## _072709 — Executive Summary ## Source details **Canonical URL:** [_072709 — Executive Summary](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2009/_072709.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2009/_072709.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2009/_072709.pdf.json) --- ### Executive summary — key findings and policy lessons - Fiscal decentralization is driven mainly by political pressures (multiple ethnicities, wide regional disparities, demand for more participatory government) and can have important macroeconomic implications. - IMF/Fund advice focuses on designing and implementing intergovernmental fiscal arrangements to minimize risks to macroeconomic stability and debt sustainability while safeguarding public service provision. - General lessons distilled from Fund engagement (including FAD TA to ten representative countries) include: - The sequencing of decentralization matters: resources should be made available to subnational governments pari passu with the assignment of spending responsibilities. - Pace of decentralization should be linked to subnational governments’ capacity; increased devolution should be conditioned on compliance with a minimum set of public financial management (PFM) requirements and can proceed asymmetrically across jurisdictions. - Subnational governments should be provided with an overall envelope of resources (own revenue and transfers) that ex ante would allow them to carry out assigned spending responsibilities at an average level of efficiency. - Control over a portion of own resources is key to promoting accountability and fiscal responsibility; assignment of own revenue must reflect economic considerations (degree of mobility of the tax base) and institutional capacity (subnational tax administration). - Because own resources typically fall short of spending responsibilities, intergovernmental transfers (including revenue-sharing) are needed to offset vertical imbalances and to moderate horizontal imbalances. - Equalization transfers are the preferred means to address horizontal imbalances, taking into account relative taxing capacities and spending needs given available data. - Transfers should be designed to minimize subnational fiscal vulnerability to cyclical fluctuations. - Subnational borrowing controls must be carefully designed to ensure fiscal discipline. - Sole reliance on market discipline is often inadvisable because many countries do not meet conditions for its effectiveness. - A combination of reasonably flexible rules with market discipline can work better if timely and reliable information on subnational finances (including contingent liabilities) is available. - Effectiveness of Fund advice has varied with political commitment of national authorities; ongoing Fund engagement and coordination with other multilateral and bilateral providers remain important. *Source: Executive Summary of IMF paper _072709 - Executive Summary.* ### Introduction — scope and approach - Paper presents an overview of Fund (particularly FAD) advice across surveillance, Use of Fund Resources (UFR), and technical assistance (TA) on intergovernmental fiscal relations. - Advice covered: - expenditure and revenue assignments; - design of intergovernmental transfers; - PFM and subnational revenue administration; - subnational borrowing. - FAD tailored advice balancing macroeconomic constraints, efficiency vs. distributional objectives, and institutional factors. - Ten illustrative countries informing the analysis: Bolivia, People’s Republic of China, Colombia, Democratic Republic of Congo, Indonesia, Kosovo, Liberia, F.Y.R. Macedonia, Mexico, and Nigeria. ### Macrofiscal implications and operational policy implications - Changes in spending assignments and financing (taxes, transfers, borrowing) affect budgetary balances and debt positions, motivating Fund engagement. - Operational implications for decentralization design: - Ensure resource envelopes accompany expenditure devolution to preserve enforceable budget constraints. - Condition devolution on PFM capacity and compliance; allow asymmetric sequencing across jurisdictions. - Assign own revenues weighing tax base mobility and administrative capacity to promote accountability and sustainability. - Design transfers to: - offset vertical imbalances; - address horizontal disparities via formulaic equalization accounting for taxing capacity and spending needs; - reduce cyclicality exposure of subnational budgets. - Design borrowing frameworks combining rules and market discipline, supported by timely and reliable subnational fiscal information and disclosure of contingent liabilities. ### Structure of Fund advice and emphasis - Advice areas: - defining spending responsibilities and sequencing devolution; - ensuring sound subnational PFM; - defining intergovernmental revenue arrangements (own revenues, revenue sharing, transfers); - mechanisms to control subnational borrowing and contingent liabilities. - Emphasis on coherent design so subnational governments face credible “hard budget constraints” and fiscal incentives align with responsibility/accountability. --- ### Box 1 — Measures of Fiscal Decentralization ### Measurement approaches and scope - Empirical measures: - Quantitative: subnational shares in total spending and revenue; measures of reliance on transfers. - Qualitative: institutional and regulatory framework indicators. - World Bank identifies 18 indicators of fiscal decentralization. - GFS database described as best source of internationally comparable data on fiscal variables by government level. ### GFS coverage and data availability (exact figures preserved) - Total IMF members 185 - Countries reporting GFS statistics 152 - of which: with published central government data 1/87 - of which: with published data of all government levels 1/60 - Note: tally covers countries with published data during 2002-06, with data for at least one year. ### Limitations and biases in quantitative measures - Available data tend to overestimate true degree of decentralization: - No clear distinction between “autonomous” versus “mandated” revenue and spending. - Expenditure undertaken under central mandate may reflect deconcentration rather than decentralization. - Subnational revenue not often distinguished between “own-source” and “shared” revenue. - Aggregation by government level can mask heterogeneity across subnational entities. - Limited coverage can cause sample selection bias: reporting countries more likely to have well-developed subnational governments. ### Need for qualitative indicators - “There is much more to fiscal decentralization than quantitative measures suggest.” - World Bank and OECD have promoted qualitative data collection on: - expenditure and revenue assignment design; - transfers; - borrowing arrangements. - Coverage remains limited (e.g., revenue assignment information available for 31 countries, of which 20 are OECD countries). ### Key policy links highlighted - Measurement issues relate to: - center’s stabilization capacity when subnationals control large spending shares; - role of central revenue for redistribution and entitlement programs; - potential weakening of fiscal discipline with decentralized spending/taxing; - importance of nationally binding rules and intergovernmental cooperation to promote stabilization and fiscal sustainability. - In practically all countries subnational spending shares exceed revenue shares, implying central transfers are commonly required to cover vertical gaps. *Source: Box 1 in IMF background studies.* --- ### Spending Assignments, Clarity, and Sequencing ### Political and economic pressures - Political/economic pressures frequently prevent orderly decentralization. - Examples of mismatches where resources and responsibilities were mistimed: Colombia (1990s), Indonesia (2000), Nigeria (recent oil-boom years), and potential risks in DRC and Kosovo. - Typical dynamic: administrative weakness limits subnational capacity initially; rising revenues later promote additional (often inefficient) spending. - Consequences: emerging central deficits without subnational surpluses; deteriorating general government budget/debt positions; unsustainable borrowing, arrears, and service-quality declines in some transitions. ### Hard budget constraints and comprehensive reform packages - Effectively enforced “hard budget constraints” are difficult to design and enforce. - Isolated changes to intergovernmental fiscal relations risk inconsistencies and fiscal imbalances. - Territorial reorganization may be required where subnational units are too many or too small to be viable (examples: Liberia, Macedonia). - Increasing local participation and staged reforms can mitigate political pressures and allow time for capacity building. ### Defining spending responsibilities — theoretical guidance and practical limits - Normative principle: services should be provided by jurisdiction that internalizes costs and benefits (Oates, 1972); national public goods remain central responsibilities. - Practical realities: - Most public outputs are not neatly national vs local; concurrent/joint assignments are common (health, education, social welfare, environment, infrastructure, water and sanitation). - Spending assignment decisions cover formulation, financing, and implementation roles. - Reforms should clarify responsibilities and identify financing sources rather than attempting full remapping. ### Problems from unclear or overlapping responsibilities — country examples - Causes: unclear constitutional/legal provisions; duplication and weak coordination; insufficient local ownership. - Accountability erosion via “blame game” illustrated by health care in Italy (regional administration, central financing, repeated bailouts). - Country examples: - Colombia: fragmented responsibilities between municipalities and center for education/health; union opposition impeded municipal control. - Bolivia: split responsibilities created unavailable information on total spending and personnel. - Nigeria: decentralized primary education without capacity led to shortfalls, eventual recentralization of hiring/management. ### FAD advice on spending assignments and sequencing - Clarify roles and legal frameworks; align assignments with subnational implementation capacity. - Use asymmetric/phased transfers of responsibilities (example: Macedonia phased approach; China centralize pensions). - Suggested sequencing per sector: transfer physical assets + maintenance; then personnel decision powers over time; final stage paying a significant share of current expenditures (e.g., part of wage bill). - Table 2 country recommendations preserved in original document for case-level guidance. ### Impact on service quality - Theory: decentralization can improve performance via efficiency and accountability. - Evidence limited and mixed due to: - lack of outcome measures; - weak accountability links among providers, citizens, policymakers; - partial decentralization undermining local accountability and service outcomes. *Source: Background studies chapter on expenditure assignments and sequencing.* --- ### Public Financial Management (PFM) and MoF Roles ### Ensuring sound PFM - Clear, transparent PFM frameworks at subnational and central levels are key prerequisites to effective fiscal decentralization. - Strengthening PFM should be a pre-requisite for increasing decentralization (examples: Bolivia, DRC, Indonesia, Kenya, Kosovo, Liberia, Nigeria). ### Critical PFM elements (findings) - Realistic budget envelope prepared timely. - Adequate budget classification system, preferably compatible with international standards (e.g., GFS2001). - Accounting framework consistent with budget classification. - Effective audit and control mechanisms with high probability of detection and penalties for misuse. - Firmly enforced requirements for timely and accurate reporting. ### Budget framework, reporting, and information systems - Budgets should use common macro assumptions, harmonized classifications, and coordinated schedules. - Problems from incomplete coverage and weak reporting observed in Nigeria, Liberia, and China. - Fiscal year misalignment (example: Tanzania corrected by 2008). - Consensus-building and framework legislation effective where center cannot impose laws (examples: Nigeria’s 2007 Fiscal Responsibility Act; Brazil’s Fiscal Responsibility Law). - Timely monitoring, common accounting/reporting frameworks, and adequate information systems (interfacing across tiers) are essential. - Effective auditing and enforced sanctions for noncompliance critical. ### Treasury Single Account (TSA) and registries - TSAs consolidate cash, minimize borrowing, facilitate asset/debt management, and improve records. - TSAs need not impair subnational autonomy; ledger accounts can substitute bank accounts for tracking. - Kosovo: municipal operations covered by central TSA aiding reporting. - Recommend subnational debt and fiscal risk registries where below-the-line information is poor (examples: Bolivia, China, Mexico). ### MoF organizational role and capacity - MoF responsibilities include issuing financial management instructions, guidance for budget preparation, collecting/consolidating/publishing subnational budgets and execution data, and monitoring rule adherence. - Recommend establishing dedicated MoF units with adequate staff and computerized support (examples: South Africa IGFR Directorate >80 staff; Tanzania and Rwanda have few staff). - Prepare early: identify job descriptions, skills, and training needs (Liberia, Nigeria recommendations cited). *Source: IMF analysis on PFM arrangements and MoF roles.* --- ### Revenue Assignments, Own-Revenue, and Revenue-Sharing ### Funds follow functions — matching principle - Resources assigned should be linked to devolved functions: subnational spending responsibilities financed by a mix of own-source revenue, shared taxes, transfers, and borrowing where allowed. - Revenue assignments should be preceded by clear definition of functions to be funded. - Risks when mismatched: DRC planned to assign ~40 percent of overall revenue to provinces while responsible for ~20 percent of overall spending (as noted in source text). ### Right financing mix and taxing autonomy - Own-revenue at the margin promotes fiscal discipline and accountability. - Significant expenditure devolution may require significant subnational tax handles to link spending to financing. - Excessive tax autonomy can widen disparities and undermine central stabilization; appropriate degree of autonomy depends on country-specific factors. ### Normative principles for tax assignment - Central government typically assigned taxes with large/elastic bases (income taxes) for stabilization/redistribution; import duties also central. - Local taxes appropriate where bases are immobile and administration feasible (property, user charges). - To avoid tax competition and erosion, local tax bases should be relatively immobile, evenly distributed, and stable. ### Subnational own-revenue instruments — properties and trade-offs - Property and land taxes: immobile base, visible, but administrative valuation challenges; typically limited yields; cross-country shares vary (examples preserved): - property tax share in subnational revenue: 5 percent in Turkey, 7 percent in Norway, 90 percent in New Zealand, 100 percent in Australia, Ireland, United Kingdom. - property taxes as share of GDP: on average about 1 percent in unitary countries and 2 percent in federal countries. - Benefit taxes and user charges: low revenue potential; politically constrained. - Excises at subnational level: can prompt smuggling/cross-border shopping if rates differ. - Subnational VAT: long-term potential but administrative/border-control difficulties; dual VAT options exist (Canadian dual VAT cited as relatively successful; Brazilian ICMS generated tax wars). - Subtraction-type origin-based VAT (Italian IRAP, German Gewerbesteuer): avoids cascading; relatively simple but may be unsuitable where local administration capacity is limited. ### Advantages/disadvantages by tax type (high-level) - Personal Income Tax: buoyant, visible, cost-effective if piggybacked; can create horizontal imbalances and distortions with mobile people. - Corporate Income Tax: mobile base, complex administration — typically central collection. - User Fees: visible, linked to benefits, but low yield. - Property Tax: immobile and stable base but administrative difficulties. - Sales/Excises: easy to administer but risk cross-border shopping. - VAT: good if well designed; complex administration. - Resource Tax: significant horizontal imbalances and volatility. ### Surcharges / piggybacking and revenue-sharing - Surcharges enable subnational revenue using central administration; limited autonomy; rates often capped by center. - Revenue-sharing akin to transfers: subnationals lack marginal control over rate structure; does not equalize unless designed to do so. - Aggregate revenue-sharing may provide more resources than own-source but must be accompanied by own-source revenue to promote accountability. - Revenue-sharing can increase macroeconomic rigidity and procyclicality if not designed with smoothing mechanisms (e.g., moving averages). ### FAD country-specific recommendations on own-revenue and revenue-sharing - General advice: clarify legal frameworks, streamline own-source revenue, eliminate nuisance taxes, strengthen property taxes where administration weak (DRC, Indonesia, Kosovo, Liberia, Macedonia, Nigeria). - In more advanced administrations, create VAT-like taxes or piggyback arrangements (Colombia, Mexico). - Selected country recommendations preserved as in source (e.g., Indonesia: include land and building tax as local tax; Mexico: piggyback IDTU; China: allow provincial surcharges on PIT; Nigeria: additional revenue sources and repeal nuisance taxes; Bolivia: assign significant own sources conditioned on sequencing). ### Natural resource revenues - Resource revenues unsuitable for origin-based sharing due to volatility and concentration. - Recommendations: avoid sharing resource revenues (DRC, Liberia, Nigeria); use production-related excises with equalization transfers instead. *Source: IMF background chapter on revenue assignments.* --- ### Transfers — Design Principles and Country Guidance ### Purposes and types of transfers - Transfers correct vertical imbalances and address horizontal disparities. - Types: - General purpose (unconditional) transfers. - Equalization transfers to correct horizontal imbalances. - Block grants for sectors. - Special-purpose (earmarked/conditional) grants for designated spending. - Transfer systems must complement spending responsibilities and revenue assignments. ### Design principles and discretion - Mandatory (entitlement) vs discretionary transfers distinction important. - Allocation mechanisms vary from simple per capita to complex formulas combining poverty, geography, spending needs, and revenue capacity. - FAD recommends transparent formulae including spending needs and revenue-capacity criteria; avoids reliance on historical spending or revenue. ### Smoothing cyclicality and coparticipation - Transfers should minimize subnational exposure to cyclical fluctuations (moving averages or GDP-based smoothing recommended). - Coparticipation (matching) can divert funds from basic spending and disadvantage poorer regions; use special-purpose grants only for monitorable programs. ### FAD country recommendations on transfers (selected highlights) - DRC: establish equalization formula-based transfers (e.g., inhabitants and surface). - Liberia: adopt simple transparent formula for county operations when decentralizing. - Nigeria: move toward general transfer system based on spending needs and revenue capacities; introduce floors financed by savings from high oil prices; limit specific-purpose grants. - Bolivia, Colombia, Mexico, Kosovo, Macedonia, China, Indonesia: specific reform recommendations preserved in source (e.g., reduce earmarking, phase-in equalizing systems, reform complex systems). *Source: IMF analysis and Table 8 recommendations.* --- ### Subnational Tax Administration and Revenue Collection ### Central vs subnational administration considerations - Choice depends on country circumstances: central administration can collect subnational taxes on agency basis where local capacity weak. - FAD TA limited but targeted where financing changes require tax administration rethinking. - Examples: Colombia (centralize while building local capacity), Macedonia (align administration to municipal reorganization), China (1994 reforms gave center more tax handles and established state tax administration). ### Revenue-sharing and surcharging to address administration constraints - Revenue-sharing/surcharging can circumvent administrative challenges for major taxes (e.g., surcharges on income taxes). - Concerns: subnational worry central administration may lack incentives to collect for subnationals — mitigated by service contracts. - Feasibility constraints: high setup costs for cadastre/information systems; complex tax assignment rules can hinder modernization (e.g., Sudan example); DRC recommendation to compensate central tax office if it collects local taxes or limit local tax administration to clearly defined local taxes. *Source: Paragraphs 73–74 and related country examples.* --- ### Subnational Borrowing — Controls, Preconditions, and Policy Advice ### Rationale for borrowing controls - Objective: avoid excessive subnational borrowing due to adverse externalities: - preempting financing and raising interest rates; - increasing risk premia on government bonds; - bailout costs. ### Four broad categories of borrowing controls 1. Reliance on market discipline 2. Cooperation between central and subnational governments 3. Rules-based controls 4. Administrative controls - Ranked from maximum subnational autonomy (market access) to maximum central control (administrative bans). ### Approaches — advantages, preconditions, and examples - Market Discipline: - Advantages: self-control; rating agency monitoring. - Preconditions: comprehensive, timely, reliable information; developed markets; no privileged financing; no bailout history. - Examples: Canada, Australia, Sweden. - Limitations: seldom met; prior bailouts weaken signals (e.g., German Länder experience). - Cooperative Approach: - Advantages: dialogue, shared responsibility. - Preconditions: constitutional/institutional underpinnings, culture of discipline, reliable information. - Examples: Australia’s Loan Council, EU Domestic Stability Pacts, Brazil debt agreements, Argentina bilateral pacts. - Mixed record; effectiveness context-dependent. - Rules-Based Controls: - Forms: quantitative ceilings on borrowing/debt/debt service (often relative to revenues); procedural budget rules. - Critical design: well-defined, transparent, flexible rules; full coverage; clear escape clauses; credible sanctions; revenue base stable over cycle. - Implementation via national legislation or subnational constitutions. - Administrative Controls: - Use in unitary states/local governments; may limit flows or debt stock; risks moral hazard if central authorization creates expectation of bailouts. - Effectiveness depends on de-politicized central decisions and comprehensive information. ### Overall assessment and complementary measures - No one-size-fits-all; rules-based approaches supported by enforcement may suit countries lacking market discipline preconditions. - Complement rules by building market preconditions: transparency, eliminate privileged financing, establish no-bailout track record. - Recommend ex post insolvency mechanisms for efficient resolution of distressed debt. ### FAD country recommendations on borrowing (selected) - DRC: define control mechanisms; tight controls to avoid unsustainable provincial spending. - Kosovo: proceed slowly relaxing municipal borrowing ban; link relaxation to PFM advances and legal framework. - Macedonia: maintain borrowing prohibition initially; consider Municipal Investment Fund for limited borrowing. - Indonesia: impose borrowing limits and strengthen reporting. - Mexico: establish prudential limits linked to fiscal rules; create full subnational debt registry; legal PPP framework. - Nigeria: restrict domestic borrowing to fiscally sound local governments; maintain no external borrowing; build DMO reporting and debt management capacity. - Bolivia, China: strengthen framework, transparency, and fiscal/fiscal-risk registries; consider asymmetric frameworks allowing advanced subnational borrowing with preconditions. ### PPPs and contingent liabilities - PPPs pose fiscal risk if legal/institutional capacity lacking; require strong project evaluation, coordination with public investments, disclosure standards, and transparent financial reporting. - Fund cautioned against PPPs in countries lacking capacity (Kosovo, Macedonia) until frameworks are in place. *Source: Paragraphs 75–84 and Table 10 recommendations.* --- ### Synthesis: Design Principles, Sequencing, and Implementation ### Overarching design considerations - Intergovernmental fiscal arrangements span multiple policy and institution-building issues requiring coordination and sequencing. - No single “right” model—arrangements must incorporate macroeconomic constraints, balance efficiency and distributional objectives, and account for constitutional/legal constraints and subnational capacity. ### Sequencing and pace - Resources must be provided pari passu with assignment of responsibilities. - Pace linked to capacity; devolution conditioned on minimum PFM requirements and can be asymmetric across jurisdictions. - Asymmetric approaches useful for capital cities and advanced regions. ### Clarity, capacity-building, and complementary reforms - Clarify overlapping responsibilities to avoid duplication and accountability erosion. - Complementary reforms in civil service and regulatory frameworks may be needed. - Significant investments in subnational PFM systems, capacity-building, and technical assistance required for efficiency gains. ### Resource envelopes and hard budget constraints - Subnationals should receive an overall envelope enabling assigned responsibilities at average efficiency. - Persistent mismatches lead to inadequate services, borrowing, or arrears. - Assessing subnational spending needs is difficult; sample studies and phased approaches recommended. ### Own revenues, transfers, and smoothing cyclical risk - Control over own revenue portion fosters accountability; assignment should reflect tax base mobility and administrative feasibility. - Transfers required to address vertical gaps; equalization transfers preferred to correct horizontal imbalances. - Use smoothing mechanisms (moving averages, rainy day funds) to reduce cyclical exposure. ### Information, registries, and crisis resolution - Timely, reliable information on subnational finances and contingent liabilities essential. - Recommend subnational debt registries and fiscal risk registries (examples: Bolivia, Mexico, China). - Reinforce debt limits with prudential regulations on lenders and develop mechanisms for debt crisis resolution. ### Political economy and effectiveness - Effectiveness of Fund advice depends on political support; external crises can catalyze reform. - Piecemeal reforms (e.g., borrowing controls) often easier to implement than comprehensive reforms. - Fund should continue macro-focused engagement and strengthen cooperation with institutions supporting capacity-building. *Source: IMF content unit _072709 (chapter/section on fiscal decentralization).* *Source: IMF document _072709 — Executive Summary and selected background chapters.* ### Executive Summary ...................................................................................................... ### _072709 - Executive Summary ...................................................................................................... ### Executive summary — key findings and policy lessons - Fiscal decentralization is driven mainly by political pressures (multiple ethnicities, wide regional disparities, demand for more participatory government) and can have important macroeconomic implications. - The Fund’s advice focuses on designing and implementing intergovernmental fiscal arrangements to minimize risks to macroeconomic stability and debt sustainability while safeguarding public service provision. - General lessons distilled from Fund engagement (including FAD TA to ten representative countries) include: - The sequencing of decentralization matters: resources should be made available to subnational governments pari passu with the assignment of spending responsibilities. - Pace of decentralization should be linked to subnational governments’ capacity; increased devolution should be conditioned on compliance with a minimum set of public financial management (PFM) requirements and can proceed asymmetrically across jurisdictions. - Subnational governments should be provided with an overall envelope of resources (own revenue and transfers) that ex ante would allow them to carry out assigned spending responsibilities at an average level of efficiency. - Control over a portion of own resources is key to promoting accountability and fiscal responsibility; assignment of own revenue must reflect economic considerations (degree of mobility of the tax base) and institutional capacity (subnational tax administration). - Because own resources typically fall short of spending responsibilities, intergovernmental transfers (including revenue-sharing) are needed to offset vertical imbalances and to moderate horizontal imbalances. - Equalization transfers are the preferred means to address horizontal imbalances, taking into account relative taxing capacities and spending needs given available data. - Transfers should be designed to minimize subnational fiscal vulnerability to cyclical fluctuations. - Subnational borrowing controls must be carefully designed to ensure fiscal discipline. - Sole reliance on market discipline is often inadvisable because many countries do not meet conditions for its effectiveness. - A combination of reasonably flexible rules with market discipline can work better if timely and reliable information on subnational finances (including contingent liabilities) is available. - Effectiveness of Fund advice has varied with political commitment of national authorities; ongoing Fund engagement and coordination with other multilateral and bilateral providers remain important. ### Introduction — scope and approach of the paper - The paper presents an overview of Fund (particularly FAD) advice to member countries on intergovernmental fiscal relations across surveillance, Use of Fund Resources (UFR), and technical assistance (TA). - Advice covered expenditure and revenue assignments; design of intergovernmental transfers; PFM and subnational revenue administration; subnational borrowing. - FAD tailored advice to country-specific circumstances, balancing macroeconomic constraints, efficiency vs. distributional objectives, and institutional factors. - The paper draws particularly (but not exclusively) on FAD engagement with ten countries: Bolivia, People’s Republic of China, Colombia, Democratic Republic of Congo, Indonesia, Kosovo, Liberia, F.Y.R. Macedonia, Mexico, and Nigeria. ### Fiscal decentralization: key issues — theoretical and practical considerations - Definition: Decentralization is the transfer of authority and responsibility for public functions from central government to subnational entities; “fiscal decentralization” refers to changes in intergovernmental fiscal relations. - Traditional (first-generation) normative theory assigns stabilization and redistribution to the center and supports decentralization for allocative/productive efficiency where local information and heterogeneous preferences matter. - Table 1 summarizes the rationale for centralization versus decentralization across stabilization, redistribution, and allocation objectives. - Political-economy (second-generation) perspectives emphasize that decentralization is largely politically driven and that politicians may be self-interested; empirical outcomes do not always match welfare-maximizing predictions. - Two main sources of distortion: - Common pool problem: subnational policymakers do not fully internalize the cost of spending when financed by central transfers or shared revenue, generating overspending and deficit bias. - Soft budget constraint problem: expectation of central government bailouts undermines subnational fiscal responsibility; credible enforcement of budget constraints is crucial. - Empirical evidence on macroeconomic impacts of decentralization is mixed: - Studies linking measures of fiscal decentralization to growth and inflation show contradictory results. - Measurement difficulties (see Box 1) and identification issues (direction of causality) complicate empirical assessment. - The design of intergovernmental fiscal relations matters more than the degree of decentralization per se for efficiency, growth, and macro-stabilization. ### Macrofiscal implications — operational implications for policy advice - Changes in assignment of spending responsibilities and their financing (taxes, transfers, borrowing) affect budgetary balances and debt positions; such macro implications motivate Fund engagement. - Key operational implications for decentralization design: - Ensure resource envelopes accompany expenditure devolution to preserve enforceable budget constraints. - Condition devolution on PFM capacity and compliance; allow asymmetric sequencing across jurisdictions. - Assign own revenues taking into account tax base mobility and administrative capacity to promote accountability and sustainability. - Design transfers to: - Offset vertical imbalances created by limited own revenues. - Address horizontal disparities via equalization formulas that account for taxing capacity and spending needs within data limitations. - Reduce cyclicality exposure of subnational budgets. - Design borrowing frameworks that combine rules and market discipline, supported by timely and reliable subnational fiscal information and disclosure of contingent liabilities. ### Structure of Fund advice and areas of emphasis (high-level) - Advice covers: - Defining spending responsibilities and sequencing devolution. - Ensuring sound public financial management at the subnational level. - Defining intergovernmental revenue arrangements (own revenues, revenue sharing, transfers). - Mechanisms to control subnational borrowing and contingent liabilities. - Emphasis on coherent design so subnational governments face credible “hard budget constraints” and fiscal incentives are aligned with responsibility and accountability. *Source: Executive Summary of IMF paper _072709 - Executive Summary.* ### Box 1. Measures of Fiscal Decentralization ### Box 1. Measures of Fiscal Decentralization ### Measurement approaches and scope - Empirical studies use both quantitative and qualitative indicators to measure fiscal decentralization. - Quantitative: subnational shares in total spending and revenue; measures of reliance on transfers. - Qualitative: aspects of the institutional and regulatory framework for subnational government finances. - The World Bank identifies 18 indicators of fiscal decentralization. - The Government Finance Statistics (GFS) database is described as the best source of internationally comparable data on fiscal variables by government level. ### GFS coverage and data availability - Total IMF members 185 - Countries reporting GFS statistics 152 - of which: with published central government data 1/87 - of which: with published data of all government levels 1/60 - Note: The exercise tallies countries with published data during the period 2002-06, with data for at least one year. - The database includes a statement of government operations and detailed tables on revenue, expenses, and transactions in assets and liabilities. - Data availability is described as relatively (sometimes severely) limited; general lack of standardized recording and reporting across government levels and among jurisdictions at the same level hampers collection of subnational fiscal statistics. - Limited coverage can lead to sample selection bias: countries reporting GFS data are more likely to have well-developed subnational governments with better defined responsibilities, and may not be representative of the majority of decentralized countries. ### Limitations of quantitative measures - Available data tend to overestimate the true degree of decentralization. - No distinction is usually made between “autonomous” versus “mandated” revenue and spending. - Expenditure undertaken by subnationals under mandate from the central government may reflect deconcentration rather than decentralization. - Subnational revenue is not adequately distinguished between “own-source” revenue (where subnationals have some degree of discretion over tax rates and tax bases) and “shared” revenue (with little or no subnational taxing autonomy). These are stated as well-known limitations of GFS data. - Data are typically aggregated by government level, possibly masking important differences within levels. Aggregation in GFS data loses information useful for assessing differences among subnational entities’ fiscal operations and capacities. ### Need for qualitative indicators - Quantitative data eschew important institutional dimensions; “there is much more to fiscal decentralization than quantitative measures suggest.” - Qualitative information is relevant but not always available. - The World Bank and OECD (under the Fiscal Federalism Network) have promoted efforts to collect qualitative information relating to: - design and management of expenditure and revenue assignments, - transfers, - borrowing arrangements. - Coverage of qualitative indicators remains limited—for example, information on revenue assignment is available for 31 countries, of which 20 are OECD countries. ### Key implications highlighted in the surrounding discussion (contextual links) - Measuring decentralization is described as “both difficult and controversial” (Treisman, 2003). - The box is presented within broader analysis that links fiscal decentralization measurement issues to policy concerns such as: - the center’s capacity for stabilization when subnationals control a large share of spending, - the role of central revenue for redistribution and entitlement programs, - the potential weakening of fiscal discipline with decentralized spending and taxing decisions, - the importance of nationally binding rules and intergovernmental cooperation mechanisms to promote short-term stabilization and medium-term fiscal sustainability. - The discussion notes that in practically all countries subnational spending shares are larger than revenue shares, implying varying degrees of central transfers are required to cover vertical fiscal gaps. *Sources: GFS database; Ebel and Yilmaz (2002); and World Bank Fiscal Decentralization website.* ### 20. However, political, and sometimes economic, pressures frequently prevent such ### _072709 - 20. However, political, and sometimes economic, pressures frequently prevent such ### Political and economic pressures on decentralization - Political, and sometimes economic, pressures frequently prevent an orderly decentralization process. - Examples of mismatches between devolution of resources and reassignment of spending responsibilities: - Colombia in the 1990s - Indonesia in 2000 - Nigeria in the recent oil-boom years - Possible current occurrences in Democratic Republic of Congo (DRC) and Kosovo as they move toward decentralization - Typical dynamic: initial administrative weaknesses limit subnational capacity to spend devolved resources (as initially in Indonesia); over time increased revenue availability tends to promote additional (and frequently inefficient) spending. - Consequences: - Emerging deficits at the central government level may not be offset by surpluses at the subnational level. - This can lead to deteriorating budget and debt positions of the general government as a whole or require additional increases in the overall tax burden. - In other cases (e.g., transition economies in the early nineties) central fiscal stress led to pushing spending responsibilities down without passing corresponding resources, prompting unsustainable borrowing, accumulation of arrears, and deterioration in service quality (e.g., health and education). ### Hard budget constraints and fiscal discipline - Key assertion: design, implementation and enforcement of policies that ensure a hard budget constraint are challenging tasks. - Without an effectively enforced “hard budget constraint” on subnationals: - Ex ante matching of spending responsibilities with resources does not ensure ex post adequate maintenance of fiscal discipline. - Subnational governments may overspend and/or slacken revenue effort, creating fiscal pressures for the general government. ### Need for a comprehensive package of measures - Fiscal decentralization reforms require a consistent and well coordinated package of measures. - Risk: changing specific aspects of intergovernmental fiscal relations “in isolation” (e.g., modifying expenditure mandates, introducing new revenue-sharing schemes, changing the transfer system) can create inconsistencies and imbalances across government levels and undermine fiscal policy effectiveness. ### Institutional and political arrangements at subnational level - Where there are too many (or too small to be viable) subnational entities, spending/tax assignments cannot be properly addressed without territorial reorganization (example noted in Liberia and Macedonia). - Increasing local participation in economic decision making outside formal channels (e.g., increased consultation with local communities) can mitigate political imperatives driving fiscal decentralization and allow more time for proper design and implementation. - Sound and viable political mechanisms are necessary to identify and express local preferences; local elections can reveal preferences and establish accountability (example: introducing local elections of mayors contributed to the success of fiscal decentralization in Colombia). ### Defining spending responsibilities (theoretical guidance and practical limits) - Normative guidance: efficiency considerations and the decentralization theorem — “each public service should be provided by the jurisdiction having control over the minimum geographic area that would internalize the benefits and costs of such provision” (Oates, 1972). - National public goods (e.g., defense, foreign affairs, macroeconomic stabilization) should be centrally provided. - Practical realities: - Most public outputs do not lend themselves to neat national vs local categorization; concurrent/joint assignments are common. - Spending assignments reflect “political jurisdictions,” not “economic clubs” (Dafflon, 2006). - Identifying “minimum geographic areas” where costs and benefits can be internalized is generally not feasible. ### Overlapping and concurrent spending assignments - Spending assignment decisions cover: - Which level should formulate a spending program; - Which level should finance it; - Which level should implement the related spending. - Concurrent assignments are common in healthcare, education, social welfare, environment, infrastructure, and water and sanitation. - Overlaps exist regardless of the degree of spending decentralization (as measured by subnational spending shares, Figure 3). - Fiscal decentralization reforms should aim to clarify responsibilities and identify appropriate financing resources, rather than attempting a comprehensive remapping of spending assignments. ### Problems arising from unclear or overlapping responsibilities - Causes of problematic concurrent responsibilities: - Unclear or inadequate constitutional and legal provisions on the role and authority of different government levels. - Duplication of tasks and weak coordination between tiers, possibly resulting in excessive spending and waste. - Insufficient involvement and ownership at lower government levels; for politically sensitive social programs, lower levels may opt for suboptimally low service levels while relying on the center as financier of last resort. - Accountability erosion: unclear assignments enable a “blame game,” weakening incentives for corrective action (example: health care in Italy—center sets standards and largely finances, regions administer service provision; repeated “ex post” bailouts occurred when regional financing needs exceeded center allocations). - Case examples of divided responsibilities and consequences: - Colombia (1991 Constitution): education and health envisaged at local/municipal level with earmarked transfers, but hiring/firing of teachers and health workers remained with the center; powerful unions thwarted moves to municipalities. - Bolivia: municipalities build infrastructure for education and health and handle operation/maintenance; departments hire teachers; central government pays for them; municipalities certify hours worked—resulting in unavailable information on total spending or actual personnel numbers and poor coordination. - Nigeria: primary education assigned to the lowest tier—districts—but many lacked capacity; transfer financing was not effectively used, causing shortfalls in teacher wage payments and eventual recentralization of hiring/management to states and financing through a special purpose or earmarked grant from the center. ### FAD advice and the need for clarity and capacity-aligned assignments - FAD emphasis: - Clarify roles and responsibilities of government levels; rationalize legal frameworks; eliminate overlaps that can lead to waste and duplication. - Assignments should reflect subnational implementation capacity; where capacity is limited or variable, asymmetric options can allow advanced regions and urban centers to take on responsibilities earlier while capacity is built elsewhere (example: Macedonia’s phased approach). - Centralize pooling of pension provisions where decentralization to very low tiers undermines risk sharing (example noted for China). - Decentralization sequencing: - Start with areas where success is likely and accountability can be easily established—“asymmetric” sectoral decentralization. - Suggested stages: transfer of physical assets (e.g., buildings) plus maintenance responsibility; then personnel decision responsibilities (e.g., recruiting teachers) over time; final stage: paying a significant share of current expenditures (e.g., at least part of the wage bill). - Sectoral pace may vary (example: education may be easier to decentralize than health). ### Table 2: Main recommendations on spending assignments in case countries (summary of presented guidance) - DRC: Aim for clear delineation of expenditure responsibilities; transfer gradually in line with provincial capacity strengthening, including PFM; warn against full and immediate transfer to largely unprepared provincial authorities. - Liberia: Precede fiscal decentralization with structural reforms and legislative initiatives; pursue increased deconcentration and improve procedures for County Development Fund (CDF) operations. - Nigeria: Subject to national legislation detailed distribution of functions for individual services; establish streamlined special-purpose transfers to achieve national objectives; ensure availability of financial resources is broadly commensurate with devolution of responsibilities. - Bolivia: Clarify and review spending responsibilities. - Colombia: Clarify legal framework and eliminate overlap. - Mexico: Clarify functional spending responsibilities for lower levels; allow full control over choice of economic inputs (wages and salaries, operations and maintenance, etc.). - Kosovo: Clarify spending responsibilities (often defined in legislation but blurred in practice); assign increased responsibilities in line with proved capacity and well-defined accountability mechanisms. - Indonesia: Clarify and review spending responsibilities; consider all levels of service delivery and limit cost shifting. - Macedonia: Transfer expenditure assignments gradually; proceed at different pace in different sectors and across municipalities (asymmetric options); begin with basic administrative tasks and progressively expand to decision-making powers. - China: Clarify and review spending responsibilities; centralize social security (pensions) to allow better pooling of risks. ### Impact of spending decentralization on service quality - Theoretically, assignment of expenditure responsibilities can improve public sector performance by generating efficiency gains and strengthening accountability. - Empirical evidence is limited: conclusive evidence on a positive impact of decentralization on quality of public services remains limited. - Reasons for mixed or limited outcomes: - Lack of measures of outcomes of subnational spending makes performance measurement difficult. - Service efficiency depends on strong accountability links among service providers, citizens, and policymakers. - Partial decentralization, where citizens cannot hold local governments accountable for budgetary allocations and outcomes, prevents successful provision of decentralized public services and undermines promised gains in service quality. *Source: "Macro-Policy Lessons for a Sound Design of Fiscal Decentralization: Background Studies."* ### 33. By their nature, issues of expenditure assignments are inextricably linked to ### 33. By their nature, issues of expenditure assignments are inextricably linked to ### Linkage with sectoral policies - Expenditure assignment issues are inextricably linked to sectoral policies. - FAD’s TA on fiscal decentralization has often required dedicated sectoral expertise. - World Bank inputs have complemented FAD TA notably on health and education (Nigeria, Macedonia, Kosovo), and on social assistance (China). ### Ensuring Sound Public Financial Management - A clear and transparent PFM framework at subnational and central levels is key to effective fiscal decentralization. - Sound PFM arrangements should be supplemented by legal provisions and institutional mechanisms governing responsibilities and financing of different government levels to generate incentives for efficient resource management. - FAD has emphasized that strengthening PFM arrangements should be a pre-requisite for increasing decentralization (examples: Bolivia, the DRC, Indonesia, Kenya, Kosovo, Liberia, and Nigeria). ### Critical elements of subnational PFM (findings) - Realistic budget envelope prepared in a timely manner. - Adequate budget classification system, preferably compatible with international standards (e.g., the GFS2001—especially for the economic and functional classifications). - Accounting framework consistent with budget classification. - Effective audit and control mechanisms, with a high probability of detection of, and penalties for, misuse of public funds. - Firmly enforced requirements for timely and accurate reporting. ### Budget framework and coverage (findings and country examples) - Budgets of all jurisdictions should be based on common macroeconomic assumptions, harmonized budget classification, and coordinated schedules. - Problems from incomplete budget coverage and weak reporting: - Nigeria: part of transfers to states and local governments is netted out against subnational mandatory spending (including for teachers’ salaries), reducing budget coverage and weakening transparency and accountability. - Liberia: budget document does not include county-level operations; FAD advised including an annex showing consolidated projected spending by county to improve transparency and rationalize spending and pave the way for further devolution. - China: central budget adopted late, subnationals do not receive allocation information until late in the fiscal year. - Fiscal year misalignment can exacerbate coordination problems (example: Tanzania’s former July-June central fiscal year vs. calendar year subnational fiscal years; as of 2008 subnational fiscal year was aligned to central government). ### Consensus building and framework legislation - Where the center cannot "impose" common framework legislation, consensus can be forged: - Nigeria: institutionalize a framework for sharing and spending of oil revenue; encourage states to approve fiscal responsibility legislation paralleling federal government; 2007 Nigerian Fiscal Responsibility Act sets transparency requirements, sanctions, compliance provisions, and guidelines for budgetary practices. - Brazil: national Fiscal Responsibility Law binding for all levels was facilitated by comprehensive consensus-building by federal government with subnational governments, federal congress, and public opinion. ### Monitoring, reporting, and information systems - Timely monitoring and transparent reporting of fiscal operations at all levels, with guidance for subnationals on content and format, are important for macroeconomic management. - A common framework for accounting and reporting—consistent with budget classification—is critical for transparency and accountability. - Adequate information management systems are essential; systems need to interface across tiers to allow comprehensive reporting. - Effective auditing and control systems with enforced sanctions for noncompliance are additional governance conditions. ### FAD emphasis and framework legislation support - FAD’s advice focuses on PFM elements critical from a macroeconomic standpoint, including strengthening framework legislation for PFM to underpin budget formulation, execution, accounting, and reporting. - Examples of collaboration: - Financial Management Legislation in Iraq (2004). - Planned reforms to framework legislation for financial management for all government levels in Bolivia and Peru. ### Treasury Single Account (TSA) guidance - TSAs consolidate government cash, minimize borrowing, facilitate asset and debt management, and establish better records of spending. - Establishment of a TSA need not impair subnational autonomy; TSA can increase transparency and improve decision-making. - Kosovo: municipal operations covered by the central TSA provide the main source for reporting on municipal finances and cash flows. - Whether subnationals should have their own TSAs or share a centralized TSA depends on institutional arrangements and political preferences; ledger accounts may substitute for bank accounts to track spending. ### Below-the-line monitoring and debt/risk registries - When above-the-line information is inaccurate or unavailable, FAD advised monitoring below-the-line subnational operations: - Mexico: recommended tracking subnational fiscal operations from bank transactions with assistance from the Bank of Mexico. - To cover build-up of arrears and contingent liabilities, FAD recommended establishing and maintaining subnational debt and risk registries in some countries (Bolivia, China, and Mexico). ### Organizational issues and MoF role - Decentralization places responsibilities on the MoF to guide and monitor fiscal relations with subnational governments. - Basic MoF responsibilities include: - Preparing and disseminating financial management instructions for subnationals. - Providing guidance for budget preparation. - Collecting, consolidating, and publishing subnational budgets and budget execution data. - Monitoring adherence to financial rules and PFM legal framework. - These tasks require full-time, dedicated staff with computerized support; ideally a special unit should be established within the MoF with clear reporting responsibilities. - Size of the unit depends on the complexity of the devolved government system and MoF role (example: South Africa IGFR Directorate has over 80 staff; Tanzania and Rwanda have only a few staff in respective budget departments). ### Preparations for dedicated MoF units (recommendations) - Begin preparations early in the fiscal decentralization process: identify job descriptions, skills, and training needs. - Liberia: advised to start after main law establishing local governments adopted. - Nigeria: recommended establishment of a Fiscal Analysis Unit to provide a common macroeconomic framework and monitor fiscal developments at all government levels. ### Defining intergovernmental revenue arrangements - Funds should follow functions: resources assigned to government levels should be linked to the scope of devolved functions ("matching principle"). - Two implications: - Subnational spending responsibilities should be financed through a combination of own-source revenue, shared taxes, transfers, and, to the extent allowed, borrowing. - Revenue assignments should be preceded by transparent and clear definition and assignment of functions they are intended to fund. - Risks when not respected: - DRC: planned decentralization strategy would assign some 40 percent of overall revenue to provinces while they would be responsible for some 20 percent of overall spending. - Liberia: risk if pressures to share large resource revenue flows before counties are ready for new spending responsibilities. - Eroded central fiscal space weakens center’s ability to undertake stabilization and redistributive policies. ### Right financing mix and taxing autonomy (analysis and trade-offs) - Allowing access to own revenue at the margin through local taxation promotes fiscal discipline and accountability by curbing perceptions of soft budget constraints. - If considerable expenditure responsibilities are devolved, access to significant tax handles at the subnational level may be needed to link local spending decisions to financing by beneficiaries. - Excessive tax autonomy can create inefficiencies, widen disparities, and undermine central stabilization roles. - The appropriate degree of taxing autonomy is a matter of judgment reflecting historical, political, and country-specific factors. ### Normative theories of fiscal federalism (principles for tax assignment) - Tax assignment derives from optimal assignment of government functions. - Principles include: - Taxes with large and elastic bases (e.g., income taxes) should be assigned to the central government for macroeconomic stabilization and income redistribution. - Allocation functions: benefit taxation should be used by various government tiers. - Local services with local benefit zones (parks, roads) should be financed with local taxes; user charges and fees for public utilities and local transportation. - Goods and services with significant externalities should be financed with region-wide taxes or transfers. - To prevent revenue losses/tax erosion from tax competition, local tax bases should be relatively immobile, evenly distributed across jurisdictions, and yield relatively stable revenue streams. - Public choice critique (Brennan and Buchanan, 1980): if governments behave as rent-seeking, then subnational taxes should be imposed on mobile factors to promote competition among "rapacious" governments; less availability of taxes would help limit the size of governments ("starve the beast" argument). - Main limitation: normative theories do not fully explain tax assignments in the real world. *Source: "Macro-Policy Lessons for a Sound Design of Fiscal Decentralization: Background Studies".* ### 47. More recent positive theories underline the political interactions among ### More recent positive theories underline the political interactions among government levels—and the incentives faced by politicians—as key factors in the distribution of taxing powers ### Political economy and practical considerations in revenue assignment - Recent positive theories emphasize political interactions among government levels and incentives faced by politicians as key determinants of tax assignments, but “no clear indication concerning the optimal tax assignment to different levels of government emerges from them” (Bordignon and Ambrosanio, 2006). - Practical guidance on revenue assignments is guided by multiple considerations: - Ensure funding adequacy. - Promote accountability. - Administrative feasibility—some tax handles suitable for subnational levels in countries with sophisticated administrations may be inadvisable where subnational administrations are weak. - Consequently, major tax handles—such as the VAT and income taxes—tend to be assigned to the central government because administration across multiple government layers presents particular challenges. - Import duties are also typically assigned to the center. - These central assignments often create vertical imbalances in favor of the center, enabling redistribution via equalization mechanisms. ### Spectrum of intergovernmental revenue arrangements - Revenue assignments rarely take the form of full and exclusive assignment to one level; instead, funding sources form a continuum with overlapping alternatives. - Three main types of arrangements (varying by degree of subnational revenue autonomy): - Own-revenue assignments: - Own-source revenue: subnational governments have legal control over definition of taxable bases and rate structure (Tax Base: Yes; Tax Rate: Yes; Administration: Yes/Possibly). - Surcharges or piggybacking on central taxes: subnational governments have limited or no control over the tax base but have (typically bounded) control over tax rate(s) (Tax Base: No; Tax Rate: Yes; Administration: Possibly). - Revenue-sharing arrangements: - Subnational governments receive shares of taxes whose bases and rates are defined (and typically administered) by the central government (Tax Base: No; Tax Rate: No; Administration: Possibly). - Intergovernmental grants: - Transfers where central discretion is higher than in revenue-sharing; subnational governments may not be able to predict amounts with confidence (Tax Base: No; Tax Rate: No; Administration: No). ### Subnational own-revenue assignments: incentives and instruments - Incentive effects: - Assigning own-revenue to subnational governments provides incentives to promote improved economic performance: if subnational governments keep a share of locally raised revenue, they have interest in promoting business- and market-friendly policies to expand local tax bases. - If locally raised revenue is largely diverted to the center or other jurisdictions, local politicians have weak incentives to boost local revenue and economic activity. - Historical/example evidence: argument used to explain contrasting 1990s growth between China and Russia, though subsequent evidence (Treisman, 2006) and later developments complicate a simple causal reading. - Typical own-source taxes and properties: - Property and land taxes: - Reasons for subnational assignment: immobile base; reflect benefit principle; enable subnational governments to determine service levels and raise revenue accordingly; high visibility enhances accountability. - Administrative challenges: determining property values is often difficult where market valuation is hindered by limited real estate market activity and/or limited information on transactions—valuation often requires judgment. - International experience: yield from property taxes is usually limited; perception that property taxes remain underexploited. - Cross-country statistics: property tax share in subnational revenue varies from as low as 5 percent in Turkey and 7 percent in Norway, to 90 percent in New Zealand and 100 percent in Australia, Ireland, and United Kingdom. As share of GDP, property taxes account on average for about 1 percent in unitary countries and 2 percent in federal countries. - Benefit taxes and user charges: - Important at the margin; low revenue potential relative to subnational financing needs due to distributional and political considerations; proliferation may create nuisance taxes. - Excises at subnational level: - Efficiency suggests destination-based levying, but differing subnational rates can prompt smuggling and cross-border shopping. - More mobile tax bases: - Increasingly recommended for subnational assignment to enhance competition and spending discipline, but constrained by administrative feasibility. - Subnational VAT: - Continued interest due to strong long-term revenue potential, but administrative and border-control difficulties (absence of border controls makes destination-based subnational VAT difficult). - Practical experiences are mixed (Perry, 2009): options include dual VAT (jointly determined base; each level chooses its rate), with the Canadian dual VAT cited as relatively successful; ICMS in Brazil generated tax wars; Indian state VAT experience too short for definitive judgment. - Subtraction-type origin-based VAT (local business tax replacement): - Levied on value-added measured as revenue minus purchases; cited examples: Italian regional business tax IRAP, German local trade tax Gewerbesteuer. - Advantages: avoids cascading, base measures entity value-added; should be levied at a single rate to be effective. - Administrative suitability: relatively simple, but unsuitable where local administration capacity is limited—central administration could administer it in such cases. ### Advantages and disadvantages by tax type (high-level synthesis from Table 5) - Personal Income Tax: - Advantages: Buoyant revenues; Visible (increases accountability); Cost-effective if piggybacked on national taxation. - Disadvantages: May create or aggravate horizontal imbalances; insufficient yield where average incomes are below threshold; different rates across jurisdictions can create distortions with mobile people. - Corporate Income Tax: - Advantages/role: Sometimes seen as a benefit tax. - Disadvantages: Mobile tax bases and complex administration, making it suitable for central collection (except small businesses). - User Fees and Charges: - Advantages: Low mobility base; no obvious horizontal or vertical imbalance problems; Visible; Linked to benefits. - Disadvantages: Generally low yield; Low cost-efficiency. - Property Tax: - Advantages: Immobile base; Visible; Stable yield; Indirectly linked to benefits. - Disadvantages: Difficult administration (cadastre setup), often resulting in low yield. - Sales and Excise Tax: - Advantages: No horizontal or vertical imbalance problems; Visible; Easy to administer. - Disadvantages: May create cross-border shopping if levied at different rates among subnational jurisdictions. - Value-Added Tax: - Advantages: If properly designed and administered, could be a good local tax. - Disadvantages: Complex administration; destination principle requires border controls between local jurisdictions; origin principle can cause tax exporting and transfer pricing issues. - Resource Tax: - Disadvantages: Significant horizontal imbalances; Difficult to administer; Excessively volatile. ### Surcharges / piggybacking on central taxes - Surcharges allow subnational revenue builds on central tax administration and are a key source of subnational revenue in many countries. - Characteristics: - Limited degree of autonomy: surcharge on tax bases defined by central government. - Potential mutual advantage: local authorities may better identify taxpayers, aiding collection. - Most typical on personal income taxes, but also used on sales taxes and excises. - Surcharge rates frequently subject to upper and lower limits set by the central government. - Drawback: may deepen horizontal disparities due to geographic concentration of revenues in richer jurisdictions—necessitating an appropriate equalization transfer system. ### FAD advice and country-specific recommendations (summary of Table 6) - General FAD advice balances normative principles with administrative capacity and political economy constraints; common themes include clarifying legal frameworks for local taxation, streamlining own-source revenue, eliminating nuisance taxes, and strengthening property taxes. - Selected country recommendations (as summarized in the source): - DRC: Clarify legal framework for sub-national taxation; streamline system of own source revenue; strengthen property taxes. - Indonesia: Include the land and building tax in the list of local taxes, with local governments allowed to set the assessment rate. - Liberia: Clarify legal framework for local taxation; strengthen design and administration of property taxes. - Kosovo: Strengthen property taxes through (i) establishing a national fiscal cadastre; and (ii) applying common and harmonized valuation procedures; exploit fee-for-service instruments (e.g., municipal parking lots). - Macedonia: Increase municipal tax collection by giving municipalities bounded control over rates; remove caps on revenue; allow municipalities to keep all property taxes; monitor horizontal adequacy of revenue bases as expenditure mandates broaden. - Nigeria: Provide state and local governments (SLGs) with additional revenue sources and some control over rates (including excises, business taxes, surcharges on utility bills); repeal nuisance taxes; transform oil revenue derivation into a royalty share and an explicit environmental excise. - Bolivia: Assign significant own sources to subnational governments (enable rate-setting for local taxes and marginal surcharges on national taxes) with appropriate sequencing tied to acceptance and performance of new responsibilities. - China: Give more control over tax rates to local governments (allow provinces to levy surcharge on personal income tax; grant power to level domestic excises; increase freedom in setting property tax rates) to compensate richer provinces losing under stronger equalization. - Colombia: Increase local own-source revenue via tax and tax administration measures; convert local turnover tax into a simplified subtraction-based local VAT; simplify departmental excises’ procedures and administration and redistribute part of proceeds among departments. - Mexico: Piggyback the IDTU (a new federal tax akin to an income-type, origin-based VAT administered with the subtraction method) to create a major tax handle for states; alternatives include piggybacking on the income tax and revamping the property tax. - Implementation guidance: - In countries where legal framework for local taxation is unclear and administration weak, FAD recommended clarifying/simplifying legal frameworks, eliminating nuisance taxes, and improving design and administration of a few “good” tax handles—most notably property taxes (examples: DRC, Indonesia, Kosovo, Liberia, Macedonia, Nigeria). - In countries with relatively advanced tax administrations, recommendations included creating new VAT-like taxes assigned directly to subnational governments (Colombia) or assigned to national government with piggybacking for subnational governments (Mexico). - In other cases, surcharges on national income taxes were suggested to bolster subnational taxation (Bolivia, China). *Source: IMF background chapter excerpt on revenue assignments and subnational own-revenue options.* ### 59. Availability of own-source revenue is a necessary, but not sufficient, condition ### 59. Availability of own-source revenue is a necessary, but not sufficient, condition ### Own-source revenue, accountability, and fiscal discipline - Availability of own-source revenue is necessary, but not sufficient, for accountability and fiscal discipline at the subnational level. - The ability of subnational governments to mobilize own-source revenue does not guarantee incentives to do so. - Own-source revenue helps mitigate the common pool problem but does not necessarily eliminate the soft budget constraint problem (illustrated by the Italian experience with regional health spending mentioned in Section III.A). - Outcomes depend critically on interactions among: - clarity of expenditure responsibilities; - scope and design of the transfer system; and - borrowing options. ### Revenue-sharing: definition and implications - Revenue-sharing can be based on either individual taxes or total tax collection. - Revenue-sharing similarities to surcharges: administrative simplicity. - Revenue-sharing is not own-source revenue in the sense that subnational jurisdictions lack control at the margin over rate structure. - Revenue-sharing generally does not provide equalization unless specifically designed to do so. - Aggregate revenue-sharing may provide greater resources than own-source revenues but must be accompanied by own-source revenues to promote subnational accountability. - Revenue-sharing makes one government tier’s revenue dependent on another’s tax-base and rate choices; it is often considered akin to transfers rather than a method of revenue assignment. ### Policy choices in revenue-sharing - Two main policy choices: - how to define the revenue (or pool of revenue) to share; and - how to allocate such share among subnational governments. - If some taxes are shared and others are not, central governments face a bias to maximize collections of non-shared taxes. - Taxes should be shared at the same rates (a certain percentage of all taxes) to avoid distortions; extraordinary revenue (e.g., sales of assets) should not be shared. - To reduce procyclicality of shared revenues, consider: - a moving average of revenue collected over the cycle; or - a fraction of a moving average of nominal GDP over the cycle (as recommended by FAD in Argentina). - Distribution methods: origin basis (derivation principle), equal per capita allocations, or other criteria. ### Natural resource revenues (Box 3) - Natural resource revenues are unsuitable for sharing arrangements because they are very volatile and geographically concentrated. - Main problems: - Origin-based sharing creates significant horizontal disparities. - Subnational governments are typically ill-equipped to handle volatility. - Specific risks and observations: - Attribution of volatile resource revenues to local governments complicates fiscal management and macroeconomic policy at the central level; e.g., Nigerian states receiving a fixed share of oil revenue built unsustainable spending during an oil boom. - Geology/geography differences should not determine distribution of national resources; origin-based sharing would create horizontal imbalances. - Resource revenues reduce local accountability and can exacerbate corruption and absorptive capacity issues. - Fund recommendations (examples cited): - Avoid sharing resource revenues (DRC, Liberia, Nigeria). - Introduce production-related excises together with equalization transfers instead of resource revenue-sharing (Indonesia, Nigeria, Bolivia). - Excises might be fully or partly assigned to lower levels as compensation for environmental or other costs, coupled with a transfer system to fund basic subnational spending. ### Drawbacks and fiscal rigidity from revenue-sharing - Revenue-sharing creates macroeconomic management rigidity: e.g., if 50 percent of revenues are shared, central government revenue adjustments must be substantially larger to achieve a target. - Distortion of incentives: high marginal transfer rates reduce incentives to raise revenue (noted case: Germany). - Origin-based revenue-sharing exacerbates horizontal inequalities and requires redistributive transfers to compensate. ### FAD advice on revenue-sharing - Revenue-sharing should be considered part of the overall financing framework for subnational governments and is conceptually akin to transfers. - Avoid sharing geographically concentrated revenues (e.g., natural resource revenues); prefer taxes more evenly distributed across jurisdictions (e.g., VAT over income taxes in Macedonia). - Revenue-sharing should be accompanied by a proper equalization transfer system. - Table 7 contains more detailed country-level FAD policy recommendations (selected examples summarized below). ### Selected country recommendations on revenue-sharing (from Table 7) - Bolivia: - Problem: Distribution of hydrocarbons revenues generates vertical and horizontal imbalances and transfer volatility, and has resulted in low incentives to exploit tax bases at the subnational level. - Fund advice: Hydrocarbons taxation should accrue to the central government level only. - China: - Problem: Revenue-sharing based on taxes stronger in richer provinces has elements of regressivity. - Fund advice: Reform transfer system to deliver a minimum standard of public services and base it on a rule-based equalization system. - DRC: - Problem: Constitution grants provinces 40 percent of government revenue collected in their territory, but does not define tax bases. - Fund advice: Revenue-sharing mechanisms should be gradual and in line with the devolution of expenditure responsibilities. - Indonesia: - Problem: Oil and gas revenues included in revenue-sharing create imbalances and volatility; a floor was imposed on transfers. - Fund advice: Avoid sharing oil and gas revenues on an origin basis. - Liberia: - Problem: No revenue-sharing mechanism yet; pressures likely to assign/share natural resource revenue on origin basis. - Fund advice: Refrain from sharing natural resource revenues; limited spending assignments at initial decentralization stages likely do not require revenue-sharing. - Macedonia: - Problem: No revenue-sharing in place; VAT-sharing preferred when revenues are more evenly distributed. - Fund advice: Share VAT rather than PIT for administrative simplicity, stability, and redistribution. - Mexico: - Problem: Complex and not redistributive system based on revenue-sharing and earmarked transfers. - Fund advice: Reform revenue-sharing and fold into a new transfer system reflecting spending needs and revenue capacity. - Nigeria: - Problem: Federally collected receipts pooled and shared; sharing oil revenue is highly procyclical and creates horizontal imbalances; states mandated to share 10 percent of receipts with local governments. - Fund advice: Repeal oil revenue-sharing and convert into a royalty share and explicit environmental excise; repeal the 10 percent sharing mandate and fold into properly designed revenue assignments (including piggybacking of PIT). ### Transfers: purposes and design principles - Transfers correct vertical imbalances between spending responsibilities and tax revenues devolved to the local level. - Types and purposes: - General purpose (unconditional) transfers: for general funding. - Equalization transfers: correct horizontal imbalances among subnational governments. - Block grants: provide funding for specific sectors. - Special purpose (earmarked/conditional) grants: mandate spending within a designated sector. - Transfer system design must complement spending responsibilities and revenue assignments. - Coparticipation (matching) requirements: - Can divert resources from basic spending when own-source revenue is weak. - May prevent poorer regions from accessing transfers and exacerbate horizontal inequalities. - FAD often recommends special-purpose grants based on costed programs and only where central government can monitor correct use. - Consider whether special-purpose grants should be counted when setting equalization grants. ### Discretion and allocation criteria for transfers - Degree of discretion: - Mandatory transfers (entitlements): rules-based, legally defined amounts/conditions. - Discretionary transfers: ad hoc, usually temporary (infrastructure, emergency). - Allocation mechanisms: - Simple equal per capita transfers achieve basic equalization but ignore differential needs and costs. - More complex formulas combine poverty indices, geographic criteria, spending needs, and revenue capacity. - Sole reliance on needs-based or capacity-based criteria has trade-offs; many countries use a combination. - Advanced systems: Australia, Denmark; recent reforms: Sweden, Switzerland. - FAD recommendations emphasize transparent formulas including spending needs and revenue-capacity criteria and discouraging criteria based on historical/actual spending or revenue. ### FAD advice on transfer design (from Table 8) - DRC: - Problem: Planned equalization fund limited to investment financing and unable to mitigate horizontal imbalances. - Fund advice: Establish a transfer system based on an equalization formula with transparent criteria (e.g., number of inhabitants and surface). - Liberia: - Problem: Transfers vertically inexistent except County Development Fund for capital projects. - Fund advice: With decentralization, adopt a simple transparent formula to fund county operations. - Nigeria: - Problem: Low correlation between transfers and states' relative needs; derivation formula benefits middle- and high-income states; no definition of minimum public services. - Fund advice: Establish special purpose transfers in lieu of direct federal investment; over time include a general transfer system based on spending needs and revenue capacities; introduce a floor for transfers financed by savings from high oil prices; limit specific purpose grants to high-priority national objectives. - Bolivia: - Problem: Excessive earmarking without equity. - Fund advice: Reduce earmarking and reform transfer system to account for spending needs and fiscal capacity. - Colombia: - Problem: Transfers excessively earmarked; at mission time transfers were fixed amounts not linked to revenue. - Fund advice: Maintain defined transfer growth framework to avoid rigidity and procyclicality; reduce earmarking gradually. - Mexico: - Problem: Complex, non-redistributive system based on revenue-sharing and earmarked transfers. - Fund advice: Freeze existing transfers in nominal terms (hold-harmless) and phase in an equalizing system based on spending needs and revenue capacities; clarify and simplify special-purpose transfers. - Kosovo: - Problem: Earmarked transfer formulas for health and education inadequate; unconditional transfers too small to provide spending autonomy. - Fund advice: Increase unconditional transfers based on equalization criteria; better link block grants for health and education to spending needs; move gradually to general transfers and include hold-harmless provision. - Macedonia: - Problem: Transfers allocated using a nontransparent system. - Fund advice: Create Municipal Equalization Fund and adopt a formula-driven allocation rule; broaden revenue-sharing and/or transfer system as expenditure mandates expand. - China: - Problem: Complex and not fully transparent transfer system. - Fund advice: Reform transfer system to deliver minimum public services; expand rule-based equalization; clarify and simplify special-purpose transfers. - Indonesia: - Problem: Oil and gas revenues included in revenue-sharing creating imbalances and volatility; a floor on transfers imposed. - Fund advice: Remove floor on transfers while keeping hold-harmless rule for the general purpose fund (DAU). ### Managing revenue and tax administration - Choice between central and subnational tax administration depends on country circumstances. - FAD TA has been limited but provided where financing changes required rethinking tax administration. - Examples: - Colombia: advice to centralize tax administration while building subnational capacity. - Macedonia: advised building capacity and aligning tax administration to municipal territorial reorganization. - China: 1994 revenue-sharing arrangements designed to give the center additional tax handles and establish a state administration of taxation separate from local administrations. *Source: _072709 - 59. Availability of own-source revenue is a necessary, but not sufficient, condition* ### 73. Revenue-sharing/surcharging may help circumvent the challenges associated ### _072709 - 73. Revenue-sharing/surcharging may help circumvent the challenges associated ### Subnational tax administration: revenue-sharing and surcharging (paragraphs 73–74) - Revenue-sharing/surcharging can help circumvent administration challenges for major taxes at the subnational level (para. 73). - Taxes that can be collected with minimal additional effort (e.g., surcharges on income taxes where subnational information may be helpful to the center) face limited resistance from the central tax administration to collect on behalf of subnational administrations (para. 73). - Collecting a VAT on behalf of subnational administrations may be more problematic for national administrations (para. 73). - Subnational governments often worry that the central tax administration may lack incentives to devote sufficient resources to collect subnational taxes; this concern can be partially mitigated by well-structured service contracts with the central tax administration (para. 73). - Subnational tax administrations can be designed and operated on modern tax administration principles; examples: - Large municipalities in Colombia have led to considerable improvements in own-revenue collection (para. 74). - Some Brazilian states have effectively administered subnational VATs, albeit origin-based ones (para. 74). - Feasibility constraints in many developing countries: - High setup costs for information systems and processes to support local tax collection (e.g., establishing a cadastre for property taxes) (para. 74; footnote 40). - Complex rules for tax revenue assignment among national and subnational levels can hinder tax modernization (example: Sudan’s dual network of tax offices hampering VAT and income tax integration) (para. 74). - In the DRC, FAD recommended either compensating the central tax administration (Direction Générale des Impôts, DGI) with a larger share of locally collected taxes if it collects local taxes, or limiting subnational tax administrations to managing only local taxes within a framework clearly defined by law if local administrations collect them (para. 74). *Source: IMF document content provided.* ### Mechanisms to control subnational borrowing — rationale and overview (paragraphs 75–76) - Essential objective: avoid “excessive” borrowing by subnational governments because of adverse externalities for other subnational governments and the central government (para. 75). - Channels of adverse externalities include: - Preempting a disproportionate share of available financing and putting upward pressure on interest rates (para. 75). - Pushing up risk premia on government bonds more generally (para. 75). - Bailout costs (para. 75). - Four broad, not mutually exclusive, categories of borrowing controls (para. 76; footnote 41): 1. Reliance on market discipline 2. Cooperation between central and subnational governments 3. Rules-based controls 4. Administrative controls - These approaches are ranked from maximum subnational autonomy (market access) to maximum central control (administrative measures, including outright bans) (para. 76). ### Approaches to controlling subnational borrowing — advantages, preconditions, and examples (paragraphs 76–81 and Table 9) - Market Discipline - Advantages: Emphasis on self-control; monitoring by credit rating agencies (Table 9). - Preconditions: Comprehensive, timely, and reliable information; developed financial markets; no access to privileged financing; no previous history of bailouts (Table 9). - Examples: Canada, Australia, (also Sweden mentioned in text) (para. 77; Table 9). - Limitations: Preconditions seldom met in practice; implicit central guarantees or bailout histories weaken market signals (para. 77). Example: German Länder with significant debt maintained very high credit ratings, attributed to bailouts of Bremen and Saarland in the 1990s (para. 77; footnote 42). - Cooperative Approach - Advantages: Promotes dialogue; enhances responsibility of subnational policymakers (Table 9). - Preconditions: Constitutional underpinnings; culture of fiscal discipline; institutions for cooperative decision-making or strong central bargaining position; reliable and timely information (Table 9; para. 78). - Examples: Australia’s Loan Council; EU Domestic Stability Pacts; Brazil’s debt agreements; Argentina’s bilateral pacts (para. 78; Table 9). - Record: Mixed effectiveness; success depends on absence of severe fiscal stress, relative economic homogeneity, tradition of cooperation, institutional negotiation framework, and information availability (para. 78). - Rules-based Controls - Forms: Quantitative ceilings on borrowing, debt and/or debt service (often specified relative to revenues, as in Brazil and Colombia); procedural rules for budget processes (para. 79). - Advantages: Transparency; avoids bargaining; potential central government control (Table 9). - Preconditions and design elements critical to effectiveness: - Sound and credible rules (well defined, transparent, flexible) - Clear coverage and full information - Clear escape clauses and credible sanctions for noncompliance - Use of a revenue base relatively stable over the business cycle to reduce procyclicality (para. 79). - Implementation: Rules may be in national legislation or subnational constitutions/laws (para. 79). - Administrative Controls - Use: More common in unitary states and vis-à-vis local governments; may focus on overall borrowing flow or debt stock rather than individual authorizations (para. 80). - Variants: Limits on external borrowing; limits on banks’ exposure to subnational governments (para. 80). - Risks: Central authorization of individual borrowing can create moral hazard (lenders may perceive implicit central backing); effectiveness depends on de-politicization of central decisions and availability of comprehensive information (para. 80). - Overall assessment - No single universally preferable approach; context matters (para. 81). - Rules-based approaches, supported by enforcement, may be preferable where market discipline preconditions are inadequate and central discretion weak (para. 81). - Complement rule-based frameworks by building market-discipline preconditions: improve transparency, eliminate privileged financing channels, and establish no-bailout track record (para. 81). - Include ex post insolvency mechanisms to manage default expectations and allow efficient resolution of distressed debt (para. 81). ### FAD advice and country-specific recommendations (paragraphs 82–83; Table 10) - FAD’s approach: eclectic and tailored to country-specific circumstances, legal/constitutional constraints, and preconditions for market discipline (para. 82). - Examples of FAD advice: - Countries advised to build preconditions before allowing subnational borrowing (e.g., DRC, Kosovo, Macedonia) to avoid undermining macroeconomic management and debt sustainability (para. 82). - Asymmetric arrangements suggested where appropriate (example: China — allow more advanced provinces to seek credit ratings and limited borrowing) (para. 82). - Where subnational borrowing is established (Bolivia, Indonesia, Mexico, Nigeria), advice focused on establishing guidelines/rules, strengthening reporting frameworks, and creating capacity for debt management (para. 82). - FAD emphasized the fiscal risk from indirect borrowing mechanisms, notably PPPs: - PPPs often use land or other assets and involve complex financial and legal arrangements that pose significant risks for inexperienced governments (para. 83; footnote 44). - Successful PPPs require strong institutional, operational, and legal frameworks; stringent investment planning and project evaluation; coordination with other public investments; clear disclosure standards and transparent financial reporting (para. 83). - Fund advice has cautioned against PPP arrangements in countries lacking legal and institutional capacity (examples: Kosovo and Macedonia) until appropriate frameworks and expertise are in place (para. 83). - Table 10: Main recommendations on borrowing arrangements in case countries (summarized highlights) - DRC: Define control mechanisms for subnational debt; tight control to avoid unsustainable provincial spending (Table 10). - Kosovo: Proceed slowly on relaxing municipal borrowing ban; link any relaxation to advances in public financial management and a well-defined legal framework (Table 10). - Macedonia: Maintain borrowing prohibition at initial decentralization stage; consider a Municipal Investment Fund for limited investment borrowing (Table 10). - Indonesia: Impose borrowing limits and strengthen reporting (Table 10). - Mexico: Establish prudential limits linked to fiscal rules; ensure full coverage of subnational debt registry; legal framework for PPPs (Table 10). - Nigeria: Restrict domestic borrowing to fiscally sound local governments; maintain practice of not allowing external borrowing by local governments; build debt management capacity and reporting to the Debt Management Office (DMO) (Table 10). - Bolivia: Strengthen framework and transparency; create a fiscal risk register to take stock of subnational liabilities (Table 10). - China: Consider asymmetric framework allowing some local governments to borrow if preconditions are met; establish sub-national fiscal risk registry at central treasury (Table 10). ### Broader lessons: borrowing controls in context of fiscal decentralization (paragraph 84) - Controls on subnational borrowing are one element of intergovernmental fiscal relations and cannot be viewed in isolation (para. 84). - Demand for greater subnational borrowing autonomy often reflects inadequate financing means for assigned spending responsibilities or lack of hard budget constraints, leading to inefficient spending and prodeficit bias (para. 84). - A comprehensive assessment of other aspects of fiscal decentralization is required when addressing subnational borrowing arrangements (para. 84). *Source: IMF document content provided.* ### 85. The design of intergovernmental fiscal arrangements is one of the more complex ### The design of intergovernmental fiscal arrangements is one of the more complex ### Overview - The design of intergovernmental fiscal arrangements spans multiple policy and institution‑building issues requiring careful coordination and sequencing. - Fund advice focuses on macrorelevant aspects, recognizes there is no single “right” model, and stresses that arrangements must take into account: (i) macroeconomic constraints; (ii) the need to strike a balance between efficiency and distributional considerations; and (iii) relevant institutional factors (constitutional and legal constraints, and subnational capacity to spend well and raise own revenues). - Fund advice on fiscal decentralization has evolved as circumstances change over time and across countries. ### Sequencing of decentralization - Resources should be made available to subnational governments pari passu with assignment of spending responsibilities. - Mismatches in sequencing can lead to problems: examples include Colombia and Nigeria (where devolution of resources initially outpaced spending responsibilities) and transition economies in the 1990s (where spending mandates were pushed down without adequate resources), resulting in debt, arrears, or deterioration in service quality. - Territorial organization and political arrangements are relevant to sequencing, though not central to the Fund mandate. ### Pace of decentralization and capacity - Pace should be linked to subnational governments’ capacity to carry out assigned functions. - Corollaries: - Increased devolution of expenditure functions should be conditioned on subnational compliance with a minimum set of PFM requirements, notably an orderly, transparent, and reasonably participatory budget process. - Asymmetric arrangements (differentiated speed based on clearly specified criteria) are feasible and desirable to reflect varying capacities across jurisdictions. - Colombia’s experience shows asymmetric decentralization can promote incentives for “good” policies. - Asymmetric approaches suit capital cities or larger/more advanced jurisdictions (recommended in Kosovo, Liberia, Macedonia, and noted for China regarding some provinces). ### Clarity of roles and overlapping responsibilities - Overlap (e.g., education and health) is inevitable, but clarity on roles of each government level is important to avoid duplication, waste, and loss of accountability. - Experiences in Bolivia, DRC, Mexico, and Nigeria illustrate costs of lack of clarity. ### Complementary policy reforms - Complementary reforms may be required in civil service and regulatory frameworks. - Without adequate geographic or functional mobility of civil servants, devolution may deteriorate service quality in certain jurisdictions or increase overall public employment. - Devolution of subsidized public services to local governments is unlikely to reduce subsidies, or may worsen services, unless local authorities can adjust prices as needed. - Allowing regulatory autonomy and increasing subnational participation in policy matters can help address decentralization pressures while allowing time for necessary institutional and legal changes. ### Capacity building and PFM systems - Realizing efficiency gains from decentralization requires significant investments in capacity-building and improvement of subnational PFM systems, especially to improve budget transparency in all phases. - Technical assistance, dissemination of good practices, and peer pressure can be useful, but central and subnational government commitment is essential for lasting success (experience in Colombia and Brazil). - Risks include spreading thin limited central capacity, particularly in post‑conflict countries (DRC and Liberia); decentralization should not distract from central PFM reform needs in such contexts. ### Resource envelopes and hard budget constraints - To implement a “hard budget constraint” on subnational governments, they must be provided an overall resource envelope that ex ante allows them to carry out assigned responsibilities at an average level of efficiency. - Persistent significant mismatches between spending needs and resources lead to inadequate service provision and/or fiscal indiscipline (borrowing or arrears). - Assessing subnational spending needs is difficult given frequent absence of reliable information on cost‑effectiveness; sample studies comparing cost‑effectiveness across similar jurisdictions can help. ### Own revenues, accountability, and tax assignment - Control over a portion of subnational resources is key to accountability and fiscal responsibility. - Assignment of own‑revenue sources should consider economic factors (tax base mobility) and institutional capacity (subnational tax administration). - Where subnational administration is weak, central administration on an agency basis may be preferable. - Potential regional own‑revenue sources: surcharges on central/federal taxes (notably the personal income tax); retail sales taxes; regional VAT forms (e.g., Italian IRAP). - Coordination of subnational taxes—especially tax base definition—is important to avoid predatory competition and reduce compliance costs (experience with Brazilian ICMS). - For local governments, movable (motor vehicles) and immovable property taxes are appropriate, but often under‑exploited; advice typically focuses on property tax design and administration. ### Intergovernmental transfers and smoothing cyclical risk - Subnational own resources typically fall short of spending responsibilities, necessitating intergovernmental transfers to address vertical imbalances. - Revenue‑sharing corrects imbalances but can expose subnational budgets to cyclical fluctuations they are less able to absorb. - Suggestions include choosing least‑cyclically sensitive taxes as own revenue handles and using smoothing mechanisms (e.g., moving averages of central revenues for revenue‑sharing). - Smoothing attenuates but does not eliminate cyclical impact; subnational governments should build cushions (e.g., “rainy day funds”) during booms. ### Equalization and earmarked transfers - Horizontal redistribution is best achieved through equalization transfers designed to account for relative taxing capacities and spending needs, subject to data availability. - Special‑purpose, earmarked transfers should be used sparingly because they impart rigidity to subnational spending and are difficult to monitor effectively (illustrated by Bolivia, Colombia, Mexico, Nigeria, and Indonesia at earlier stages). ### Subnational borrowing, market discipline, and controls - Limits to subnational borrowing are often needed to ensure fiscal discipline. - Effective market discipline requires preconditions (notably a history of no bailouts) met in few countries (e.g., Canada and the United States). - Choice of control methods (administrative, rules‑based, cooperative) depends on country circumstances; blends may be most effective. - In countries with less developed credit markets or inadequate information, borrowing should be conditional on well‑identified preconditions (possibly including certification of minimum PFM requirements), as recommended for China, Kosovo, and Macedonia. ### Information, debt registries, and crisis resolution - Availability of relevant information is crucial to enforce controls, including on floating debt and contingent liabilities (guarantees and PPPs). - Recommend establishing a subnational debt registry (recommended in Bolivia and Mexico) or a fiscal risk registry (recommended in China) by consolidating information on below‑the‑line financing to subnationals. - Important to progressively develop subnational debt management capacity. - Debt limits should be complemented by strengthened prudential regulations on financial intermediaries lending to subnationals. - Sound mechanisms for resolution of subnational debt crises are crucial. ### Effectiveness of Fund advice and political economy - Effectiveness of Fund advice on fiscal decentralization has varied across countries and over time. - A fundamental factor is the degree of political support for reforms involving difficult trade‑offs and distributional consequences. - Political commitment has sometimes been enhanced by external factors (e.g., financing crises). - Piece‑meal reforms (e.g., tightening borrowing controls) have generally been easier to implement than comprehensive reforms that could yield more efficient and lasting outcomes. ### Looking ahead and institutional cooperation - As decentralization advances, including in currently centralized countries, the Fund should remain engaged on macroeconomically relevant aspects of the process. - The Fund should strengthen cooperation with other institutions (World Bank, other multilaterals, bilateral donors) that can support subnational capacity‑building, since capacity improvements are often key to successful fiscal decentralization. *Source: IMF content unit _072709 (chapter/section on fiscal decentralization).* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2009/_072709.pdf_