegro and Croatia. Local governments here all control revenues equal to between 5.3 and 7.0% of GDP and between 14 and 18% of public revenues. This is very low in comparison to the EU average. But in Turkey 9 It should be noted that in Moldova, Romania, and Macedonia local governments have very little control over the wage components of their education subventions and in many ways have just assumed the payroll function of the national government for these services. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 25 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 and Montenegro local governments do not pay the wages of any social sector workers whereas in Croatia, RS (of BiH), Serbia, and Slovenia they pay the wages of preschool teachers, a signiicant expense, at least in urban jurisdictions with high enrollment rates. Chart 8 shows local government revenue as percentage of a GDP in 2006, 2009 and 2014 for all NALAS. With the notable exceptions of Turkey, Macedonia, and Kosovo, local government revenues across the region have stagnated or declined since 2006. But while local revenues increased in Turkey without the devolution of new responsibilities, in both Kosovo and Macedonia they have risen signiicantly because over the period local governments became responsible for paying the wages of social sector workers. Finally, it is important to recognize that our speculations about “balanced and unbalanced” of-loading are complicated by the fact that we do not have data about the allocation of revenues across local governments. This is important because in some places local revenues They have fallen most in Montenegro, RS (of BiH), Serbia are skewed toward large municipalities in general and and Moldova. In Montenegro, a real-estate boom pushedcapital cities in particular. As a result, the relatively up local revenues before 2006. But their fall has been made comforting macro-picture suggested by high local rev- harder by national government restrictions on municipalities’ enue in comparison to total public revenue may be mis- right to tax businesses (in the name of improving economic leading at the micro-level because of the overfunding “enabling environment”). Similar things could be said about RS of the few against the interests of the many. (of BiH), Croatia, and Serbia where national governments also made some cuts in grants in response to the Great Recession. Chart 8 Local Government Revenue as a Share of GDP in 2006, 2009, 2014 26 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Chart 9 shows the per capita revenues of the consolidated public sector and of local governments in EUR in 2014. The Chart is a useful reminder of how little revenue the governments of most of South-East Europe have to work with. It also shows how much variation there is across the region in the “relative weights” of the local public sector. It is particChart 9 ularly striking that local governments in Moldova, Kosovo, and Macedonia pay for teachers’ wages on per capita revenues of less than 250 EUR, while Croatian and Slovenian municipalities bear little of these costs and have per capita revenues 3 to 4 times higher. Consolidated Public and Local Government Revenue (EUR Per Capita-2014) 27 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 The Local Fiscal Autonomy and the Basic Composition of Local Revenues Our data on the composition of local public revenue is problematic because diferent places account for different revenues in diferent ways, and because in some places accounting classiications have changed over time. The classiication of shared taxes is particularly problematic. In most places, only shared PIT is considered a Shared Tax, with shared Vehicle Registration and Property Transfer Taxes misclassiied as Own-Revenues. Because these are important sources of revenue in many places, this misclassiication signiicantly overstates the iscal autonomy of local governments. In Turkey, some shared PIT revenues are accounted for as Unconditional Transfers while in Slovenia some Unconditional Transfers are accounted for as shared PIT. MeanChart 10 Composition of Local Revenue in 2006 28 (NALAS) Network of Associations of Local Authorities of South-East Europe while in Croatia, some of what is accounted for as shared PIT should be recorded as an own-source revenue because it comes from locally imposed surcharges on personal income and not just from the centrally set shares. Finally, in most places we cannot separate Conditional Grants for speciic investments or programs from Block Grants for social sector functions. Despite these shortcomings, however, the data is still informative. Chart 10 and 11 show the change in the basic composition of local revenue between 2006 and 2014 as an average for all NALAS members. The share of Unconditional Grants as a percentage of total revenue has remained stable over the period. But Own-Revenues and Shared Taxes have declined, while Conditional Transfers have increased as from 16 to 26% of the total. Local borrowing has also increased, but remains extremely low. Chart 11 Composition of Local Revenue in 2014 (NALAS) Network of Associations of Local Authorities of South-East Europe Charts 12 and 13 present the same information for individual members of the group ordered by local governments’ share in total public revenues. They help explain what is driving this increase in Conditional Grants. Reading from left of Chart 13 (for 2014), we ind Kosovo, Romania, Moldova, Macedonia and Bulgaria. They are alChart 12 Composition of Local Government Revenue 2006 ready familiar to us as the ive places that have devolved the most signiicant social sector functions to local governments. And not surprisingly, they are the ive places where local government revenues are now highest as a share of total public revenues. We can also see that the revenues of all ive are dominated by Conditional Grants, with much less coming from shared taxes and own-sources. Chart 13 Composition of Local Government Revenue2014 29 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 This is very diferent than situation in 2006. Kosovo and Macedonia have moved from the far right of the chart to the far left: In fact, between 2006 and 2014 they journeyed from being the least decentralized countries or entities in the region to being the most. And this journey transformed the structure of their revenues, which no longer come mainly from own-revenues and shared taxes but from conditional and unconditional transfers. Moldova and Bulgaria are also interesting in this respect. In neither country, were major new functions devolved to local governments between 2006 and 2014. Nonetheless, in both, Conditional Grants have increased at the expense of shared taxes. This is because early in their decentralization eforts both Bulgaria and Moldova devolved schooling to local governments, but tried to inance it through PIT sharing. 30 This proved disastrous for rural areas that had weak tax bases and hence low PIT revenues, but high education costs because of lots of small schools with small classes. Indeed, in both Bulgaria and Moldova these imbalances distorted their intergovernmental inance systems for many years. In 2008, Bulgaria completed a process of replacing PIT sharing with sectoral block grant that unlike PIT shares can be allocated according to objective measure of need, a process which Moldova began in 2014. This in shift towards Conditional Grants in Bulgaria, Moldova, Kosovo, and Macedonia explains the change in the composition of local revenues that we saw for the region as whole in Charts 10 & 11. It also nicely illustrates a well know paradox in intergovernmental inance: As countries devolve social sector functions to local governments, local governments typically become more inancially dependent on their national governments than before10. Or put more bluntly, decentralization often leads to a reduction in the “iscal autonomy” of local governments. 10 See Blochlinger and King, “Less than you thought: The Fiscal Autonomy of Sub-Central Governments” OECD, 2006 (NALAS) Network of Associations of Local Authorities of South-East Europe This paradox is important to appreciate because too often advocates of decentralization see the principle measure of success in terms of how much local governments inance themselves. This clearly is wrong-headed, at least to the degree that we think social sector functions should be devolved to local governments. Equally importantly, this paradox should disabuse us of the idea that the inancing of local governments can ever by truly separated from the inances of national governments. This is a bitter pill to swallow for local government oicials who want to be the masters of their own houses. But the fact remains, that local governments almost always need signiicant transfers from their national governments and the size of these transfers increases with the devolution of social sector functions. As a result, there will always be struggles over how much money local governments need to pay for these functions and the goal of reformers should not be to try to eliminate these struggles by making local governments “iscally autonomous”. Instead, the goal should be to civilize the struggle through informed intergovernmental dialogue that allows both sides to reach reasonable and dynamic compromises over time. Unfortunately, this commitment to intergovernmental dialogue is weak across much of the region. Here, however, it is also worth noting that while many countries have recognized that the inancing of major social sector functions is best achieved through the use of Block Grants and not shared taxes or own-revenues, it is also true that in many places central control over these Block Grants remains excessive: It is one thing for national governments to want to make sure that monies earmarked for health and education are actually spent on them, and quite another for them to control exactly how local governments use these funds within a sector. After all, the entire logic of devolving these functions is defeated if central regulations make it impossible for municipalities to use their knowledge of local conditions and needs to improve the efectiveness of how these monies are spent. (NALAS) Network of Associations of Local Authorities of South-East Europe But returning again briely to Charts 12 & 13, it is worth highlighting that of all members of our group, Montenegro has the highest share of own-revenues in total revenues. On the one hand, this is possible because Montenegrin municipalities have no social sector responsibilities. On the other hand it is because they have a particularly broad palette of own-revenues that includes not just the property tax, asset sales and rentals, land use and development fees, but most interestingly, local PIT surcharges. Nonetheless, between 2006 and 2014, Montenegro moved from being a country in which local governments received a fairly large pie of the iscal pay, to one whose share is now relatively modest. It is also worth noting that local governments in Turkey, Slovenia, Croatia, and Moldova receive no income from Unconditional Transfers. This absence of Unconditional Transfers raises questions about the equity of these countries’ intergovernmental inance systems because it is generally through such transfers that central governments provide additional money to poorer jurisdictions. Nonetheless, Chart 14 equalizing funds can be provided in many ways and the absence of unconditional transfers does not mean that no equalization is taking place. Indeed, Croatia, Slovenia and Turkey do at least some equalization through other mechanisms. Moreover, the simple existence of Unconditional Transfers tells us little about how efectively equalization is being carried out. In fact, in many places there is good reason to believe that they have only modest efects on the horizontal equity of their systems. But again we lack the data to systematically address this question. Finally, between 2013 and 2014 the revenue position of many local governments in the regions changed quite signiicantly. Montenegrin municipalities lost the most because of restrictions on Land Development Fee and the elimination of the Land Use Fee while gains elsewhere were due to a variety of factors. In FBiH (of BiH) and Slovenia, local governments saw their PIT shares increase; Bulgaria introduced a new national program of investment grants inanced largely by EU funds; and in Romania grants for healthcare services went up. Growth/Decline of Local Government Revenue between 2013 and 2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 31 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 The Composition of Own-Revenues and the Property Tax Unfortunately, the accounting of own-revenues difers substantially across the region. In some places it is quite detailed and contains more categories than are presented in our charts. Others use only two or three categories and it is diicult to say what they contain. For example, local govChart 15 Composition of Own-Revenues in 2006 ernments in Croatia, Turkey, FBiH (of BiH) and RS (of BiH) derive signiicant revenue from Land Development Fees and quasi-iscal Construction Permits, but record them as Communal Fees. Revenues from asset sales and rentals, fees for the use of public space, and ines and penalties are also lumped into this category. Meanwhile, Slovenia records revenue from the Property Tax and the Land Use Fee together. Chart 16 Composition of Own-Revenues in 2014 32 * More than 90% of what is recorded as property tax in Slovenia comes from the Land Use Fee. (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe These accounting issues make it hard to come to any general conclusions about the nature of own revenue in the group. But a few observations are worth making. Charts 15 & 16 present the composition of local government ownrevenues in 2006 and 2014 for all members of the group, ranked by the share of own-revenues in total revenues. Montenegro is the outlier with own-revenues equal to about 80% of total revenues in 2006 and 66% in 2014. Over the period, and without Montenegro, the average share of own-revenues to total revenues fell from about 38% in 2006 to 34% in 2014. Surprisingly, these shares of own revenue to total revenues are in line with the average for the OECD. The also almost Chart 17 Absolute Growth of Own Revenues 2006-2014 certainly overstate the real revenue raising powers of local governments in the region because of the misclassiication of many shared taxes and fees as own revenue11. But again, the point is that almost everywhere own-revenues are much lower than the literature suggests is optimal because of the grants and transfers that local governments receive –particularly, for social sector functions. Indeed, it is the decentralization of these functions in Kosovo and Macedonia that is largely responsible for lowering the share of own-revenues in total revenues for the group. But across the group there is much variation in both the growth of own revenues over the period, and their yield in EUR per capita. This can be seen in Charts 17 and 18. Chart 18 Absolute Growth of Own Revenues and their per capita Yield in EUR 2014 33 11 See again, Blochlinger and King. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 In Slovenia, the sharp decline in own-revenues was the result of slow economic growth combined with 2014 elimination of the Land Development Fee. The restriction or elimination of this fee also had a strong impact in Serbia and Montenegro. But in both sharp increases in property tax collection (c.300 and 400% respectively) have softened the blow. The stagnation of own-revenues in Albania is the result of the progressive restriction and ultimate elimination of a local tax on small businesses, as well as a moratorium on new construction that has reduced the yield of the Infrastructure Impact Tax. More generally, national governments throughout the region have been reducing local governments’ ability to tax businesses –a policy frequently supported by the World Bank and USAID. 34 In Macedonia, the sharp increase in own-revenues –from a very low base— has come largely from a 600% improvement in property tax collection. Growth in Turkey has also been driven by the almost doubling of property tax revenues, and by the substantial growth of revenue from asset sales and rental, an important and often underappreciated source of local revenue. Improved property tax collection has also helped in Kosovo, Moldova and Romania. Nonetheless, the lion’s share of growth in all three places has come from “Other” sources. Throughout the region national and local governments have made substantial investments in the technical infrastructure for property taxation. Donors have strongly supported these eforts which are almost always backed by recommendations from the International Monetary Fund, and the American literature on local inance which argues that the property tax is the single most appropriate tax to devolve to local governments12. And over the last ten years the tax has indeed become one of the two or three most important own-source revenues in most of the region. 12 The North American literature on iscal federalism stresses the importance of the property tax for local governments. The Nordic countries —the most decentralized unitary states in the world – however, base their local government inance systems on giving municipalities’ wide control over personal income tax rates. (NALAS) Network of Associations of Local Authorities of South-East Europe This growth is impressive. There is also room for improvement. Nonetheless, there are limits to how much the property tax alone can be used to bring local government revenues in line with their service responsibilities. This can be seen from Charts 19 and 20, which shows that though the yield of the tax has increased considerably, it still accounted for only 7.6% of total local government revenue in 2014. Moreover, as a share of GDP, the average for the region remains a modest 0.46%, up from 0.36% in 2006. To be sure, in the US and Canada, the property tax generates revenues equal to between 2.5 and 3.5% of GDP. But this is exceptionally high and there are very few countries that manage to get more than 2.0% of GDP out of the property tax. Indeed, the average for the EU is only 1.1%. It is also important, to recognize that in many countries a disproportionately high percentage of the tax comes from businesses (often over 70%). Finally, it is worth remembering –though the Charts do not show it— that within the SEE, own-source revenues tend to be disproportionally concentrated in capital cities and very strongly tied to the real-estate market through asset sales, land development fees, construction permits, the Property Transfer Tax and to a lesser degree the Property Tax. (NALAS) Network of Associations of Local Authorities of South-East Europe Chart 19 Property Tax as % of GDP and Total Local Revenue in 2006 Chart 20 Property Tax as % of GDP and Total Local Revenue in 2014* 35 * Data for Slovenia include revenues from the land use fee. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 The Composition of Expenditures and Investment Spending Chart 21 shows the composition of local government expenditures by economic type for each member of the group, as well as the average for the group as a whole (SEE); the average for the EU (EU28); and the average for the seven postcommunist countries that joined the EU in 2004 (NEWEU7)13. Chart 21 Composition of Local Government Expenditure in 2014 36 13 As with revenues, there are inconsistencies in the way expenditure data is reported. For example, some places treat capital transfers to public utilities as investment expenditures while others record them as subsidies which cannot be distinguished from transfers to individuals or grants to non-governmental organizations. Similarly, in many places debt repayment is not accounted for separately but included in the category “Other”. Czech, Estonia, Hungary, Latvia, Lithuania, Poland and Slovakia. (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Local governments in the ive members of the group that have decentralized social sector functions (*) spend a higher percentage of their expenditures on wages. This is to be expected. More surprising, is that local governments in most of South-East Europe spend a larger share of their budgets on investment than their counterparts within the EU. Moreover, investment rates would probably be higher if capital subsidies to municipal utilities were properly accounted for. Given the generally poor inancial condition of municipalities in South East Europe it is diicult to fully explain these high investment rates. Part of the explanation may lie in the fact that local governments in the region often pay for investments that elsewhere in Europe are inanced through utility tarifs. And part of it may be the greater decentralization of social sector functions within the EU, function whose high operating costs depress the share of expenditure going to investment. But for whatever reason, the diferences in the average investment rates for the three groups (SEE, EU28, and EU7) Chart 22 have been remarkably consistent over the last 8 years. This suggests that local governments in South-East Europe, like those of the EU7 are playing an extraordinary game of catchup, spending as much they can to modernize the run-down infrastructure they have inherited. In short, municipalities in South-East Europe seem to be working harder than their counterparts in most of the EU to build new infrastructure because they are spending higher proportions of their income on investment despite receiving signiicantly lower shares of public revenue measured either as a percentage of GDP or of total public revenue (Chart 6). But here too it is important to remember that we don’t know how much these rates are being driven-up by the spending of capital cities and other wealthier jurisdictions. Moreover, investment spending has been both volatile and falling in a number of the members of the group, most notably Montenegro, RS (of BiH), FBiH (of BiH), Serbia and Croatia. This can be seen in Chart 22, which presents local government investment spending in EUR per capita in 2006, 2009, and 2014. Local Government Investment in 2006, 2009, 2014 (EUR per capita) REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 37 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 In Kosovo, on the other hand, investment levels have not only risen but are surprisingly high in per capita EUR given that it is one of the poorest members of the group. Meanwhile, Albania, despite relative high shares of investment spending in total expenditure (Chart 21) and positive investment rates since 2006 is still spending less than 40 EUR per capita on local infrastructure. Finally, it is worth adding that EU Structural Funds have kept investment higher than would otherwise have been in Romania, Bulgaria and Slovenia. Indeed, in Bulgaria EU structural funds have accounted for almost all local government investment spending in recent years (see the country report for Bulgaria). Chart 23 Chart 23 shows the average public investment by level of government as shares of GDP for the period 2005-2014. As can be seen from the Chart total public spending for all members of the group —with troubling exception of FBiH (of BiH)— has been higher than the average for the EU28. And in most of the region, central government expenditure accounts for the lion’s share of total public investment, as it does in the EU. Nonetheless, local government investment as a share of total public investment has exceeded both the average for both the EU28 and the New EU 7 in RS (of BiH), Slovenia, Romania, Kosovo and Montenegro. Total Public Investment by Level of Government as shares of GDP (average 2006-2014) 38 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Chart 24 Investment Spending in Million EUR 2014 Chart 24 shows local government investment in million EUR for all members of the group in 2014. Again, it is a useful reminder of how much variation there is across the group in both relative wealth, and in local revenue and expenditure patterns. 39 But the situation elsewhere is not so positive. Local government investment as a percentage of GDP is extremely low in both Albania and FBiH (of BiH) and is not much higher in Macedonia and Serbia. The picture is better in Bulgaria, Croatia, Moldova, and Turkey. Nonetheless, local government investment as a share of GDP remains lower than the average of the new EU7. This seems lower than what might be reasonably expected, given the fact local governments in South-East Europe have huge deicits in basic urban infrastructure that can only be overcome through high levels of sustained investment. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Local Government Borrowing In most of the region, local government borrowing is a new phenomenon. In some places however, its development is being constrained by high levels of central government debt. Chart 25 shows total public debt and annual deicits of all NALAS-member countries/entities in relationship to the Maastricht Treaty’s guidelines for total public debt and annual budget deicits (less than 60% and 3% of GDP respectively). Chart 25 As can be seen from the chart, most members of the group have total public debt and deicit levels below the Maastricht limits. But Croatia, Slovenia, Albania and Serbia have exceeded the more important level for total public debt (though not as much as is the average for the EU) while Montenegro is very close to the line (58.5%). This is important because under Maastricht, total public debt includes the debt of both national and subnational governments (though not the debt of publically-owned but commercialized utilities). As a result, when total public debt is close to or above Maastricht limits, not only is there pressure to reduce overall borrowing but local governments compete with their national governments for “debt space”. Public Debt and Budget Deficits in SEE Region in 2014 40 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Chart 26 shows that the vast majority of this debt space is already taken up by the national government. Albania and Serbia are already above the Maastricht limits despite the fact that local government debt represents a negligible fraction of total public debt. Nonetheless, it is unlikely that the national governments of either country will look favorably on new subnational borrowing. Meanwhile, in Slovenia, Montenegro and Croatia, local government borrowing has been more substantial. But because total public debt is also at or above Maastricht limits, here too national governments are likely to constrain new local borrowing. Given the infrastructure deicits facing local governments across the region this is unfortunate and eforts should be made Chart 26 to ensure that municipalities in Albania, Slovenia, Montenegro and Croatia have at least some access to debt capital. In other members of the group, debt remains well below the Maastricht limits and local governments should confront fewer policy obstacles in borrowing. Data for Turkey includes unpaid liabilities to private contractors or government agencies. These equal at least half of the total. Public Debt by Level of Government as Share of GDP 41 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Chart 27 shows the increase in total outstanding local government debt between 2006 and 2014 in per capita terms. As can be seen in from the chart, local borrowing has increased substantially in all countries or entities in the region except for Moldova, Albania, Macedonia, Kosovo and Serbia. Growth has been particularly striking in Croatia, Montenegro, and the RS (of BiH). Not all of this growth has been prudent, however and it seems that during the recession some local governments at least in Montenegro and the RS borrowed less to build new infrastructure than to avoid making painful cuts in operating costs. In any case, some municipalities in all three countries are having trouble meeting their debt service payments. In many places, the overall adequacy and predictability of local government revenues will have to be increased if municipalities are to prudently incur debt. Indeed, given the dependency of local governments on transfers, the rules Chart 27 regulating intergovernmental inances need to be clear and stable if borrowers and lenders are to be conident that municipal governments will be able to pay of their debts. Local governments will also have to do a better job collecting own-revenues, particularly with respect to setting higher tarifs and then forcing utilities to collect them. Local governments will also have to radically improve their ability to prepare, plan, and cost-out complex, multiyear investment projects— particularly in the water and solid waste sectors. This sort of planning, however, requires money and time that many local governments in the region do not feel they have. Scarce investment funds tend to be spent on pay-as-you-build road projects and not on debtinanced, pay-as-you-use environmental facilities because planning roads is simpler; construction can be delayed if money runs out; and because the beneits are more likely to be visible to voters before the next election. Increase in Local Government Debt 2006-2014 (EUR per capita) 42 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Gender Mainstreaming in Fiscal Decentralization G ood governance aims to ensure policies and institutions put the needs and interests of citizens at the forefront. Yet, the fact citizens are not homogeneous but a rather diverse group whose needs are deined by gender, location, age, ethnicity, religion, requires governance practices and models that efectively take those diversities into an account. With the assignment of iscal, political and administrative powers, local authorities emerged as an important actor to promoting good governance and ensuring adequate representation of citizen’s needs and interests as means to ensure equality. Namely, the decentralization process has given local authorities the responsibility over key policy areas which have direct impact on the quality of citizen’s lives i.e. health, social welfare, infrastructure. Also an opportunity to use the proximity, increased participation and improved knowledge of individual/group priorities to tailor governance practices and models, policy planning and budgeting, in a way that better responds to citizens and their diversities. NALAS has long-standing commitment to promoting decentralization and gender equality in South East Europe. Through the Strategy on Gender and Youth, NALAS aims to directly contribute to the implementation of international and national gender equality commitments. With involvement as partner in the UN Women regional programme “Promoting gender responsive policies in South East Europe and the Republic of Moldova” NALAS also aims to work on opening venues for the systematic integration of a gender perspective into decentralization processes and local governance models. IV Indicators for Gender Responsive Budgeting in SEE As a irst step NALAS has introduced a set of indicators as part of this cycle of monitoring of the progress of the iscal decentralization in South East Europe to understand the extent of integration of gender into decentralization and governance. Following a training on gender responsive budgeting, with the support of UN Women – the United Nations Entity for Gender Equality and the Empowerment of Women, the members of the NALAS Task Force on Fiscal Decentralization developed four indicators to be monitored within this report for the irst time: 1) Level of adoption of a Law on gender equality; 2) Existence of policies for gender responsive budgeting for local/central level; 3) Degree to which the Budget Circular on central or local level integrates gender responsive budgeting (GRB) indicators and 4) Policy and budgets analysis of municipal programmes. The data collection was done through a survey illed by representatives of the NALAS Task Force on Fiscal Decentralization with the support of the NALAS Focal Points on Gender and Youth. The questionnaire was distributed to all NALAS members and the analysis is based on eleven responses and it should therefore be noted that they might provide a limited view on the situation in the diferent countries. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 43 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 The irst indicator explores the existence of a Law on Gender Equality/equal opportunities in the countries/entities where NALAS members operate in a form that requires gender mainstreaming in policies and budgets at central local/level. The inding is that that there is an adopted Law on Gender equality in eleven countries/entities and Romania is the only country that has not adopted a Law yet. In most of the countries that have adopted such law there is a special provision emphasizing that the local and regional governments and their local institutions and public enterprises are obliged to assess and evaluate the efects of their acts, decisions or actions on the position of women and men in their community. Such assessment is required in all stages of planning, adoption and implementation of legal acts, decisions and actions with a goal of achieving real equality between women and men. States have adopted Law on Gender equality (equal opportunities) which requires mainstreaming in policies and budgets at central local/level 44 YES Croatia Serbia Macedonia Turkey Montenegro Kosovo BiH Republic of Srpska BiH Federation of BiH Bulgaria Albania NO Romania Chart 28 Existence of a Law on Gender Equality/Equal Opportunities (NALAS) Network of Associations of Local Authorities of South-East Europe Regarding the second indicator on the existence of policies for Gender Responsive Budgeting at central/local level the indings say that only 27% of the countries have adopted such policy, while 73% have not yet approached that level. States have adopted policy for Gender responsive budgeting at central/local level YES Turkey Kosovo BiH Republic of Srpska NO Croatia Serbia Macedonia Romania Montenegro BiH Federation of BiH Bulgaria Albania Chart 29 Existence of Policies for Gender Responsive Budgeting at Central/Local Level (NALAS) Network of Associations of Local Authorities of South-East Europe Related to the third indicator, the Degree to which the Budget Circular (central/local) has been revised to include gender provisions (e.g. indicators) the majority of countries 80% do not have a revised budget circular, while only two countries, Turkey and the entity of Republic of Srpska in Bosnia and Herzegovina have conirmed such practice. States that have revised budget circular (central/local) to include gender provisions (e.g indicators) YES Turkey BiH Republic of Srpska NO Croatia Serbia Macedonia Romania Montenegro Kosovo BiH Federation of BiH Bulgaria Albania Chart 30 Budget Circular (central/local) includes gender provisions Croatia has emphasized that although systematic solutions to assess gender implications of policies and budgets do not exist, various research and studies of the situation of women in the labor market, political participation, combating violence against women, the media, education and other areas were carried out, including the irst scientiic research on “Perception, experiences and attitudes about gender discrimination in the Republic of Croatia”. The fourth indicator refers to the embedded practice of GRB Analysis as an essential step in identifying impacts of national and local policies. In 40% of the countries the local governments are undertaking gender budget analysis of their policies and 10% stated that it has happened on in very few areas or cases, while in the majority of countries such initiatives do not exist. LSG’s are undertaking gender budget analysis of local level policies YES NO Only a few Croatia Romania BiH Federation of BiH Serbia Montenegro Macedonia Kosovo Turkey Bulgaria BiH Republic of Srpska Albania Chart 31 Embedded Practice of GRB Analysis as an Essential Step in Identifying Impacts of National and Local Policies In Republic of Srpska (BiH), GRB is being partially applied in 8 local governments, and the process has started in another 12. Their gender responsive analyses take into account the gender component in the planning of the budgets and deinition of similar long term priorities, such as: support to victims of domestic violence, awareness campaigns on gender equality, small grants for non-governmental organizations that are focused on achieving gender equality and agriculture etc. The major reasons related to the absence of gender-responsive budgeting are lack of inancial resources, lack of speciic knowledge and skills, and the Law on Budget System of Republic of Srpska (BiH) which in fact does not recognize this obligation. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 45 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Findings and Recommendations The fact that eleven countries have adopted a Law on Gender Equality suggests that there is a high-level political commitment among decision-makers and institutions in South East Europe to advance gender equality at central and local level. The majority of those laws require institutions to integrate gender in planning and budgeting, which not only opens an important venue for integrating gender perspectives into decentralization processes, but also places an obligation on local governments for integrating a gender perspective into the areas of decentralized competence. 46 To efectively deliver on the political commitments, the local governments could irst beneit from an improved understanding what are the negative implications of local decisions and policies, of not taking into account women’s or men’s gender speciic needs on their performance and on the community. They could also see how gender blind distribution of public resources or budget allocations leads to ineicient use of resources, deepening of gender inequalities or widening of the existing gender gaps at local level. The local governments could inally beneit from support to strengthen institutional capacities and mechanisms to apply gender sensitive public policy tools such as gender responsive budgeting. According to the survey indings, gender responsive budgeting tools are currently applied only in Kosovo, Republic of Srpska (BiH) and Turkey. The extent of application or possible form of institutionalization at central/local level in South East Europe require further and more in-depth research. The reported small number of practices and formal policies to introduce gender responsive budgeting may suggest that the relationship between gender equality commitments and governance at local level could still be rather unclear to the local/central government actors or that such practices are unknown to the associations of local authorities. (NALAS) Network of Associations of Local Authorities of South-East Europe Where policies and practices exists, local governments and their associations could more actively engage in exploring and promoting their results as means to both encourage others and make the local actors more visible in the process of advancing gender equality. The fact that 40% of the countries stated that local governments are undertaking some kind of gender budget analysis of local level policies and 10% that have done gender responsive budgeting analyses suggests there is a political will to engage and understand the implications of local decisions and policies. What is lacking is acting more systematically. To facilitate the systematic integration of gender perspective into planning and budgeting Turkey and Republic of Srpska (BiH) have already revised the budget circulars to include gender equality provisions. However, from the survey data it can be noted that the majority of the countries have not undertaken such step yet. Future research should qualitatively look into the type of revisions of the budget circular and collect information on revision of any other document (such as Budget law or methodologies for strategic planning), which could help to detect alternative routes countries have used to integrate gender equality into local governance and decentralization processes. Building on those indings, with the support of its decentralization specialists and associations of local government, NALAS has an opportunity to explore the possibilities to carry out a comprehensive gender assessment on the gender perspectives of decentralization processes in South East Europe. Such assessment could provide more precise overview of the commitments, policies and practices in the region and could facilitate the development of network’s strategies to positively inluence those processes and the areas of decentralized competence from a gender perspective. (NALAS) Network of Associations of Local Authorities of South-East Europe Country Reviews of Fiscal Decentralization Trends and Developments V 47 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Albania Intergovernmental Finance System Albania’s Organic Law “On the Organization and Functioning of Local Governments14”, deines three types of transfers, shared-taxes (which have never been created), unconditional transfers and conditional transfers. The law does not deine how the size of unconditional transfers to local governments should be determined. Nor does it specify the formula that should be used to allocate them. It does however clearly state that iscal equalization should be the primary objective of unconditional transfers. 48 The legal regulation of the transfer system is based on the following rationales: (i) provision of adequate revenues to local budgets, in addition to local taxes; (ii) assistance to lower levels of government; and (iii) inancial equalization to compensate for inequalities between central and local governments (vertical) and among local government units (horizontal). The Unconditional Transfer was introduced in 2001 and provides local governments with funds to execute their exclusive functions. The size of the transfer was initially based on the historic cost of the services that were transferred to local governments and are not based on standardized measures of costs. Since 2002, both the size of the transfer and the formula used to allocate it have been repeatedly changed by amending the national governments Annual Budget Law. 14 In 2015, Albania reduced the number of it local governments from 373 to 61 and passed a new Organic Law. The new law however, does not fundamentally change the basic deinition of local government revenue or the legal rules governing the allocation of grants and transfers. These issue are expected to be address in a new law on Local Government Finance anticipated in 2016. (NALAS) Network of Associations of Local Authorities of South-East Europe In 2014, Municipalities and Communes received 91% of the Unconditional Transfer pool, while the regions or Qarks received 9%. The following factors are used to allocate the grant to municipalities and communes: (i) an equal lump sum payment for all communes and municipalities to ensure that even the smallest jurisdictions can pay their administrative stafs; (ii) 70% of the remaining pool distributed on straight per capita basis; (iii) 15% of the pool distributed to the communes on the basis of their surface areas; (iv) and 15% of the pool distributed to municipalities (with the exception of Tirana) distributed on a per capita basis after adjustment by special coeicients for iscal distress and mountainous terrain. Once the Unconditional Transfer has been calculated, then a separate set of calculations are made for those local governments whose total per capita revenues are more than 25% below or above the national average. Local governments whose total per capita income is less than 75% of the national average and whose grant is less than 91% of the of last year’s grant are then compensated for the diference. Conversely, local governments whose total per capita income is 25% greater than the national average and whose grant is more than 93% of last year’s grant must give up the diference to help pay for the compensation of the others. Finally, the law ensures that inal value of the grant should be no less the 2000 lek per capita for communes and 2960 lek per capita for municipalities, again with exception of Tirana. As a result, Tirana is probably the only capital city in the region that has a lower share of total local revenues than its share in the country’s population. Conditional transfers come from two sources. The irst is from appropriations from line Ministries that are allocated to local governments through decisions taken by the Committee of Regions. The second is from an increasingly large Regional Development Fund. Indeed, since 2009 conditional transfers have constituted the single largest source of local government revenue. The extensive use of conditional transfers has substantially reduced the iscal autonomy of local governments and has led to allegations that they are being allocated for political purposes which do not relect clear developmental goals. (NALAS) Network of Associations of Local Authorities of South-East Europe The Property Tax Until recently, the most important local taxes have been a tax on small businesses (SBT) and an Infrastructure Impact Tax calculated as a percentage of the value of new construction. Since 2006, the base of the SBT has been repeatedly reduced and in 2014 it was transformed into a shared tax that is now collected by the national government. At the same time, the yield of the Infrastructure Impact Tax has fallen because of a centrally imposed moratorium on new construction permits. Local Governments are responsible the administration of the property tax, including valuation, billing and collection. For a fee, they may also choose to use the IPRO as a billing agent, and a few have included property tax charges on the bills citizens receive for water and sewage services. Local governments may grant exemptions and abatements for poor families and for properties damaged in natural disasters. They can also deny citizens services for non-payment, but this is rarely done. The recurrent Property Tax is regulated by the Law on the Local Tax System. The tax is a local government revenue and local governments are responsible for its administration and collection. Between 2006 and 2014 the yield of the property tax declined from 0.29% of GDP to 0.18% of GDP. In 2014, the tax represented about 15% of own revenue and about 7% of total revenue. The national government is currently considering reforms to expand the base of the tax and to move assessment closer to a market values. 49 Agricultural land and urban buildings are subject to the tax but urban land is not. Valuation is established through square meter rates set by the national government. These rates are diferentiated according to three types of local governments, with separate values for commercial and residential properties. Base valuations are adjusted for the age of the building, but not for location within the municipality. Until 2006, local governments were allowed to adjust these rates by up to 30%. Only owners of properties are liable for the tax. Properties owned by national and local governments are exempted from the tax, as are properties owned by religious institutions. Agricultural land used for orchards and vineyards are also exempted from the tax for the ive years after their initial planting. The national government is responsible for maintaining a national cadaster of all properties through its Immovable Property Registration Oice. In theory, the IPRO has the legal obligation to refuse the registration of any property that has outstanding property tax liabilities. In practice this is rarely done. About 85% of agricultural land is registered with the IPRO, but only 25% of urban buildings. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of Local Government Finance in Albania 2006-2014 Albanian local governments receive less revenue as both a share of GDP and of total public revenue than all their counterparts in the region. Worse, this share fell from a high of 3.2% of GDP in 2008 to 2.5% in 2014, still well below the level achieved in 2006. They also receive the lowest share of total public revenues in the region. Chart 32 Albania Local Government Revenue as a Share of GDP and Total Public Revenue 2006-2014 50 Between 2007 and 2014, local government revenues fell faster and rose slower than the revenues of the national government. This suggests that the national government is not committed to sharing the beneits and burdens of economic growth with local governments. Chart 33 Albania Revenue Fluctuations of the General Government and Local Governments 2006-2014 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe A very high but unstable share of local revenue comes from Conditional Transfers, subjecting local budget planning to large degrees of uncertainty and political patronage. Chart 34 Albania Composition of Local Government Revenue 2006-2014 51 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 After peaking in 2008, local revenue in per capita EUR fell steadily through 2013, before increasing slightly in 2014. Ownrevenue fell because of progressive reductions in the base of the Small Business Tax and because of a moratorium on new building permits that reduced revenues from the Infrastructure Impact Tax. Chart 35 Albania Composition of Revenues 2006-2014 (EUR Per Capita) 52 Despite the low share of local government revenues in total public revenues, investment spending as a share of total local expenditures has been relatively high. This is due in part to the high share of Conditional Transfers in the system. Chart 36 Albania Composition of Local Government Expenditures 2006-2014 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Wages, Investment, and Property Tax revenues are all low as shares of the GDP. Chart 37 Albania Investment, Wages, and Property Tax as Share of GDP 2006-2014 53 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Bosnia and Herzegovina Federation of Bosnia and Herzegovina osnia and Herzegovina (BiH) has an extremely complex administrative and iscal structure. According to the preliminary results of the recent census of October 2013, BiH has a population of 3,791,662, a decline of 13% in comparison with 1991. In the Federation of Bosnia and Herzegovina (FBiH) there are 2,371,603 people and in the Republic of Srpska (RS) 1,326,991. 7.7% of the population (291,422) lives in Sarajevo, the capital of BiH, which is also the capital of the FBiH.15 B In FBiH, the entity’s share of indirect taxes is allocated to municipalities and cantons according to ixed percentages set in the Law on the Allocation of Public Revenues. These shares are given to municipalities and cantons as Unconditional Transfers and are allocated by formula. The main criteria for allocating the transfer is population (70%). But there are other coeicients for surface area, school age children and relative wealth —as measured by the yield of the Personal Income Tax— that have equalizing efects. In 2014, the Unconditional Transfer constituted about 27% of municipal revenues. The Intergovernmental Finance System In recent years, the Unconditional Transfer has fallen because of the rules governing entity debt. These rules require that debt service payments to foreign creditors be paid directly and immediately from each entities’ share of indirect revenues. As a result, the pool of revenues that would otherwise go to cantonal and municipal governments is automatically reduced by the debt service payments of the Federation government. Because 60% of all indirect taxes are earmarked for cantons and municipalities they are effectively paying 60% of all debt incurred by the Federation. 54 Despite its size, Bosnia and Herzegovina has three almost separate iscal systems: FBiH, RS, and the Brcko District. Indirect taxes are the most important source of revenue for all levels of government. They are collected by the State of BiH and then divided between the State of BiH, the two entities –FBiH and RS—and the Brcko District according to a formula stated in the Law on Indirect Taxation in BiH. The allocation of indirect taxes within each entity, as well as the regulation of direct and other indirect taxes are regulated by entity legislation. 15 Unfortunately, these census results are still preliminary. Lack of an accurate census has impeded policy development at all levels of government including the adoption and adjustment of rules governing the allocation of grants and transfers. http://popis2013.net/index. php?docid=1042 (NALAS) Network of Associations of Local Authorities of South-East Europe During the economic crisis of 2009, FBiH took a loan from the IMF for over 250 million EUR to inance current expenditures. Now the entity government must pay back the loan. But because much of the debt-service cost is being born by cantons and municipalities, they have seen their revenues from indirect taxes fall substantially, despite an overall improvement in the economy. To address this problem, the FBiH Parliament created a Fiscal Coordination Body in 2104. (NALAS) Network of Associations of Local Authorities of South-East Europe This new institution will be responsible for determining the status of the entity’s debt obligations and for taking measures to ensure that debt service payments can be met in fair and equitable way. The Fiscal Coordination Body includes the Federal Minister of Finance, all cantonal Ministers of Finance and a representative of the Association of Municipalities and Cities of the Federation of BIH. The Law on the Allocation of Public Revenues also requires cantonal governments to share a speciied percentage of PIT with their municipalities on an origin basis. In 2012 the FBiH Parliament increased the share of PIT that cantons must share with municipalities from 34.46% to 41% in all cantons except Sarajevo Canton. Municipalities within Sarajevo Canton were also give the right to receive their share of indirect taxes directly from the entity government, and on the same basis as municipalities in other cantons. In 2014, about 17% of local government revenue came from shared taxes. Another other 17% comes from conditional grants which municipalities receive from either the entity or, more frequently, the cantons. Most are for speciic investment projects Since 2008 about 40% of local government revenues come from own sources, principally land use and land development fees. Unfortunately, there is no federal level accounting of local government own revenues and data about the nature, type and composition of these revenues are accounted for diferently in each canton. The Property Tax In the FBiH (of BiH), the recurrent property tax is regulated by the ten cantonal governments and there is no entitywide legal regulation of the tax. As a result, FBiH (of BiH) has the highest number of property tax laws in the region. In all cantons the tax is a cantonal levy, regulated and administered by cantonal authorities. In some cantons, however, its yield is shared with municipalities. Municipalities do not play an active role in levying the tax and its revenue potential is not a major concern of authorities at any level of government. Between 2006 and 2014 the yield of the tax (including the Property Transfer Tax) decreased from 0.42% of GDP to 0.34% of GDP. In most cantons, only buildings are subject to the tax. And in most, owner-occupied structures are tax exempt, including commercial properties used by their owners. Only owners are liable for the tax. This has created serious administrative problems because war and migration has left the ownership of many properties unclear. There are also exemptions for properties used for diplomatic and consular services, religious purposes and for the provision of public goods. War Veterans and their families are also exempt from the tax. Valuation is done according to square meter rates set in cantonal law, and cantonal legislatures set property tax rates. So far eforts to harmonize property tax legislation across the Federation or to decentralize it to municipalities level have failed. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 55 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of Local Government Finance in FBiH (of BiH) 2006-14 Local government revenues in FBiH (of BiH) peaked as a share of both the GDP and public revenues in 2008. Since then, they have declined substantially despite the fact that the economy has slowly recovered and total public revenues as a share of GDP have increased. Local government revenues as a share of total public revenues have fallen from 13% to under 10% since 2008. Chart 38 FBiH (of BiH) Local Government Revenue as Share of GDP and Total Public Revenue 2006-2014 56 Local government revenue fell much faster than that of the general government during the economic crisis of 2008-2009. The have also recovered more slowly suggesting that a disproportionate share of the burden of the downturn was placed on local governments. Chart 39 FBiH (of BiH) Annual Fluctuations in the Revenues of the General Government and Local Governments (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe All categories of local government revenue declined in 2009 in the face of the global downturn and –with the exception of own revenues- have yet to return to pre-crisis levels despite an improvement in the overall inancial picture of local governments in 2014. Chart 40 FBiH (of BiH) The Composition of Local Govrnment Revenue 2006-2014 (EUR Per Capita) 57 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Wage spending remained stable despite the economic downturn. Investment, however, fell from a peak of 37% of total expenditure in 2008 to 24% in 2014. A high share of expenditure consists of subsidies to utilities, grants to NGOs and transfers to individuals. Chart 41 FBiH (of BiH) The Composition of Local Government Expenditure 2006-2014 58 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe The yield of the property tax –which in some cantons is a local tax but in most remains controled by cantonal governments—is low and has fallen since 2008. The accounting of local government debt remains problematic, but it remains under 1% of GDP. Chart 42 FBiH (of BiH) Local Government Wages, Investment, Property Taxes as % GDP 2006-2014 59 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Bosnia Herzegovina – Republic of Srpska 60 s in most of the region, Republic of Srpska (RS of BiH) went into recession in 2009. Weak growth returned in 2010 and 2011 only to be followed by a second downturn in 2012 and the resumption of very modest growth in 2013 in 2014. The poor performance of the economy has negatively impacted local inances. A to four criteria: the total per capita revenues of registered businesses (35%); the per capita budgetary revenues of the municipality in the previous year (25%); population density (20%); and the unemployment rate (20%) Finally, municipalities are eligible for conditional grants from the entity government, most of which are for investment. Conditional grants accounted for between 5 and 10% of local budgets for most of the 2006-2014 period. The Intergovernmental Finance System Municipalities derive own revenue from property taxes, a local business registration tax, a hotel tax, land use and land development charges, other communal fees, asset sales and rentals, fees for the issuing of oicial documents, and interest, ines and penalties. Unfortunately, the available data for own-revenue is poor and most of it is accounted for under the title “Communal Fees and Charges”. Local governments in Republic of Srpksa (RS of BiH) derive their revenue from an Unconditional Transfers, Shared Taxes, Conditional Grants and Own Revenues. Since 2006, the size of the Unconditional Transfer has been set as a percentage of the entity’s share of indirect taxes (24%) and allocated by formula. 75% of the formula is allocated on a per capita basis, 15% on the basis of the territory of the municipality, and 10% the basis of the students in secondary schools. While the share of indirect taxes used to fund the transfer has been stable, the formula for allocating it has been repeatedly changed. The Unconditional Transfer accounted for between 50 and 60% of local revenue between 2006 and 2014. Municipalities also receive 25% of the Personal Income Tax (PIT) generated in their jurisdictions. These revenues are freely disposable and have accounted for between 6 and 12% of local budgets since 2006. There is also a Transfer for Underdeveloped and Extremely Underdeveloped municipalities. The amount of this Transfer is set in the annual budget law and allocated according (NALAS) Network of Associations of Local Authorities of South-East Europe In 2011, revenues from the business registration tax were signiicantly reduced by amendments introduced to the Law on Business Registration and a court decision made it impossible for municipalities to collect land use fees from certain tax payers. Between 2006 and 2014, own revenues have constituted between 25 and 35% of total local government revenues. (NALAS) Network of Associations of Local Authorities of South-East Europe The Property Tax In 2012, the entity government introduced a new law on Property Taxation. This Law centralized property tax registration and collection at the entity level, but left rate setting at the local level. Problems with establishing the central cadaster led to a drop in the yield of the tax in 2012, but collection has improved since. The law has been amended twice. The most recent amendments introduced a two-rate system of taxation: Up to 0.10% of market value for commercial properties used for production purposes, and up to 0.20% of market value for all other properties. Real-estate owners are required to ile property tax registration with the regional oices of the entity’s tax authority when they acquire or begin to use properties. These forms describe the physical attributes of their properties, including any improvements. Local governments forward these forms to the entity’s Tax Administration Oice, along with information on the property tax zones they have established, and their tax rates. The Tax Administration Oice determines the square meter value of all properties, applies the selected rates and exemptions, issues bills and collects the tax. Taxes are paid in two installments and the irst cannot be less the 50% of the liability. Owner-occupants receive exemptions equal to the value of 50 square meters of residential real-estate, as well as 10 square meters for every family member living with them. Property used for diplomatic and consular services, religious purposes, as well as properties owned by all governments in BiH are exempt from taxation. Tax payers can appeal their valuations to the Tax Administration Oice within 15 days of receiving their bills. The entity-level Authority for Geodetic and Property Afairs has to give municipalities and the Tax Administration Oice unrestricted access to its real-estate data. The yield of the property tax has fallen as percentage of the GDP from 0.33% in 2007 to 0.26%. It currently constituted about 10% of local government own revenue and 4% of total local revenue. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 61 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of the Finances of Local Governments in RS (BiH) Local government revenue as a share of GDP declined from a peak of 8% in 2007 to 6.1% in 2014. Local revenue as a share of total public revenue fell from 20% in 2007 to 14% in 2014. In short, the inancial positon of municipalities in RS (of BiH) has deteriorated quite substantially. Chart 43 RS (of BiH) Local Government Revenue as Share of GDP and Total Public Revenue 2006-2014 62 Local government revenues have declined faster and risen slower than the revenues of the entity government suggesting that the entity government has placed a disproportionate share of the burden of economic downturns on local governments. Chart 44 RS (of BiH) Annual Fluctuations in the Revenues of the General Government and Local Governments (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Municipalities are heavily dependent on the Unconditional Grant, which now constitutes 50% of their revenues. Own Revenue declined as a share of total revenue between 2006 and 2012, but have since recovered to pre-crisis levels. In part, this is due to an improvement in property tax collection which fell dramatically when the entity government recentralized billing, valuation and collection in 2010. Chart 45 RS (of BiH) Composition of Local Government Revenues 2006-2014 63 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Over the last few years, investment as share of total expenditure has fallen while spending on wages has increased. Like their counterparts in FBiH (of BiH), RS (of BiH) municipalities spend signiicantly on subsidies to municipal utilities, grants to NGOs and transfers to individuals. Until recently, Conditional Grants played a marginal role in the system. Chart 46 RS (of BiH) Composition of Expenditures in 2006-2014 64 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe The overall deterioration of the inancial position of local governments in RS (of BiH) can be seen in the sharp drop in investment spending between 2006-2014. Wages have also declined while property tax revenue have increased but still remain under 0.5% of GDP. Chart 47 RS (of BiH) Investment, Wages, and Property Tax as Share of GDP 2006-2014 65 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Bulgaria The Intergovernmental Finance System Prior to 2003, the Bulgarian intergovernmental inance system was heavily dependent on sharing personal income tax with local governments on an origin basis. This system was problematic because municipal governments were expected to cover the full costs of pre-tertiary education from the yield of their PIT share, something that was virtually impossible for many local governments to do given the weakness of their tax bases. 66 In 2003, amendments to the Municipal Budgets Act (MBA) introduced a clearer division of the responsibilities for inancing local government own and delegated functions. These changes were accompanied by the development of a set of block grants for delegated functions –principally education—based on centrally determined service costs. At the same time, PIT sharing was phased out, and the Education Block Grant became the largest single source of local government revenue. Between 2006 and 2014 it has accounted for about 45% of total local revenue. Block Grants for social sector services are supplemented by a freely disposable transfer for equalization. The size of the equalization grant pool cannot be less than 10% of the ownrevenues of all municipalities in the previous year. It is allocated by criteria determined jointly by the Ministry of Finance and the National Association of the Municipalities of Bulgaria (NAMRB). These criteria should relect the objective disparities among municipalities due to external factors and should not act as a disincentive for local revenue mobilization. The criteria for allocating the equalization subsidy have been changed repeatedly. Currently, the allocation formula has two (NALAS) Network of Associations of Local Authorities of South-East Europe components. The irst provides local governments whose per capita own-revenues are less than the national average with 90% of the diference between their per capita revenues and the per capita national average. Since 2008, this amount of the grant can be reduced by up to 25% if a municipality’s tax rates are below the national average. The second component allocates the remaining funds in the grant pool according to a separate calculation of expenditure needs. These needs are calculated on the basis of costing standards for preschools and homes for the elderly, as well as a municipality’s surface area. Municipalities whose per capita expenditures on these functions are less than 100% of the national average are entitled to 100% of the diference. Municipalities whose expenditures are higher than the national average, receive 50% of the diference. NAMRB’s has argued that despite eforts to improve equalization, the current methodology has serious shortcomings because it is not based entirely on criteria that are fully independent of local decisions, and thus can be “gamed”. Since its introduction, the Equalization Grant has accounted for about 5% of local revenue. Local Governments can also receive Conditional Grants for speciic investments and government programs. Since joining the European Union in 2007, most Conditional Grants have been for investments and have been funded by EU monies. In total they have received over 5 billion EUR in EU grant money, mostly for projects to improve their environmental, social, and technical infrastructure. Municipalities receive over 60% of all EU inancial support to Bulgaria and almost all of their investments come from this source. Within the public sector, the efects of the economic downturn were felt most profoundly by local governments. In 2010 –the worst year of the crisis in Bulgaria- the national government severely cut most transfers to local governments. The yield of the Property Transfer Tax -a major local revenue- also declined signiicantly because of the sharp decline in private investment. As a result, local revenue fell sharply between 2010 and 2013, leaving municipalities with 25% less revenue than they had in 2008. Municipalities accumulated payment arrears of about (NALAS) Network of Associations of Local Authorities of South-East Europe 100 million EUR (10% of own-revenue) while trying to cover the costs of underfunded (delegated) social sector functions. These account for 60% of local expenditure and should be fully covered by state transfers. About 40% of local governments had signiicant operating deicits and had trouble meeting their co-inancing requirements for EU funded projects. .In 2013, NAMRB signed an agreement with the newly elected national government to correct some of these problems. This agreement created a new public investment program called “Growth and Sustainable Development of Regions”. As result, and for the irst time, local governments and ministries competed for investment resources according to publicly deined selection criteria. The application process started in 2014, and 70% of municipalities have been granted 145 million EUR for nearly 400 investment projects. Equally importantly, the agreement called for restarting the decentralization process in accordance with a two-year roadmap that the government adopted in February 2014. The roadmap outlines the following reform measures: †† Transferring a portion of the personal income tax (PIT) to the local level. As a result, citizens would pay a 7% PIT rate to the national government plus up to a 3% rate to their local governments based on their tax policies. This change should go into efect in 2015 and should double local government tax revenues while keeping the overall iscal burden on citizens the same. †† Introducing a facultative municipal sales tax on the consumption of luxury goods and services (similar to the American sales tax). †† Introducing a local tax on agricultural land (currently nontaxable). †† Introducing a new way to calculate the waste disposal fee so that it relects not property values but the actual generation of waste. †† Reshaping the equalization subsidy. The Property Tax Since 1951, Bulgaria has had a property tax whose yield was given to local governments. The current tax is regulated by the 1998 Local Taxes and Fees Act. In 2007, constitutional amendments granted local governments tax powers for the irst time. As a result, local governments now have the right to set property tax rates within lower and upper limits determined by law. Zoning, which afects assessed value, is also under municipal discretion. Since 2006, the total yield of the tax has increased from about 0.03% of GDP to 0.06% of GDP and it now constitutes about 30% of local government own revenue, and about 12% of total local government revenue. Despite this growth, statutory tax reliefs and exemptions reduce the yield of the property tax by estimates as high as 60%. Parliament sets maximum and minimum tax rates limits for all taxable properties. Local governments than determine what rate they wish to apply within these limits. Rates must by uniform for all types of real estate and all types of owners. The constitution also determines that only parliament can grant abatements and exemptions. These exemptions are extensive and include the following: †† 50% for owner-occupied houses or apartments; †† 75% for handicapped owners; †† 100% for up to 10 years for several types of achieved levels of energy eiciency; †† 5% for those who fully pay their tax bills in one, early installment. The base of the property tax consists of land lots and buildings within settled areas. There is currently no tax on agricultural land. But this is being considered. The law stipulates two methodologies for assessing tax value –one for buildings and for empty land lots. The basis for both is a square meter price adjusted by coeicients for location, available public infrastructure, development zone, area of plot and the REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 67 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 nature of the building itself. For residential property, 60% of the valuation depends on the location coeicient. For commercial properties, valuation is based on whichever is higher, the property’s book value or its value as determined by the rules used to assess residential properties. Municipal Tax Oices are responsible for collecting tax declarations from tax payers. These declarations contain the information necessary for valuation. They are also responsible for collecting the tax. Municipalities have free access to the national government’s national this cadaster but only 50% of all properties are currently registered in it. Municipal Tax Oices issue the tax bills which can be paid in cash or via bank transfers, POS terminals and e-pay platforms. Local Governments can enforce collection by engaging public or private debt execution agents, garnishing wages, and by blocking the ability of delinquent taxpayers to sale their properties. 68 There are administrative and legal options for appealing valuations. But there is no legal schedule for revaluing properties. Nonetheless, revaluation is required when: †† The basic square meter value of properties is changed by law. †† The municipal council changes the local ordinance on property tax zones †† Owners declare they have improvements in their properties made substantial Owners are liable for the property tax. National and local governments are liable for the tax for properties they own, but which are not being used for public purposes. Other tax exempt properties are: Facilities providing diplomatic or consular services; buildings of the Bulgarian Red Cross; schools, academies, churches and other religious institutions, museums, galleries, libraries, and properties valued at less than 840 euro. (NALAS) Network of Associations of Local Authorities of South-East Europe In 2015, the “Program for Decentralization” was signed between the government and NAMRB. As part of this agreement, NAMRB prepared a concept for a new Local Taxes and Fees Act which was presented to MoF in July 2015. The concept stresses the following areas for improvement: the taxation of agricultural land, introducing municipal PIT, limiting the scope of tax exemptions, increasing local discretion on the tax assessment, and adding new fees for street lightning and city-center congestion. The policy dialog on developing the draft will commence in 2016. (NALAS) Network of Associations of Local Authorities of South-East Europe Statistical Overview of Local Government Finance in Bulgaria 2006-2014 Local government revenue as a share of GDP declined from a high of 7.8% in 2008 to a low of 5.8% in 2012, before rebounding to 6.4% in 2014. The local share of total public revenue also fell from 21% in 2008 to 16% in 2012, before jumping back to 18% in 2014. Both shares however have to be considered very low given that Bulgarian local governments are fully responsible for inancing all pre-tertiary education. Chart 48 Bulgaria: Local Government Revenue as Share of GDP and Total Public Revenue in 2006-2014 69 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Local and General Government revenues declined in tandem with the economic crisis. But General Government revenues increased much faster during the recovery. This trend then reversed in 2013. Chart 49 Bulgaria Fluctuations in the Revenues of the General Government and Local Governments In 2006-7, Bulgaria replaced PIT sharing with an expanded set of Block Grants for social sector functions. Since then the composition municipal revenue has been dominated by own-revenue and conditional grants, almost 80% of which are for education. 70 Chart 50 Bulgaria Composition of Local Government Revenue 2006-2014 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Since 2008, the share of local public investment in total public investment has risen from about 30 and 40%. Much of this is due to the inlux of EU funds. Chart 51 Bulgaria Public of Investment by Level of Government and as a % GDP 2006-2014 Since 2006, local governments have doubled the yield of the property tax as a percentage of GDP. Investment dropped sharply with the economic crisis of 2010 and has yet to fully recover. Wage spending fell less sharply and in 2013 returned to pre-crisis levels. The outstanding debt of municipalities has risen and is now above 1% of GDP— due largely to local governments borrowing to coinance EU-funded investments. Chart 52 Bulgaria Local Government Wages, Investment, Property Taxes and Outstanding Debt as % GDP 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 71 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Croatia The Intergovernmental Finance System 72 Croatia’s intergovernmental inance system is heavily dependent on the origin-based sharing of Personal Income Tax. Local Governments receive from 56.5 to 90% of the PIT generated in their jurisdictions, depending on their development index and the functions they perform. As such, the rules governing PIT sharing also constitute the backbone of Croatia’s equalization system. Local governments are also allowed to impose a surcharge of up to 18% on the amount of PIT taxpayers owe to the national government. The surcharge currently constitutes 10% of all local PIT revenues. Taken together, PIT revenues have constituted more than 50% of total local government revenues since 2006. In 2010 the rules regulating the Personal Income Tax were changed. The number of tax brackets was reduced from 4 to 3 and the base rate was lowered from 15% to 12%. Since the income tax is jointly shared between municipalities, cities, counties and the national government the reduction of these rates had a signiicant negative efect on local budgets. About 10% of local government revenues come from Conditional Grants for speciic programs or investment projects. Since 2006, about 30% of local budgets come from ownsources. Most own-source revenue comes from Land Use and Land Development Fees, with the former known locally as the “Communal Fee”. Croatian local governments also derive a signiicant amount of own-revenue from the sale and rental of municipal assets. But despite years of discussion, Croatia has yet to develop a local property tax (see below). (NALAS) Network of Associations of Local Authorities of South-East Europe The economic crisis reduced local government revenues, expenditures and investments signiicantly. Many of the 555 local governments (without Zagreb) increased their budget deicits and turned to borrowing. In 2010, measures aimed at improving the eiciency of the use of public revenues were implemented. One of these is the Fiscal Responsibility Act which sets limits on national and local government spending, strengthens the legal and functional accountability of budgetary resources, and introduces stronger controls for inancial reporting. Measures to improve tax compliance were also introduced. In late 2012, the Fiscalization Act for Real Cash was adopted. Its main objective is to monitor cash transactions and to increase tax collection. The Tax Administration now has internet access to the accounts of all taxpayers who are dealing in cash and is in a much stronger position to reduce evasion. This has contributed to an increased awareness of the need to pay taxes and to an improved balance in public inances. In 2012, a fee was also introduced for the legalization of illegal buildings. Building owners are now required to pay a fee for the legalization of structures built without proper permits. 50% of the fee goes to the national government, 20% to the competent body issuing the permit, and 30% to the local government in which the illegal construction is located. Also, in 2013 and as result of changes in EU regulations, a Law on Sustainable Waste Management was introduced. Local governments are now obliged to inance the recycling and sorting of solid waste from their own sources and through the tenders of the Fund for Environmental Protection and Energy Eiciency. (NALAS) Network of Associations of Local Authorities of South-East Europe The Property Tax Croatia does not have a market-based property tax and is still using a quasi-property tax called the “Communal Fee” which has not changed much over the last decade. Draft legislation for a property was prepared in 2012, but was only submitted to Parliament at the end of 2015. This text describes the current framework for the Communal Fee. The Communal fee is an area-based charge imposed on residential and commercial buildings, garages, land in commercial use, and construction land. The fee is calculated by multiplying its base value by coeicients for property type and municipal zone. Municipalities can set the base of the fee at any level they like and they draw up the zones used to adjust it. Parliament however, has imposed some restrictions on the coeicients municipalities can use for diferent property types, mainly to protect businesses. Municipalities are allowed to exclude properties of interest to the community from payment of the fee and also to set general criteria for (annual) case-by-case exemptions. The range of exemptions varies signiicantly among municipalities, but the most common are for municipal property, the poor, public kindergartens and schools, religious facilities, and sport and culture facilities. Municipalities issue tax bills and payment orders and are free to decide the frequency of collection. Generally, payment is made in monthly installments. Municipal Finance Departments are responsible for collection and enforcement and the most common form of enforcement is to garnish money from people’s bank accounts. Owners or occupiers of properties are liable for the fee. They are legally obliged to register their ownership with their municipality within 15 days of their acquisition of the property. They also must inform the city if they make any substantial changes in the use or physical characteristics of the property. Since many people do not do this, municipalities also make eforts to independently update their property registries. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 73 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of Local Government Finance in Croatia 2006-2014 Local government revenues as a share of GDP contracted signiicantly after the economic crisis of 2009. The local share of total public revenue however remained reasonably stable, suggesting that the national government did not try to push the costs the recession onto local governments. Chart 53 Croatia: Local Government Revenue as a Share of GDP and Total Public Revenue 2006-2014 74 The revenues of local governments and the General Government declined in tandem during Croatia’s long recession. In 2013, local revenue increased faster than the revenue of the General Government, but this trend reversed in 2014. Chart 54 Croatia: Fluctuations in the Revenue of the General Government and Local Governments (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Between 2006 and 2014, the composition of local revenue changed little and remains dominated by shared taxes. Croatia has yet to introduce a local property tax, and local governments have relatively little control over other fees and charges. They can however impose PIT surcharges. Chart 55 Croatia: Composition of Local Government Revenue 2006-2014 75 Local government revenue per capita has yet to recover to its pre-crisis levels. Unlike in many other places in the region local governments have not responded to the economic downturn by increasing the collection of own-source revenues. Chart 56 Croatia: Local Government Revenue 2006-2014 EUR per capita REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Chart 57 Croatia: 76 Composition of Local Government Expenditure 2006-2014 Local investment fell substantially in 2010 and has yet to recover. Wage spending has increased slightly but remains low. The total outstanding debt of local govenrments is equal to about 1% of GDP. Through 2009 there were strict limits on local government borrowing because total public debt had exceeded the Maastricht limit. In 2010, debt space was created for local governments, and the space was quickly used. This space was expanded in 2013 when borrowing to coinance EU projects and for Energy Conservation Companies (ESCOs) were exempted from the limits. Chart 58 Croatia: Local Government Wages, Investment, Property Taxes and Outstanding Debt as % GDP 2006-2014 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Total public investment has fallen dramatically since 2007. But the local government share of it has remained fairly stable at between 39 and 45%. Chart 59 Croatia: Shares of Public of Investment by Level of Government and as % GDP 2006-2014 77 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Kosovo Pupil numbers are used to determine the amounts for salaries, goods and services, and building maintenance in accordance with class size norms of 1 teacher to 23 students in majority communities and 1 teacher to 14 students in minority communities. The Health Grant is also allocated by formula according to population. The formula is based on the assumption that each person visits primary health care facilities 2.5 times year at a cost 4 euro per visit, and that they receive 3.5 services a year at 3.9 euro per service. The Intergovernmental Finance System 78 In 2009, responsibility for managing and inancing all preuniversity education and primary health care was decentralized to local governments in Kosovo. As result, Kosovo became one of the most decentralized countries or entities in the region. In 2014, Kosovo local governments derived 40% of their revenues from block grants for Education (31%) and Primary Health Care (9 %). They also receive a General Grant which in 2014 constituted 40% of their revenues. Of the rest, about 15% comes from own-revenue, and 15% from the shared Property Transfer Tax. The size of the General Grant is deined by law as 10% of the total operating revenues of the central government. All local governments receive a lump sum payment of 140,000 euro, minus one EUR per capita for all local governments with populations greater than 40,000. Municipalities with populations greater than 140,000 therefore do not receive any lump sum payment. The remainder of the grant pool is then allocated to municipalities by formula: 89% by population, 6% by square kilometers; 3% by the number of ethnic minorities; and 2% for municipalities in which the majority population is a national minority. The size of the Education and Health Grants is determined by a National Grant Commission in accordance with a Medium Term Expenditure Framework (MTEF). The Education Grant is allocated to local governments on the basis of a formula that takes into account the wages of teachers, administrators and support staf, goods and services, building maintenance, and speciic education policies. (NALAS) Network of Associations of Local Authorities of South-East Europe The most important own-revenues are the Property Tax (described below) and revenues from Construction Permits. Municipalities have been using Construction Permits as quasi-iscal infrastructure impact fees, a practice that the national government has been trying to stop –with mixed success—in order to improve the “business enabling environment.” They are also allowed to collect fees for health and education services. Municipalities receive 100% of the national government’s property transfer tax. The global inancial crisis of 2009 did not precipitate a recession and while growth has slowed it remains positive. It also has not afected intergovernmental iscal relations: Transfers to local governments have increased, as has the collection of own source revenue. In 2013, an agreement was signed between the governments of Kosovo and Serbia to regulate the status of the four Serbian-majority municipalities in the north of Kosovo. Under this agreement, these municipalities have enhanced powers and are now responsible for providing secondary health services and university education. A special fund was also established to help them. The Fund will be inanced from customs duties from the border with Serbia. To date 400,000 EUR have been placed in the fund. Some communities are interested in becoming separate municipalities but there have been no recent changes in the Law on Territorial Division and there are still 38 municipal governments. A separate law for the Capital City of Pristina is however, being considered. (NALAS) Network of Associations of Local Authorities of South-East Europe The Property Tax The Property Tax has been decentralized since 2003 and is regulated by the Law on Immovable Property. Now only buildings are subject to the tax, but the government is planning to introduce land taxation in 2017. Municipalities are responsible for property registration and valuation as well as for collection. The Kosovo Cadaster Agency issues tax bills based on the information provided to them by municipalities. The initial registration of buildings was done by the national government in 2004, and donors funded a second wave of registration in 2010. The Government of Kosovo has also recently instituted a performance grants system in an efort to improve collection. Between 2006 and 2014, the yield of the property tax remained stable and generated revenues equal to about 0.32% of GDP. In 2014, the property tax constituted about 30% of local government own-revenue and about 3.5% of total revenues. Valuation is done on the basis of centrally determined square meter rates for residential properties. Municipalities may establish up to four location zones to adjust these rates. Rates are also adjusted by coeicients for building quality and use. Commercial properties are valued according to procedures determined by local governments in accordance with regulations issued by the Ministry of Finance. Municipal councils determine property tax rates within minimum and maximum rates set by the national government. At present the minimum rate is 0.15% of assessed value and the maximum rate is 1.0% of assessed value. Taxpayers may challenge their assessments irst with the municipality, then with the Ministry of Finance, and inally with the Courts. properties used for diplomatic and consular services, by international aid organizations, and by religious institutions. Properties under construction receive a 60% abatement and the irst 10,000 EUR of value are exempted from the tax for owner-occupied residencies. The Kosovo Cadastral Agency is run by the central government and maintains a central registry of all properties. Local governments are responsible for providing the Agency with information about all properties in their jurisdictions, their value for tax purposes, and the tax rate as set by the municipal council. The Agency is responsible for issuing tax bills and for maintaining a central registry of property tax payments, unpaid liabilities, and fees and ines for late payment. It can also audit the registration and valuation practices of local governments. Municipalities are legally required to revalue properties every three to ive years. Properties with outstanding tax liabilities cannot be legally sold until their debts are been cleared. Public tenders cannot be awarded to legal entities who owe property taxes. Municipalities can seize the property of delinquent tax payers as well as to deny some services to citizens for non-payment. Thus far, however, these powers are rarely used. Natural and legal persons who own property or have the legal right to use it are liable for the tax. If the owners or those with legal use rights cannot be identiied, then the physical or legal persons occupying the property are liable for the tax. Properties owned by the national government and by local governments are exempted from the tax as are REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 79 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of the Finances of Local Governments in Kosovo 2006-2014 In 2009, schools and healthcare clinics were decentralized to local governments. As a result, local revenue as both a share of GDP and total public revenue increased sharply making Kosovo one of the most decentralized governments in the region. Municipalities receive almost a third of all public revenues and are getting a remarkably fair share of the overall iscal pie in comparison to many of their counterparts in the region. This share has also increase since 2009. Local spending on education and healthcare however remains heavily controlled by the central government and municipalities have yet to be allowed to borrow. Chart 60 Kosovo Local Government Revenue as Share of GDP and Total Public Revenue 2006-2014 80 Unlike in many other places in the region, there is no consistent pattern in the relationship between local and central government revenues: In some years local govenrment revenues have increased while central government fell, while in others the opposite has been true. In no year, however, has local revenue fallen below that of the previous year. Chart 61 Kosovo Fluctuations in the Revenue of the General Government and Local Governments (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Since 2009, the size of the Unconditional Grant has increased in line with national budget revenues. Own revenues and revenues from the shared Property Transfer Tax have also increased while the size of the health and education grants have remained stable. Chart 62 Kosovo Composition of Revenues 2009-2014 81 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 The two most important own-source revenues are the property tax and building permits. In 2011, legislation was passed to eliminate the quasi-iscal use of building permits. Income reported under this category declined in 2012, but in fact local governments simply classiied it as other revenue. In 2013, the restrictions on the pricing of building permits were loosened and revenue in the category increased. In 2014, the central government again tried to tighten up on building permits, but it appears that local governments responded by classiing the revenue under fees and charges. Chart 63 Kosovo 82 Composition of Own Revenue 2009-2014 (EUR per capita) Rather remarkably, local governments have devoted almost 35% of their total expenditures to investment, despite spending more than 50% of their budgets on wages. Chart 64 Kosovo Composition of Expenditures in 2008-2014EUR) (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe In recent years, investment and wage spending has remained stable as shares of GDP. The same is true of the yield of the property tax, despite signiicant investments by the central government into Kosovo Cadaster Agency to improve registration and billing, and substantial increases in the minimum property tax rates that municipalities can impose. Chart 65 Kosovo Investment, Wages & Property Tax as a Share of GDP 2006-2014 83 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Macedonia The Intergovernmental Finance System 84 As a result of the Ohrid Agreements of 2001, and as part of the country’s efort to accede to the European Union, Macedonia has pursued an incremental decentralization strategy. The process began in 2005 with the consolidation of 124 municipalities into 85 (then to 81 in 2013). In 2007, municipalities that had cleared their payment arrears and met other criteria for good inancial management were allowed to enter the so-called Second Phase of Decentralization. At this point, they became responsible for inancing and managing all schools, as well as a number of other cultural and social welfare institutions and were given Block Grants to inance these new functions As a result, municipalities are now responsible for the maintenance and improvement of local infrastructure, water and wastewater treatment, public hygiene, public lighting, local public transport, ire protection, pre-school, primary and secondary education, local cultural institutions (Cultural Houses, libraries, and museums) and care of the elderly. Since 2011, they have also assumed responsibility for managing state land. In accordance with the Law on Local Government Finance, municipal revenue consists of: †† Own Revenues, include the Property Tax, other local fees, charges and taxes, asset income and income from ines, penalties and donations; †† Shared Taxes, in particular a share of the income tax coming from artisans; †† A General Grant deined as a percentage of the national yield of the Value Added Tax and allocated by formula; (NALAS) Network of Associations of Local Authorities of South-East Europe †† Block Grants from the national budget for primary and secondary education, culture and social welfare; †† Earmarked grants for special programs or speciic investments; †† Debt Finance and donations. The size of the Grant is anchored by law at 4.5% of the national yield of VAT. The criteria used to allocate the grant are deined by an annual ordinance. According to the ordinance: †† All jurisdictions receive a lump sum payment of 3 million denars. †† These payments are then deducted from the grant pool and the residual is divided between the capital city of Skopje and its composite jurisdictions (12%) and all other municipalities (88%). †† The funds for municipalities outside of Skopje are divided by a formula which allocates 65% of the pool on the basis of population; 27% on the basis of square kilometers; and 8% on the basis of the number of settlements. The allocation of the Block Grant for Education is also determined by an annual ordinance. The main criteria in the formula for allocating the grant are enrollment, employment, and –since 2009— the number of children entitled to free school transport. The formula for determining per pupil payments are publicly available, but the amount of money that municipali- (NALAS) Network of Associations of Local Authorities of South-East Europe ties receive through the grant is insuicient and often requires substantial contributions from their general budgets. The allocation of the block grant for preschool education is also governed by an annual ordinance. The formula contains variables for the number of pupils, the type of heating system and the duration of the heating season, the number of teachers in the school, and the utilization rate of the facility. Municipalities that have cultural institutions receive a block grant for culture based on the number of employees working in the institutions covered by the grant; the total square meters of the buildings; and coeicients for the particular cultural services these institutions provide. †† The municipal share of income from the sale state land was increased to 80%; †† The municipal share of income from mineral concessions was increased to 78%; †† The municipal share of revenue from other concessions (e.g. water) will be increased from 25% to 50% in 2016; †† Revenue from fees for washing and separating gravel are now split 50%/50%. †† Revenue from fees for legalizing illegal structures built on agricultural state land are now split 50%/50%. Every year and in accordance with the Local Government Finance Law, the Ministry of Finance provides municipalities with a Budget Circular informing them of about their block grants. In theory, municipalities are autonomous in managing the funds they receive through the block grants. In practice, the situation is much more complicated. Based on criteria approved by their Councils, municipalities allocate block grant funds to schools and other institutions on a monthly basis. †† Revenue from fees for legalization properties will now go The iscal decentralization process can best be seen through the expansion of local government revenue as percentage of GDP between 2005 and 2012. In 2005, it equaled only 1.9% of the GDP while by 2012 the share had more than tripled to 6.5% of GDP. It has however fallen signiicantly since then and in 2014 was only at 5.4% of GDP. Thus, despite the radical increase in their revenues, Macedonian municipalities still face profound inancial challenges and are clearly underfunded for the functions they perform. which allocates money to regions according to a formula contained in the Law on Regional Development. By Law this fund should be equal to 1% of the GDP, but so far this has not been the case. entirely to municipalities. †† Starting in 2015, 10% municipalities will receive 10% of concessions on agricultural land, a share that will be increased to 50% by 2018. †† There is also a fund for balanced regional development In order to strengthen their inancial position, the municipal association, ZELS has lobbied the government to make amendments to the Local Government Finance Law. This has resulted in the following recent changes: †† The percentage of the national yield of VAT earmarked for the general grant was increased from 3% to 4.5%, and will be progressively increased to 6%. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 85 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 The Property Tax The recurrent property tax was introduced in Macedonia in 2005 as part of the larger decentralization program. There are three property-related taxes, the Property Transfer Tax, the Inheritance and Gift Tax, and the recurrent Property Tax. The introduction of the recurrent property tax was accompanied by the elimination of the so-called Land Use Fee, a lat, square meter charge that users of properties paid to help compensate municipalities for public services. The property tax is paid on all land and buildings except agriculture land. Until 2012, values were determined by municipal commissions. Since 2012, they are determined by trained municipal employees or by an appointed appraiser using the criteria set in law. Recently ZELS has proposed changes in the valuation methodology. Between 2006 and 2014 the yield of the tax increased from 0.06% to 0.22% of GDP. It now represents 13% of local own-revenue and 4% of total revenue. 86 The base of the tax is the market value of land and buildings. The market value is determined on a square meter basis according to a methodology set by the national government, but approved by ZELS. The square meter market value is determined by adjusting base rates set in the Methodology for Assessment of the Market Value of Immovable Property by coeicients for macro and micro locational zones, as well as building age, quality, and use. Municipal Councils are responsible for setting the boundaries of both macro and micro locational zones. The Ministry of Finance conducts periodic audits of how local governments determine of market value. Property owners may appeal their appraisal for review to the Commission for Appraisal and Audit of the Chamber of Appraisers. Taxpayers who reside in their properties are entitled to a 50% reduction in their tax bills. All property owners are liable for the property tax, but the tax can be imposed on the user of a property if its owner cannot be identiied. Legal and physical persons using (NALAS) Network of Associations of Local Authorities of South-East Europe municipal or state property are also liable for the tax. If a property is owned by several persons, each of them is responsible for the portion of the tax corresponding to their ownership share. Municipal and state property are tax exempt. Properties used for diplomatic and consular missions, religious purposes, environmental protection, mining and the processing of agricultural products are also exempt, as are cultural landmarks. The national government is responsible for determining the base of the tax, the methodology for assessing its value, and the maximum and minimum rates that local governments can impose. Local governments set tax rates within these limits and administer the tax. At present, local governments can set the tax rate at between 0.1% and 0.2% of assessed value. Rates are determined by the municipal council. Mayors can issue abatements and exemptions and determine schedules for the payment of arrears, within limits set by national regulation. Interest of 0.05% is paid for each day of delay in payment. Municipalities may not withhold services for the non-payment of property taxes. (NALAS) Network of Associations of Local Authorities of South-East Europe Statistical Overview of Local Government Finances in Macedonia 2006-2014 Since 2007, Macedonia has progressively devolved major social services to local governments. This has substantially increased local revenue as a share of both GDP and of total public revenue. Even in 2012, however, when local government revenue peaked at 20% of total public revenue and 6.3% of GDP, Macedonian local governments appear to be underfunded given their responsibilities. Moreover, in 2013 and 2014 the positive trend local revenues was reversed. Chart 66 Macedonia Local Government Revenue as a Share of GDP and Total Public Revenue in 2006-2014 87 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Macedonian local governments derive modest shares of their revenues from shared taxes and unconditional grants. Block grants are their largest source of revenue. Chart 67 Macedonia 88 Composition of Local Government Revenues 2006-2014 Since 2009, local governments have done an impressive job mobilizing own source revenues. Though the overall yield of the property tax remains modest they have increased collection ive times. They are also more argressively collecting Land Development Fees, Lighting Fees and other communal charges. Chart 68 Macedonia Composition of Own Revenues 2006-2014 (mln EUR) (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe The share of local expenditures going to wages has increased steadily as local governments have assumed responsibility for primary and secondary education. The investment rate is modest, with a large share of it probably coming from the capital city. Chart 69 Macedonia Composition of Expenditures in 2006-2014 89 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Wage and investment spending as a share of GDP have expanded over the last eight years, while the property tax has increased more modestly and still only yields revenue equal to 0.2% of GDP. Chart 70 Macedonia Investment, Wages, & Property Tax as Shares of GDP 2006-2014 90 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Local governments are responsible for about 25% of total public investment which in 2014 amounted to a little over 4% of GDP. Chart 71 Macedonia Public Investment by Level of Government and as a share of GDP 2006-2014 91 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Moldova The Intergovernmental Finance System On paper, Moldova has a highly decentralized public sector with raions and municipalities responsible for all pretertiary education and 25% of all public expenditures. This picture, however, is misleading because of the subordination of municipal governments to raion authorities, and the subordination of raions to the national government. 92 In 2012 the Ministry of Finance –with the support of the UNDP— prepared draft legislation designed to eliminate the inancial subordination of lower-level local governments to higher ones. The legislation: †† Preserves the existing division of total public revenue between levels of government and is broadly speaking iscal neutrality; †† Requires the national government to fully inance delegated functions. †† Requires the national government to provide raions and municipalities with separate transfers, ending the inancial dependency of municipalities on raions. †† Requires the separation of Conditional Grants from the General Grant; †† Deines local governments’ right to speciic percentages of shared taxes. †† Eliminates disincentives for local revenue mobilization by basing the equalization system on shared taxes and not on locally collected taxes and fees. (NALAS) Network of Associations of Local Authorities of South-East Europe Unfortunately, after Parliament approved the draft legislation the government reversed itself out of fear of losing political, administrative and inancial inluence overs mayors and local oicials postponed the implementation of the law until after the 2014 elections. Making matters worse, the government has continued to politicize the already non-transparent allocation of national funds for local infrastructure investments while capping all local taxes. The attempt to cap local taxes however was contested and declared unconstitutional by the Constitutional Court. (NALAS) Network of Associations of Local Authorities of South-East Europe The Property Tax The Property Tax is regulated by Section VI of the Tax Code. In 2007, the land tax and the building tax were uniied into a single tax on real-estate. The new system is being progressively implemented and the tax still does not generate much public revenue. Between 2006 and 2014 the yield of the property tax decreased from 0.54% of GDP to 0.32% of GDP. At present it represents about 30% of local government own-revenues and about 3% of total revenues. Prior to 2007, land and buildings were taxed on the basis of centrally imposed norms. Under the new system, a central government agency with branch oices throughout the country is tasked with determining the market value of all properties using computer-aided mass valuation techniques. Since 2004, about 85% of the properties in the country have been registered, but only 12% of them have been assessed for property tax purposes. Most of these are in rural areas. Properties that have not been assessed by the new techniques are valued using the old centrally established norms. The national government is responsible for property registration and valuation. Local government councils have the right to set property tax rates within limits deined by national legislation. The maximum and minimum rates for residential properties and garages are, respectively 0.02% and 0.25%. The rate for commercial and industrial properties is 1%. Local governments are responsible for billing and collection, and most taxpayers pay in cash at municipal oices. Property sales require certiicates stating that the concerned properties do not have any outstanding tax liabilities. Local governments can garnish the accounts of taxpayers with outstanding liabilities. Ultimately, they can seize properties for non-payment. But these actions require court orders that are in practice hard to get. After six years all tax debts are extinguished by law. 93 All land and buildings are subject to the tax, with exemptions for properties used for diplomatic and consular services and religious purposes, as well as for state and local government property used to provide public services. Property owners are liable for the tax, as well as those who have acquired the right to use public property for commercial purposes. There are statutory abatements for owner occupied properties used for residential purposes as well as for certain classes of individuals (disabled, elderly). REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of Local Government Finances in Moldova 2006-2014 On paper, Moldova has a highly decentralized system of public administration. Local government revenues equal about 10% of GDP and 25% of total public revenues –levels close to the EU average. In reality, however, the situation is quite diferent because of the political and economic subordination of municipal governments to raion governments and to the line ministries of the national government. Chart 72 Moldova Local Government Revenue as a Share of GDP and Total Public Revenue 2006-2014 94 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Moldovian local governments derive most of their revenues from conditional grants. There are no unconditional grants in the system and thus no clear mechanism for horizontal equalization. The share of shared taxes in the system has decreased, as Moldova, like Bulgaria before it, backs out of trying to fund social sector functions with shared taxes whose distribution is highly-skewed. Own revenue as a share of total revenue is low. Chart 73 Moldova Composition of Local Government Revenue 2006-2014 95 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 In EUR per capita terms, local revenue increased sharply in 2012 and have held more or less steady since. Own-revenues, which showed substantial growth between 2006 and 2012, have fallen over the last two years. Chart 74 Moldova 96 Composition of Local Government Revenue 2006-2014 (EUR per capita) Despite the growth in local revenue, local government investment spending declined signiicantly between 2006 and 2012, though it rose sharply in 2014. Wage spending jumped in 2009 because of state mandated increases in teachers’ salaries but have since fallen as a share of both local budgets and the GDP. Chart 75 Moldova Composition of Local Government Expenditures in 2006-2014 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe The yield of the property tax has declined as a share of GDP. Local wages as a share of GDP remain high, though they have fallen since 2009. Investment is low, but rose in 2014. Chart 76 Moldova Investment, Wages, and Property Tax as Shares of GDP 2006-2013 97 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Montenegro The Intergovernmental Finance System 98 Montenegro’s intergovernmental inance system is unique the Personal Income Tax, 10% of the national yield of the to the region, and thus informative. Most striking, is that lo- Property Transfer Tax, 100% of the national yield of Vehicle cal governments have derived more than 70% of their total Tax and 40% of the yield of concession fees from games of revenues from true own sources for the entire period. This chance. In 2014, the Fund equalled 29 million euro or 13% has been possible in part because on the expenditure side of local revenues. municipalities are responsible for no social sector functions and thus have less need for grants or transfers. And in part Municipalities whose per capita revenues from own sources it is because, Montenegrin local governments control of a –excluding the land development fee— and shared taxes for variety of instruments to tax land and buildings in an envi- the last three years are lower than the national per capita ronment where land —particularly coastal areas— was ex- average from these same sources are entitled to receive ceptionally valuable in the early 2000s. grants from the Fund. The Fund is allocated in two phases. In the irst, 60% of the Fund is allocated on the basis of the Even more interesting, is that while own-revenues continue diference between an individual municipality’s per capita to be the main source of municipal revenue, their compo- revenue from own-sources and shared taxes, and the nasition has changed substantially. This change has come in tional per capita average for the same revenues for the last the context of overall hard times and the bursting of the three years. The per capita diference is then multiplied by land bubble of 2005-2007. But it has also been driven by the number of inhabitants and coeicents based on the central government policies which have been pushing local population of the municipality. The coeicient for municigovernments to make greater use of local PIT surcharges, palities with less than 3,000 inhabitants is 2; for municipaliand the property tax –as opposed to land use and develop- ties with populations between 3 and 6,000, 1.5; and for the ment fees, and charges to local businesses. Historical Capital 2.5. For all others it is 1. Equally importantly, Montenegro has as reasonably robust and evolving equalization system which provides about 13% of local revenue. In recent years, reforms have tried to ensure that equalization monies are allocated not on the basis of what municipalities actually collect in own-revenues but what they could collect given their tax bases. In accordance with the Law on Local Government Finance, iscal equalization is performed through the Equalization Fund. The Fund is formed from 11% of the national yield of (NALAS) Network of Associations of Local Authorities of South-East Europe In the second phase, the remaining 40% of the Equalization Fund is allocated to local governments based on their estimated budgetary needs using the following procedure: 20% of the fund is allocated equally to all local governments entitled to equalization; 60% of the remainder is allocated on the basis of a municipality’s area in relationship to the area of other municipalities entitled to equalization; and 40% on the basis of its share of the total population of municipalities entitled to equalization. (NALAS) Network of Associations of Local Authorities of South-East Europe Amendments to the Local Government Finance Law introduced in 2011 changed the way in which iscal capacity is calculated. Earlier, iscal capacity was determined using the own-revenues that municipalities collected, efectively rewarding municipalities for non-collection. Since 2011, the Ministry calculates an “estimated amount” of own-revenues for each municipality and uses this both to determine who is entitled to receive equalization funds and how the Fund is allocated. The Law on Local Government Finance also requires periodic consultation with the Committee for Monitoring the Development of the System of Municipal Fiscal Equalization. The Committee monitors the implementation of the criteria for iscal equalization, gives recommendations for improving the system, and issues opinons on documents prepared by the Ministry of Finance related to the allocation of the Fund. Municipalities can also receive conditional grants from the State Budget for inancing investment projects that are of special interest to one or more local govenrments. These grants can be used to co-inance donor funded projects. In order to receive a conditional grant, municipalities must have adopted a multiyear investment plan. The maximum amount of a conditional grant cannot exceed 50% of the anticipated cost of the project. The amount of conditional grants that a local government can receive also depends on the level of per capita revenues they generate from the land development fee in relation to the national per capita average in the proceeding year. Conditional grants have proved to be very useful instruments for co-inancing investment projects that are also being supported by EU funds. Nonetheless, conditional grants represented only 1.3% of total local government income in 2014. The “great recession” of 2009 had a strong impact on Montenegro’s public inances. Eforts to limit the efects of the crisis through countercyclical spending led to a sharp increase in the level of the public debt. The economic crisis also afected local governments. Like the national government, they borrowed heavily to limit the impact of the crisis. They also accumulated payment arrears to suppliers and contractors. So they are now in a period of retrenchment, struggling to reduce overstaing, collect revenues and decrease ineiciencies. Since 2008, a number of local own-revenues have been reduced or eliminated. These include fees for transmission towers, telecom facilities, and TV and radio receivers; the Land Use, Land Development, and Business Sign (registration) Fees; and the gambling tax. In 2011, to compensate local governments for the loss of these revenues, amendments were introduced into Law on Local Government Finance which increased the municipal share of PIT from 10% to 12%; the share of the Property Transfer Tax from 50% to 80%; and the share of concessions and other fees from 30% to 70%. They also increased the size of the Equalization Fund and changed the criteria for allocating it. But the amendments came too late and the loss of ownrevenues compounded the efects of the crisis. Indeed, it is one of the reasons that crisis produced such a sharp increase in local government debt and payment arrears. Moreover, they did not fully compensate local governments for the own revenues that had been lost, revenues which were both more stable and robust than shared taxes. As a result, local government budgets have not recovered to pre-crises levels. Most municipalities have reached their legal debt limits and many do not have enough revenue to inance all their obligations to banks, suppliers and the state budget. Indeed, payment arrears have risen from 27.9 million EUR in 2008 to 167 million in 2014. Investment spending has fallen from 53% in 2008 to 23% in 2014, while debt service payments have increased from 6% of local expenditures in 2008 to 28% in 2014. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 99 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 The Property Tax The Property Tax was decentralized in 2003. Following its decentralization, the collection of the tax signiicantly increased. According to the 2003 Property Tax Law, local governments have the right to set the property tax rate at between 0.08 and 0.80% of market value. In 2010, the national governement increased the minimum rate to 0.10% in efort to help municipalities compensate for the loss of other revenues. The tax base was also expanded. In 2015, further amendments were introduced into the Law. These will go into efect in 2016 and raise the minimum rate to 0.25% of market value. This brings the minimum rate up to just about the average rate that local governments have used for the last few years (0.26%). The increase is part of the national governments plan to abolish the Land Development Fee in 2020. By abolishing the fee the government hopes to improve the “business enabling environment. 100 Municipalities however, fear that the property tax will not generate enough money to compensate them for the revenues they have already lost, to say nothing about the very signiicant revenue they will lose from the elimination of the Land Development Fee —their most important capital revenue. Between 2006 and 2014, the yield of the property tax increased from 0.46% of GDP to 1.18% of GDP. In 2014, it accounted for 28% of local government own-revenue and 19% of their total revenue. The property tax should be paid on all land and buildings except those explicitly exempted from taxation by the law. Exempted properties include facilities used for diplomatic and consular services, religious institutions, schools, and properties owned by the national government. The owners of land and buildings are liable for the tax. However, if the owner of a property is not known, the occupier or user of the property must pay the tax. The tax is value based, and local governments are responsible for valuation. In determining property tax values, Local Government Tax Authorities use data from the State Tax (NALAS) Network of Associations of Local Authorities of South-East Europe Authority and/or the State Statistics Oice on the market value of a square meter of property in their jurisdictions. If these institutions don’t have such data, municipalities can engage a court expert to deine market value but this solution is very expensive and is used very rarely. The “Regulation on the detailed Criteria and Methodology for the Determination of the Market Value of Properties” deines the nature of the valuation process. Market value is calculated by multiplying a base square meter rate by a number of coeicients that adjust for: the use of the property; its location, its quality size and a number of other elements that could inluence its value. Municipalities revalue properties every year. There is no external oversight of the valuation process, but taxpayers have the right to appeal their valuations. In the appeal process, if either local governments themselves (irst instance) or the courts (second instance) ind any mistakes, new tax bills have to be issued by the municipality. The National government is responsible for determining the legal framework regulating the property tax and for maintaining a central cadaster of all properties. Local Tax Authorities use data from the Cadaster Oice to determine who is liable for the tax in their jurisdictions. Relations with the Cadaster Oice however, have been problematic and most local governments maintain their own cadasters. As of 2016, they will be able to use information from the courts about the transfer of properties in their jurisdictions to update their cadasters. Through local ordinances, municipalities determine the tax rates for each type of property and each type of owner within the limits set by the law. Most abatements and exemptions are also set in the Law. There is a 20% reduction for owneroccupied residential properties, a reduction that is increased by 10% for every family member, up to a maximum reduction of 50%. Municipalities have the right to introduce some other exemptions and abatements for special purposes, such as creating business improvement districts. (NALAS) Network of Associations of Local Authorities of South-East Europe Local governments are fully responsible for issuing tax bills. Tax bills must include: the name of the tax payer, the number of the cadastral parcel, the market value per square meter, it’s adjustment, the tax rate, any abatements or exemptions, the dates when taxes must be paid, and the account number on which payment must be made. Bills are usually delivered to taxpayers by post, though some local governments deliver them by hand. Local governments are also responsible for collecting the tax. Taxpayers should pay their obligations in two equal installments on a special bank account set up for property tax purposes. Local governments and taxpayers can set up payment schedules that include more than two installments. Currently, local governments have very weak powers to enforce collection, though these will be strengthened under the amended property tax legislation in 2016. Municipalities will be able to garnish wages, block accounts and block sales of properties that have unpaid tax liabilities. 101 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of Local Government Finance in Montenegro 2006-2014 The impact of the global economic crisis on the Montenegro was particularly strong. In 2007, local government revenue as a share of both total public revenue and GDP was extraordinarily high (22% and 11% respectively) given that Montenegrin municipalities have no major social sector responsibilities. Since 2007, however, local government revenues have fallen signiicantly and now stand at 14% of total public revenues and only 6.3% of GDP. Chart 77 Montenegro Local Government Revenue as Share of GDP and Total Public Revenue 2006-2014 102 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Montenegrin municipalities are unique in the region in that they derive over 70% of their revenues from own sources. Indeed, in the middle of the decade, own revenues accounted for more that than 80% of total revenues, and were being driven up by a real estate boom that increased income from asset sales, land development fees and other property related fees and charges. Chart 78 Montenegro Composition of Local Government Revenues 2006-2014 103 Until recently, the Land Development Fee was the largest source of local own-revenue. But the central government has been imposting constraints on it, and the Fee is scheduled for elimination in 2020. If so, this will have a serious impact on municipal inances. Meanwhile, the Land Use Fee was eliminated in 2009. Local governments have tried to replace the lost income by makinig greater use of the Property Tax. Chart 79 Montenegro Composition of Own Revenues 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Local government investment has dropped from 268 EUR per capita in 2008 to 80 EUR per capita in 2014 while debt service payments have more than doubled from 54 EUR per capita to 112 EUR per capita. Chart 80 Montenegro Composition of Expenditures in 2006-2014 (EUR per capita) 104 Local governments have responded to the economic downturn and the policy changes discussed above by reducing wages, raising the property tax, lowering investment and increasing borrowing. Chart 81 Montenegro Investment, Wages, Debt Service and Property Tax as Share of GDP 2006-2014 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe The economic downturn led to a sharp contraction in public sector investment between 2008 and 2013. But while total public investment recovered somewhat in 2014, the share coming from municipalities continued to decline. As a result, municipalities now account for only 27% of total public investment, down from close to 60% in 2010. Chart 82 Montenegro Public Investment by Level of Government and as a Share of GDP 2006-2014 105 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Romania The Intergovernmental Finance System 106 In Romania, public sector revenues account for only 33% of GDP. This is low by EU standards. In terms of expenditure, they are below 35% the lowest in the EU. Nonetheless, local governments play a very important role in the country’s public sector. Their revenue as a share of GDP is above 9%, which is high for comparable European countries. As a result, local governments have been targeted for many of the iscal adjustment measures taken in the wake of the inancial crisis of 2009. These included a reduction in their share of the personal income tax, a reduction in grants for social sector functions, wage cuts, layofs and a tightening of debt limits. By the end of 2013, 56% of all public employees were paid for by local governments. This includes almost 300,000 teachers, over 100,000 social service employees and since 2010, health workers. Indeed, over the last six years local governments at once added 120,000 hospital employees to their payrolls while shedding 140,000 employees from other local services – a net reduction of about 20,000 people. Local governments have full expenditure control of about 50% of their revenues which come mostly from shared income tax and property taxes which they collect on their own. Grants from the national budget account for another 30%, and grants from the EU for 7%. The iscal adjustment program has led to 4% reduction in state transfer for social sector functions. It also led to a reduction of the local share of personal income tax from 82% in 2010 to 71.5% in 2012. (NALAS) Network of Associations of Local Authorities of South-East Europe Most local government expenditure is for education (c. 20%), health (13%) and social welfare (10%) and most is for recurrent expenditures (c. 65%). Nonetheless investment spending is high by European standards, especially if one adds EU grants, which are generally for investment (14% + 10%). Expenditures on debt service however remain low –though rising—and account for only 4% of total spending. In 2009 and 2010, new limits were set for local debt and both borrowing and investment spending declined. There are however exceptions for loans incurred to co-inance EU funded projects. Romania’s intergovernmental inance system tries to equalize local government revenues both vertically and horizontally. Vertical equalization is achieved by sharing Personal Income Tax (PIT) on an origin basis. The shares vary according to the type of local government: municipalities get 41.75%, counties 11.25%, the city of Bucharest 44.5% and its six districts 20%. Horizontal equalization is carried out at the county level from funds created by 18.5% of the PIT collected in a given county plus an equalization grant from the state budget. Since 2006 horizontal equalization has been managed mainly through a mathematical formula. Until then discretionary allocations by county councils and central government were prevalent; since the adoption of the formula, discretionary transfers have been drastically reduced, but still continue to be a feature of the system. Figure 1 below shows the formation of the horizontal equalization pool at the county level. The pool is created by a share of the income tax collected within the county (18.5%) and an equalization grant from the state budget (so-called “VAT sums for equalization”). The latter arrives by formula to each county. The county pool is split between the county council (27%) and the municipalities (73%). In the latter case, most is distributed by a two-step formula. (NALAS) Network of Associations of Local Authorities of South-East Europe 11.25% for county budget County budget (41) 27% to the county budget 18.5% for equalization County PIT (71.5%) + VAT equalization grant County PIT equalization pool 73% to municipalities 41.75% returns to the cities, towns and communes Figure 1 – Financial lows of the Romanian equalization system The variables of all formulas in the equalization system are based on income tax, population, county area and urbanized area. No weight is given to any indicators of expenditure need like population density, geographical position or development level. Income tax per capita is used in most formulas as a proxy for economic development. The indicator is designed to allocate more funding to poorer municipalities whose per capita revenue from shared PIT is below the county average. In contrast, population and area are employed as proxies for expenditure needs providing more money to local governments with large populations or which service large territories. Overall, the most important indicator in all formulas is income tax per capita. The system does not contain any “Robin Hood” mechanism whereby richer municipalities are taxed to help cover the costs of equalization. The formula allocations are uniform and unbiased. But the discretionary allocations on top of the formulas provide county councils with signiicant leverage over poor municipalities. The current equalization system has a series of drawbacks which should be corrected. The most important of these are: 80% by formula in two phases taking into account population, area (phase I) and income tax per capita (phase II) 20% for arrear payment and local development projects (decided by the county council) Municipalities (102 cities, 217 towns and 2,861 communes) †† It is unclear how well vertical equalization performs because local government expenditure needs have not been thoroughly measured; †† Income tax is shared on the basis of the tax payers’ place of work, not their place of residence. Because many people work in big cities this increases iscal inequalities; †† The signiicant weight of discretionary transfers from the national government and county councils make the system unnecessarily unpredictable, non-transparency and subject to political bias; †† The formation of 41 separate county pools exacerbates the diferences in per capita revenues between similar local governments from diferent counties; Despite its laws Romania’s equalization system manages to reduce the wealth gap between local governments even in the current setup. Some scenarios were tested with a view to improving current resource allocation and achieve better outcomes. The best results were obtained with the formation of a unique national equalization pool. Such a solution would be technically feasible, but politically dificult to sell to would-be losers: county councils, Bucharest districts and wealthy counties. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 107 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Against this backdrop, changes to Romania’s equalization system were recently enacted through the 2015 state budget law. Without any prior notice to local governments, the Parliament adopted provisions which in efect suspend the application of the statutory equalization system in 2015. Instead, a diferent system is being used. It is based on revenue thresholds calculated for each category of local governments – communes, towns, cities and counties. These thresholds include own revenues, shared PIT and equalization, and the new provisions of the budget law guarantee all local governments the attainment of the respective thresholds, regardless of their population, through equalization allocations to cover the deicit. Once these equalization allocations are made, whatever remains in the pool of funds earmarked for equalization is then distributed to all local governments based on a number of criteria, of which population is the most important. An impact analysis carried out by the Association of Communes reveals major drawbacks in the new system: 108 †† Half of local governments are losing money compared to 2013 and half are gaining; †† On the losing side, are over 500 local governments in the poorest two quintiles of local governments -mostly heavily populated but poor communes and towns; †† Winning are almost 500 well-of local governments in the richest two quintiles; †† Over 1,800 local governments are subject to a major variation (+/- 50%) in their equalization revenues as compared to 2013, half of them on the negative side †† The coeicient of variation of local governments’ per capita discretionary revenues after equalization has deteriorated when compared to 2013, which means the 2015 system equalizes less than the statutory one. (NALAS) Network of Associations of Local Authorities of South-East Europe This is an example of opaque and hasty policy decision that was not proceeded by an impact analysis and has had unforeseen consequences. Hopefully, the system will not be implemented beyond 2015, otherwise we fear a signiicant change in local governments’ behavior (e.g. reduction of tax collection eforts, break-up into smaller units) . (NALAS) Network of Associations of Local Authorities of South-East Europe The Property Tax The property tax was decentralized to local governments in 1999 and is fully administered by them. In 2016, a new Fiscal Code was adopted. The Code did not introduce structural changes in the tax. Instead, it increased some of the coeicients used to determine tax value, equalized tax rates for buildings owned by natural and legal persons, and eliminated incremental tax rates for multiple properties of the same owner. It also expanded local governments’ right to increase the tax rate from 20% above the base rate to up to 50%. In addition, the tax rate for unfarmed land and abandoned buildings may be increased ive-fold. In 2014, the property tax yielded revenue equal to 0.80% of GDP and accounted for 31% of own-source revenue and 9% of total revenue. Property tax valuation in Romania is not strictly tied to market value. Instead, for buildings and land owned by natural persons the value is determined by multiplying the area of each with diferent coeicients relating to their physical characteristics, use, and location. For buildings owned by legal persons the tax value is either the book value, the construction value or the transaction value of the property, subject to particular conditions. The calculation of tax value is uniform across the country, though some coeicients are adjusted to account for local government status, population and location. Property valuation is performed every year by local governments, who are also responsible for billing taxpayers. Oversight is carried out by the Court of Accounts. The Ministry of Finance and Ministry of Regional Development and Public Administration provide technical support. Taxpayers may appeal their tax bills at the tax administration or at the Administrative Court. licly owned property which is rented to a legal or natural person, it is the occupant who pays the tax. The legislation provides for numerous exemptions. As a rule, public property is not taxed unless used for economic activities. In addition, public infrastructure of any kind, educational, religious and healthcare facilities, as well as residences of disabled and impoverished persons are tax exempt. Most tax exemptions are set by law but local governments may issue abatements for historical buildings, buildings occupied by social services providers, and some other facilities. An abatement of 10% is provided by law if the tax is paid before due date. National legislation regulates the property tax quite tightly and until recently local governments had limited power to set rates, or issue exemptions and abatements. The national cadaster contains only 15% of all properties, but local records of properties are quite good. In case of non-compliance, local governments issue summons and if this is not efective begin forced execution procedures. This includes the garnishing of bank accounts, and the seizure and eventual the sale of the concerned property. The sale of property requires certiication that all taxes against it have been paid. Payment methods include cash or credit cards at the tax administration oices as well as online payments and bank transfers. There is a national portal for online payments, which has failed to become popular because of its user unfriendliness. The tax rates are set by the local governments within intervals provided by the law. The intervals vary based on the property destination and location, but not the legal status of the owner. The latter only inluences the tax value. Taxpayers are owners of buildings or land. In the case of pubREPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 109 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of Local Government Finance in Romania 2006-2014 Local government revenues in Romania both as a percentage of GDP (9%) and a share of total public revenues (29%) have been both high and stable over the entire period. Given the overall modest size of the total public sector, this suggests that the national government has been treating local governments reasonably fairly and predictably. Chart 83 Romania Local Government Revenue as Share of GDP and Total Public Revenue in 2006-2014 110 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Since 2006, the share of own revenues in local budgets has increased from 20 to 30%. Part of this increase has been due to a doubling of property tax revenues whose yield is now equal to 0.8% of GDP, one of the highest in the region. Chart 84 Romania Romania Composition of Revenue 2006-2014 With the decentralization of hospitals in 2010, hospital fees have also become an important source of own-revenue. But these revenues must be spent in the health sector. Chart 85 Romania Composition of Local Government Own Revenue 2006-2014 (mln EUR) REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 111 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Local governments’ initial response to the economic crisis was to slash expenditure on investments and on goods and services, and to reduce –though to a lesser degree—wage spending. By 2011, however, investment spending as well as spending on goods and services increased while wage spending declined through before rising in 2013 and 2014 as investment expenditure fell. Chart 86 Romania Composition of Local Government Expenditures in 2006-2014 (EUR per capita) 112 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Chart 87 Romania Investment, Wages, and Property Tax as Shares of GDP 2006-2014 Since 2010, local governments have accounted for over 50% of all public investment. As in Bulgaria and Slovenia much of this investment is being facilitated by EU grants. Chart 88 Romania Public Investment by Level of Government and as a Share of GDP 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 113 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Serbia The Intergovernmental Finance System In 2006, Serbia passed the Law on Local Government Finance. This law set the share of the personal income tax local governments receive on an origin basis in framework legislation for the irst time. It also decentralized the administration and collection of the property tax to local governments and set the total pool of funds to be used for both vertical and horizontal equalization at 1.7% of the GDP. 114 The irst call on this pool of funds is for horizontal equalization. Local governments whose per capita revenues from shared taxes are less than the national average –calculated are entitled to an equalization grant. Their grants are equal to a percentage of the diference between their per capita revenue from shared taxes and a percentage of the national average multiplied by their populations. The remainder of the pool is allocated by formula to all local governments as an Unconditional Transfer. The allocation of the transfer to individual local governments is determined in accordance with uniform criteria set in the Local Government Finance Law. These criteria include metrics for population, territory, number of classes in elementary and secondary schools, number of elementary and secondary school buildings, number of children attending preschool and number of pre-school buildings. The general transfer thus has an equalizing efect, independent of the equalization grant. The economic crisis of 2008-9 had extremely negative consequences for the Serbian economy in general, and local government budgets in particular. In 2009 the GDP declined 3.5%, and the real-estate market collapsed, leading to a sharp decline in shared taxes and own-revenues associated with property transactions. But the situation was made much worse by (NALAS) Network of Associations of Local Authorities of South-East Europe the government’s suspension of the Law on Local Government Finance between 2009 and 2011 which led to a dramatic fall in the unconditional transfer. In 2011, amendments were introduced into the law that radically changed its character. The share of the wage tax that local governments retain (on an origin basis) was increased from 40% to 80% for all municipalities except Belgrade, whose share was raised to 70%. But at the same time, the amount of unconditional transfers was reduced, and a smaller pool of grant funds was allocated to municipalities in accordance with a complicated development index that divided them into four groups. Municipalities in the fourth group continued to receive 100% of the transfers they received before, while those in the third group got 10% less, in the second group 30% less and in the irst group received 50% less. In 2012, the Law was amended again, this time signiicantly limiting some local communal fees like the business sign tax and eliminating others like the local motor vehicle fee. Meanwhile, the national government raised all taxes that accrue to the central budget, including VAT, the capital income tax, excises, and social contributions. In June 2013, the government reduced the rate of the wage tax from 12% to 10% while increasing the threshold for non-taxable income. These changes led to a direct loss of local revenue of about EUR 200 million. At the same time, the government increased the rate of payroll taxes for social contributions from 22% to 24%, basically transferring what it had taken away from local governments to the National Pension Fund. Finally, on January 1, 2014, the government eliminated the Land Use Fee, the second most important source of own-revenue while passing a new Property Tax Law that will go into efect in 2015. (NALAS) Network of Associations of Local Authorities of South-East Europe The 2011 amendments also created a new transfer called the Solidarity Transfer which all municipalities are entitled to except the City of Belgrade. The size of the Solidarity Transfer is equal to 10% of the wage taxes of the City of Belgrade. It is allocated to local governments through the use of the complicated coeicients for development that now divide municipalities into four groups. Unfortunately, the 2011 introduction of the Solidarity Fund, and the adjustment of all transfers by the development index have rendered the Serbian intergovernmental inance system in general, and its equalization mechanism in particular extremely non-transparent. As a result, the Standing Conference of Serbian Towns and Municipalities is trying to once again amend the Law on Local Government inance in order to return to its original principles. Eforts to do this began in the irst half of 2014 when the Ministry of Finance created a working group to redraft the law. These eforts were postponed when the Ministry of Finance resigned but were resumed in 2015. The Property Tax The Property Tax in Serbia is regulated by the Law on the Property Tax, the Law on Tax Procedure and Tax Administration, and the Local Government Finance Law (LGFL). Until the LGFL came into force on 1 January 2007, the tax was assessed, collected, and enforced by the national government, but its yield was returned to local governments on an origin basis. With the passage of the LGFL, local governments were made responsible for administering the tax and were given the right to set tax rates within limits set by the law. Local governments, however were given two years to establish their own local tax oices and to fully assume responsibility for administering the tax. The Law on the Property Tax deines the types of properties subject to taxation, who is liable for the tax, as well as the rules governing exemptions and abatements. The Law on Tax Procedure and Tax Administration regulates the grounds and manners for assessment, collection and control of public revenues and regulates rights and responsibilities of taxpayers, their registration, tax ofences and sanctions. This law also regulates other types of tax obligations administered by the local government. Although, the laws regulating the property tax have been reformed several times since 2009, local government oficials have identiied a number of gaps in the legislation. Between 2006 and 2014 the yield of the property tax increased from 0.27% of GDP to 0.70% of GDP. Much of this growth came in 2014 and is associated with a grant program that provided additional revenues to local governments who increased property tax collection. The property tax now represents about 30% of local government own revenue and about 11% of total revenues. For legal entities and individual entrepreneurs keeping business accounts the base of the property tax is the book value of the real estate as presented on last day of the business year. This value should be calculated according to the fair value method as deined by International Accounting Standards. Since 2013, municipalities can assess the market value of business properties if commercial tax payers fail to abide by these standards. REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 115 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 For taxpayers who do not keep commercial books, the base of the tax is the value of the real estate as determined by local governments in accordance with the Property Tax Law. Local governments must calculate the average square meter market price for speciic types of property in each of the locational zones they establish. This value is then reduced by a depreciation rate of 1% per year for the building, up to a maximum reduction of 40%. Owner-occupied residential properties are entitled to a 50% tax abatement, up to a maximum limit of 20,000 RSD. Nonurban properties of less the 60 square meters occupied by persons over 65 years or age are entitled to a tax abatement of up 75%. 116 Legal and physical persons owning property are liable for the property tax, except in cases where the property is owned by the Republic of Serbia, but used by other entities for commercial purposes. In these cases, the user of the property is responsible for the payment of the tax. If publicly-owned property is used by the state authorities, local authorities, or users established by them, these properties are tax exempt. Properties used for diplomatic and consular services, for religious purposes, and for the provision of public utilities are exempt from the tax, as are historical landmarks, roads and railways, and, for ive years agriculture and forestry land that is being restored for its original use. The Geodetic Agency of the national government maintains a cadaster of all properties registered in the Republic of Serbia. This Agency should share its data with local governments, but historically cooperation has been poor. There are also problems with the data. As a result, municipalities maintain their own property tax cadasters. Local Governments are fully responsible for registering the base of the tax, valuing properties, setting rates, issuing tax bills and collecting the property tax, but within the limits set by national legislation. The maximum tax rate on commercial properties is 0.40% and for residential properties 0.30%. Local governments have limited powers to enforce tax collection and cannot deny citizens service for non-payment of the tax. (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Statistical Overview of Local Government Finance in Serbia 2006-2014 Local revenue as shares of total public revenue and GDP fell sharply between 2009 and 2011 as the national government dumped some of its iscal problems on to municipalities. Between 2012 and 2013 they recovered somewhat only to fall again in 2014. Chart 89 Serbia Local Government Revenue as a Share of GDP and Total Public Revenue 2006-2014 117 Local revenues have declined faster than those of the general government during the economic crisis. The rose faster than those of the general government immediately before and after the 2011 national elections. Chart 90 Serbia Fluctuations in the Revenues of the General Government and Local Governments REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Until 2012, about 40% of local revenue came from own-sources, 40% from shared taxes, 15% from unconditional grants, and about 5% from conditional grants. In 2012, this balance was changed by a sharp increase in the local PIT share. Since 2012, reductions in the base and rate of PIT have reduced the yield of the tax for local governments and pushed the system back towards its earlier composition. Chart 91 Serbia Composition of Local Government Revenue 2006-2014 118 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Own-revenue from Communal Fees and Charges has declined sharply since 2009 because the rate of the Property Transfer Tax was cut in half and caps were put on the Business Sign Tax and the Land Use Fee before the latter was eliminated in 2012. The inancial situation of local governments will worsen if plans to eliminate the Land Development Fee go forward. Local governments have almost doubled the yield of the property tax since 2006. Chart 92 Serbia Composition of Own Revenues 2006-2014 (mln EUR) 119 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Local government investment as a share of total expenditure remained stable during the worst years of the crisis because of large infrastructure projects in Belgrade. But they have fallen sharply since and are now under 15%, low for the region. Serbian local governments also spend a large share of their budgets on transfers to individuals and organizations (14%) and subsidies to public utilities (12%), some of which is for capital investment. Debt service payments now account for about 5% of total expenditure. Chart 93 Serbia Composition of Local Government Expenditure 2006-2014 120 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Despite the inancial diiculties of local governments, local wage spending has remained remarkably constant over the last eight years. Investment spending has fallen considerable since 2012. 2014 saw a very sharp increase in property tax collection, due in part to a grant program that incentivized collection. Chart 94 Serbia Investment, Wages, and Property Tax as Shares of GDP 2006-2014 121 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Slovenia The Intergovernmental Finance System 122 The Slovenian intergovernmental inance system is built around the origin-based sharing of personal income tax. Revenue from this source has provided local governments with more than 50% of their budget income since 2006. Unusually for the region, local governments receive virtually no unconditional grants from the central government. There is, however, a complicated mechanism for equalization that works by computing for every local government a “lump sum” per capita expenditure need that is supposed to represent the costs of its statutory tasks. Those local governments who PIT share is insuicient to fund this measure of need are given additional increments of PIT. Own-revenues constitute a falling share of total local government budgets and now account for just over 20% of local revenues. Meanwhile, the share of conditional grants has risen, as has income from borrowing. In the irst years of the inancial crisis, Slovenian municipalities didn’t sufer from the overall down-turn. But In 2011, municipal revenue declined 5.5% and total expenditure fell 9%. In 2012, because of the persistence of the crisis, Parliament adopted austerity measures which also afected municipalities. On the revenue side, the national government reduced the lump sum per capita need calculation used to determine the share of PIT local governments receive by 3.7%. It also froze the national government’s share of investment co-inancing to the already reduced levels of the previous year. On the expenditure side, the austerity measures included a reduction in public sector wages. But there was also an increase in some social transfers. As a result, municipal expenditures decreased by less than 1%. (NALAS) Network of Associations of Local Authorities of South-East Europe In 2012, the Government and the municipal associations signed an agreement for 2013 and 2014 to further reduce the lump sum per capita needs indicator used to calculate shared taxes, essentially forcing municipalities to lower expenditures. In 2013, additional iscal consolidation measures placed new expenditure burdens on municipalities. These included an increase in the VAT rate, a rise in social transfers, and a further reduction in the co-inancing by the national government of local investments. Only the state-mandated reduction of public sector wages worked in the opposite direction. At the end of 2013, the national government passed a new Law on Real Estate Taxation. This Law called for elimination of the Land Use Fee, a charge that had been completely under municipal control and which generated 9% of local revenue. It also transformed the Property Tax into a shared tax that will be fully administered by the national government, and whose yield will be divided 50/50 between local governments and the state. Municipalities would no longer have the right to determine the base of the tax, or to issue exemptions, though they would retain the right to set the rate within centrally set limits. As such the new Law on Real-Estate Taxation, would have signiicantly reduced the iscal autonomy of municipalities. But the Constitutional Court ruled against the law and as a result is was never put in force and the previous Land Use Fee still remains valid. The iscal pressures generated by the inancial crisis have also led to proposals to consolidate local governments in order to improve the economic eiciency of the public sector. The Ministry of the Interior, the competent authority for local governments, has stated that there are too many small municipalities with limited governance capacities. In the summer of 2013, the Ministry proposed a territorial reform that would have eliminated all municipalities with fewer than 5,000 inhabitants in 2014, reducing the number of municipalities from 212 to 122. After protests by mayors and criticism of the proposal by municipalities, the associations, independent experts the proposal was withdrawn. Instead, the Ministry promised to develop a more strategic approach to territorial reform that would include objective analysis, wide discussion, and consultation. This strategic approach is expected to be completed by 2018. (NALAS) Network of Associations of Local Authorities of South-East Europe The Property Tax The Property Tax in Slovenia was decentralized to local governments in 1995 and is regulated by the Law on Property Taxation since 1998. Local governments set property tax rates within limits determined by law and are responsible for registering and valuing properties. But due to the fact that the tax is paid only by natural persons for residential properties, and that the valuation and registration of these properties is poor, revenue from the tax is very low and it represents an unimportant share of local revenues (less than 0.5 %). (Local governments however can impose a Land Use Fee on all urban land. This fee is an original local tax and represents an important share of local revenue (about 9%, see above). and the self-declarations of tax payers. Local governments register properties, value them, and set rates within limits determined by law. The Ministry of Finance however, is responsible for issuing tax bills. Compliance is high, with about 95% of taxpayers meeting their obligations regularly. Simpliied court procedures make it easy for local governments to garnish the wages of delinquent taxpayers. Between 2000 and 2005, the Ministry of Finance developed a computer aided system of mass valuation and in 2013 promulgated a new law to regulate the property tax. This law sharply raised valuations and recentralized the administration of the tax. It was however, iercely resisted by local governments and oicially abandoned by the Ministry of Finance in 2014. 123 Only buildings are subject to the property tax in Slovenia. The value of residential properties are determined by local governments on the basis of a methodology set by the national government. Municipalities adjust square meter base rates by coeicients for location, building age and building quality. Owners and the oicial holders of use rights to state property are liable for the property tax. Property taxes used for diplomatic and consular services and religious purposes are exempt from the tax, as well as state and local government properties used for the provision of public services. There are exemptions for owner occupied dwellings. The Surveying and Mapping Authority of the national government maintains a central registry of all real-estate holdings in Slovenia and this data is shared with local governments. Local governments however, have built up their own iscal cadasters using data from the utility companies REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of Local Government Finance in Slovenia 2006-2012 The overall size of the local government sector in Slovenia increased from about 5% of the GDP in 2006 to close to 6% of the GDP in 2009 and has remained at about this level since then. This suggests that the national government has been distributing the costs of the economic adjustment reasonably fairly between levels of government. Chart 95 Slovenia Local Government Revenue as a Share of GDP and Total Public Revenue in 2006-2014 124 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe In the initial years of the Great Recession, the national government protected local governments’ share of public revenues. Since 2010, however, it has shifted a disproportionate burden of iscal stress onto their shoulders. Chart 96 Slovenia Fluctuations in the Revenue of the General Government and Local Governments Slovenian local governments are heavily dependent on PIT sharing for most of their revenues. There is no general grant in the system and local governments with weak tax bases are given additional increments of PIT. It is unclear how efective this is in reducing horizontal inequities. Recently the share of conditional grants in the system have increased while the share of own-revenues have fallen. Chart 97 Slovenia Composition of Local Government Revenue 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 125 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 The fall in the share of local own-revenues has been accompanied by signiicant changes in its composition: The Land Use and Land Development Fees have been eliminated in favor of the Property Tax. As result, the share of own revenues coming from the tax has increased, though absolute yields have remained stable. Slovenian municipalities also derive a large share of their own revenues from asset sales and rentals, as share that has also increased with the central roll-back of other own-revenues. Chart 98 Slovenia 126 Composition of Own Revenue 2006-2014 Between 2009 and 2013, the investment rate of Slovenian local governments dropped from 45% of total spending to 35% before rising to early levels in 2014. Chart 99 Slovenia Composition of Expenditure in 2006-2014 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Slovenia’s has combined robust local investment with low wage spending. Revenue from the Land Use Fee and the Property Tax is better than most countries relying solely on the property tax but still modest. Chart 100 Slovenia Investment, Wages, Property Tax as Share of GDP 2006-2012 127 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 they cannot complete themselves. They typically focus on water-supply, sanitation and roads to urban centers. Turkey The Intergovernmental Finance System 128 The revenue entitlements of provincial administrations and municipalities from the national budget are deined by Law number 5779, passed in 2008. According to this law, different types of local governments are entitled to diferent percentages of national taxes. Thus, 6% of national taxes are earmarked for metropolitan municipalities, 4.5% for district municipalities, 1.5% for other municipalities and 0.5% for special provincial administrations. Depending on the type of local government between 60 and 70% of these shares are returned to them on an origin basis. The remaining 30-40% are gathered into grant pools speciic to each type of local government and redistributed according to two criteria, population and a development index. Eighty percent of these pools are then allocated to local authorities on a per capita basis and 20% according to the development index. This index divides local governments into ive groups, with the least developed group getting 23% of the pool and the most developed group gets 17% of the pool. Unfortunately, all of these revenues are classiied as shared taxes, instead of being divided into shared taxes and unconditional grants. Together they account for between 40% and 45% of local government revenues, with revenues from own sources accounting for a similar share and conditional grants making up the diference of about 15%. Conditional grants are generally used to help poorer jurisdictions. For example the Koy-des Program provides additional support for villages and the Bel-des Program provides support for small districts. These Programs help villages and districts complete investment projects that (NALAS) Network of Associations of Local Authorities of South-East Europe Turkey’s macroeconomic journey over the past decade can be divided into three phases. The irst came after the currency and banking crisis of February 2001 and lasted until the global crisis of 2007-08. Most macroeconomic indicators improved during this period. The public debt-to-GDP ratio was halved from a post-crisis peak of 75%, while inlation dropped from around 70% to single digits. Major reforms of the banking sector afected all sectors and credit lowed back into the economy. GDP per capita rose from about $4,000 to almost $11,000 (in current U.S. dollars) in 2013. The second phase began with the global inancial crisis of 2008, during which Turkey’s economy contracted by 5%. But recovery came remarkably quickly. Signiicant policy easing and an exceptionally low interest rate environment at home and abroad allowed for growth to average 9% over the next two years. Now the country has entered a third phase. Growth has visibly slowed and the economy seems driven by the ups and down of the Eurozone crisis and the decisions about quantitative easing taken by the United States Federal Reserve. Public spending has quickened while private investment remains sluggish, suggesting that the private sector-led growth that Turkey’s government once liked to boast about is losing momentum. On March 30, 2014, Turkey held local elections for metropolitan and district mayors, as well as for municipal councils in cities, and muhtars (village leaders) and “elderly councils” in rural areas. The governing Justice and Development Party took 43% of the vote, winning 818 of 1395 municipalities and 11,309 council seats. The opposition Republican People’s Party came in second with 26% of the vote, winning 232 municipalities and 4,320 council seats. With this election, Turkey now has two distinct types of local government structures: First, the old system continues in provinces in which there are no cities whose populations (NALAS) Network of Associations of Local Authorities of South-East Europe are larger than 750,000 inhabitants. In these provinces, there are three basic types of local governments: small cities, special provincial administrations, and villages. Second, in the 30 provinces where there are cities with populations larger than 750,000, these big cities became metropolitan cities while special provincial administrations and villages were eliminated. As a result, the number of metropolitan cities increased from 16 to 30, and in 30 provinces where they exist there are only two forms of local government, metropolitan cities, and the district cities underneath them. The Property Tax In Turkey, there has been a property tax since the time of the Ottoman Empire. Until 1937, the Turkish Republic used the Ottoman law on property taxation. It then adopted new legislation, legislation which has not been amended since 1970. The tax is one of the most important taxes collected by Turkish municipalities. Between 2006 and 2014 the total yield of the property increased from 0.25% of GDP to 0.31% of GDP. In 2014 the tax accounted for 13% of local government own-revenues and 6% of total revenues. The base of the property tax is the square meter value of urban buildings and land adjusted for location, use and building quality. Municipalities are legally required to value properties every four years. Municipalities also must value properties when agricultural land is rezoned as urban construction land. Valuation is done on the basis of tax declarations iled by the owners and users of property. Municipalities are also responsible for setting property tax rates within the limits set by the national government. These limits are currently 0.1% and .3% of assessed value except in metropolitan municipalities within large agglomerations like Istanbul. In such metropolitan areas minimum and maximum rates are double what they are elsewhere. the property is liable for the tax. There is however, a long list of properties and types of taxpayers who are exempted from the tax. These include: all agricultural land; properties owned by national and local governments; public universities; water, electrical and natural gas facilities; religious institutions and cemeteries; embassies and consulates and slaughterhouses. Pensioners, widows, orphans, disabled people, single mothers, war veterans, and relatives of martyrs are also exempted from the tax. The national government determines the methodology for valuing property and sets minimum and maximum rates. It also maintains a cadaster of all properties in the country. This cadaster is managed by a department within the Ministry responsible for the Environment and functions well. To oicial register the purchase of property and to receive a deed for it, a transfer tax equal to about 2% must be paid to the national government. All municipalities have access to the national cadaster and can use it to identify properties and owners within their jurisdictions. The national government will also block the sale of properties until property taxes are paid. It will also deduct property tax liabilities from any payments that individual or irms are scheduled to receive for work done under government contracts. In 2011, the national government forgave 50% of penalty interest on unpaid tax liabilities and allowed delinquent tax payers to spread out the payment of their debts over 36 monthly installments. Municipalities are responsible valuing property, setting rates with the limits determined by the national government, and for collecting the property tax. The tax is paid in two installments but there is no individual billing. Instead tax payers are required to come to city hall to determine how much they owe, and to pay the tax. Municipalities do not have the right to deny citizens services for the non-payment of the tax. The legal owners of land and buildings, as well as those with usufruct rights are liable for the property tax. If these taxpayers do not ile declarations, whoever is actually using REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 129 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Statistical Overview of Local Government Finances in Turkey Since 2005, local government revenue as a share of GDP has increased by about 1%, while it has remained more or less stable as a share of total public revenues. This growth was not afected by the economic downturn of 2009. Local government debt, including unpaid liabilities to suppliers, has also been stable at about 3% of GDP. Chart 101 Turkey Local Government Revenue as a Share of GDP and Total Public Revenue in 2006-2014 130 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe The composition of local revenues has changed little between 2006 and 2014 except for a very recent fall in the amount of conditional grants they receive for investment purposes. Chart 102 Turkey Composition of Local Government Revenue 2006-2014 131 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 Local governmnet revenues have, however grown sharply in EUR terms since 2010. Since 2006, local governments have increased own-revenue collection by 65%. Chart 103 Turkey 132 Composition of Local Revenue 2006-2014 (EUR per capita) Local investment as a share of total expenditure declined slightly in 2009-10 rose again in the last two years and is again above 35%. Wages as a share of total expenditures have also declined while expenditures on good and services have increased. This suggests that many local governments are outsourcing the provision public goods to commercialized providers. Chart 104 Turkey Composition of Expenditure in 2006-2012 (NALAS) Network of Associations of Local Authorities of South-East Europe (NALAS) Network of Associations of Local Authorities of South-East Europe Neither wages nor the yield of the property tax have increased as a percentage of GDP. Local public investment has recently risen to about 2% of GDP, while outstanding debt has again risen to over 3% of GDP. But this is due more to unpaid liabilities to suppliers and contractors than it is to bank debt. Chart 105 Turkey Investment, Wages, Outstanding Debt and Property Tax as Shares of GDP 2006-2014 133 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 REPORT | Fiscal Decentralization Indicators for South-East Europe: 2006-2014 PARTNER ORGANISATIONS Organizations, institutions and companies that have given signiicant support to NALAS and its Member Associations are recognized as NALAS Partners. Their support may include, but is not limited to lobbying for NALAS and its members, expertise support and inancial support. In addition, NALAS proved to be a valued asset for many of these partners, by providing regional experience, guidelines or coordination of activities conducted in the member countries Implemented by: Institutional Strengthening of the Network of Associations of Local Authorities of South-East Europe (NALAS) Swiss Agency for Development and Cooperation Internet: www.sdc.admin.ch